Return to Continuing Maintenance


"Status of the 1401 Data Center - 2022"

Hosted by Carl Claunch from November 2018 to July 2021 (off to Florida ;--))
Now continued by others

I (ed@ed-thelen.org) announce updates
using e-mail addresses from Team Bios
- Restoration Team, - Docents, - Remote
+ Poppy Haraldson, Kate McGregor,
     Jessse Nichols and Luis Valencia of CHM.
     Suggestions accepted.

Contents

Some relaxed pictures taken Wednesday, March 30, 2022
2022 Maintenance
For 2018 through 2021, click here.
2022 Demonstrations
For 2018 -2019, click here.
Jan - 5, 12, 19, 26
Feb - 2, 9, 16
Mar - 16, 20, 22, 23, 26, 30
      Group Pictures
Apr - 9, 13, 20, 27 Apr - 30
May - 4, 11, 17&18, 24&25, 31 May - 7, 14, 18, 21, 28
Jun - 1, 7, 14&15, 21&22, 29-A.C. Jun - 4, 11, 22, 25, 29
Jul - 6, 13, 20, 27 Jul - 2, 6, 9, 13, 16, 30
Aug - 3, 10, ?, 17, 24, 31 Aug - 3, 6, 7, 13, 17, 19, 20 !!, 24, 27
Sept - 7, 14, 16, 29, 29+!, Sept - 3, 7, 8, 10, 14, 21, 28,
Oct - 4 :--((, 5 , 5a , 5d , 8, 12, 19, 26, Oct - 5c, 12, 29,
Nov - 2, 5, 8, 9, 16, 20, 23, 30, Nov - 2, 5, 9, 16, 23, 26,
Dec - 7, 14, 21, 28, Dec - 3, 10, 14, 17, 28,
TOPICS
1402 Card punch 10/05/22, 10/12/22
DE breakers & power 10/05a/22, 10/12/22,
Equipment Status
Scheduled Maintenance
  729s
    Discussion 11/02/22, 11/04/22,11/08/22
A Thank you for class demo
Wire Relays, cleaning indefinitely??
2023 Maintenance
For 2018 through 2021, click here.
2023 Demonstrations
For 2018 -2019, click here.
Jan - 4,11, 18,25, Jan - 7,11, 21,

Equipment Status - as heard from miles away
001 - up
026 - all up
083 - up - now good enough for serious sorting
1401 - CT - up - 1402 reader shaky or worse
1401 - DE - up - #1 729 tape drive sick
P.A. System for docents - sick, battery charger, volume control


Maintenance - January 25, 2023 - from Robert Garner, David Clementson & Ken Shirriff
from Robert Garner
1401ers,

Here’s the 1401 restoration report for Wednesday, Jan 25th.

Focus this week was on DE 729#1’s faulty neon bulbs; DE 729#2’s refusal to load; and fixing DE 729#3 refusal to load/unload caused by deteriorated oil on its right-side capstan. The CT 1401 utterly behaved itself.

Len Shustek dropped by to borrow some pre-1950, unit-record items for the Stanford Continuing Studies “The History of Computers” class that he’s teaching: https://continuingstudies.stanford.edu/courses/professional-and-personal-development/the-history-of-computers/20222_HIS-134 
The Youtube sessions are being broadcast from the CHM (mostly in Lovelace, but sometimes in the lobby theater) on Monday evenings 6-8 pm. He graciously invited us to just show up to watch/participate in person if we’d like.
(Len borrowed Franks’ Type 001 hand keypunch normally attached to the railing, 1402 card brush read assembly, two small control panels/plugboards, contact relay, 405 electromechanical decade counter, and a couple of reference punched cards.)

Here are the detailed reports for this week, followed by the attendance/hours-volunteered table.

From: David Clementson Subject: Restoration status report 2023/01/25

  • I started the day with the intention to replace the "failed" +62V PSU diode in the DE#1 729 tape machine with a modern 1N4004. I figured that if that fixes the issue, we can then discuss what kind of diode we want in the unit long-term. I took a soldering station from the workshop over to the unit, but found that the AC power was off for the unit and the courtesy outlet was not live. No problem, probably safer to work that way. I found an extension cord and began the search for a working AC outlet. It turns out the room doesn't actually have any, so I had to use the orange extension cord emerging from the floor beneath the sorter. I used an outlet tester and checked that it was live and wired properly. More on that tester later...
  • Replacing the diode had no effect on the problem. Ken and I proceeded to look at the rail waveform on the scope and found an interesting thing: the rail is not actually DC, but AC. No wonder the DVM recorded only a few volts DC. We compared the waveform with the +62V rail on the CT#1 machine. It had a half-wave rectified waveform. After deep diving into the schematics, we found that the DE machine does not actually have any path from the diode's cathode to ground. In contrast, the CT machine's cathode has a ~6k resistor to ground. So on the DE machine, when the transformer voltage reverse biases the diode, the cathode is free to float below ground; there is nothing to hold it at ground potential. Adding an external 6k2 load resistor to the DE machine resulted in the same half-wave rectified waveform.
  • So the conclusion Ken and I arrived at is that this is how IBM intended the system to work, since there is no harm to having the +63V rail go below ground. However it is a little weird, which is maybe why the load resistor was added on later models. Just because the documentation calls the rail "+62V," don't assume it will actually measure 62 VDC.
  • So I think it is safe to emerge from the +62V rabbit hole and get back to confirming the neon indicators.
  • I selected a random neon bulb from the indicator array and attempted to remove it to see what the replacement effort might be like. No problem: just pull the two pins from the back of the socket and bush the bulb out the front. Reassembly is just as easy. So I will buy a bag of neon bulbs from Digikey and replace any that are dodgy or outright broken.
  • While I had the bulb out, I thought I should try to test it by measuring its turn-on threshold. I went back to the workshop in search of a Variac or a DC supply that could produce a variable voltage in the ~80V range to "margin" the bulb candidates. The Variac looked pretty sketch and didn't have cabling that would be friendly for connecting bulbs to, but there was an HP 6299A that should do just the trick. I fired it up to find that the meter switch was clearly flaky. So I wasn't able to tell if the supply itself was working.
  • Meanwhile, Frank grabbed the Variac and attempted to test the neon bulb, although without any ballast resistor. About 10mm of the bulb's connection wire was vaporized, which is a much better outcome than Frank being vaporized.
  • I was able to partially disassemble the 6399A enough to access the back of the meter switch, which I doused with Deoxit. Meter fixed! After reassembling the 6399A and adding the requisite ballast resistor, I was able to check one of the spare neon bulbs I brought from home. I was able to show that 82 volts is a typical turn-on voltage for the bulbs I had. I checked a few more bulbs, and found pretty consistent results. So I reinstalled a known-good bulb into the 729.
  • I then reinstalled the original diode into the 729, so it remains period-correct. And I made sure to use period-correct Sn60/Pb40 solder with period-correct Kester 44 flux.
  • As mentioned before, I had my AC outlet tester handy, so I checked the 729's courtesy outlet and found that it does not conform to NEC standards. Neither did the CT machine. Ken did a deep dive into the schematics and discovered that the courtesy outlet voltages are actually generated by isolated windings from the main power transformers in the 1402s. But they are not truly floating, since the receptacles' ground sockets are indeed grounded. So even though they don't conform to the electrical code (neutral is not bonded to ground), the courtesy outlets seem fine to use.
  • Ken was working with DE#3 729 which was refusing to load/unload. He noticed that the right capstan was not retracting properly. We checked the shaft and found that the oil for the bronze bushing had dried to a VERY sticky honey-like substance. We were able to clean the shaft with isopropanol and re-lubricate with oil from the IBM repair kit. That fixed the load/unload problem nicely. So we should probably do this cleaning/oiling maintenance on all of the capstan bearings. I do not know exactly what oil IBM recommends for this application, but I imagine a high quality light spindle oil would be appropriate. I will bring in some Mobil DTE Light, which I use for my lathe headstock.
  • AC outlet tester:
  • HP 6299A repair:
  • Neon bulb under test:

From: Ken Shirriff Subject: Re: Restoration status report 2023/01/25

One other problem: DE drive #2 is refusing to load although it was working last week.
I'd blame the relays like DE #1 except DE #2 has the NOR (SMS) logic.

Ken

1401 Restoration Hours: 1/25/2023

Clementson, Dave 5
Garner, Robert 4
Howard, John 4
Jelsema, Dale 0
Kim, Mariane 5
King, Frank 4
Menendez, Iggy 0
Paddock, Stan 4
Shirriff, Ken 4
Szolyga, Tom 6
Thelen, Ed 0
Verdiell, Marc 0

Demo - from Stephen Madsen, - January 21, 2023
Hi,
I was lead for the 11:00 am demo and Peter Chang was the Operator. Karen Sun and Jack Ghiselli were also there as helpers/observers. We had 32 visitors for the demonstration; two were from England.
We used CT, and all equipment worked fine. My thanks to the restoration team for fixing the 1402 card reader.

Stephen H. Madsen

Maintenance - January 18, 2023 - from Robert Garner, Ken Shirriff, David Clementson, & Mariane Kim
from Robert Garner
1401ers,

Here’s the 1401 restoration status for Wednesday, Jan 18th.

This week the focus was on testing the incalcitrant relays in the CT 1402 and on diagnosing non-functioning neon indicators in the DE 729 #1 tape drive and the DE 1401 TAU panel (the former due to a bad power supply and the latter due to a failed transistor). Here are the detailed reports, followed by the attendance/hours-volunteered table. (A welcome surprise: Ed Thelen made it in!):

 1401 Restoration Hours: 1/18/2023

Clementson, Dave 5
Garner, Robert 3
Howard, John 0
Jelsema, Dale 4
Kim, Mariane 6
King, Frank 6
Menendez, Iggy 0
Paddock, Stan 6
Shirriff, Ken 5
Szolyga, Tom 0
Thelen, Ed 4
Verdiell, Marc 0

from Ken Shirriff
I finally tracked down the problem with the B reg A-bit light not working on the German TAU. As you can see below, the A light is not lighting up, but it's not a problem with the bulb.

The problem turned out to be the DHC hex inverter card in 00XC C11: the inverter output on pin L was stuck, which turned out to be a problem with transistor T9 (lower left).

I replaced the card with the one spare in the filing cabinet. That one had several notes about replaced transistors, but it seems to be functioning. I didn't swap transistors because the card takes type 102 transistors and I couldn't find any in the workshop. Based on transistor replacement on other DHC cards in the machine, this card type seems a bit prone to failure.

After replacing the card, the B reg A light works, showing both signal levels are getting processed (B reg is low and A reg is high). The TAU is still showing a bunch of errors from the tape (skew and R/W VRC), so the drives will need more adjusting.

Ken

from David Clementson Subject: Restoration status report 2023/01/18
  • I began the day by dismantling a relay wiring harness that was extracted from some poor donor machine [as provided by Dale Jelsema?]. The goal was to extract enough relay connector pin wires to allow Stan to finish the modifications to his relay tester to allow it to test all types of relays. I spliced some wires to extend their length, and delivered everything to Stan. He should be unblocked from finishing his tester.
  • I next turned my attention to the 729 DE #1 tape machine. Last week, I had left the machine with its FUSE light lit. When I tried powering it up today, there were no front panel lights at all! I opened the clear side panel to measure the power supply voltages, whereupon I discovered that all of the power supply rails were zero. After a bit of sleuthing, I discovered that Circuit Interrupter (CP) #2, the one associated with AC input to the DC power supply, was open even though its reset button did not indicate a tripped condition. With the power off, I manually set and reset the protector several times and was eventually able to get it to close and stay closed. This brought back the power supply rails. So CP #2 could probably do with a rebuild, although it currently seems to be working. The machine now properly shows its front panel lamps, and the FUSE lamp is OFF.
  • I next looked into the missing +62V rail that could be affecting the neon indicators. I was able to confirm that the DC voltage of that rail was very low - less than 3 volts - and that the AC voltage of the rail measured 52 VRMS. I was able to isolate the rail from its load and measure the DC resistance looking back into the supply. It measured around 40k ohms, much higher than expected if the diode had failed short. I was also able to use a DVM diode test on the suspect rectifier, and it measured okay. The only explanation I have for these observations is that the diode is functional but has huge reverse leakage.
  • I discussed the situation with Ken and we agreed that the diode should probably be replaced, although we may disagree on the appropriate replacement part. Ken favors using a "period correct" diode, whereas I favor using a modern, reliable 1A 100 PIV diode, 1N4002 or equivalent. Thoughts? DC

from Mariane Kim Subject: Re: Restoration status report 2023/01/18
Hi all, 

I'll piggyback off Dave's message to give our update for this week.

This week, Frank and I spent most of our time working on the 1402 reader, with some time spent on the 026 and 083. 

083 card sorter:

Frank had previously noticed that the time for which the machine runs after processing the final card in a sort deck is quite long, running for seconds after the final card had been dropped in the farthest 9 slot. This duration can be modified using the run-out potentiometer. We modified the potentiometer such that the run-out duration is approximately 1s.

1402 reader:

  1. A report from Paul Laughton on 2023/01/11 mentioned that CT refused to load any program decks.
  2. When we powered on the CT and attempted to run a deck through the 1402, the punch "chips" light went off. 
    We believe this light should go off when the container of discarded punched chads exceeds a certain weight. We opened the panels and inspected the bin containing punched chads, it was less than 50% full, so we believe the light was on by error. A mere restart of the system "fixed" this issue (possibly more on this later)
  3. Next, we re-attempted to load a program deck. This time, we noticed that the 1402, though running, was not processing cards. No error lights turned on on the panel, either. Frank immediately thought to test the reader relays because they have proved fussy in the past. 
  4. Using Stan's (AMAZING) Permissive Make Relay Tester, we tested relays 1-13 which were associated with the reader. Results included below for future reference:
    • Relay # Highest Resistance Contact (ohms) Notes
      1 1.14
      2 1.7
      3 2.3
      4 0.77
      5 0.41 The connector for this relay had a bent #1 and R pin 
      6 0.37
      7 0.6
      8 0.39 The connector for this relay had multiple bent pins
      9 0.42
      10 0.53
      11 0.66
      12 0.89
      13 0.72
    • We manually pushed any bent connector pins back into place to the best of our ability.
      • Note that based on the wiring diagrams: Relay 5 is connected to Punch Chip Stop function (possibly related to the error in point 2 we had seen)
      • Relay 8 is connected to Run - Delay - Read function (possibly the culprit for the card reader failing to load programs)
  5. On our next attempt to load a program, the cards were being processed as expected, but we did run into a Reader Check on Column 32. We checked the problematic card against the data stored by the CPU and found that it had been read correctly, this told us that it was a problem detected with the first set of brushes. Frank and I took out the first set of brushes and inspected them visually. They seemed OK. So we attempted to finish loading the program and found that it loaded correctly. 
  6. We concluded that the relay connectors must have been the culprit here, and once fixed, the reader worked as expected. 
026 card punch:
  • We noticed that one of the card punches was making a rattling noise. Frank showed me that this usually occurred because the clutch had failed to latch properly after a card was released. In the future, this can be easily solved by releasing a card; this will cause a rotation and therefore a new latching event. 
  • RE: Ink ribbons. No promises are being made, but I have decided to attempt to create an automatic ribbon re-ink contraption. (Note that in my research, I also found 026 keypunch ribbon for sale here: https://www.aroundtheoffice.com/IBM-026-Keypunch-Ribbon/productinfo/RP-520-IBM/ , let me know if you feel we should purchase a sample or two).
Let me know if you have any questions, or want further information about the above points.
Mariane

Demo - from Paul Laughton - January 11, 2023
1401 Demo, 12:30, 1/11/23
Samuel Planefield and I gave a private demo to  a group of 24 high school students (11th and 12th grade) from The Bay School of San Francisco.

We used the three 029, the 083, CT for tape operations and DE for running programs.
CT refused to load any program deck. 

Paul

1401 Demo, 3:00 PM, 1/11/23
Karen Sun and I gave the regularly scheduled demo about a dozen guests. 

We used the three 029s, the 083, CT for tape operations and DE for running programs.
CT refused to load any program deck. 

Karen shadowed this demonstration performing all the required tasks to qualify as a 1401 Demo Operator. 

Paul

Maintenance - January 11, 2023 - from Robert Garner, Ken Shirriff, & David Clementson
from Robert Garner
1401ers,

Here’s the 1401 restoration status for Wednesday, Jan 11th.

Today’s focus was on relay testing; neon bulb/light & fuse indicator issues in DE 729 #1 and the DE 1401’s TAU control panel. (as it's done before, DE tape drive #1 mysteriously and inexplicitly returned to life, satisfactorily performing tape loads. Relays!&^*%$)

Detailed reports and the attendance table:

In the workshop, Tom continued designing and assembling his Arduino-based interface to the 1401’s serial I/O port, photo below:

from Ken Shirriff

The DE #1 tape drive was unexpectedly working today after its intermittent problems. Apparently, Frank has the magic touch and restored it to operation, David and John are investigating two neon bulbs that aren't lighting on the back panel.

I looked into the TAU problem with the B register A bit light not lighting. (This corresponds to the "low" A bit signal.) This happens with multiple drives, so it's a TAU issue, not a drive issue. I traced the signals and the problem appears to be early in the signal processing. This problem is likely to cause bad data when reading so it's worth fixing.

On the oscilloscope, the signal levels from the drives vary widely from 4.88V p-p to 8.16V. The A bit is in the middle at 5.84V, so that doesn't seem to be the issue. The signal goes through a DZA amplifier/clipper card and the low output looks maybe .5V lower than the others, but it's hard to tell if this is the problem. Next, the signals go through YCC peak detector/integrator cards. This is generating spike signals but they look 4 volts too weak. Later in the processing chain, the low A signal disappears entirely and the light doesn't light.

My conclusion is that the input from the tape drive is okay but bit A low deteriorates after the first two cards. My plan is to swap out those cards to see which one is causing the issue.

Ken

from David Clementson

  • I started the morning working with Stan to source some AMP crimp pins he needs for his relay tester project. The pins are needed to connect the relay sockets to the tester circuitry. Unfortunately, I was unable to find a suitable modern replacement for them, so it looks like we will need to harvest some from another machine. There was discussion about going to the Yosemite warehouse to raid pins from a 088 collator there, and I believe Tom agreed to get some pins the next time he went to the warehouse.

  • At lunch, a new volunteer, John Howard (I hope I got that name right) joined us and he and I agreed to work together on the 791 DE #1 tape unit. Talking with Ken, he suggested we focus on validating the neon indicators. After much discussion and head scratching about where "ground" is on a 729, we concluded that the chassis ground and circuit ground are the same. Next, we discovered that the +62V voltage rail that the neon indicators use measured only about +2.5 VDC. We manually tripped the +62V circuit protector and observed that the +62V rail was pulled by the neon ballast resistors to -48V as we predicted. When we reset the circuit protector, the rail returned to the fault value of +2.5V. I looked at the schematics and realized that +62V is generated by a half-wave rectifier off a main transformer tap. So I would guess that the rectifier diode has failed short and needs replacement. We should be able to confirm that by looking at the AC voltage of the +62V rail. Replacing that diode looks like it may entail some surgery, however, since the power supply is pretty deep inside the machine.

  • One troubling thing is that somehow during our debugging, something caused the FUSE indicator light on the front panel to illuminate. I power cycled the machine and the FUSE light remained lit, so we may have broken something else somehow.  @Ken, is there any known issue with FUSE light coming on for this machine?
DC

Demo - from Paul Laughton - January 7, 2023
Steve Madsen and I gave the demo to about 40 people. We used CT, the 026s, and the 083. All equipment worked A-OK. Most of the guests were from the Bay Area with a few from outlier states like Wisconsin. There were also a couple of computer experts from Israel with some fascinating stories. 

There was also a demo at 1:00 PM with a different team. I did not see the demo but I am guessing that they had a lot of guests from the number of cars in the parking lot.

Paul Laughton

Maintenance Report from Stan Paddock - Wed Jan 4, 2023
People who attended the CHM, and hours, were:

Frank King   6.0  
Stan Paddock   6.0
Tom Szolyga   6.0  
Robert Garner   4.0

Demo - from Paul Laughton - December 28, 2022
Tim Robinson and I gave the last demo of 2022 this Wednesday, 12/28/22.

We had about 30 guests in the actual demo with another half dozen or so before and after the demo.

The CT 1401, 083, 029s and 1403 worked perfectly.

The CT 1402 card reader not so much. It would load and run a Big Print program fine once. A few minutes later it would refuse to read the first card. It did not matter if the first card was a FC of the demo deck or a Trojan card. We solved that problem three times by power cycling the 1401. The power cycling worked three times ... but not the fourth time. The 1402 would, however, load and run the Powers Of Two with no problems. Go figure.

The microphones seem to be going from bad to worse. Today, they were over amplified even at very low console settings. The sound was badly distorted, The guests had a hard time understanding either Tim or me.

Another issue. I took the mic batteries out of the charger but they quickly failed. I used Jack's supply of non rechargeable batteries for the demo.

If you put batteries into the charger and the charger lights do not come....unplug the charger from the extension cord and plug it back in. The lights should come on.

If you leave batteries in the charger without the lights on then the next team that works with those batteries will be working with uncharged batteries.

Happy New Year!
Paul

Maintenance - December 28, 2022
from Robert Garner - 1401 Restoration Volunteer Hours - 12/28/22
Clementson, Dave 7
Make, Brenda 7
Menendez, Iggy 6
Shirriff, Ken. 5
Szolyga, Tom 5

from David Clementson 083 Sorter - December 28, 2022
Just me working on the unit today:
  • Per last week's observation, I adjusted the Cam #5 "make" event angle to the value listed in the Service Manual. This is about two degrees earlier than before adjustment.
  • *Note* The cams are held in place with spline-type setscrews, so restorers need to be careful to use the correct spline wrench on them. Fortunately, we have the correct wrench in the lab toolbox.
  • After the cam adjustment, the problems with sorting pockets #4 and #7 appear to be fixed. I ran the test deck through about 10 times without a single reject. So I think we can cross this project off the to-do list.
DC

from Ignacio Menendez - German tape 1 activity - December 28, 2022
Brenda, Ken, a Gentleman that I didn’t get his name, and I worked on the intermittent ‘fail to load’ problem on the German #1 729 tape drive.

After trying many hunches that appeared to work briefly, we all kind of agreed that this intermittent problem was caused by most possibly a wire relay contact problem.

Brenda took it upon herself, to replace all the contact wires, on all the relays affected in any way, with the intermittent failure to start the vacuum pump, to proceed with the load operation.

Stan is modifying his Arduino relay exerciser/tester to check out all these 729 relays after the contact wires are replaced.

I showed Brenda some of the quick things to check, clean, and oil to prevent problems.

We ran out of time, and I am happy and grateful to have spent some time with these excellent CHM folk.

I think my VA appointment is sometime around February 2023, and I hope that you let me in to work with you.

Iggy

from Brenda Make - Re: German tape 1 activity - December 28, 2022
Hi,

Nine, relays 119-127 have been cleaned, not yet checked.

Ken and Stan located the bottle of shiny silver-plated relay wires!

(These replacements vary from the originals only by a slight manufacturing marks on the side from perhaps from an overzealous feed roller.)
I am very happy to see NOS wires, as some of the originals have wear marks in the silver from the contacts.

A relay appeared to have already had its wires replaced. I gave them minor polishing with #2000 paper.
As for the many of the others, the wires looked pretty bad. While tarnished silver is conductive, the tarnish approaches that which can be felt by a finger nail,

Old wires were put in a bag for later cleaning.
     (Ultrasonic? Dip? Inspection)

Generally, of those surfaces worked on, the aim was to keep as much silver on the parts as possible.

For each relay:

  • Removed wires
  • Cleaned the common contacts( which aren't usually the problem), by rubbing fairly firmly across the embedded contacts with a swab with Deoxit
  • Polished socket pins with folded #2000 until they looked decent, but not perfect, as perfect removes more silver than decent.
  • Using the magic burnishing tool, (which is about #1500-#2000 grit), I took about 5 light passes across each contacts, keeping the burnisher level. I flipped the relay to do each side, keeping the burnisher reasonably level. The tricky bit is making sure the center contacts are also done.
  • The relay was blasted with spray alcohol, and then air.
  • Deoxit was put on socket pins.
  • Wires were inserted, in pairs, which makes it a lot easier than doing them one-at a time. A dental scraper helps whilst coaxing the pins through the relay "Wire Bracket".
  • Wires are checked to make sure wire hooks are parallel, centered
For reference, #2000 paper is generally the last sandpaper before buffing. Thanks Ken, for getting some!

[Also, as I read the manuals, I'm becoming curious about the timer, too.]

Thank you Iggy for your help, today!

Take Care,
Brenda

from Brenda Make - Documentation Concerns - December 28, 2022
Hi,

We seem to have a few issues with documentation, which is important for maintaining these systems.

  1. We seem to be missing some debugging/flow charts for the German 729-#1
  2. Several our the scanned pdf's are illegible. With hand-drawn schematics, and ALD's/ILD's with ribbon patterns on it--even the source is not good, but we have to be able to read the numbers/letters.
I understand that people have taken a lot of effort the scan the documentation we have. Though, much of it was scanned in not in grayscale, but black/white. Because there is a lot of documentation, the decisions were likely good for then, to put reasonable limits on documents source scans and PDFs.

Now storage is cheap, so the idea of spending 100-500MB for the source scans may seem jarring, but the source for 1,000 documents could fit on on 1TB.

- Does the museum have one of those document scanners that doesn't break the binding?
- Is there a camera to photograph the large books?
- Is there an area for photographing the books?

[As an experiment, I downloaded the source 223-6988-3 Instruction for 729, II, IV, VI from archive.org, which looks really great. I am cropping all the pages. Adobe Reader broke my Acrobat X, but what I'd like to make a lightweight PDF, but also fine preservational PDF, if I can get Acrobat going again.]

Take Care,
Brenda

from Robert Garner - 1401 Restoration Report for Weds, 12/21/2022
1401ers,

Here’s this week's Christmas Eve 1401 restoration status for Wednesday, Dec 21st.
Overall, Brenda and Ken struggled with an unreliable relay in the obstinate & unwilling-to-load DE 729 #1, and Mariane and David worked on the 083 sorter's inconsistent handling of 4 and 7 card holes. It looks like the cam timing is slightly late.
Their detailed reports and the attendance table:
HAPPY HOLIDAYS !!!!

1401 Restoration Volunteer Hours - 12/21/22
Clementson,, Dave 7
Garner, Robert 4
Kim, Mariane 7
Make, Brenda 7
Shirriff, Ken. 6
Szolyga, Tom 7

from David Clementson - Subject: 083 Sorter restoration report for 2022/12/21
It was just Mariane and me working on the unit today, still debugging the occasional sort rejection of cards destined for the 7 and 4 pockets.
  • Per the prior week's plan, we used a contact burnisher on timing cams #5 and #6. We followed that with a card strip and isopropanol cleaning. After exercising the machine with a few decks, its behavior was about the same.
  • We then set out to examine the thyratron grid waveforms. The schematics we have been using are for a later variant, one that uses a printed circuit board for the thyratron matrix. Our machine uses a "steckeinheit" (plug-in unit) construction, and the relevant schematics are in German. We discovered that the thyratron cathodes appear to be grounded on the PCB variant, whereas they are connected to -48V on our steckeinheit version. Nevertheless, there is a ground reference terminal on the steckeinheit assembly, so we were able to use that as an okay ground reference for the scope.
  • Please note that page 2 (most of the thyratron matrix) is missing from the German schematic PDF online labeled "IBM_Sortiermaschine_Type_083_336584_C.pdf," so it should be rescanned. The page-to-page overlap is also pretty minimal, so it would be good if that could be increased in the rescan.
  • We captured a bunch of waveforms for the thyratron anode, grids, and the sorter brush: all of the relevant signals to determine the root cause of the thyratron misfiring. Sure enough, we were able to see a pocket 7 misfire due to the grid 1 signal having insufficient amplitude to fire the tube.
  • During the 3:00 demo we went to a conference room to discuss our findings. We were able to convince ourselves that the likely cause of the runt grid pulse is a late "make" connection on cam switch #5. The theory is that the punched hole has passed and the brush signal is already beginning to decay by the time the cam switch closes. This causes the thyratron to miss the brush event and the card gets rejected as if it had no punches.
  • After the demo, we measured the entire card cycle to be 65.5 ms. The theoretical cycle is divided into 32 11.25 degree slots each lasting 1.8 ms. The expected brush events are supposed to fall within those 1.8 ms windows. Looking at the timing of a functional pocket (#8) the grid pulses are indeed about 2 ms, so we have some confirmation of our understanding of the timing.
  • We scrutinized the cam #5 timing statically and found that it does indeed appear to be late by a little bit. But it was getting late, so we agreed to adjust that cam next time.
  • I will hopefully be able to make the session next week, but Mariane will not. Frank is also out until January. So I will attempt to adjust the timing myself. The set screws locking the cam to the shaft are easily accessible, so it should be a pretty simple process.
DC

from Ken Shirriff - 12/22/2022 at 10:19:29 AM PST
Brenda and I worked on the 729 and made negative progress :-) We fixed DE 729 #1 a couple of weeks ago by replacing relay #119. However, the drive failed again, refusing to load. Brenda tested relays and didn't find a good one so she disassembled one and tediously cleaned the wires. We put the cleaned relay in the 729 and it still didn't work. I swapped relays #119 and #120 (which caused the drive to work last week) and the drive still didn't work. So I'm suspicious that the drive has encountered a new problem. Brenda is making a relay extender so we can probe the relays and see what is happening.

We also looked at a bad light on the DE TAU control console: the register B bit A light doesn't illuminate. Brenda replaced bulbs last week, fixing one failed light, but B bit A still didn't work. We probed the SMS cards with an oscilloscope and found that the driver card is not getting a signal, so the problem is in the TAU circuitry, not the bulb. The demos were starting, so we will continue the investigation next week.

Ken

from Brenda Make - Relay Extender Circuit - 12/22/2022

Demo @ 11:00 AM Dec 17, 2022 - from Stephen Madsen - dated 12/17/2022 2:59 PM
Hi,
Paul Laughton, Michelle Yu (trainee) and I demonstrated the IBM 1401 to 35 Boy Scouts and 25 adults.
We used CT and everything worked fine. Thank you, restoration team. Michelle did a great job.

Stephen H. Madsen

Demo @ 1:00 Dec 17, 2022 - from Paul Laughton - dated 12/17/2022 9:16 PM
I was the Lead. Pat Buder was titular Operator while Jim Maurer, the Operator Trainee, did the actual Operator tasks. Jim did those tasks well and is now eligible to be labeled Operator (aka 1401 Docent).

Just before the demo there were only about seven guests in the room. A few were from Israel. The rest were from California. About a femtosecond before the demo start time the dam broke and about 30 more people flooded in.

We used CT, the 083 and all three 026s. All worked "demo" perfectly. The crowd was pleased judging from the applause.

It has been a good week for new Operator qualifications with more great candidates in the pipeline.

Oh...and the microphones were working to our satisfaction. Thank you Jon Plutte.

Paul

from Stan Paddock 1401 Restoration Report for Weds, 12/14/2022 & resignation
Brenda Make 6.0
Dale Jelsema 6.0
Dave Clementson 6.0
Frank King 6.0
Ken Shirriff 6.0
Robert Garner 4.0
Stan Paddock 6.0
Tom Szolyga 6.0
Mariane Kim 6.0

A tribute to Ken Shirriff, Brenda Make, Mark Verdiell who all contributions for bring the first tape drive on the German system back to life.
On day one, when power up the tape drive would not do anything.
After some testing, it was found the -48 volt circuit breaker tripped every time power was applied to the tape drive.
This would be easy to find if -48 volt did not go every where through all parts of the tape drive.
Early efforts of divide and test did not find the problem and took a lot of time.
Mark Verdiell brought in a DC amp meter where the probe is touched to insulation of a wire and will give the relative current flowing through the wire.
Using this cool tool, it was found that the heavy current was coming from the top control panel of the tape drive.
When the control panel was opened, it was a mess.
Several switches had broken lose from there mounts and had shorted to neighboring switch.
Now that the problem was found, the question became how to fix the problem.
The switch mounts were 50+ year plastic parts.
In the warehouse we have some tape drives that will never come to life again.
Tom Szolyga spend his Thursdays at the warehouse.
We showed Tom what was needed and the week, we had the necessary assembly to fix the tape drive.

Dave Clementson, Mariane Kim, and Frank King continued their work on the IBM 083.
While if you power up the machine and put cards in the hopper,
It will feed the cards and flash the cards and put the cards into pockets.
While this makes the Docents happy, the fact that some cards went in to the wrong pocket does not make the 1401 repair time happy.
The sorter will fail on 7 and 4 when it feel like it.
Work on this problem continues.

NOTICE ===========================================================

A combination of old age and health issues force me to resign my position as "1401 Status Report Editor"
If you would be interested in taking this function please contact the old "1401 Status Report Editor".

from Robert Garner 1401 Restoration Reports for Weds, 12/14/2022
1401ers,

Here’s this week's 1401 restoration status for Wednesday, Dec 14th. Overall, Brenda, Stan & Ken worked on the obstinate & unwilling-to-load DE 729 #1, and David, Mariane & Frank worked on fixing the 083’s inconsistent sorting of 4 and 7 holes. Their detailed reports as follows:

From: David Clementson
Subject: 083 Sorter restoration progress report for Dec. 14, 2022
Date: December 14, 2022 at 6:39:47 PM PST
Mariane, Frank, and I met this morning at the exhibit to talk about next steps for the sorter restoration project:
  • The resonance and delay caps have been installed and tested okay, so we can check those off the to-do list
  • The initial status is that the unit does sort cards, but it improperly and randomly rejects valid cards, mostly for pockets 4 and 7.
  • We were concerned that the timing cam contact cleaning job we did earlier was insufficient and that contamination remained on some contacts. So we first cleaned each contact pair thoroughly using strips of punch card soaked in isopropanol. We cleaned each contact until there was no visible residue being left on the cleaning cards.
  • Powering up the unit after the cleaning at first caused many cards to be rejected, but after a few sorts of our test deck, the unit settled into a pretty reliable pattern of mostly sorting correctly but occasionally rejecting valid 4 and 7 cards.
  • Next we tried using some test equipment to analyze the thyratron firing for pockets 7 and 4. We tried the Tek digital storage scope at first, but the user interface for the storage function proved inscrutable. Next we tried my HP 5316A counter, but it registered thousands of counts per card instead of just one. I'm sure the problem was excessive noise pickup on the probe ground.
    Correction: the 083 thyratron cathodes are actually grounded, so it should be possible to get a good look at the plate and grid waveforms with a scope probe referenced directly to the cathodes. Hopefully we will be able to see from the grid signals why the affected tubes aren’t firing. Next week.
  • Finally we used the (very nice) Rigol scope from the lab. We were able to view the ground noise pickup (explaining the counter's confusion) but we could also clearly see the thyratron firings. Using the scope's zoom function we were able to confirm that the thyratron firing count agreed with the pocket contents. If a card was rejected, it was visible as a missing pulse on the scope. So we can safely conclude that the root cause of the sorting issue is "upstream" of the thyratrons.
  • Next, I ran the sort deck repeatedly to collect some stats. The number of false rejects seemed to drop, and several sorts worked perfectly, but it was not 100%. All of the rejects were either pocket 4 or pocket 7. We concluded that the problem is most likely with the two cam switches (C5 & C6) associated with pockets 4 and 7.
  • Frank and I manually checked the static timing of C5 and C6. The make and break angles were all very close to the target values. Only the C6 break time was out of spec, but only by 1 degree.
  • From here we theorized that C5 or C6 must be dodgy and should have their contacts burnished. I am not a fan of burnishing plated contacts, since abrasion will eventually destroy the plating. But Frank indicated that contact burnishing was a standard maintenance technique.for these machines, and the official toolkit was flush with burnishers. Maybe the contacts aren't plated? Who knows. Anyway, a crowd was assembling for the 3:00 demo, so we had to close the unit up. We will need to defer the burnishing until next time.
DC

Dec 14, 2022 - from Ken Shirriff - dated 12/16/2022 10;34 AM
Status:

Brenda, Stan, and I worked on DE 729 #1.
Brenda and Stan replaced some broken bulb holders in the drive.

In the 1401 TAU controller, two indicator lights didn't work so Brenda replaced them. One now works but the other is still dead, so there may be a problem with the SMS driver board.
Unfortunately, DE 729 #1 stopped working again, not loading as before. Then it started working again, and then failed again. The problem before was relay 119. We swapped relays 119 and 120 and it worked, so the problem must be relay 119. Cleaning the contacts didn't help. Brenda disassembled the relay for a thorough cleaning.

The neon for relay 119 doesn't illuminate on the back of the drive. However, the neon didn't illuminate when we swapped the relays, so it may be an independent failure. But it would be a strange coincidence for relay 119 to be bad and the light indicating relay 119 working to be bad too.

Ken

Dec 14, 2022 - from Paul Laughton - dated 12/14/2022 7:42 PM
I was the Lead. Jack Ghiselli was scheduled for Operator but could not make it due to illness. Mariane Kim was scheduled as the trainee Operator Shadow. Due to the circumstances, Mariane took on the full Operator role. She did the job in an outstanding manner. To add to that, she gave an impromptu, 12 minute full demo to five delighted post demo guests.

We started with CT for the demo but the CT 1403 could not do line feeds or carriage restores. We quickly shifted to the standby DE for the Big Print. We shifted back to CT for the tape demo.

The 083 operated nicely for the guests. My thanks to the restoration team including Mariane for their 083 repair efforts.

Frank King provided us with a print train that we could use for the demo.

The demo microphone system had been worked on; however, the output volume of both mics are now lower than either one had been before. We cranked up the master volume to make ourselves heard but this also resulted in more feedback zones.

The 083 is now not powered by the Keypunch breaker. It is powered by breaker #6.

Paul Laughton

Dec 10, 2022 - from Paul Laughton - dated 12/10/2022 6:03 PM
I was the Lead on two different 1401 Demos today. One at 11:00 AM and one at 1:00 PM

At the 11:00 AM demo, Jack Ghiselli was the Operator. We had 25 guests including some from Italy, India, France and Germany.

At the 1:00 AM demo Kathleen O'Brien was the Operator. Samuel Plainfield was the shadowing operator trainee. Samuel accomplished all the required tasks in excellent manner and is thus now qualified to be an official Operator.

We had over 30 guests attending this demo. Most of them were from the Bay Area. I was surprised at the number of people and their enthusiasm at this additional Saturday Demo. It looks like the 1:00 PM Saturday demo is going to be a big hit. CT, the 083 and all three 026s participated for both demos. All worked as required.

Paul Laughton

PS: Some work had been done on the mics. Mic #1 is now as loud as Mic #2. Two new, different, Mics had been delivered to Jesse this week. However, we were not able to use them as Jesse was not there and no one knew where they were.

Dec 7, 2022 - via Robert Garner - dated 12/9/2022 5:44 PM
1401ers,

Here’s this week's 1401 restoration status for Wednesday, Dec 7th.
Overall, Brenda, Ken & Stan successfully addressed a misbehaving relay in DE 729 #1; Mariane & David tuned up the 083 sorter; and also David proposed an ingenious debugging/tuning scheme. Their detailed reports as follows:

- from Brenda Make - DE 729 #1
- from Mariane Kim - 083 sorter
- from David Clementson - 083 sorter

from Brenda Make - DE 729 #1
German Tape Drive #1 appears to work for a preliminary bit-write test!

(Last week, we had corrected shorted lamp and accessory power cable pin, and replaced the -48v relay which was out for short tracing, but we did not have time to test the drive before the demo.)
Today, Tape Drive #1, Accessory power cable appeared working, but Drive #1 would not load the tape.

Ken located the relay suspects might be affected, which is quite a feat with IBM's non-standard and as-built schematics.
We removed the #119 relay for inspection.
(The relay extender cable appears mislaid.)

Stan wanted the relay be out for testing, and good thing.
One set of contacts was not making connection, so Stan cleaned it.
We retested the relay, finding the contacts making connection.

The relay was reinstalled.
The drive loaded the tape!

Ken did a bit-write test on the drive, using half of the tape. German Tape Drive #1 appeared not to show any errors on the TAU Panel!

Issues Were Found:

  1. German Processor/Tape Adapter Unit: Showing Errors, but Not Halting on Errors (on Tape #3, below).
  2. German #3 Tape Drive: Reporting Errors on Check-Bit during Write-bit test.
  3. German #1 Tape Drive: 1 Rear Neon status lamp not lighting when should.
  4. German Processor/Tape Adapter Panel: Has 2 burned-out bulbs: LRCR 2-BIT, and BREG A-BIT.
Noting: no terminator was installed to German Tape #3, but there seems to be a question if that could have created the errors.

Also performed on German Tape Drive #1, just before the demo:
     Front tape area was vacuumed.
     Some vacuuming was done inside unit to remove dust and abrasive rust particles.
     Covers were replaced and fastened.

Take Care,
Brenda

- from Mariane Kim - 083 sorter
This Wednesday, we continued our work on the 083 card sorter.

Power Supply:
As a reminder, the original capacitor labeled C2 in the wiring diagrams had 1000uF 50V specifications. Last week, we put in a temporary fix using a 2500uF 50V capacitor, and the demonstration team reported that the sorter not only powered on but also sorted correctly.
This week, after finding shortcomings with three different 1000uF 50V capacitors (one arrived as an axial capacitor instead of radial, one required soldering, and one was too short to use existing wiring), we decided to continue using the 2500uF 50V capacitor. We installed this into a bracket Dave provided and confirmed that the machine powered on as expected. The only expected change in behavior with a higher capacitance was a slightly longer startup time for the machine.
I am attaching a photo of the assembled power supply circuitry (prior to cleaning), to this email.


Resonant Capacitor:
Once C2 was permanently installed, we checked the voltage across the resonant capacitance circuit and found it to be approximately 415V. This indicates that the three capacitors currently in use are providing an appropriate capacitance for the circuit to resonate, so we re-installed a plastic barrier to alert anyone to the high voltage. This is consistent with the original resonant capacitor which had a High Voltage label (per the 083 Parts Catalog).


Actual Sorting Functionality:
Finally, once the machine was turning on and starting reliably, we turned our attention to the sorting. It seems that the machine could reliably sort most rows with the exception of 4 and 7. Using the service index that Robert provided, we were able to determine that both these rows were connected to cams 5 & 6, so we did a visual inspection and found that the interior of the machine was filled with dust and lint from the cards. We vacuumed away as much of the surface dust/lint as we could and cleaned all 16 cam contacts using Deoxit and alcohol.

Following this cleaning, we noticed a reduction in the number of rejected 4 & 7 cards, but not the 100% success rate we were hoping to achieve. We believe there is a poor contact somewhere in this circuit (perhaps caused by switch bounce), and have planned to diagnose this next week using one of the scopes.

Marc mentioned to us that since this 083 sorter came to CHM, it has never known a 100% successful sort rate. We are looking forward to fixing that :]

-Mariane

from David Clementson - 083 sorter
Thank you for the excellent report Mariane. One further observation is that not only were the cam contacts found to be surprisingly dirty, but I think it is likely they are still dirty. I propose that next time we do a much more thorough job with the cleaning, wiping each contact set until there is no evidence of any dirt being removed. Last time we only did one wipe per contact, so it is likely some dirt was left behind.

We also talked with Frank and Stan talked about using the lab's storage oscilloscope to observe the cam switch timing. We would pick one switch as a cycle-start reference to trigger the scope, then observe the other contact closures in temporal reference to it. The trick with that technique is to get the cams to show consistent waveforms regardless of the data contents of the deck. That would allow us to differentiate cam switch inconsistencies from deck data variations. Jumpering the sort brush directly to the contact roller would do that, although loading a homogeneous deck with all holes punched would do so as well.

I also think that using a digital event counter to count thyratron anode firings and comparing them to the known deck contents would be useful. For example, if we sort a deck that we know has ten "7" cards, then we should see ten firings of the "7" thyratron. If we can't correlate the card count with the thyratron firing count (or if the firing count is inconsistent from run to run of the same deck), then we know the problem is logically "upstream" of the thyratrons. If the counts and are correct and consistent, then the problem must be downstream from the thyratrons, and we can turn our attention to the sort knives and their relays. Comparing the counter results with the final pocket counts will also proof the circuitry downstream of the thyratrons. I have an HP 5316A Universal Counter and a 10X scope probe which should do the trick. The 5316A can also do time interval measurements between cam closure events which we can then translate to angular values for checking against the cam timing table. But the storage scope will show any intermittent or fluctuating timings whereas a counter would not.

DC

Dec 03, 2022 - from Paul Laughton - dated 12/5/2022 12:27 AM
Duane Sand, an Operator Trainee supervised by JackGhiselli, and I gave the demo to about 40 people. Duane did an excellent job and is now qualified to be a regular Operator.

The group was mostly local folks with the exception of some from Sweden and Italy.

Thanks to excellent work of the Restoration Team, the 083 sorter worked and worked perfectly. All the cards dropped into the correct slots.

We used CT which worked perfectly.

We did have a new problem with audio equipment. There are new audio feedback zones around the 1402.

Paul

Nov 30, 2022 - via Robert Garner - dated 12/1/2022 2:08 PM
1401ers,

Here’s this week's 1401 restoration status for Wednesday, Nov 30th. Overall, Brenda & Ken successfully repaired defects in DE 729 #1 and Mariane & David in the 083 sorter, their reports as follows:

- from David Clementson
- from Mariane Kim
- from Ken Shirriff

- from David Clementson
Mariane and I spent the morning reviewing the schematics to confirm the identities and wiring of the two capacitors that have been receiving attention of late:
  • The filament transformer resonance capacitor. This is the one that Frank fashioned a temporary replacement for using a series/parallel network of motor start caps. And,
  • The power-on delay capacitor C2. This is the large orange electrolytic cap, 1000uF at 50V, whose purpose is to delay the operation of the start-up relay until the tubes have come to temperature and their anode bias supplies have stabilized.
When Mariane first opened the unit, she found that the resonance capacitor had had one if its leads disconnected from the transformer. Since it was pretty obvious how the lead should be reconnected, we did that.

Next, we studied the circuit to see if we could find a similar temporary cap to use until the correct replacement part arrives tomorrow. We dug through the capacitor tote in the lab and found a 2500uF 50V cap that might work, although with a longer turn-on delay. After discussing the temporary fix with Robert and Stan, Mariane wired in the new cap.

After some further checking, we powered the unit up. No smoke! After a rather long time, the start relay finally closed and the green power light lit. We loaded a deck and it sorted, although we were unable to confirm that the sort was correct. We will do that next time.

Next, we powered it down, connected a DVM (Fluke 112) across the resonance cap, powered it on, and recorded 415 VRMS across the cap. I asked Stan if there was a spec for this voltage and he said 415V was "pretty good." Only after reading the (newly scanned and posted) Theory of Operations Manual did I realize that the proper way to check this circuit is probably to measure the filament voltage and ensure that it is 6.5 VAC +/- 5% over an input line voltage range of +/- 10% (manual page 20.)

I then powered down the unit and disconnected the resonance cap and measured its capacitance with the DVM. The measured value was 1.65uF. I am wondering if this agrees with the value printed on the original cap. I also wonder if anyone has ordered a proper replacement cap, or if y'all would like me to do it. If so, what value should I order?

That's it for today.

DC

- from Mariane Kim
This week, Dave Clemenston and I worked together to understand the 083 power supply circuitry, and ultimately get a "temporary fix" running for the demonstrating Docent Team.

We focused our attention today on the missing C2 delay start capacitor (as Frank and I had previously found that it did not have the expected 1000uF capacitance).
Since I have not yet received the correct 1000uF 50V electrolytic capacitor from eBay, we used a 2500uF 50V/60V electrolytic capacitor we found in the restoration lab as the "temporary fix". Please find image attached.

Using this temporary capacitor, we were able to power on the machine (green power light came on) with about 420V across the resonant capacitor circuitry. The machine also was able to run cards, with cards falling into distinct slots. Note that, at the time of my departure, we were not able to determine whether they were properly sorted. That might be a challenge for future weeks! :)

Next week we should be able to install the correct C2 as a more permanent solution.

When I left, the machine was plugged in and handed over to the Docent Team for their 3pm demonstration.

Kindest regards,
Mariane

- from Ken Shirriff
Summary: We found the short in the DE #1 729 tape drive. We'll power it up next week and see if it works.

Details: The tape drive had a short in the -48V circuitry that we've been trying to track down for a while. Marc brought in his vintage HP DC current meter (originally Robert's) and we hooked up a current-limited power supply. This helped Brenda and me narrow down the path of the short to a few branches.

We had earlier ruled out the front panel lights as the problem, but I stared at the panel for a while and eventually noticed the short,
(annotation by Robert):

Brenda fixed the wiring and re-installed the -48V circuit breaker. We ran out of time so we'll test it out next week and hopefully this fixed the problem.

Ken

Nov 26, 2022 - Demo
from Stephen Madsen - dated 11/27/2022 10:11 AM
Paul Laughton, trainee David Riazati (who did a good job), and I demonstrated the 1401 to 39 people. The 083 sorter was not working, so we talked about it. We used CT. It worked fine during setup, but it refused to read cards during the demo. We switched to DE to run BIGPRINT. It was helpful to have three people for the demo, so one could get DE ready while the other two continued with the presentation. DE ran fine during and after the demonstration. Also, after the demonstration, CT was restarted, and it then read cards OK.

Stephen H. Madsen


Kate McGregor is actively soliciting ideas and suggestions for improved (longer lasting ) audio equipment for demonstrators.
Also coordinating with the CHM Media team, which oversees A/V for the museum.

Nov 23, 2022 - Demo
from Pat Buder - dated 11/26/2022 2:39 AM
On Wednesday Paul Laughton and I gave the demo to a holiday crowd of 58, with another 25 coming before and after the main demo. 1401 Operator class member Larry Hara also assisted.

We ran on CT, which generally performed well. Tape #2 refused to unload, and the 083 sorter and 026 #2 are still down.

Larry was doing an official "shadow" of the Operator role. This means he did most of the Operator functions, including speaking parts. He did a good job in all areas except describing the tapes, which he elected not to do this time. Other members of this Operator class will be doing similar shadowing over the coming months.

Our main problem was the audio/visual system. One of the microphones was unable to be worn as a headset as designed. There is a very small, fragile plastic piece that connects the mic boom to the main frame that had broken. This failure mode is chronic on this mic model. The problem is due to the design of the unit, not abuse. Due to the failure, the mic had to be held in one hand rather than being placed on the head. Given that the Operator is loading cards and doing other jobs that require two hands the impact was high. After the demo Paul replaced the failed mic with one he had purchased. This has also been necessary on multiple occasions in the past.

Pat


Nov 26 - There have been several e-mails today between Kate McGregor (CHM staff) and the docents clarifying the mic/audio situation -
"Major issues like this should not be happening, and our volunteers should not be (or feel) responsible for replacing essential equipment for our interpretive programs."

An hour after the italics just above

Kate,

Just to be clear about the core issue here, it's my opinion that the microphones we have are the wrong design. They are made of very thin wires and do minimum obstruction of the wearer's face. That could be appropriate for a singer or a media personality doing face closeups on camera. The trade off is reduced sturdiness. I would much rather have a design with stronger construction even if it's more noticeable.

We do not abuse the current mics. They are difficult to adjust even when new. With time they become bent and looser. When the flimsy, tiny plastic part eventually breaks it's game over as reported last week.
Given the noise level in our room the audience can appreciate that we must use mics and amplification. I'm sure they would much rather hear us clearly than worry about the cosmetics.

I know that mic designs that are more suitable for our needs exist. We would be happy to provide feedback on potential replacements if other solutions were to be considered. The audience and demonstrators would certainly appreciate it.

Pat

Nov 23, 2022 - Work on the 729 continues.
- from Brenda Make, Work on the 729 continues - dated Nov 25, 2022 - 11:59 AM
- from Ignacio Menendez, Thanks - dated Nov 25, 2022 - 12:55 PM

- from Brenda Make, Work on the 729 continues - dated Nov 25, 2022 - 11:59 AM
Hi,

(Work on the 729 continues. ) - revised Nov 26

--------------
If it exists, another 729 power cable should brought from storage and/or the cable should be redressed with a new pin, as the pin only has 1:4 of its barbs on it. It may work for short-term use. I hope to bring some yellow tape to mark the cable.

The accessory power cable had a shorted pin that as Ken had mentioned (kudos), might have been causing issues with the power converter. This was a serious issue as it dealt with mains power.
Kudos to Marc for noticing the bent pin on the connector by feel and/or the dim light at the base of back of the 729 tape drive. (Also, thank you Stan for covering up the power-converter voltage selector switch for the German 1401 system.

I straightened the pin, and polished it with #600 sandpaper, noticing that some of the barbs that held it on were missing. My attempt at straightening the remaining barb failed. Marc found a microscope in back, straightened the barb better, and inserted it in the connector. This time, the pin held under pressure enough for a usage attempt.

The pin which appears to be a replacement pin, longer than the original. The longer pin was snipped, and the end filed round'ish. Upon reassembling the connector, the ground lug failed from fatigue. A new connector was crimped, dripping-wet soldered, with heat-shrink tubed.
The ground-lug clearances are tight with respect to the housing. Marc reconnected the cable, as I wanted someone who had previous feel for it to do it. I labeled the connector with the problem.

The interesting chronology about this problem:

  1. The cable was previously damaged/repaired.
  2. A substitute pin was used, which longer than the original. (What seemed to be a pin that was pulled out was really a longer pin.)
  3. Not noticing/knowing that one pin was longer than the other, the cable was likely inserted using normal technique.
  4. Because the longer pin would require an extremely careful axial alignment for insertion, the pin became folded, shorting to another.
I feel that the if we knew the there was an issue with the cable, this might have been mitigated.
For this reason, I request we ID tag name or number 1401 system assets wherever we can, and keep a log/notes for each.

Take Care,
Brenda

- from Ignacio Menendez, Thanks - dated Nov 25, 2022 - 12:55 PM
Congratulations on your thorough troubleshooting and report ! THANKS,
Iggy

A Thank you for class demo - from Prof Ronald Mak, SJSU, Nov 2022
My students and I had a wonderful visit to the Computer History Museum last Saturday. [Nov 19, 2022] Many thanks to Jack and Paul and the computer operator whose name I didn’t catch for helping me present the IBM 1401. There were about 20 students, including one student’s boyfriend, another student’s wife, and a student’s father. All were amazed and enlightened by witnessing the operation of 60-year-old computer technology.

Here’s a short video I made: http://www.cs.sjsu.edu/~mak/CHM/printer480.mov

I hope to do this again in the coming spring semester.
I forgot to mention and thank the gentlemen who helped the students with the keypunch machines. I should have gotten their names, too.

— Ron

Ronald Mak
http://www.cs.sjsu.edu/~mak/

Department of Computer Science
Department of Applied Data Science
San José State University
San José, CA USA 95192

November 16, 2022 - from Pat Buder - dated 11/21/2022 8:38 AM
On Wednesday Tim Robinson and I gave the demo to 26 people from Australia, Nepal, and other places. Another 14 came for mini-demos before and after the main event. 1401 Demo Operator student Larry Hara also assisted.

We began on CT, which couldn't finish loading the BigPrint deck. We quickly moved to DE, which couldn't either. At that point I had the audience remain at DE for show & tell while Tim returned to CT to see if he could get BigPrint to load. When he informed me that a different deck did load successfully we moved back to CT for the rest of the demo. No other problems except that the 083 sorter is still down. Jack, it looks like it's time for running VOBJ on the demo decks again.

At the end of the demo I answered questions while Tim and Larry took care of audience BigPrints. The Australian visitors seemed to have as much fun doing arm exercises with the big cable as they did taking a photo with the it and the USB cable we also give them for contrast.

Pat

November 20, 2022 - Tools, Voltmeters, etc

- from Brenda Make - Wish List
- from Robert Garner - Wish List
- from Marc Verdiell - excellent Fluke for $44
- from Robert Garner - Fluke 101

from Brenda Make - Wish List - 11/19/2022 12:13 PM
Hi Robert,

Hi please disseminate this to parties involved. I would like input from others. There is no I in team. Thank you. : )

Multimeters:

Apparent situation: It seems that we have about a half-dozen analogue meters. I strongly feel that the two Simpsons are worth keeping-even worth repairing if need be, because they are big, classic, and show voltage swings well. As for the other meters, perhaps they should be gone through. New or used, and aside from mine, I feel that we need at least one more safe auto-ranging digital meter. Restoration members may also have older yet workable meters, which may be cheaper yet.

I don't yet have a grasp of the larger financial situation, noting that"

  • We will always need parts, parts, and more parts.
  • If we have a pile of SMS cards to repair, I would rather some money saved for a desoldering gun, as to better preserve the phenolic single-sided boards.
  • We need a tip for the soldering iron, if it's worth putting a new tip on. Hmm.
Tom had an inexpensive meter picked out, likely one he had good experience with, so I hope that his choice is considered, and one of those could be afforded.

As for the Brymen meters, Dave from EEVBlog just stamps his name on them and imports them. I have no idea why Brymen does not import their meters.
Aside from the Dave/EEVBlog cultive personality, they still review elsewhere well for the money, though returns/repairs will be cumbersome, and their probes seem only satisfactory, but those can be swapped out. Greenlee seems to rebrand Brymen meteres, too, but they seem pricier.

Fluke does make a safe meter, and they do warranty their stuff for life. I have a 87v Max, but it's extra toughness makes the meter clumsy to work with, so the standard 87v would do just as well, for cheaper. Fluke has some less expensive meters, but their product line is confusing and mixed with electrician- and automotive- only meters. You might ask others about the lower priced Flukes, and other meters.
$$$$ Fluke 87v https://www.testequipmentdepot.com/fluke/dmm/handheld/heavy-duty-industrial-multimeters/87v.htm
Other helpful Items:

  1. Coated-cloth wire markers such as these, might be a small help our tape drive problem, as to some degree it has to be reverese engineered.

    https://www.amazon.com/3M-ScotchCode-Pre-Printed-Wire-Marker/dp/B001CGX20Q

  2. To keep the nice vintage look to the inside of the machines, wire-tying string like this would be nicer than zip-ties, if we don't have any.

    https://www.amazon.com/Mmei-Sewing-Thread-Stitching-Leather/dp/B06XQ19G2Q/

  3. A bag of cotton rags would be helpful for working on mechanical stuff. The cotton ones, I feel are better than the synthetic microfiber. We have used the same cotton rags for 15 years, and I have been pretty hard on them.

    https://www.amazon.com/dp/B0815RVGKP/

  4. Sewing Machine (light) oil is handy to have, as it is very resistant to gelling and gumming, yet has an old/plastic compatible chemistry. In a pinch it can also be used to remove old grease without solvents or worry for rusting.

    https://www.amazon.com/Singer-2131E-All-Purpose-Machine/dp/B000YZ1Y06

    [Transmission is perhaps even better, but it's unfortunately red. I have some "Fluid Film," it supposed to work quite well as a rust preventer, but because it has lanolin, it doesn't smell nice.]

  5. We will need AAA batteries for the flashlights, disposable or rechargeable. I have some AA, but none AAA.
Take Care,
BrendaEM

from Robert Garner - Wish List - 11/20/2022 9:38 PM
Brenda,

Thanks for your wish list of germane and helpful 1401 lab supplies! :))

As you rapidly noticed, we really do need a real/contemporary multimeter. (Our previous one walked…)
The BM786 looks like quite a capable multimeter and would nicely match our everyday needs (esp its continuity beeper :)
I’ll go ahead and purchase one.
(We could always use a 2nd one, say the one Tom was eyeing.)

Regarding the helpful items: Would you be willing to procure and then submit reimbursement requests against the CHM’s 1401 account??
(A reimbursement request email to George Holmes < gholmes@computerhistory.org >, with a scan of the invoice should suffice.)

> I don't yet have a grasp of the larger financial situation ...

The 1401 fund can cover all the items in your list (and more :).
A desoldering gun would be a nice addition.

Thanks for all your help and expertise — it’s so much appreciated! :))

Cheers,

— Robert

from Marc Verdiell - excellent Fluke for $44 - dated 11/20/2022 10:55 PM
You can have an excellent Fluke for $44:
https://www.amazon.com/Fluke-101-Multimeter-Resistance-Capacitance/dp/B00HE6MIJY
I got myself one and bring it with me so I don’t have to run around for one…

If you want the original feel, last Wednesday I repaired two of the original IBM multimeters, complete with original IBM leads. I hear the IBM electrons feel better when measured with IBM meters
Marc

from Robert Garner - Fluke 101 -- dated 11/21/2022 8:28 AM
Marc,

Thank for getting a Fluke 101 for the lab!

Its basic features and easy-to-hold size make it a good match for our normal needs, esp its continuity buzzer.

— Robert

p.s. I also wonder how IBM electrons/EM waves feel after being sampled and projected into digits. :)

November 16, 2022
- from Ken Shirriff - DE tape drive #1 - Nov 16, 2022- 4:59 PM
- from Robert Garner - DE tape drive #1 - Nov 16, 2022- 5:17 PM
- from Brenda Make - DE tape drive #1 - Nov 16, 2022- 9:08 PM
- from Tom Szolyga - DE tape drive #1 - Nov 16, 2022- 10:03 PM
- from Grant Saviers - DE tape drive #1 - Nov 17, 2022- 7:41 AM
- from Ignacio Menendez - DE tape drive #1 - Nov 17, 2022- 8:06 AM
- - with comments by Allen Palmer, Bill Worthington, Ed Thelen

- from Ken Shirriff - DE tape drive #1 - Nov 16, 2022- 4:59 PM
Summary: DE tape drive #1 still has a -48V short. Also, the power cable is messed up.

First, the power cable. Frank suggested removing the cables so we could move the 729 more easily. I couldn't get the cable fully seated when I put it back. Marc noticed that two pins were bent together and appeared to have welded themselves together. Marc thinks this short may have contributed to the power converter issues.

From the documentation, these appear to be the pins for the 115V AC outlet:

Do we have any spare 729 power cables?

As for the -48V problem, Brenda carefully disconnected a bunch of wires from the relay board (which was not easy). We found a bunch of places where the short isn't, but haven't found where the short is. Hopefully we will find it next week.

Ken

- from Robert Garner - DE tape drive #1 - Nov 16, 2022- 5:17 PM
Ken,

> Do we have any spare 729 power cables?

We should have one (or perhaps two) in our Yosemite 1401 depository.

— Robert

- from Brenda Make - DE tape drive #1 - Nov 16, 2022- 9:08 PM
Hi,

Your (Kens) pic is better than mine, but both pics show minimal welding, though the pin it's laying on looks pushed down through the connector. I think we may be able straighten the bent pin. The other is more worrisome. A resistive connection would be bad.

~

Indeed, the density of the wires and screws make working on the -48 power block and the relays quite difficult. Some of the press-fit pins took nearly 40lbs of force to pull. I have not yanked a wire out, but I did scratch one. It was heat-shrinked just in case. Between the physical issues--and the as-built schematic discrepancies, they want to drag a methodical approach into a desperate search.

Noting: When reattaching wires to a relays, a proper width screwdriver can interfere against the dense terminals causing a false indication that the screw is tight.
This is easier to feel when the screw is loose. It clicks. Though, when the screw is starting to tighten, it may not be. Fortunately, Ken caught it. (I am sorry for my mistake.)

~

Also noting, there are 2 diodes hidden under the timer control.

I could come another day to help.

Take Care,
Brenda

- from Tom Szolyga - DE tape drive #1 - Nov 16, 2022- 10:03 PM
I was researching documentation today and discovered that 729 tape drives could be ordered/configured to connect to two different computers at the same time. Inside the tape drive was a relay panel. There were two sets of connectors for cables, one set for each computer. The relays switched the tape drive electronics between the connectors.

When a 1401 was used as a front end for a 7044 or 7094, input or output was to a tape. The 729 was switched to the other computer and the tape information was taken off the tape. The operator did not have to unload a tape, move it to another drive and reload it. Job turnaround became faster.

Apparently one of our 729 drives has this feature.

Best regards,
Tom


Allen Palmer e-mailed:

A slight correction "….. that 729 tape drives could be ordered/configured to connect to two different computers at the same time …."

Not at the same time. A drive could be physically connected to two different system at the same time but electrically ( signal connection ) to only one system at any time.
You could transfer the connection from one system to different system via the bank a relays as you say [Now Above].

Allen Palmer


Bill Worthington e-mailed:

The bank where, in 1961, I learned to program 1400 Autocoder had two 1400 systems -- a 4k 1401 and a 60k 1410. The two systems shared a pool of tapes using a tape switch device. Individual tape drives were controlled by an "A/B" switch of non-IBM vintage. As I recall we had a total of 8×729s: 2 were attached to the 1401 and the 4 to the 1419. The remaining two could be attached using the A/B switch to the 1401 or the 1410 using the A/B switch–but not both Both systems had to be stopped before the tape could be switched to the other system. (It was not a dynamic switch.)

The 1410 had a 1419 MICR reader sorter and ran 7×16 (no third shift)) sorting for the Fed bank in Boston. The 1401 had a 1412 MICR reader sorter and did the more mundane "bank stuff you'd expect to find in a commercial bank.

Bill

- from Grant Saviers - DE tape drive #1 - Nov 17, 2022- 7:41 AM
One of the early projects was to clean all of the wire contact relays in that 729. We removed all the wires and contacts and VERY lightly air blasted the contacts with very fine aluminum oxide media. The wires were ultrasonically cleaned IIRC. This cleaned them without removing the gold plating.

I have a collection of photographs & micrographs I can post if they aren't already in Ed's collection of reports.

Grant Saviers


Ed Thelen replied:

> I have a collection of photographs & micrographs I can post if they aren't already in Ed's collection of reports.

There is text and pictures of contact cleaning at
http://ibm-1401.info/Sched2006May.html#GS-04

Is this what you are thinking about ?
If not, I could post more text and photos from you - to aid this challenging problem.

Dirty or intermittent contacts are a curse on the world ;--))

Good luck to us all
-Ed Thelen

- from Ignacio Menendez - DE tape drive #1 - Nov 17, 2022- 8:06 AM
Great progress gal and guys !

My 2 cents worth…

The bent pins on the power cable connection is very common; any time that you have locking down the cable, probably is because of a bent pin.

It is a good practice that anytime that you have these cables disconnected, one should check for bent pins; the same goes for the huge cable’s connectors, and terminator, check for bent pins there as well.

Where the ALD shows the 48 vdc going to prolay power components, some drives have a metal box, hanging on the left side, upper left frame member. This box can be disconnected, and removed to a bench, for trouble shooting and repair. Perhaps the short may be there.

If all goes well, I may visit CHM on Wednesday, December 28. I hope that you allow me to do some work on the 729 tapes.

Iggy

Nov 9, 2022 - Demo - dated 11/12/2022 2:20 AM
from Pat Buder

On Wednesday Bill Worthington and I gave the demo to 22 visitors from Ukraine, UK, and other locations. Another 8 came by before or after the demo. Art and Sam from the 1401 Operators class attended.
Frank [King] left during the demo, so we had the audience give him a round of applause on behalf of the entire restoration team.

We ran on CT, which performed perfectly. We did not need to power up the DE machine. The 083 Sorter is still down.

We unloaded the tape drives before shutdown and powered off all equipment before locking up.

Pat

November 9,
- from Ken Shirriff 729 tape drive update:

from Ken Shirriff 729 tape drive update:
Summary: DE tape drive #1 has a short that trips the -48V circuit breaker.

Details: The symptoms are that the tape drive was completely dead except for the unit number light. Last week, we determined that the -48V circuit breaker was tripping. This week, Brenda removed the circuit breaker and tested it on the bench. It is rated at 2 amps. Putting 3 amps through it causes the breaker to trip after some delay. So the circuit breaker seems to work okay.

With the circuit breaker removed, we measured the resistance between -48V and ground was .5 ohm, so there must be a short. Fortunately, there's a terminal strip TB130 so we can disconnect the -48V branches and track down the short.
Unfortunately, the terminal strip is inaccessible under a thick bundle of yellow wires.

We found other places to disconnect -48V branches and ruled out the panel lights and much of the relay logic as the location of the short. Inconveniently, the wiring diagram is semi-fictional and the wiring doesn't always match what it says.

By the way, -48SP is a second -48 volt supply present on some machines to power the logic panel separately. (I'm not sure what SP stands for: special? separate? service pack?) This tape drive includes the -48SP supply; you can tell because there is a separate circuit breaker #10.

It seems like the problem isn't in the relays; we disconnected -48V at relay R2 and the problem didn't seem to be downstream. So the clutch filter box, the tape switches, or the RC modules seem to be the remaining possibilities for the short. Next week we'll check these.

Ken

November 8, 2022 - from Grant Saviers, re the PM of 729's discussion
My 1401 operating experience was to thoroughly clean the full tape path for 729's every shift, using no lint cloths and solvent cleaner. So that would imply about monthly for CHM. An IBM CE had one dedicated shift per week (all other shifts were production) to do complete PM, most of which was spent on the 1402 and 729's.

It seems that worn out carbon brushes causing faults in the 729's is a good indicator that more frequent PM's are needed. Worn out belts also.

Grant Saviers

Nov 5, 2022 - 11:28 PM - White Board
from Robert Garner

Brenda,

There’s a status white board mounted on the side of the Liebert unit.
But we’ve haven’t been disciplined at keeping it up to date. :(

There are also record books for each of the 1401 systems and subsystems, but over time the team slowed in their use. :(
It would be good to dust them off. :)

I have encouraged the team to use and update them.
Any effort to keep them updated would be welcome and desirable encouragement/kick-in-the-pants. :)

Best Regards,

— Robert

p.s. Flashlights are always appreciated. :)
I believe there’s an LCR meter in the workshop. There’s definitely one or more capacitance meters

Nov 5, 2022 - Demo - Nov 5, 2022 2:07 PM
from Paul Laughton

Steve Madsen and I gave the demo. We had four(4) people at the start and twelve (12) by the end. The sorter was not working so we skipped it. We used CT for the demo and all (mostly) worked well.

Notes:

  1. When I came into the room at 10:00 AM the light on top of CT was on and thus the CT circuit breaker was on.

  2. The Green Bar paper on the CT had been replaced by 1/3 full box of old plain White paper. Since the paper was old, we had the paper separate during the demo Big Print. This gave us the opportunity to demonstrate replacing paper on the 1403 during the demo. Moral: If there is not new Green Bar loaded into the printer before the demo, change it before the demo.

  3. Arthur Morgan, one of the people training to be a 1401 Demo Operator, donated his father's IBM THINK sign for permanent display on the CT 1401. Photo attached.
Paul Laughton

Scheduled Maintenance, 728s 11/04/22
from Robert Garner - Nov 4, 2022 4:39 PM
Iggy

> From memory, and I was working in the 60’s, on 2401’s, very similar to 729’s, it was once a week.
>I think PM once a month would be recommended.

I relish and look forward to regular 729 PM (per your checklist below)!
I’m also wondering about the best PM interval, based on our lower up time/utilization compared to original usage.

From your 1960s memories/experience, were the 2401s generally running around the clock, 24/7?
(As were our 1401s during their first ten years of 24/7 service at insurance firms.)

Any one of our 729s only runs about 14 hours a week (or less), for a utilization of just 14/168 = 8.3%, that implies a scaled-PM interval of 1/0.083 = 12 weeks = 3 months.
Or perhaps lack of 24/7 use (and frequent power cycling) accelerates the need for 729 PM?
Perhaps some 729 PM items make sense once a week, and others less often?

- Robert
From Iggy - Nov 3, 2022, at 2:59 PM
From memory, and I was working in the 60’s, on 2401’s, very similar to 729’s, it was once a week.

In our case, the machines are so old, that materials and contact surfaces deteriorate much faster.

If we want to keep all these 729’s in good running condition, and able to run from CPU programs, I think PM once a month would be recommended.

Iggy

From Robert Garner - Nov 3, 2022, at 2:46 PM
Brenda,
Thanks for taking on 729 PM!
(It’s only really been done when Iggy was in or a 729 was acting up...)

Thanks for all your help, expertise, and contributions! :))

— Robert
p.s. Iggy has said he’ll try to get back in a couple of months or so...
From Brenda Make - Nov 3, 2022, at 12:08 PM
Hi All,

(Iggy, thank you finding the cryptic location for that test bus. : )

In general, I fancy the idea of doing preventative maintenance, whenever time constraints permit.
In the head assembly section, it's stated that the heads should be cleaned every 8 hours of service or 10 reel passes.
I would also like to clean the inside of the tape area, when time permits.
It would seem that shutting the drive doors, nightly would help.

Interestingly, there is a routine 48v voltage check.

This following is from one of the (223-6988_729_CE_Mar62.pdf) 729MKII Manuals:

Take Care,
Brenda

Nov 2, 2022 - Demo - Nov 7, 2022 12:19 AM
From Pat Buder

On Wednesday Bill Worthington and I gave the demo to 28 guests and mini-demos to 15 more before and after the main demo, including 8 from a senior group after their Restoration tour. Several visitors were from the Hamburg area of Germany. Like most German visitors, they enjoyed reading the labels inside the DE 1403 and hearing about the other working 1401 in Boblingen.

We started the demo on CT, which failed in the mode it has often shown recently. After power on at about 1:00 pm, it loaded BigPrint and read name cards fine for about two hours before the demo, but refused to load BigPrint during the demo. We made a hasty switch to DE (thanks Bill for moving our stuff), which ran programs fine.

We returned to CT to run the tapes via the TAU. All four drives were loaded with about 1/4 of the reels read. Newly repaired Drive #1 just stopped reading at one point. Pressing Start caused it to resume spinning. Drive #2, which often displays a multisecond pause before executing a rewind, refused to go into fast rewind at all. Drives #3 & #4 ran fine.

The 026's and 001 worked well. The 083 sorter was inoperable, per the Restoration team.

At about 4:30 after the demo and late visitors, we shut down all systems and power and left the room open with the video playing, the standard practice. Everything was locked and all power was off, verified by no "pilot" lights lit on the two CPUs and the unit record gear.

Two students from the demonstrator class were present and got to witness the real-time CPU switch and other recovery actions we sometimes need to take.

Pat

Nov 2, 2022
- from Stan Paddock, DE 1402 punch switch - Nov 1, 1:27 PM
- from Ken Shirriff, 729 tape update - Nov 2, 3:45 PM
- from Ignacio Menendez, 729 tape update - Nov 2, 4:03 PM
- from Ken Shirriff, 729 tape update - Nov 2, 4:25 PM
- from Ignacio Menendez, Tape Drive's Status - Nov 2, 4:28 PM
- from Ignacio Menendez, 729 Preventative Maintenance Checklist - Nov 2, 4:54 PM
- from Stan Paddock, Wednesday Maint Report - Sat Nov 5, 19:20 PM

from Stan Paddock, DE 1402 punch switch - dated Nov 1, 2022 1:27 PM
German Card Punch:
A UPS truck stopped by my house yesterday.
He had a small package for me from Greece.

This is a new micro-switch for the punch in the German system.
Brenda will be installing it in the German system tomorrow.

Stan

from Ken Shirriff, 729 tape update - Nov 2, 3:45 PM
Summary: DE tape drive #1 has a problem with the -48V circuit breaker tripping.

Details: Iggy said that there was a power supply problem. The symptoms were that the tape drive was completely dead except for the unit number light.
Brenda measured the voltages to the SMS cards: +6, -6, +12, -12 were all okay. At the circuit breakers, the voltages were okay except the +62V was very low, around 26V, and the -48V was dead. (The separate -48SP supply for logic was okay.) More investigation showed that the -48V breaker trips when you turn the drive on.

We measured at the power supply and the -48V was good (about -53V). We suspect there is a short somewhere that is tripping the breaker. Frank suggested that the circuit breaker itself might be bad. Next week, we can remove the circuit breaker and check it on the bench.

If there's a short, it may be tricky to track it down. The -48V is distributed by TB 130-COM, a 13-pin terminal block. We could remove connections from the terminal block to narrow down the location of a short, except we can't find that terminal block. We found many other terminal blocks, but not the -48V distribution. Any ideas where it is?

Ken

from Ignacio Menendez, 729 tape update - Nov 2, 4:03 PM
Ken, just before leaving I found a location shown on page 09.04.2,
Brenda was still there and I showed her the location shown, in the relay gate.
Iggy

from Ken Shirriff, 729 tape update - Nov 2, 4:25 PM
Great! For reference, here's the diagram from 00.04.2:

Here's a file photo of those components. The terminal blocks must be on the back:

from Ignacio Menendez, Tape Drive's Status - Nov 2, 4:28 PM
Connecticut tapes:

PM’d all 4 drives.
Drive 4 was dropping ready in go-reverse, pulling tape out of the right column… adjusted right shaft reverse clutch carbon contact assembly, repositioned it closer, and replaced a holding screw, which would not tighten correctly.
Ran from TAU fwd/reverse, all 4 drives OK.
R/W only drives 1 & 2 are OK;
3&4 need some looking into.
Drive 3 needs ALL 4 motor to clutch belts replaced, have old style ‘cog’, but we have in stock ‘V’ belts, which are OK… will run for a while, but need replacing.

German tapes

Tape 1, see Ken’s report, power problem

Tape 2 was missing idler shaft retaining wire…. Made one with a piece of solid, but small diameter wire, works OK.
Drives 2&3 OK for go-fwd/bkwd.
These 2 drives intermittently detect end of tape at the same time, and light up the indicator on the top panel; no end of tape sticker on either one…. Need to investigate further.
Had a few errors R/W, but may be the tapes, or a need for adjustments.


All 7 drives appear to be at a point that the carbon brushes need to be replaced…
Last time, some 5 years ago, Mark made some for us in his lab.
Dave Bennett found a shop nearby that made many of these, perhaps he can get some more made up.

Iggy

PS… It was an honor and a pleasure, as always, working with you guys.

from Ignacio Menendez, 729 Preventative Maintenance Checklist - Nov 2, 4:54 PM
Robert asked Iggy for a checklist for our tape drives PM...
(Perhaps go through these about once every 4 months?)

Robert, you asked me about what quick checks can be done:

With all power off to each or all drives:
->>>(It is very easy to get a bad electric shock, if power is present!)-<<<<

  1. make sure all vacuum ports at the bottom of both columns are free from debris, like pieces of tape.
  2. the capstans should be able to retract, when manually extended, with little or no effort… if not, try adding a drop of oil on the shaft itself, front and back.
    The idlers on the prolays, should not bind, and rotate freely… if not, the shaft needs to be removed, cleaned and a very small amount of oil applied… then reassemble, making sure the retaining small wire is inserted on the shaft and screwed down.
  3. the sense Micro switches should transfer when the flywheels on the rear fwd and back, and not in physical contact. If not, need adjustment.
  4. if equipped with an eddy current mercury switch on the back of right shaft, should move freely, by hand and return by itself when released. (Some drives have a timer to determine when the transfer from high to low rewind motor speed.
  5. clean with isopropyl alcohol the R/W head and the cleaner blade to the left of the head, with lint free cloth, if available, if not paper towel.
    Make sure that the alcohol does not contact the tape.
Then, with power on, tapes loaded and ready, run from TAU GO-FWD/BKW all drives and make sure that the tape is kept between the portholes in the columns, and not bottoming down or pulling out of the columns.

Test each individual drive for error free write and read, from TAU, or with a card program from the 1401. Drives need to be in optimal condition to do this.

Iggy

From Stan Paddock, Wednesday Maint Report - Sat Nov 5, 19:20 PM
Frank King 6.0
Stan Paddock 6.0
Tom Szolyga 6.0
Robert Garner 4.0
Brenda Make 6.0
Ignacio Menendez 6.0
Dale Jelsema 6.0

Stan suggested that the result rated a gold star The punch has been sick from before covid -

Tom Szolyga project on the German 1401 System

I assembled the biscuit connector into the Serial I/O connector housing. It plugged into the 1401 without error. Next I connected the wire for “1401 running” signal output coming from the connector to the wire for “I/O EXT” signal input going into the connector. If the connector and pins worked properly, pressing the 1401 RUN and STOP buttons on the console should turn the I/O EXT light on and off. IT DID! It works.

Next I wired up the 1401 line receiver to TTL signal circuit. I placed a LED on the TTL output. When connected to the “1401 running” signal, the LED should behave exactly like the light on the console. IT DID! It works. Receiving 1401 output signals into the Arduino is now possible.

Frank King, Brenda and Dale Jelsema

The IBM 083 sorter is down with a power supply problem and is not plugged in.

Brenda took pics of the caps in the 083 power supply that was in the museum.

This should help fix the problem with the IBM 083 sorter in the Demo room.

Other items here, here

October 29, 2022 - Demo - from Paul Laughton
I was the Lead. Jack [Ghiselli] was my charismatic partner.

There were 26 people in the main demo, plus 8 who wandered in later. International visitors from Germany, The Netherlands, Turkey, and UK.

  • The 083 Sorter is completely dead. Won't even power on. Jack, however, made a fine show of doing a pretend sort. The audience appreciated the effort.

  • The CT 1402 gave a few random read failures, but we were able to recover, and then they sort of went away.

  • CT 729 Tape Drives 1-3 work. Tape #4 seems to work with the TAU, but fails with the Tape Exerciser program.

  • DE 729 has only one working 729 Tape Drive

  • The DE 1402 Punch appears to be fixed.
Our demo vacuum tube rolled off of the demo table and broke.

There was a previous report that one of the three bottles of magnetic cores was missing. It looks like the two remaining bottles mated and thus created a third bottle. There are now three bottles in the cart.

Demo mic battery charger: If the lights do not come one when used batteries are put into the charger, unplug the charger from the extension cord and then plug it back in. The lights will come on.

October 26, 2022
- from Ken Shirriff, DE 1402 punch
- from Ignacio Menendez, Status of CHM TAPES
- from Stan Paddock, People Present, DE Power, 083


from Ken Shirriff - DE 1402 punch - dated Oct 28, 11:58 AM
Stan suggested that the result rated a gold star The punch has been sick from before covid -
Brenda, Frank, and I worked on the DE punch and got it to operate successfully. We took out the punch brush card lever microswitch, cleaned the contacts, put the switch back, adjusted the switch so it closed solidly, and we were able to duplicate the powers-of-two deck and the bigprint deck.

Details:

The symptom was that the punch would stop with a punch stop error. We determined that the problem was due to the punch brush card lever. I put an oscilloscope across the microswitch and it's clear that the contacts were bouncing wildly. The yellow line is the -20V signal from a cam during one card cycle. The blue line is the other side of the switch. Since the switch is closed, both lines should match. (I verified that this is the case on the CT punch.) Note that the blue line jumps around, indicating that the switch is rapidly opening and closing.

We took the microswitch apart and Brenda burnished the contacts. The secret to opening the switches is the pin in the middle and the matching pin on the other side. If you push the two pins in, the switch case separates.

Inside the switch is a buckling spring mechanism, not what we expected.

We put the switch back in but still had problems until Brenda adjusted the switch to close more tightly. After that, the punch worked reliably and we could duplicate multiple card decks. The new decks successfully loaded and ran.

We still saw a few punch stops, but they seemed to be genuine problems due to feeding cards. These problems went away, but there could be lurking problems with another switch. We also got a few punch check errors due to non-blank cards in with the blanks. But overall the punch seems to be working successfully. Stan has ordered a replacement microswitch from Greece in case this one fails again.

Ken

from Ignacio Menendez - Status of CHM TAPES
Connecticut system:
Tape 1
  1. was missing taper pin on right hand reel hub, causing hub latch to allow tape spinning on hub, and tape dumping.
  2. tape pulling out of left column during forward motion: found sparking on left forward clutch, commutator to carbon contacts; re-aligned brush block closer to clutch rings, by about .020”
All 4 tapes working OK for demo, mechanically only.
Did not have enough time to do data W/R tests.

German system:
Tape 1:

Has power problems A/C comes up, but DC +12 v missing and tape does not complete power up.
Attempted to locate +12 v wiring from SMS logic down to power supply cage below. I could not find wiring diagrams on this ALD book, that has this tape s/n
Will take a bit of time to trouble shoot.
CPs set and reset Ok, none was tripped.

Tapes 2 and 3 going forward stop at the same exact time with ‘tape indicate’ always at same time and place going forward; there is no end of tape sticker on either one.

Looking at tape 2, noticed right prolay’s idler shaft has moved 1/2 way out, missing retaining clip, and screw is laying there by prolay and head cover.
Will replace or make a clip.

Only tape 3, is OK mechanically; tape 2 will be. As soon as the idler shaft retaining clip is replaced.

It nice seeing you all, enjoyed working there again at CHM.

Iggy

from Stan Paddock, People Present, DE Power, 083 - dated October 28 8:50 PM
People who attended the CHM were:
Frank King 6.0
Stan Paddock 6.0
Tom Szolyga 6.0
Ken Shirriff 6.0
Marc Verdiell 6.0
Robert Garner 4.0
Brenda Make 6.0
Ignacio Menendez 6.0

German Power Supply:

Frank, Robert and Stan took further measurements on the three phase power lines between the Power panel and Pacific power system.
The labels on the Pacific will be updated when the data is provided.
Robert, Iggy and a third person replaced the main cover over the power panel.

IBM 083 Card sorter:

While not helping visitors using the IBM 001, Frank is trying to fix the IBM 083 card sorter.
It appears that the ferroresonant part of the high voltage power supply is not resonating.
Some time in the past, a part of this supply was re-wired and now it does not work.
We have access to another IBM 083 to inspect as how the power supply is supposed to be wired.

Stan

October 19, 2022
- from Ken Shirriff - DE 1402 punch - received Oct 19, 4:05 PM
- from Marc Verdiell - IBM 029 keypunch - received Oct 19, 5:07 PM

from Ken Shirriff - DE 1402 punch - received Oct 19, 4:05 PM
Frank, Brenda, and I worked on the DE 1402 punch today; Brenda is a new member of the 1401 team.
Summary: the "punch brush card lever" microswitch seems flaky and probably needs to be cleaned or replaced.

Background:

The 1402 punch has multiple card levers that detect a card's path through the punch. If a switch isn't activated at the right time, the punch assumes there is a jam and triggers a punch stop. We are encountering problems with the punch brush card lever that causes Punch Stop errors when punching.

Brenda adjusted the microswitch's position so it would open and close properly, but that didn't help. We looked at the signal across the switch with an oscilloscope:
This shows the input to the switch (yellow) and the output from the switch (blue) for part of one card cycle. The yellow line shows the -20V signal when the cam closes. The switch should be closed at this point, so the blue line should track the yellow line. However, the blue line drops down to the yellow line repeatedly, along with 100 V spikes, presumably inductive kickback from the relay coil that the switch drives. The switch should be solidly closed at this point, but it looks like it's continuously bouncing, probably from vibration.

My guess is that the contact is corroded/pitted/bad so vibrations are causing the contact to bounce open and closed. Next week we plan to look at the same switch on the CT machine, to verify if the DE switch is bad, or if this is the typical behavior. Maybe the contacts can be burnished on this microswitch. Looking at the photo right now, I'm not sure if the wire on the left is screwed down all the way; that's another thing to check.

Ken

from Marc Verdiell - received Oct 19, 5:07 PM - IBM 029 keypunch
Brenda,
It was great to meet you! Welcome to the fun!

And my update on the 029 punch. It had a bit of a regression after I got it working last time, with the I key failing again but in a different way. It’s been corrected, and now things seem to be holding up. I punched many “all characters” cards and they were all perfect. The 029 is all buttoned up, but Stan suggested we re-test it again next week before we move it to the Babbage room or wherever we want it, to make sure things are stable. Seems like a good idea to me.
Marc

October 12, 2022 - demo - from Pat Buder
On Wednesday Tim Robinson and I gave the demo to 28 visitors. Another ten people came by before or after the demo. In addition, we had several members of the upcoming 1401 Demo Operator class observing.

After discussions with Marc and others we decided to run on DE because the CT 1402 was having trouble reading cards.

While readying the DE tape drives we heard a loud metal snap from drive #3. A lever near the rightmost vacuum column had dislodged. We were able to point this out to Frank.

The 083 was not putting any cards in one column (still). This week it was the ones column. This isn't a big deal for our quick sorter demo. I think Dale is still working on the 083.

Other than these minor issues, the demo ran smoothly.

We inadvertently left CT tape #3 loaded when powering off. We had prepped the CT tapes before the demo when we thought we were going to run on CT.

Unfortunately we lost one of the vials of cores and the USB cable. We had passed two core vials but only one came back. This day, as always, I mentioned that we are only passing the objects and that they are not souvenirs.

Pat

October 12, 2022
- from Robert Garner, -
DE 1401 Power Draw
- from Ken Shirriff, - Update DE 1402 Punch
- from Stan Paddock, - Status Report
from Robert Garner, - DE 1401 Power Draw
Stan,

Here are the measurements I took today of the power INTO the DE 1401 (i.e., output of the Pacific Power unit).

Wire Idle +729s(3) +1403 Power
Red 5.4A 9.0A 9.3A 2.1 kVA
Blue 7.8A 7.9A 8.7A 1.9 kVA
Black 5.7A 5.9A 6.4A 1.4 kVA
Total: 24.4A @220V = 5.4 kVA.

Idle = everything powered up (but no tapes spinning or printing).
+729s = TAU maintenance panel reading from all three tape drives.
+1403 = printing powers-of-2 with tape drives spinning.

Next Weds, I’ll (re-)measure the current INTO the Pacific Power unit from the circuit breaker panel.

After that, let’s formulate a single label to attach to the Pacific Power unit.

— Robert

p.s. These were Ron Crane’s April 6, 2004 measurements, 5.75 kVA, with no 729s attached. I’ve put in the implied currents (dividing by 220V). http://ibm-1401.info/Sched2005April.html
2.8 kVA on PhA (46%) => 12.7 A
2.4 kVA on PhB (40%) => 10.9 A
.55 kVA on PhC (9%) => 2.5 A

I suspect PhA and PhB were higher then as this was before Ron added the caps to fix the poor power factor (too inductive, current lagging voltage).

from Ken Shirriff, - Update DE 1402 Punch
October 12, 2022
Frank and I worked on the DE 1402 punch today. We adjusted a microswitch and got it to punch about 50 cards successfully, but then it flaked out. So the good news is that the punch finally kind of works; the bad news is that it doesn't work reliably.

Detailed notes:

The 1402 has multiple card levers that detect a card's path through the punch. If a switch isn't activated at the right time, the punch assumes there is a jam and triggers a punch stop. Last week, Stan and I fixed the stacker card lever that was causing Punch Stop errors after punching a single card, but the problem remained.

Today, the Punch Stop occurred with a card at the punch brush card lever, after the punch and before the second brushes. Frank is pointing to the lever that detects the card. You need to remove the brushes to see it.

This lever is connected to a microswitch underneath at the back that closes when the card passes under the lever. Frank determined that the switch wasn't closing. The hook at the left behind the wire lifts up the microswitch's lever to close it.

Marc adjusted the microswitch by loosening the screws and rotating the switch slightly. The movement due to a card is very small, and it is difficult to get the switch to open and close correctly. Frank manually cranked dozens of cards through the punch so we could adjust the switch. After multiple tries by Marc and me, the punch successfully punched about 50 cards and then got flaky. I suspect the switch shifted slightly.

It seems like the switch is excessively temperamental, with approximately no position between staying open and staying closed. I don't know if this is the normal behavior, or if the switch should be replaced, or if something needs oiling to move better. We can try to adjust the switch to the exact magic spot and then tighten it firmly, but hopefully there is a better solution. (The manual says nothing about adjusting this switch, by the way.) Maybe I'll look at the CT machine next week to compare with the behavior of its switch.

Ken

- from Stan Paddock, - Status Report
People who attended the CHM were:
Frank King 6.0
Stan Paddock 6.0
Tom Szolyga 6.0
Ken Shirriff 5.0
Marc Verdiell 6.0
Dale Jelsema 4.0
Robert Garner 4.0

Ken Shirriff and Frank King continued work on the German 1402 card punch.
Ken found documentation on the card sense levers.
There are FIVE of these little buggers and any one or more of these can cause a PUNCH STOP.

Frank and Ken got the punch to WORK! (Sort of)
They tried to punch a copy of the BigPrint deck.
It punched 200~ before the punch got a PUNCH STOP.
Testing showed that card test #5 was random.
Next week we will try and find a new micro switch.
If that does not work, we will replace the micro switch with one from E-Bay.
If that does not work, we will attempt to overhaul the existing micro switch.
Why the last option sounds cool, the options of success is not very good.

------------------------------------------------------------------

Marc fixed all of the problems with the IBM 029.

This IBM 029 was set out for full access to the public and a member of that public beat up on the keyboard.

To prevent any more damage to the IBM 029, it was decided to store it in the Baggage closet.

It will be moved there next week.

------------------------------------------------------------------
BLUE BOX II.
This is a part of the Tom is continuing his working on a cable from the IBM 1401 serial port and the IBM PC to 1401 data link like the same functionally that exist on the Connecticut system.
-------------------------------------------------------------------
Dale Jelsema is learning about how an IBM 083 works.
When you run the sorter, while it looks good, it puts cards in the wrong pockets.

------------------------------------------------------------------

Stan Paddock


October 8, 2022 - from Stan Paddock
In late September, 2022 the docents were having a problem turning on the primary to the German IBM 1401.
The problem was identified to be a worn 208 three phase 90 ampere circuit breaker.

The problem was identified to Luis Valencia, CHM facility contact.

Luis contacted an electrician to replace the suspected breaker.

After the replacement, the Pacific Systems power converter started acting up.

Luis contacted the electrician to investigate the problem.

The electrician determined that the socket where the new breaker was inserted was faulty.

The electrician was called back for the third and moved the circuit breaker and wires to an vacant socket in the power panel.

After the transplant of the circuit breaker, everything was powered up including the German IBM 1401 system.

All problems identified were corrected.

Thanks to everyone involved in solving these problems.

Stan Paddock

October 5(d), 2022 - from Stan Paddock
People who attended the CHM were:
Stan Paddock 6.0
Tom Szolyga 6.0
Ken Shirriff  
Marc Verdiell 6.0
Dale Jelsema 4.0
Robert Garner 4.0

Ken Shirriff and Stan continued work on the German 1402 card punch.
Ken found documentation on the card sense levers. There are FIVE of these little buggers and any one or more of these can cause a PUNCH STOP .
We have checked card levels 1-3 and decided to check number 5.
It was stuck open!!!
We removed the stick and cleaned the contacts.
While Ken and I were in the 1401 room, the power to the German machine was disconnected due a circuit breaker problem so we could not check the problem.

------------------------------------------------------------------

Marc got the keyboard for the public keypunch reassembled.

After fixing a couple keyboard problems, Marc punched all valid punch values,

At the end of the day, inspection of the card, what should be a Z printed as a M.

On the good side, this is not a keyboard problem but a printer problem.

-------------------------------------------------------------------

Tom is now working on a cable from the IBM 1401 serial port and the BLUE BOX II.
This is a part of the IBM PC to 1401 data link like the same functionally that exist on the Connecticut system.

-------------------------------------------------------------------

Tom Szolyga is learning about how an IBM 083 works.
When you run the sorter, while it looks good, it puts cards in the wrong pockets.

-------------------------------------------------------------------

Thursday, October 6 Repair of the circuit bracer for the German system.
After testing various parts of the power links of the power going to the German system, it was identified the circuit socket in the Power Panel was defected.

Luis Valencia called in the electrician to move the circuit breaker to a vacant location in the power Panel.
Power was applied to the German system and the system worked as before.

Ken, Marc and I were hoping that the punch would come back to life, ( sigh )

A review of the problem was a lack of documentation for the German three phase power.
I made two labels that identifies the input and output for the Power Converter and attached them on the back of the Power Converter.
 
(Click to enlarge.)

Stan

October 5(c), 2022 - demo, from Paul Laughton
1401 Demo Report, Wednesday, 10/5/22

Pat and I gave the demo to about 20 people. Most of the quests were local with a few from Germany. DE was still circuit breaker limbo thus we depended upon CE to do the job.

There have been reports of CT not loading Bit Print during the demo. I decided to play it safe. Big Print was loaded before the demo. We periodically ran Big Prints before the demo to exercise the card reader. We did not reload Big Print during the demo but demonstrated it as already loaded.

There were three(?) candidates for 1401 Demo Operators to whom Pat talked while I set up the demo. My wife, Kathleen, another 1401 Demo Operator candidate was also present. She stepped in to help with the guest key punching for Big Prints after the demo.

Problems:

CT: No problems with CT but we did not challenge it to load programs during the demo.

Mic #1: Its headset failed shortly before the start of the demo. I managed to tear open the spare headset and get it installed just before the demo started. I have ordered a new spare headset.

Demo Battery Charger:I noticed that one of the mics had one non rechargeable batteries installed. I assumed that there was a problem with the charger. There was. The charger was plugged into the wrong socket.

When you put used batteries into the charger, the lights should come on (see picture). If they don’t, ensure that the charger (White cable) is plugged into the right socket (see picture). The left socket does not, for some reason, supply power. The charger plug sometimes mysteriously gets moved from the right socket to the left one.

Paul

Wednesday October 5 (a), 2022 - closing in on the DE wall circuit breaker
from Robert Garner
Luis, Jennifer, et al,

Reviewing the circuit breaker situation with Dale and Stan today, it’s clear that we do not need to replace the breaker panel, nor increase the breaker capacity from 90 to 100 A, nor change the breaker type, nor increase the wire gauge to the Pacific Power unit.

The intermittent DE 1401/Pacific Power breaker connection was due to a loose bolt that (is meant to securely) attach the breaker’s metal tab to the buss-bar, photos below. Jogging the whole breaker set is what caused the sparking last week at that bolt (photo below). Perhaps 17 years of on-off cycling caused the bolt to loosen (the Pacific Power unit was procured in 2005). The bolt socket/hole seems to be permanently damaged (we don’t know why/how).

Our proposed straightforward solution (for the electrician) is to just move the current (recently installed) set of three DE breakers upward three positions in order to bypass the damaged bolt hole (while also moving upward the CT breaker set). It looks like there may be sufficient spare wire length, so this could be a painless alteration. The damaged and (otherwise remaining) unused breaker positions can be covered by filler tabs.

The rationale:

  1. Per Ron Crane’s 2016 measurements, the DE 1401 Pacific Power’s input currents are 24A for phase-1, 25 A for phase22, and 26 A for phase-3 (total of 75A). This current load is well below the 90A-per-phase breaker ratings (total of 270A). We have no plans to ever increase the loading.
  2. Ron Crane also measured a Pacific Power current spike of 600 A. According to its spec (below), our breaker — Siemens Type BL, HACR — can properly handle current spikes under 700A.*
  3. The electrician apparently deduced that the Pacific Power current draw was 100A (per phase) from the (exceptionally faded penciled) sticker on the back of the unit/rack:
According to the Pacific Power 390-G spec…
http://ibm-1401.info/pictures/PacificPower-390-G-Operation-Manual_PN_114250-F.pdf
… the 100A/phase is the max amperage rating, assuming a full 28-kVA input supply (for a max 18-kVA output load):

Our (fixed) input supply to the Pacific Power is about half that, 15.6 kVA (= 208V * 75A).
(The power load to the DE 1401 system is likely about 10 kVA, or about 15A per phase (= 10,000 kVA / 220V.)

The Pacific Power 390-G unit was previously used at the IBM Cottle road site in their EMC testing lab, along with a second unit, so perhaps it may have been maximally loaded there. We should consider altering the sticker to our current usage, say a conservative 30A per phase (not 100!).

Stan is talking with Jennifer to communicate our information and proposed plan to the electrician.

— Robert

p.s. Photo of the breaker panel. The 3-phase DE breakers are lower left and 3-phase CT ones above them. Note the many empty positions, particularly above the DE & CT breakers. Also the wire gauge to the main breaker is only slightly bigger than those to the DE Pacific Power unit:

Pointing at the loose buss-bar bolt of the middle phase of the 3-phase Pacific Power breaker :

* Looking up the trip curve for our Siemens Type BL, HACR Type, beakers…
https://assets.new.siemens.com/siemens/assets/api/uuid:d6e2e881-6e9c-4488-af64-f5a76394f644/sie-tc-qp-bq-bl-45-125a-2-3p.pdf … our measured peak inrush current of 600A is just below the breaker's trip point of 700 - 1000A:

Also, the ratio of our peak 600A spike to the breaker’s rated value of 90A is 6.6x. From the breaker’s trip curve chart, that 6.6x multiple is within the breaker's “no trip” zone on the left side of its chart:
(Note that this 6.6x value should be used on the x-axis of the trip curves, not the 24x value I referred to in my email).

Wednesday October 5, 2022 - DE 1402 punch stoppage
from Ken Shirrriff
Stan mentioned that the DE 1402 punch was producing punch stop errors. I took a look and we found a switch in the card path that was stuck open. We fixed it but weren't able to power up the computer to test it out. I'm pretty confident this will fix the punch stop error, but there may be additional problems after this one.

Detailed notes:
A punch stop indicator happens if there is something wrong with the card motion, detected by the various card lever microswitches. This is different from a punch check, which happens if the data read from the card doesn't match the punched data. Punch stop is computed using relay logic in the 1402. (Unlike punch check, which is generated by the 1401.)

The following diagram from the 1402 CE service index shows the card path through the punch, and the various card levers. A punch stop is triggered if a card gets stuck somewhere, doesn't end up where it should be, and fails to trigger a card lever. Note the stacker card lever, after the second punch brush and before the stacker.

The following are the specific checks that can trigger a punch stop:

  • There are two checks if the punch clutch is impulsed at the wrong time, causing a punch stop.
  • If the hopper card lever is triggered and the throat card lever is not, then it's a punch stop. I.e. the card didn't feed from the hopper correctly.
  • If the punch brush card lever is triggered but not the die card lever (or vice versa), then it's a punch stop. I.e a jam in the punch mechanism.
  • If the throat card lever is triggered but not the die card lever, then punch stop. I.e. a jam before the punch mechanism.
  • Finally, if the punch brush card lever is triggered but not the stacker card lever (unless the stacker was inhibited), it's a punch stop. ** Remember this case.

This logic diagram from the 1402 CE Service Index represents the cases above. Note that the gates don't match modern symbols; triangles are AND gates and semicircles are OR gates.

This logic is implemented with relays. The card levers trigger various relays, and the relays are connected in series and parallel to implement the AND/OR logic. The relays are on page 11 of the schematics, shown below. For instance, the DIE CL DELAY relay and PCH BR CL relay are in series, implementing the fourth AND gate above. (PUNCH CYCLE PL-7) is the cam at the left below, in series with the relays.)

We looked at the card path in the punch and Stan found a switch that wasn't closing. The photo below shows the offending switch, the stacker card lever (STK CL). The switch's left contact was bent out about 30 degrees and caught on the metal frame, so the switch couldn't close. I touched it and it popped back into position. Stan burnished the contacts and checked the resistance. The failure of this switch explains the behavior that Stan saw: the card would get punched and then the punch check would occur. This switch ensures that the card enters the stacker after getting punched, so if it is stuck open, the punch doesn't see the card reach the stack and thinks there was probably a jam.

Ken

Tuesday, October 4, 2022
DE Power :--((, e-mails from Luis Valencia, Robert Garner, Dale Jelsema
From Dale Jelsema
CHM mgrs & employees, and 1401 team members...
      PLEASE, slow down, understand and read what the current PROBLEM is with the DE 1401 system .......

      Within the main wall mounted circuit panel for the various 1401 and other equipment, is a fairly new (4-7 wk old) 90 amp, 3 phase circuit breaker that consists of 3 individual breakers that are ganged together. Now, unlike a normal 15 or 20 amp breaker that "clips in" to the buss bars, this breaker has 3 input lugs/legs that go from the breaker block and are "BOLTED" to each of the 3 phase busbars.
      The PROBLEM is, one of the 3 bolts holding this breaker to the 3 vertical buss bars (for 3 phase power to the 1401s) was either not tight, worked its way loose (due to this breaker being TURNED OFF/ON so much ??), was cross threaded, whatever.
      Anyway, during last Wed's testing, I asked that one of the #4 gauge output wires be pushed toward the breaker block. This resulted in the breaker block moving just a little, with the resulting sparks, arcing, and other excitement. This is because the breaker block had one leg/lug loose connection to the busbar. The bad / burnt bolt was removed, and is laying at the bottom of the main panel. The electrician tried another bolt, but a new bolt would not tighten down (i.e. the threads on/in the busbar may be stripped), therefore you have another loose input connection. Now maybe this past weekend that was fixed, I don't know.
      In my original note, I commented that maybe the main panel can be powered down, and that one mounting hole can be rethreaded, and a larger size hold down bolt be used.

      HOWEVER, a better idea, I believe there are other unused circuit spaces in that main panel that are NOT being used. If there are 3 inline vertically together spaces, maybe the current DE 1401 breaker can be moved to a new location - - AS LONG AS ALL 3 PHASES REMAIN THE SAME ON THAT Moved BREAKER BLOCK. This would remove having to retap (?) that one hole in the busbar,
ALSO, going to #2 gauge wire or a new 100 amp breaker WILL NOT FIX this current issue / problem of getting the DE system up and running.- - i.e., you would still have a stripped mounting bolt location. As has been shown, there is not a continuous 80+ amp load going to the 60 cycle to 50 cycle converter.

I have two comments, ....

  1. Does anyone know "why" the DE 1401 breaker was turned on, when there are TWO strips of tape over that breaker that says "DO NOT TURN ON" ?
  2. This is a 3 phase 90 amp "CIRCUIT BREAKER', being used as a "ON/OFF" switch. I understand that usually there is no load going through this breaker when it is turned on or off, however, there is still a lot of internal wear within the breaker block over the years.
My background - 1963 to 1966, US Navy Radar and electric motor tech, using 3 phase 100 and 200 amp power circuits,

Sincerely,
      Dale Jelsema
            Gilroy, Calif

From Robert Garner, responding to Luis Valencia
Luis,
> I am waiting on proposal to replace the breaker and wiring with 100AMP.
Thanks for the update on the DE 1401 breaker status!
> according to any Electrician, machine pulls nearly 100amps + when it first turns on
Thanks for sharing the electrician's concerned about inrush current (in our case generated when the Pacific Power converter unit turns on).

In our 2016 posting and scope photo (below), Ron Crane measured a peak inrush current of 600 A for what looks like less than a millisecond, with the overall impulse lasting about 8 milliseconds (i.e., half of 60-Hz cycle). That 600A peak value is 24x the steady-state measured value of 25A (per phase).

I’m not sure how upgrading from 90A to 100A addresses the inrush problem.
We may need another type of breaker (designed for power supplies like our Pacific Power unit, see footnote below).

Could you let us know when the electrician will be back in?
So that one of us could meet with him, possibly turn on the Pacific Power unit, etc.

Thanks,

— Robert

p.s. The manual for our Pacific Power 390-G unit is here:
http://ibm-1401.info/pictures/PacificPower-390-G-Operation-Manual_PN_114250-F.pdf
However, it doesn’t say anything about its inrush current (only talks about its ability to handle 10x inrush motor loads).
I’ll contact Pacific Power to ask about inrush current and any circuit breaker recommendations based on our configuration.

This circuit breaker suppliers' web site…
https://www.wolfautomation.com/blog/a-quick-guide-to-circuit-breakers/
… suggests that our system could use a “D-type” breaker.
This "trip curve” chart shows that a D-type breaker doesn't trip as long as the inrush spike (of 24A in our case) is under about 3 milliseconds (which is likely the case based on Ron’s scope trace below):

click to enlarge

Info from Ed Thelen
I fully concur with Dale Jelsema and Robert Garner (above)

About the inrush current - this diagram is a part of page 60 of the Pacific Power 390-G unit (mentioned above).
     Note the multiple large capacitors ( C1, C2, ... ) at the outputs of the input rectifiers (multiple 543004s and 543005s ). These cause the very large but brief inrush current when the unit is first turned ON.
     The resistors (R1, R2, ...) are called bleeder resistors to slowly drain the charge from the capacitors when the unit is OFF - to prevent shocks to unwary finger pokers.


click to enlarge

I think that it absolutely vital for the specifying electrician understand that this is the cause of the very brief, very high current at unit turn on. This very brief current will not appreciably heat the currently installed #4 wiring. A licensed electrician should understand this, IF the information is correctly presented!!
     Most, if not all of the restorers, (if present when the electrician is present) can provide this information correctly.

Wednesday, September 28, 2022 - from Pat Buder, 1401 Demonstrations,
- troubles along the way -
Last Wednesday Bill Worthington and I gave the demo to 42 enthusiastic guests. Another 14 came by before or after the main demo. We had visitors from Germany, Romania, Switzerland, and France. The demo ran very long due to the serious problems described below.

Lately I have had several visitors ask about the "pilot" lights on top of the 1401's and the 083. I explained that their purpose is to show our dedicated restoration team which boxes have live power.

Before the demo Robert Garner was escorting a group of visitors who had hardware backgrounds. They punched name cards but declined getting them printed on the 1403.

Before the demo CHM facilities staff and electricians were present working on the DE power circuit breaker panel. We were running on CT. I asked them if that was a problem given that the CT power circuit breakers are adjacent to the DE breakers. They said no. At five minutes before demo start time they wanted to involve me in their discussion. I declined due to the time and expertise needed. Luckily restoration tam member Dale Jelesma was present and was able to help them. See his report for details.

As always, we ran BigPrint before the demo started to verify satisfactory operation. It ran fine. However, during the demo the CT 1402 would not load the BigPrint deck. Bill and tried these actions:

  1. Checked for obvious card problems (e.g. damaged cards, etc.).
  2. The deck had a "trojan" card (No Op) before the actual first card. Removing it had no effect.
  3. Ran about 30 seconds of NPRO several times.
None of these actions had any effect.

I then asked Bill to attempt to power up the DE machine so that we could move the group there. While he did that I carried on with the components part of the demo (SMS cards, core memory, big cable). Bill then reported that the DE was unusable. The electricians had left tape across the circuit breakers to indicate that it should not be powered up.

Some time had passed since the last card read attempt, so I decided to try to read another deck (Powers of 2). Fortunately it loaded fine. After explaining the 1403 operation we loaded the same BigPrint deck that failed previously, and it ran fine. Hallelujah! The tapes and the rest of the demo went without a hitch.

I know the restoration team is aware of this issue, and that intermittent problems like this are very difficult to diagnose. This problem as been happening for a long time. Here is my best description of the way it has been failing recently. Note that it doesn't fail every time.

  1. Turn on 1401. Run 30 seconds of NPRO on the 1402 reader. Within 5 minutes run a program, usually BigPrint. It runs fine. Run Powers of 2, which also runs fine.
  2. Machine sits idle for 30 - 60+ minutes until demo starts.
  3. Attempt to load a program. The motor starts but no card is read. At this point, NPRO doesn't help.
The docent team has a lot of experience in keeping the demo going in spite of the problems. Some of the problems are operator errors, which are the easiest to correct. Having a working second 1401 is sometimes a key recovery tactic. The entire 1401 team does what we can to make sure the "show must go on." This day was difficult but we soldiered on. The audience was happy.

As always, kudos to the restoration team. Without them there would be no live demos.

Thursday, September29, 2022   +! - from Luis Valencia, CHM Facilities Manager
Hello All

FYI this issue will have to be taken care on one day Museum is close so we can actually disconnect power to the panel completely, Electrician will be here to take care of the panel issue.

Electrician mentioned that if we resolve the panel problem, there is actually another issue with the existing wiring and breaker connected from the panel to the machine, Machine pulls 100AMPs, Breaker is rated for 90AMPs and wiring is size 4AWG which is not rated for 100amps, that being said if we do not replace the breaker and wiring with the right size/power rating , the existing breaker will keep on tripping.

Exiting breaker needs to be replace with a 100amps and 2AWG wiring.

I will ask Electrician to give me a proposal to replace wiring, breaker and panel repair.

Let me know if there are any questions .

Luis Valencia
Facilities Manager

Thursday, September29, 2022 - from Dale Jelsema via Robert Garners
A continuing tale - ...
Update on the subject of the 1401 German main wall panel circuit breakers.....,

I was onsite at CHM today (Wed, Sept 28) working on the 083 sorter. The German 1401 was powered down. About 3 pm CHM Facilities had an electrician re-look at the new DE main panel circuit breakers (3 breakers ganged together) that was installed a few weeks ago..

The new German 1401 circuit breaker is bolted to the 3 main 100 amp buss bars within the main wall panel. When the breaker was turned on, all seemed okay, until one of the 3 output lines was pushed toward the new breaker (which caused the whole breaker block to move a little). Then there were sparks and arcing where the center breaker input lug is bolted to the buss bar. Ends up, that bolt looked like it was cross threaded when installed, as the bolt was almost burnt thru. ( Frank King was also connected by speaker phone during this time -- Thank You Frank for talking to the electrician).

Discussion with the electrician and CHM staff that maybe parts of, or the complete inside of the wall mounted panel would need to be replaced. (which would most likely cost $$$). Instead, I suggested that maybe the circuit breaker be removed, power to the whole panel be shutdown, and the damaged breaker mounting hole be retapped to the next size bolt thread.

Tomorrow, the electrician and CHM facilities will power down the main wall panel, remove the DE breaker, and determine what arc cleaning (or damage repair) is needed to the panel busbar.

- - - Bottom Line - - -

Early on, I did power up the Pacific Power 390-G converter twice, to confirm that voltage was getting to the converter, but it was off during the time when the main panel DE breaker threw sparks and arcing,

The DE 1401 system was never powered up. My feeling is the current main panel DE 1401 breaker itself is okay, It is just the one mounting lug that ties the center (?) breaker itself to one of the 3 main panel busbars.

If anyone has additional questions, contact me. I will be onsite CHM next week Wed.

Sincerely,
     Dale Jelsema
         Gilroy, Calif

Wednesday, September 21, 2022 from Bill Worthington
Pat Buder and I led the 1401 demonstration last Wednesday for about 15 visitors. There were about another 8 who joined us later. With its being a smaller group, we found there were more questions from the visitors.

CT performed admirably. Big thanks to the restoration team.

Bill

Friday, September 16, 2022 - from Stan Paddock
? occasional missing phase from DE circuit breaker ? !!
Stan Paddock 5.0
Bill Newman 5.0

Background from recent status reports.

  1. Problem switching the circuit breakers for the German system.
  2. CHM called in an electrical contractor to replace the three wire circuit breaker.
  3. The following is a report from Frank King on the next work day.;
    "Hi, Stan,

    The German 1401 power converter acted up for the first time I have ever seen.
    When I tried to power on the German 1401, the power converter went down immediately; it has an internal circuit breaker that will reset on its own.
    After opening the 1402, looking at the 30 amp circuit breakers, I tried it again (It had reset by this time).
    Then, Mason and I took the tape circuits off line; that didn’t help.
    Mason then suggested we trip the 1402 circuit breakers one at a time.
    When we tripped the inner most 30 amp breaker the converter didn’t go down when powering up the German 1401.

    We then reset the breaker and tried it one more time.
    It stayed up.
    So we restored tape power and tried it again.
    It stayed UP.

    Frank King"
  4. The next work day, everything seemed to be OK.
  5. Bill Newman and Stan Paddock decided to come in on a Friday work day.
    All of the problems shown in number 3 above showed up again.
    Bill brought out a electrical meter and checked the power line from the circuit breakers on the power panel and the power convert.
    When the system was acting up, there was a missing leg from the circuit breakers.
  6. The problem was traced down to the three wire circuit breakers installed in number 2 above.
  7. We would request the CHM to have the electric contractor come back and fix the problem.
Stan Paddock

Wednesday, September 14, 2022 - from Pat Buder
SCHEDULED DEMO

At 3:00pm Tim Robinson, Bill Worthington, and I gave the demo to 42 visitors. We ran on CT. It and the rest of the equipment that we used performed well.

SPECIAL DEMO

In the evening we did a special demo for a group of 50 Arizona State University Electrical Engineering alumni. ASU is Robert Garner's alma mater (go Sun Devils). Last week Jack Ghiselli used his revised and improved banner program to print two welcome banners for the group.

We ran on CT, which performed perfectly. Robert was in the machine area to help. I gave a slightly shorter than usual demo (about 25 minutes) due to the late hour. It was well received.

As usual, thanks to the restoration team for providing working CompuSaurs.

September 14, 2022 - from Stan Paddock
Mason King 6.0
Frank King 6.0
Stan Paddock 6.0
Tom Szolyga 6.0
Marc Verdiell 6.0
Dale Jelsema 4.0
Ed Thelen 3.0
Robert Garner 6.0

Frank King, Mason King Dale Jelsema and Mason King continued work on the German 1402 card punch. Several circuits were checked and unfortunately the circuits checked worked as designed.
Frank and Dale

Mason King learning
how the 1402 punch works.
Dag Spicer, the Museum’s senior curator paid us a visit.
A large number of visitors wandered into the 1401 room.
Dag played docent and to the random visitors told them about the IBM 1401 and answered questions about the CHM Museum..
Marc got the keyboard for the public keyboard reassembled.
Unfortunately the keypunch rejected the implant of the keyboard.
We are looking for success next week.

Stan

September 10, (Saturday) 2022 - from Jack Ghiselli
Paul Laughton and I gave the 1401 demo on Saturday, 9/10 to 28 people, and mini-demos to 8 more who wandered in before or after. We had international visitors from Argentina and Switzerland. We had one visitor who had worked on Unit-Record Equipment and 1401 and 7090 computers in the Navy at Mare Island beginning in 1959 and was happy to see our demonstration lab.

We ran on CT, which performed well. We got a few read check errors on the CT 1402 Reader immediately after power-up, but these disappeared after some minutes of warm-up. Tape Drive #1 is still marked Out of Service, so we used Tape Drives #2, #3, and #4. We didn't see any problems with #3 as Pat had seen last Wednesday.

All three 026 Keypunches worked OK. The Amazing Restoration Team apparently fixed the poor printing on Keypunch #3 (closest to the door) and the failure to feed on Keypunch #2. Having all three Keypunches made the BigPrint punching go quite quickly.

Apparently, when I did the demo last Thursday, I somehow knocked the A/V battery charger cord from the wall, so the rechargeable batteries didn't get charged. We loaded disposable alkaline batteries into the radio microphones for the demo. Paul found that one of the power outlets in the A/V closet didn't power the charger, so he switched to the other. The charger should show red lights when it is charging. Rechargeable batteries should be ready for the next demo.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

September 8, (Thursday) 2022 - from Jack Ghiselli
Mike Albaugh and I gave a special 1401 demo today to 14 students and staff from St Helena High School in northern California. After the main demo, 22 additional visitors came by and we gave mini-demos. We had several visitors from the Economics conference upstairs, and foreign visitors from Brazil, China, Estonia, and Ukraine.

We used CT which generally ran well.

026 Keypunch #2 is still INOP (doesn't feed). #3 (nearest the door) was printing too light to read. We tried reversing its ribbon and that seemed to help a little, but it probably needs a new ribbon. So only 026 Keypunch #1 (farthest from the door) was usable for the demo. #1 generally worked well, but it still has the problem where cards sometimes jam up when transferring from the Read station to the output stacker. Usually we can pull them out by hand, though.

The 083 Sorter worked OK for the demo. It still seems to handle input cards roughly, crumples some, and sometimes the last few cards of an input deck don't feed. It's also making some unusual mechanical noises.

The 001 Manual Keypunch, which was INOP a few days ago due to card fragments jammed into it. Somebody apparently found a card saw and cleared out the fragments. The card saw is on the CT 1402. We saw some clues that it might have been Stan Paddock. Whoever it was, thank you.

We found CT 729 Tape Drive #3 loaded with a bunch of loose tape dumped in the vacuum columns. So, for the demo we used only Drives #2 and #4, which ran fine. After the demo, we manually rewound the loose tape back on the the reels of #3, and were able to get it to rewind/unload. Not sure if it has problems or not. Drive #1 is still marked OUT OF OPERATION, so we didn't try to use it.

After the demos, we punched some object decks on the CT 1402 Punch, which worked beautifully. Thank you, Restoration Team.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

September 7, 2022 - Demo - from Pat Buder
On Wednesday I gave the demo to an engaged group of 24 people. No assistant showed up. No other guests appeared before or after the main demo.

I ran on CT, which worked well. I did not prep DE because Frank & Mason were still working on the 1402 punch. It was not needed.

The only CT problem was that drive #3 (2nd from right) dumped tape into the right column when doing a rewind. Drives #2 & #4 worked fine.

026 keypunches #1 and #3 ran fine, as did the 001. I didn't try 026 #2, which looked like it was being worked on. With the small group two keypunches were sufficient.

September 7, 2022 - from Stan Paddock and Robert Garner
Frank King 6.0
Mason King 6.0
Stan Paddock 6.0
Robert Garner 2.0

Frank King and his grandson Mason spent the day working on the German 1402 card punch.
A lack of +20 volts has been traced as PUNCH CHECK being on.
While two errors lights have shown, progress has been advanced.

Stan Paddock worked on the PC to IBM 029 in the Leibert.
Progress in the code is getting close.

Stan


Aug. 24, 2022

Sept. 7, 2022
Robert Garner took before and after pictures of the DE system main breakers.
"… it appears that they replaced all 3 with identical units. (It’s hard to discern the subtle differences.)"

September 3, 2022 - from Jack Ghiselli
Steve Madsen and I gave the 1401 Demo today to 24 people, plus mini-demos for 10 who wandered in later. Our only foreign visitors were from Mexico City. Lots of questions from this group.

We used CT which generally worked well. We did have a problem with what seems to be a bad spot on the CT 1403 printer ribbon. We had tested BigPrint and Powers of Two before the demo and printing looked good. However during the actual demo, our first BigPrint printed so light it was mostly unreadable (see photo below). We re-ran it three times and no improvement. Kind of embarrassing, but we skipped that and continued with the demo. To recover, we left the CT1403 print chain engaged, so that the ribbon would keep moving while we talked about other things. Eventually, I guess the ribbon got to a good spot, and we were able to print successfully later. Or perhaps we don't have Paul Laughton's clever touch at slamming the carriage closed firmly, and it's not really a ribbon problem. Unsure. Suggest following demo'ers check for this.

The 001 Manual Keypunch is INOP. It has a fragment of a card jammed into the mechanism and we were unable to clear it. Sorry for my ignorance, but does the 001 need tools to clear a jam? We left a sign on it and hopefully visitors won't damage it further.

For the microphones, we used Paul Laughton's rechargeable batteries. When we put the old batteries into the wall charger, we don't see any blinking lights. Is the charger still working? Next demo please check whether we got the batteries re-charged correctly.

Some continuing minor problems continue, none of which affect the demo:

  • Keypunch #2 is INOP. Doesn't feed or register cards. We left a sign on it. Important only if we have a very large group.

  • CT Tape Drive #1 is marked INOP so we didn't use it.

  • The 083 Sorter moves cards sufficiently to demo. However, it doesn't REALLY sort correctly, so you cannot put a deck in sequence. As a result the sample Ken-Ross-style sort data decks I produced last Wednesday have lost my carefully marked First and Last cards and top markings (artist's "sigh" here). Unfortunately, the only card stock I'd found last Wednesday was completely unprinted, so it's difficult to find the correct orientation for 12-Edge, 9-Edge, and Front/back. Again, this is NOT a big problem for the demo.
Sarah from the front desk came in with a query she'd received apparently from an ex-IBM'er who wanted to know where he could buy IBM punch card stock. We referred her to Bob Garner or Dag Spicer. Ha! We ducked that one!

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

August 31, 2022 - from Robert Garner
Stan Paddock was not in, he asked Robert Garner for a status report. Stan's report (below) received Sept. 8
Stan,

I was in for 2 hours.

Frank and Mason worked on the DE 1402 punch.
(Mason told me that the Pacific Power breaker tripped when he first turned on the DE 1401.
Tried it again w/breakers in 1402 off, then back on, then DE powered up the 2nd time.)

Yes, Dale Jelsema was in today! < dale . jelsema @ gmail . com >
Photo here: http://ibm-1401.info/TeamBios.html#JelsemaDale (I’ll ask him for his bio.)
He asked if he could borrow some of the 1401 manuals to get up to speed. (like the 1401 CE manual). I replied no problem.

I didn’t see Marc. Tom [Szolyga] was in.

Cheers,

— Robert

Stan's report (below) received Sept. 8
Report from Frank King;

Hi, Stan, the German 1401 power converter acted up for the first time I have ever seen.
When I tried to power on the German 1401, the power converter went down immediately, it has an internal circuit breaker that will reset on its own.
After opening the 1402, looking at the 30 amp circuit breakers, I tried it again (It had reset by this time).
Then, Mason and I took the tape circuits off line; that didn’t help.
Mason then suggested we trip the 1402 circuit breakers one at a time.
When we tripped the inner most 30 amp breaker the converter didn’t go down when powering up the German 1401. We then reset the breaker and tried it one more time. It stayed up.
So we restored tape power and tried it again.
It stayed UP.

Because we had a 1402 down and no schematic for the power converter we decided to work on the 1402.
We found a circuit breaker (number 2) of the set that latches and only runs when the clutch is tripped.
It was failing to open, so we took it out lubed it and put it back. It was in the sequence of keeping up.
With card movement, our test showed no change. ;-(

Mason and I left around 15:00 Dale left a little before us Tom was still there.

--------------------------------------------------------------------
From Tom Szolyga;

Hi Stan,

Missed you today! I worked on the following items:
  • Created a real schematic for the Blue Box line driver and line terminator circuits.
  • Ran a simulation of these two circuits connected to each other.
  • Collected the parts for the next version of the German Project breadboard and assembled them. (See attached) I has the ATmega 2560 on the bottom, the board with terminal strip screws for each signal is in the middle and the diagnostic display is on top. There is a plug board for test circuits too. Everything is mounted on a plastic plate so the breadboard circuit will not fall apart.
I will be out of town next week, so I will see you in 2 weeks.

Best regards,

Tom

August 27, 2022 - from Jack Ghiselli, Stan Paddock, Paul Laughton, Robert Garner
Saturday demo, blocked breaker, demo batteries, circuit breaker cycling ON/OFF
From. Luis Valencia, Facilities Manager, CHM - Mon 8/29/2022 2:00 PM
Hello All
Breaker was replaced last Friday after hours, Electrician assessed the issue before replacing the breaker, no overloading or shortcuts were found.

Thank you,

Luis Valencia

-----------------
It appears that the Saturday Demos were run using correct power components.

from Jack Ghiselli, - Saturday demo

Hi 1401 demo and restoration teams,
Hooray, Stan, for the temporary fix. When Steve Madsen, Paul Laughton, and I came in to demo Saturday 8/27, the "1401 GERMAN" ganged circuit breakers were ON. Stan's email mentioned he had taped them on. We didn't see any tape externally so we assume it was internal. Anyway, we WERE able to power-on the Pacific Power 60-to-50 Hertz converter, so that's good. As it happened, everything on the CT 1401 was working well for the demo, so we did NOT need the DE 1401, and didn't try to power it on. But it was reassuring that we COULD have used DE if we'd needed it. Per Stan's email, when we powered everything down, we left the "!401 GERMAN" breakers ON. My understanding is that with the Pacific Power unit OFF, all the outlets and boxes are safely un-powered. Please advise if and when the breakers get replaced.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

from Stan Paddock, - blocked breaker

Somebody indicated that if the German system was needed for a demonstration, the circuit breakers would not switch on.

To make sure the German system was ready for the Saturday demonstration, I turned the three German circuit breakers on and taped the three circuits in the on position.

I put a sign on the door to the Libert room with instructions to use the main power switch for the power converter as the only off/on switch for the German system.

The current connection is still 100% safe.

This current will go back to the original configure when the electrician comes in and replaces the three circuit breakers.

Stan

from Paul Laughton, - demo batteries

There are now new sets of rechargeable AA batteries and a new charger. There are two sets of four. One set has black band at the top. The other has no band. All four of the batteries in a set should be swapped before the demo. The ones in the charger going into the mics. The ones in mics going into the charger.

Some non rechargeable batteries will remain in the drawer to use in case of problem.

Paul

from Robert Garner, - circuit breaker cycling ON/OFF

Ed Thelen noted that the Circuit Breaker Box was likely from Seimens, based on the "catalog number" in the picture.

I also wonder if there’s a grade for higher duty on/off cycling?

We cycle the circuit breakers ON and OFF at least 100 times a year. This failed breaker was cycled at least 1500 times - not its usual expected service?

1401 Demo, Wednesday, 8/24/22 from Paul Laughton
The volunteer signup page did not show ((full) not indicated) that anyone had signed up as the Assistant. I recruited Kathleen O'Brien, a retired computer science professor and my wife, to be my beautiful assistant. When we arrived on site, Bill Worthington was there. He said that he had signed up weeks ago. I decided to split the Assistant role. That worked out well.

The DE circuit breaker refused to be "unbroken" as previously reported. Emails from Robert Garner and others indicate that the problem is being investigated.
(See report below this report)

The CT 1403 printer was left in an inoperable state. The Ink ribbon was stuck at the end of the reel. The paper was jamming as it moved. Robert Garner, Frank King and Bill teamed up and got it fixed before the demo.

We had about 50 guests with visitors from the UK, Germany, Columbia, Egypt, Georgia (the nation) as well as from the USA.

Just before the demo started, I noted that the Mic #2 battery was at 2/3rds. I replaced the batteries.

The 1/2 hour demo went well with virtually all guests remaining in the room through to the end of the demo. We spent about 15 minutes doing Big Prints after the demo.

A big thanks to Bill, Kathleen, Robert and Frank for their able assistance.

Paul

August 24, 2022 - from Robert Garner and Stan Paddock
German circuit breakers solidly not engaging, and Status Report
from Robert Garner
The (90-Amp) circuit breakers for the DE (German) 1401 adamantly (after >10 attempts) do not engage/remain on, per the following video:
Editor's Note: I included this frame copy from the movie, instead of a link to the 20 MByte movie. The movie shows the (90 amp 1401 German) circuit breaker being changed from OFF position to ON position and immediately going to the TRIPPED position as shown.

It appears that we’re now unable to turn on the DE 1401.

Hoping an electrician can replace ‘em.
(I wonder if there's a circuit breaker model spec’d for more frequent cycling, say several 100 on-offs per year?)

Thanks(!),

- Robert

PS. The Pacific Power unit was turned off during the attempts to engage the circuit breakers.


Jennifer Alexander e-mailed "Facilities Manager Luis will follow up on this."

from Stan Paddock

People who attended the CHM were:
Frank King 6.0
Stan Paddock 6.0
Tom Szolyga 6.0
Marc Verdiell 6.0
Robert Garner 2.0
Mason King 6.0

Frank and Mason continued there work on the German 1402 card punch.
They traced the problem to permissive relay 26.
Frank and Mason pulled all of the permissive relays related to the punch and brought them to the workshop.

We have a permissive relay tester that we have not used for years.
When we tried to use it, we found that the power supply for the HP laptop was defunct.

I contacted Dennis Cassar of the CHM IT department to see if he could supply a replacement power supply.
In less than 10 minutes, he brought a replacement power supply to our workshop.
Thanks Dennis.
I showed Mason how to run the permissive relay tester.
Mason took off running the tester as if he had been using it forever.
Relay 26 was totally bonkers and was replaced.
Another relay was also totally bonkers and was replaced.
All relays were replaced into the German IBM 1402 and tested again.
Damn thing punched card!
Now if did the punch without a PUNCH CHECK and a PUNCH STOP error, we would done.
Way to go Frank and Mason.

Stan

August 20, 2022 - from Jack Ghiselli
IMMEDIATE PROBLEM FOR CHM FACILITIES: - with many added comments
IMMEDIATE PROBLEM FOR CHM FACILITIES:
WE NEED TO REPLACE SOME BAD CIRCUIT BREAKERS IN THE LIEBERT-ROOM PANEL IN THE 1401 DEMO LAB.

It appears that at least one of ganged circuit breakers 35-37-39 (labelled "1401 GERMAN") is faulty. Recommend replacing them all. Until this is fixed, it is impossible to power-up the DE 1401.

On Saturday, 8/20/2022, Steve Madsen and I gave the 1401 Demo to 48 visitors, and then a mini-demo for 2 visitors afterwards. We had visitors from all over the USA, and foreign visitors from the Czech Republic (near Brno), France (Toulouse), Japan (near Tokyo), and the UK (north of London). This particular group had lots of technical questions after the demo was over, so we spent quite a bit of time simultaneously answering and running BigPrints. We demonstrated the 083 Sorter, and ran the demo on CT, using 729 Tape Drives 2-4, since #1 had been switched NOT OPERATIONAL. During pre-demo setup, we worked around several hardware problems (described below), and the demo itself went well.

First, DE Power:
Pat Buder's report from Wed 8/17 suggested that the CT 1402 was refusing to read cards, and recommended using DE for the demo. When we powered on the breakers for "1401 GERMAN", they appeared to turn fully on, but we were unable to power-on the Pacific Power 60-to-50 Hertz power converter. A few green LEDs came on, but we couldn't turn on the three voltage displays at the top. This is exactly the symptom Paul Laughton reported in our email exchange of 8/13. I continue to believe that the problem is in the wall circuit breakers, not the Pacific Power unit. I turned the "1401 GERMAN" breakers off and then firmly back on. This apparently worked, because we were then able to turn the Pacific Power unit on normally.

Unfortunately, when we tried loading decks on the DE 1402, it refused to read cards. We tried using NON PROCESS RUNOUT to exercise the 1402 mechanism, as suggested by the Restoration Team. But, no luck. So, we decided to try CT, although both Pat Buder (Wed 8/17) and Paul Laughton (Thurs 8/18) had reported that the CT 1402 was refusing to load decks. For us, the CT 1402 worked perfectly. I suspect this is due to the Amazing Steve Madsen, who gave it a "stern talking-to".

The Amazing S. M. also demonstrated his extensive IBM training by clearing the card jam left in the 083 Sorter as an exercise for us. It was an ugly multi-card jam, and you should see the Amazing S. M. go to work with needle-nosed pliers. He got that 083 puppy working perfectly in time for the demo. Yay, Steve!

One minor issue:
After the demo, while we were rewinding CT tapes, 729 Drive #2 dumped tape into the bottom of the input vacuum column. We were able to Unload and manually wind the tape back on the take-up reel, and then it was OK. Possibly an intermittent vacuum-column problem?
Happened only once.

After the demo was over and visitors had left, we decided to try loading decks on DE. Guess what?
The DE 1402 loaded decks just fine.
Perhaps this means DE needs some warm-up time? It had been powered up for 40 minutes while we ran the demo on CT.

More about DE Power:
Now, we powered down the DE1401, the Pacific Power unit, and the "1401 GERMAN" circuit breakers. Just for grins, we decided to try power-on one last time, and I'm glad we did. This time, we found that the ganged 34-37-39 ("1402 GERMAN") circuit breakers would not turn on. When you flip the breakers on, they immediately bounce back about half-way, and give no power to the Pacific Power unit. Just to be clear, now they are WORSE than before. As noted above, I'd recommend that CHM Facilities replace all three breakers. Until then, DE is unusable. It would be nice to fix this before the Restoration Team and Demo Team come in Wednesday 8/24.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051


From Robert Garner, 8/20/2022 5:04 PM
The three ganged circuit breakers that supply the Pacific Power 60-to-50-Hz power converter for the DE 1401 are tripping.
Either the circuit breakers are degraded, or it’s possible that the Pacific power 390-G unit is tripping them.*
While we can perform a few diagnostic test on the Pacific Power unit,** it might also be helpful if we could get an electrician in to swap out the circuit breakers, just to eliminate them as a possible source of the problem.

Thanks(!),

— Robert

* The 390-G comprises three independent power converter units, one for each phase:
http://www.ibm-1401.info/pictures/PacificPower-390-G-Operation-Manual_PN_114250-F.pdf
One of them could be tripping its circuit breaker, but because they’re ganged, all of them go belly up.

** With the DE 1401 turned off (no load), we could disable each of the three converters one-by-one to see if one of them is tripping the breakers. But it may only be happening under full load, which would involve further diagnostic testing: looking at the phase current waveforms as power comes up.


from Stan Paddock, 8/21/2022 12:02 PM
If the power converter is ON when the GERMAN circuits are turned on, frequently the circuit breakers,

If the Power converter is always turned off before turning the GERMAN circuit breakers off, this problem will never happen.

Stan


from Robert Garner, 8/21/2022 3:24 PM
Stan, et al,

> If the Power converter is always turned off before turning the GERMAN circuit breakers off, this problem will never happen.

Jack’s message implied that the Pacific Power converter was indeed off before he turned on the circuit breakers (i.e., the conventional DE 1401 bring-up order):
“When we powered on the breakers for "1401 GERMAN", they appeared to turn fully on, but we were unable to power-on the Pacific Power 60-to-50 Hertz power converter. A few green LEDs came on, but we couldn't turn on the three voltage displays at the top. This is exactly the symptom Paul Laughton reported in our email exchange of 8/13.”

Re: Frank’s note…
"The circuit breakers for the DE machines converter have tripped when turning them on for a couple of years at least. A push back reset and then on, has always worked for me. I have never had it fail. It generally trips off about 1/2 the time or less.
This is something to consider. The main circuit breaker for the converter is always off when this has happened. i.e. there is no load on the tripping circuit breakers?”
… reaffirms that we’ve had this issue for some time. As Jack noted, it could indeed be an issue with the circuit breakers, since the power converter is typically off before the breakers are turned on.

Or perhaps our Pacific Power unit (or the 3-phase AC wiring) has an AC leak/low-impedance path to ground or between the phases (that’s tripping the breakers as they’re first turned on)? Recall that Ron Crane modified the Pacific Power unit to help reduce the reactive power (added caps I presume). Perhaps that mod is upsetting the breakers? An electrician could help determine the issue.

Cheers,

— Robert

p.s. I may have confused the discussion by postulating a turned-on Pacific Power converter when the circuit breakers are first turned on. Sorry about that… (After all, as Frank noted, why are they be tripping with no load?)

August 19, 2022, - from Paul Laughton and Marc Verdiell
Flaky CT - not LOADing programs
from Paul Laughton
was giving a demo to a private party Thursday around 4:00 PM.
CT would not LOAD any programs. Pressing LOAD caused the CPU CHECK RESET light to turn on without reading any cards.

Paul

from Marc Verdiell
I experienced the same as Paul. Was refusing to read any cards. Then after a while it started to work all of a sudden. CT system also, several occasions.
Marc

17 August 17, 2022, - from Stan Paddock
People who attended the CHM were:
Robert Garner 3.0
Stan Paddock 6.0
Tom Szolyga 6.0
Marc Verdiell 6.0
--------------------------------------------------------------------
From Tom:

Yesterday I accomplished:

  • Soldered together new breadboard. The breadboard plugs into the Arduino board. The LCD panel plugs into the breadboard. The breadboard has screw terminals for each Arduino signal. This config allows signal connections to the Arduino while keeping the LCD display diagnostic capability.

  • I Created LTspice XVII setup and analysis of 1401 cable driver and terminator. (See attached) In the bottom pane is the circuit for the IBM cable driver connected to the IBM cable terminator. In the top pane are the waveforms for the voltages at the Input, on the Cable and the Output. I chose a transition period of 1s to easily see the switching points. According to the IBM ADL pages, the Input (on the left side) varies between 0V and -12V. The Output (on the left) varies between 0V and -12V. The simulation shows these values. The simulation predicts the voltage on the cable varies between 1V and -1V.
    Click to enlarge
Best regards,

Tom

--------------------------------------------------------------------

Marc had taken the 029 keyboard home and see if he is able to fix it there.
Today he continued his work on the keyboard in the shop.
He is getting closer but the "A" key on the keyboard is still not repaired.
Taking the keyboard structure apart, fixing the problem and putting the keyboard again is not easy to do.

--------------------------------------------------------------------

Stan attempted to find the problem with the punch problem on the GERMAN 1401 system.
Going through the GERMAN ALDs but did not get far.
I will continue next week.

Stan

August 17, 2022 - from Pat Buder
Problems - Sorter, CT read
On Wednesday Bill Worthington, Paul Laughton, and I gave the demo to 50 visitors from Slovakia, Switzerland, and other places. We did mini-demos for another 15 before and after the main demo.

We ran on CT, which had the problems described below. All tapes and keypunches worked fine.

Sorter problem
Initially all sorted cards dropped into the reject bin. Choosing a second column had the same results. I reached over to verify the settings of the front panel switches, selector knob, and buttons. I was not in front of the machine so I didn't notice that the SORT/TEST switch was not up. It is not uncommon for visitors to reach beyond the rail to play with these switches, which is what had happened here.
During the struggles, I managed to cause a card jam in the sorter - sorry.

CT Problem
The CT 1402 read and ran BigPrint successfully before the demo. During and after the demo, there were several times when the motor would start but no cards would be read. As Jack, Paul, Marc, and I have recently noted this is a chronic intermittent problem.

Recommendations
I would recommend that docents do not allow visitors to press the 1402 LOAD or START buttons during the demo because this presents an example of unacceptable visitor behavior at times when the lab is unstaffed. We should be encouraging never touching anything beyond the fence.

While CT is closer to the keypunches and more convenient for demos, I plan to use DE until the CT 1402 intermittent problem is fixed.

Pat

August ?, 2022 from Robert Garner and Marc Verdiell
Repair of our IBM 029 keyboard "A" key at Marc's home
Note: from the August 10th report
> Sometime when the 1401 room was un-attended , the "A" key on that machine was broken.
> Marc volunteered to take the keyboard home and see if he is able to fix it.

An idea of the IBM 029 keyboard complexity cam be viewed on Marc's Youtube series -
IBM 029 Keypunch Restoration 6 - Insanely Complicated Keyboard
For more on the IBM 029 keypunch click here.

Material from e-mails is below

Marc (unexpectantly) noticed that several of the “Interlock Disks” were misplaced from their proper locations (they look like small donuts).
Camera, microphone and lamp are for creating Marc's wildly popular Youtube series.

I wonder if that could have contributed to the “A” key becoming stuck (i.e., a missing interlock disk)??

(click to enlarge)

from http://bitsavers.org/pdf/ibm/punchedCard/Keypunch/029/225-3357-3_29_FE_Maint_Man_Nov70.pdf
The diagram is on page 4-41 of that document.

The A-key is repaired, but I think I need to revise my story about the interlock disks, I might have caused that. Anyhow, I fixed that mess and did not lose any disk, but in doing so I created another one, now having to reseat all the springs on the code bars, for the next session I guess. I did not lose any spring either. At least not yet :--))
Never a dull moment.

Marc

August 13, 2022 - from Steven Madsen, Jack Ghiselli and Paul Laughton
from Jack Ghiselli
Just asking. In the 1401 Demo Lab, is it a good idea to have the extra 029 Keypunch outside the barrier where visitors can touch it? We already have the 1902-vintage Model 001 manual keypunch and encourage visitors to play with it. So, they'll probably assume it's OK to bang on the 029, which is easier to break. Marc and Stan reported at least one damage incident. Normally we don't need or use the 029 during demos (we use the three 026 Keypunches inside the barrier). Should it be removed? That 029 is in pretty good shape, and has been used in the past for supervised demos elsewhere in the CHM or off-site (e.g. the Maker Fair), so maybe it should be reserved for that purpose? Decisions like this are above my pay grade, so I'm just asking.

response by Marc

I'll surmise that it's a very bad idea. It's too complex and precious of a machine for the average visitor to understand, much less to operate safely. The only thing they'll manage to do is to break it.
Marc
comment by Grant Saviers
I recall the keypunch room from my student days. 30x 026's about 1/4 of which were always inoperative, even though the IBM FE came once a week to fix them and some of us staff students could undo simple problems.

A not broken program drum was usually unobtanium.

Use restricted to Docent demos sounds like a good idea to me.

Regards,
Grant Saviers

from Stephen Madsen to Paul Laughton
Hi,
Paul Laughton and I demonstrated the IBM 1401 to 40 people. We had visitors from Germany, Toronto, Hong Kong, and China. While running Bigprint we had a reader check and an operator error (pressed LOAD instead of START). That deck was marked because it might have some cards out of position.

We did not start the DE machine, because the Power Converter was not working. All other equipment worked fine.

After the demo, a group of 30 students from Japan came in. We let them punch name cards and ran Bigprint for them.

Regards,
Stephen H. Madsen

from Jack Ghiselli
Hi Steve,
Great demo report! I'm not exactly sure what you meant by "Power Converter was not working". I have noticed that the "1401 GERMAN" circuit breaker set (ganged breakers 35-37-39) are a little flaky. Sometimes when you flip them on, they bounce back a little and DO NOT TURN FULLY ON. In this case, of course, when you try to turn on the DE Power Converter, nothing happens, since it has no input power. My solution has been to turn the "1401 GERMAN" breakers back off, then forcefully turn them on, holding firmly and verifying that they go ALL THE WAY ON. See if that solves the problem.

When I get back in, I'll run VOBJ to analyze the suspect (marked) BIGPRINT deck and fix it if necessary.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

Comment by Paul Lauughton
I noticed the problem with the flak DE circuit breaker. The problem that I noticed was not that exactly. The converter was getting some sort of power. The three phase activation section at the top was not responding to the start button press.

Aug 10, 2022, from Stan Paddock
People who attended the CHM were:
Robert Garner 2.0
Frank King 6.0
Mason King 6.0
Marc Verdiell 6.0

--------------------------------------------------------------------

Frank continued his work on the German 1402 card punch, with the help of his grandson Mason, who clearly inherited the debug-anything genes.
While they tested a number of functions of the 1402 punch, all of those functions seems to be operating correctly.
The story of the 1402 punch continues on.

The docents cited an error with keypunch #1 not feeding cards into the stacker correctly.
Frank using his 50 years experience working on IBM keypunches, fixed the problem quickly.
--------------------------------------------------------------------

From Marc:

I repaired the last of the 3 faulty breakers on Tape drive 4 of CT, and also reconnected its broken wire that was starving a bunch of SMS cards from the -48V they deserved. Some more lights reappeared on the back neon debug panel as expected. I'm not sure if it fixed anything else. I tried to repeat the no-unload fault reported by the demo team, but it did unload (and load) normally. I don't know if I repaired it as a consequence of the breaker or if it's another intermittent fault that will soon reappear. I did not have time to look at CT tape #1 with the reported lazy reel, so that one was left to out of service.

We thought it was a good idea to place a keypunch outside of the barrier's in the 1401 room to be used by visitors when needed.

Sometime when the 1401 room was un-attended , the "A" key on that machine was broken.
Marc volunteered to take the keyboard home and see if he is able to fix it.
--------------------------------------------------------------------

Aug 7, 2022, from Pat Buder, and (CT Tape addendum) from Jack Ghiselli
Sunday was the second day of the annual Vintage Computer Festival. We had a special Sunday demo at 11:00 with 44 visitors. Paul Laughton, Jack Ghiselli, and Mike Albaugh assisted. We were open all day and did mini demos for another 100+ guests from Switzerland, Netherlands, India, China, Ukraine, and the U.S.. The pace was somewhat slower than Saturday, but we still needed all four docents (there were five of us on Saturday).

We began the demo on CT. The BigPrint for one audience volunteer part worked well. However, we couldn't get Powers of Two to run. We switched to DE, but same problem. At one point, all four of us were pursuing fixes. We finally got it to work. It was a missing last card. We switched back to CT for the tape part of the demo because that was where we had prepped the tapes. We ran on drives #2 & #3. See Jackk's email for a detailed description of the problems on #1 & #4. One or two drives are sufficient for the demo.

The two 026's nearest the door ran fine. The 026 furthest from the door still has the stacker feed problem that we have documented previously. We didn't use the keypunch outside the rail. The sorter worked well.

We had a few reader checks on CT, but all resolved when the card(s) were read again.

At one point the CT 1403 again had the problem where one half of the page printed much more faintly than the other side. Firmly reseating the gate latch seemed to fix it.

During a slack period after the main demo, Jack ran his VOBJ program on several demo decks. It found missing cards, etc. which he fixed.

Unfortunately this was one of the most problematic demos I've seen. Many of the problems were actually operator errors. During the heat of running a demo with the associated time pressure it's easy to lose track of things. I would make these suggestions to help forestall these types of problems.

  1. Prepare both machines for potential use. This includes having a card weight, positioning the tapes, etc. After the demo the backup machine can be powered down.

  2. Have two verified copies of each demo program (BigPrint and Powers of Two). Create a separate date card for each BigPrint Place one of each program on the backup 1402 in case it is necessary to switch to the alternate machine.. Run all four of these decks before the demo.

  3. Remove all copies of the demo programs from the top of each 1402 except for the ones listed above.

  4. I asked Jack to begin marking the date each deck was verified on the first card rather than on a slip of paper that can easily become separated from the main deck.
VCF is by far our busiest event of the year. We have learned that five docents for Saturday and four for Sunday are needed. At non-peak times one of us is often elsewhere, e.g. gone to lunch, to visit the VCF, etc. Despite the long days and problems, VCF is very enjoyable. The technical sophistication of this group is greater than average, so the questions and interaction are better. We always look forward to VCF.

Pat

Aug 7, 2022, (CT Tape addendum) from Jack Ghiselli
Pat Buder was Lead on Sunday's demo, and will be submitting a report, expectantly upcoming soonly.

I volunteered to make this addendum report on the CT tape drives. Given all the hard work the Amazing Restoration Team has been doing on these tape drives, I'm embarrassed to report any problems, but here goes.

On CT 729 Tape Drive #1 (closest to the CPU), the tape media had a 3-inch long crinkly section a few feet following the start-of-tape marker. I cut the damaged length off the tape, and put on a new start-of-tape reflective spot. Problem fixed.

CT 729 Drive #1 also seems to have a problem with the feed reel. We use the TAU to command the tape to run, and it runs OK for a short while. Then, we observe that the feed reel sometimes doesn't keep up. When the free tape in the left-hand (feed side) vacuum column reaches the top vacuum hole, the feed reel is supposed to start moving and feed more loose tape into the vacuum column. However, sometimes we observe that feed reel is slow, so that the input tape goes farther and farther ABOVE the top vacuum hole. Eventually the entire feed loop is gone from the vacuum column, and the drive stops. The right-hand (take-up) side vacuum column is working correctly. We switched this drive to OUT OF SERVICE.

CT 729 Tape Drive #4 (closest to the entry door) refuses to unload. We switched it to OUT OF SERVICE, but after the demo we had to power down the CT 1401 with Tape #4 still loaded.

The demos ran just fine using 729 Tape Drives #2 and #3, so these are not serious problems.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

Aug 6, 2022, from Jack Ghiselli
We gave the 1401 Demo Saturday 8/6/2022 during the Vintage Computer Fair, to a large group of 83 people. I was Lead and Pat Buder, Paul Laughton, Bill Worthington, and Steve Madsen were Assistants. After the main demo, we gave mini-demos to about 150 people, who kept coming by all afternoon. We had foreign visitors from Austria, Brazil (São Paulo), Canada (Toronto), Germany (Stuttgart), India (Delhi), Japan, Poland (Warsaw), and Russia.

We started using CT, but there were so many visitors we powered up DE as well. At first we couldn't get any decks to load on the DE 1402. Stan Paddock looked at it, waved his magic wand, and it healed, so we were able to use it. We had so many visitors we really needed all the docents -- they were all busy most of the time -- and both 1401's. We even used the 029 keypunch in the aisle.

The 1403 Printer on CT seemed to run out of ink, so the printeouts faded away. We fooled with it and the problem went away. Paul Laughton thinks the gate was not fully closed, so that the paper was not tight against the ribbon. We'll watch this tomorrow. The CT 1403 ran out of paper so we put in a new box, which turned out to be pure white, without bars; not a problem, just a surprise. 026 Keypunch #1 (farthest from the door) seems to have some problems moving cards to the output stacker -- Sometimes they hang up.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051


Ed Thelen here -

Jack's report (above) reminds me of stories of the last half of WWII in Europe -
Frequently, when adequate weather, the American Air Force would declare a "Maximum Effort Day".

If a plane could fly,
    It was in the sky,
        Over Germany

Looks like "we" are demoing "our" Vintage IBM 1401s
    "Maximum Effort"
        To the Vintage Computer Faire :--))


From Marc Verdiell

In my experience the CT does this inconsistent card loading consistently. It won’t pick anything, then all of a sudden it starts to pick the cards and it’s all good from there. I don’t think we ever figured out the problem. We should.
Marc

Aug 3, 2022, from Stan Paddock
Robert Garner 2.0
Frank King 6.0
Stan Paddock 6.0
Tom Szolyga 6.0
--------------------------------------------------------------------
Frank continued work on the German 1402 card punch.
Frank thinks it is switch 41 on cam shaft 4.
He will test his theory next week.
--------------------------------------------------------------------
Stan is continuing to work on the PC to IBM 029 keypunch in the Liebert room.
The first test passed two blank cards but no data.
Next week should bring significant progress.
Stan
--------------------------------------------------------------------
Tom Szolyga
Today I did:
  • Figured out how to get ROPE3 to work with the LaserJet printer. Now I have it printing the ROPE listing in plain B/W, making them readable.
  • Sorted through the wires we harvested from the IBM cable/bisque connector to use in the harness for the NEW “Blue Box” I am building. My plan is to use twisted pairs for every signal between the 1401 and the Arduino. There will be a DB25 connector for input and another one for output of the Arduino. Each signal will use a pair of pins on the DB25. If the signals use the same bit ordering, it should be easy to find signals on the connectors. The extra grounds should improve reliability.
That’s all. See you in two weeks.
Best regards,
Tom
--------------------------------------------------------------------
Stan

Aug 3, 2022 from Pat Buder
On Wednesday Paul Laughton and I gave the demo to 25 visitors. Another 15 came by before or after the demo. We ran on CT. All equipment ran well except as noted below.

CT Intermittent print problem:
Once when running BigPrint a person's name with a "D" printed as a "4" (missing zone punch). I checked that the card was punched correctly and it ran fine the second time. The card was punched on an 026, not the 001. I've seen this problem before but not recently. I discussed the problem with Frank King, who was there that day. He opined that it was not likely a printer problem or a reader problem due to the sum of punches error checking. No harm dome, but the guest wanted to keep both the erroneous and valid printouts. Extra special souvenir. I told him that the 1401 was very reliable back in the day and that we have to allow our Compusaurs a little leeway now.

Keypunch problem:
The 026 furthest from the door often only partially lifts the card into the stacker after it leaves the read station. Usually the card stops at an angle. It's easily extracted and handed to the guest. I'd say it happens 60-70% of the time.

On 8/6 before the crush of VCF visitors started I was duplicating "Decode Cards" (card with alphabet, digits, and most printable special characters) and had an opportunity to observe the problem a little more closely. I never use AutoFeed, but I do press REL/REG or REL/FEED rapidly in succession. I tried slowing down these steps, which seemed to help a little but not entirely eliminate the problem. The other two 026's worked fine.

As usual, thanks to Frank and the entire restoration team for keeping everything humming.

Pat

July 30, 2022, from Jack Ghiselli
Bill Worthington and I gave the 1401 demo today (Saturday 7/30) to 38 visitors, plus mini-demos for another 15 who came in later. We had foreign visitors from Switzerland (Basel) and the UK (London), plus various places in the US, including Hawaii, St. Louis MO, and Poughkeepsie NY (family of ex-IBM'er).

Kate, as you predicted in your email yesterday, your friend and donor Paula Golden apparently didn't show. However, Kate has lots of friends, so last Wednesday, your other friend Yi Ying, the artist from San Francisco did arrive and Pat and I had a pleasant talk with her. Yi taught me an important factoid: Where do Emoji's come from? Well, Yi apparently created artwork for an Emoji showing Boba Tea (if you don't know what this is, you're over 30). She then formally submitted it to the Unicode © Consortium, who accepted and registered it. It should appear on your smart phones soon.

We ran on CT which functioned perfectly, as did the 083 Sorter and all three 026 Keypunches. We didn't see problems with CT Tape #3 as Pat Buder had reported last Wednesday. The Restoration Team rocks!

Paul Laughton came in to take some photos for the upcoming 1401 Demo Docent training materials. We hope to give a new docent training session in the near future, if CHM can find some candidates.

A/V microphone #2 ran out of battery power at the end of the demo. We accidentally reloaded EBL batteries, and the mic went completely dead. We realized our mistake and re-reloaded disposable Energizer batteries and Mic #2 now works fine. All 1401 docents please note: The rechargeable "EBL" batteries do not work, and neither does the recharging unit. We have put two good-sized packages of disposable Energizer AA batteries in the A/V drawer; please use these for the microphones.

We had hoped to shunt 1401 demo visitors over to the PDP-1 SpaceWars competition starting at 12:00. Unfortunately, some hardware problems on the PDP-1 display forced cancellation. Hopefully it'll be rescheduled in the future.

We still need a 1401 Demo Assistant for this coming Wednesday 8/3. Then, Saturday and Sunday 8/6-8/7 we expect lots of visitors to 1401 demos during the Vintage Computer Fair. We could use at least three docents during those days. Poppy said she'd open up extra slots in the signup system.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

July 27, 2022 - from Stan Paddock
Frank King 6.0
Stan Paddock 6.0
Tom Szolyga 6.0
Ken Shirriff 6.0
Marc Verdiell 6.0
Robert Garner 4.0
--------------------------------------------------------------------
Ken and Marc
Ken started working on the CT tape read signals. Some channels are bad and we are getting errors from all the tapes.

Meanwhile I took apart and repaired the second breaker on tape #4. I think it had been worked on previously but the auxiliary contacts had been reassembled back together wrong. However nothing was broken, so reassembling it properly was all that was needed.

When I went to remove the third and last faulty breaker, I discovered that it was connected to a broken wire, with no matching other broken end for that wire attach to!

After some digging into the schematics, we figured out it was for a supplemental -48V power supply that is only present in the Phase II ans Phase III supplies. So at first we were not sure if it was actually needed. But we eventually figured out that we have a Phase III supply in the machine, therefore it’s needed. Ken figured out that it powers a section of SMS cards that drive the neon indicators on the panel at the back. So it does not affect the drive very much, just our ability to debug it.

I eventually found the other end of the wire, it had broken right at the bundle of cables that runs from the breakers to the rest of the machine, right at the sheath. I was able to unwrap a small section of the bundle and extract enough wire so we can reconnect it cleanly without running another wire. To be done next week.


Marc
--------------------------------------------------------------------
Frank continued work on the German 1402 card punch.
Frank brought in his grandson Mason to help with the punch.
Rumor is that Mason did most of the work while Frank visited with visitors to the room.
He got the punch unit timed except of the shaft that drives PC CBs. It is close but not all the way.
--------------------------------------------------------------------
Tom continued his work on the German PC connection in the workshop.
--------------------------------------------------------------------
Stan continued his work to restore the PC to keypunch system in the Liebert Room.
Getting closer.

Stan

July 20, 2022 - from Stan Paddock
Frank King 6.0
Marc Verdiell 6.0

Frank continued work on the German 1402 card punch. He got the punch unit timed except of the shaft that drives PC CBs. It is close but not all the way.

A large number of visitors wandered into the 1401 room. While Frank is not comfortable talking to strangers, he played docent and told them about the IBM 1401.
[Ed Thelen here - Stan must be "pulling our leg" !! This is not the Frank King I know !! There is no way a person can call Frank "not comfortable talking to strangers".    ;--)) ]
--------------------------------------------------------------------

This Wednesday, Frank mentioned that the breakers can be opened up and repaired, and that they had done so with many of them previously with Joe Preston. So I took one out (that's the hard part, as they have been soldered in), and indeed, you can open them up easily and repair them. I repaired the one that wouldn't trip (just freeing up and lightly oiling a lazy shaft), and put it back just in time for the 3:00 demo. Since I need to take the two other breakers out anyhow before changing them, I'll try to see if I can repair them too. I suspect it will be even easier, as they both trip but something is wrong with the low voltage contacts. Probably just cleaning the contacts will do. I'll let you know how it goes.

Marc

July 16, 2022 Demo, Saturday, from Jack Ghiselli
Bill Worthington and I gave the 1401 demo today to 38 visitors, and then mini-demos to 12 additional people who wandered in afterwards. We had foreign visitors from Australia, Canada, Korea, Russia, and Turkey. We also had some visitors from Omaha Nebraska, where I'm going to go visit next week (so I can't do the Saturday 7/23 demo), and they gave me some tips on what to see. Omaha is foreign lands to me.

We ran the demo on CT, using four (count, 'em, four) tape drives, thanks to the hard work from the Restoration Team. All CT, keypunch, and sorter hardware functioned perfectly.

Yesterday, I stopped by and repaired and verified three (3) BigPrint demo decks. These are in the Demo Tray, and have been verified 100% identical. Please use these. There are several other BigPrint decks kicking around (NOT in the demo tray) which don't work, and appear to be lacking cards at the end. Please don't use these until they've been repaired. There is one (1) Powers-Of-Two deck in the Demo Tray which seems to work.

I also duplicated several "Trojan cards" and put them in the Demo Tray. After repeated use, the first card of a deck sometimes gets a little frayed around the edges, and then won't load. You can put one or more Trojan cards at the front. They don't do anything except go to the next card, but they'll take the wear and tear away from the first card of the real deck.

Bill Worthington kindly helped me with researching exactly how the 1402 "End-of-Card" and Sense Switch A interact. Bill not only knows everything, he also knows where to find it in the manuals.

I managed to break the vacuum tube we've been using as a demo prop. I've been awarded "Klutz of the Day" for that. I found two smaller vacuum tubes which will probably suffice and put them in the demo box. If anybody has a (non-working OK) big vacuum tube, that would be great, since we try to show the size difference between the vacuum tube and a transistor.

Messing Up & sizes of vacuum tubes in archaic computers

Robert Garner has put a complete vacuum tube logic module (I think from an older IBM computer) in the 1401 demo box. It's very cool, but things get so busy during the demos that I'm deathly afraid I will damage this rare artifact. We'd all do well to handle it VERY carefully.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

July 13, 2022 from Stan Paddock
Frank King 6.0
Stan Paddock 6.0
Tom Szolyga 6.0
Ken Shirriff 6.0
Marc Verdiell 6.0
Robert Garner 5.0
Bhushan Mohan --

Bhushan Mohan is a skilled IBM 1401 technician who visits upon occasions.
While he was here, He helped Marc and Ken work on the CT IBM 729 #4 as well as helping
Frank to debug the German IBM 1402 card punch.
Bhushan is welcome to come back and help us whenever he back in the neighborhood.

--------------------------------------------------------------------

HI Stan,

Today I worked on 2 things:

New ribbon for Connecticut 1403

- German project

New Ribbon: I started by asking the status and the process for re-inking ribbons. First I learned no ink is put on the ribbon. Instead, the ribbon is sprayed with a liquid to bring ink in the ribbon cloth to the surface. Then the ribbon can print more pages. I learned that the guy who worked on the process passed away. We don’t seem to have the jig he created for the process or the exact recipe for the liquid. Next Robert found that we had a box of new ribbons. Stan pointed out it was no longer necessary to re-ink the ribbons. Thus a new ribbon was installed in the printer in time for the 3:00 PM demo today. The pages printed looked much better.

German project: I continued to develop the new PCwriter app. This app is in the PC and works with an app in the Arduino. On the PC, I implemented the command entry and decoding process. On the Arduino I implemented the command decoding process. In addition I put a looping structure around the app in both machines. This implements the shell for the rest of the apps implementation. The next step is to implement command execution processes in both apps. The PC app is using C++ under Microsoft Visual Studio. The process of translating portions of the existing app from C to C++ for the new app is ongoing. Since the SW is proceeding well, soon it will be time to build HW.

Best regards,

Tom

--------------------------------------------------------------------

Marc and Ken teamed up on the #4 IBM 729 tape drive on the Connecticut system.
    

Now if you want to know what they did on Wednesday, view the following video.
Because the video file is large, it is slow to start.
http://www.paddockdrayage.com/OneAtATime/MarcVideo.mp4
Now if you want to know what they did on Wednesday, view the following video.
Because the video file is large, it is slow to start.
http://www.paddockdrayage.com/OneAtATime/MarcVideo.mp4

--------------------------------------------------------------------

from Ken Shirriff, 7/16/20022

For future reference, here's how to diagnose the capstan switch problem from the indicator lights on the back of the tape drive. Either the Cap Out switch or the Cap In switch should be lit up (except while the capstans are moving). But we ended up with both lights off, indicating the capstan switch was messed up.

--------------------------------------------------------------------

Stan continued to bring the Keypunch driven by a PC back to life.

--------------------------------------------------------------------

Robert Garner helped getting things in order.

Stan

Wednesday July 13, 2022 - Demo Report from Mike Albaugh via Bill Worthinngton, an apology, two notes
Well, that was a fiasco. I did the demo solo, and not well. The sorter dumped all card into Reject, for several columns that clearly had punches in the selected column. It occurred to me driving home that "somebody messing with the row buttons" was perhaps a better explanation than "Evil Spirits in the clutch housing.

Tape problems were my own doing. Hwen I did the load/rewind, it ran off the end (was apparently already "to the right" of BOT. Then I muffed the re-threading... Even after I got it ok, I did not let it run long enough to hit the threshold for fast rewind.

Could not figure out how to keep the Mic from rotating down below my chin. Gave up and just yelled through my mask.

Did not pass artifacts through crowd. Enough problems already.

Did do Big Print for a kid who had "done it the hard way"

I feel like I'm in a spy novel and have returned to my desk to find a revolver with one cartridge, as encouragement.

OTOH, at least three people thanked me and raved about the Museum.

Again, apologies, although now I have given you several more ways to mess up the demo.

-Mike


from Bill Worthington

Dear 1401 Demo Folks,

Mike very nicely did not mention that I was a strong contributor to the state of the demo today.
Mike, my apologies to you for missing being the 1401 Demo Assistant for today.

In a very partial defense,

  • Lame excuse #1: I am dealing with poison oak and am not wearing socks for another week. (I didn't want to be "out of uniform" for the demo.)
  • Lame excuse #2: I was researching a 1401 question for Jack Ghidelli.
  • Lame excuse #3: None.
Again to Mike, I'm sorry to have hung you out with no assistant. :-(


a note to all from Poppy Haralson,

All – Just wanted to let you know that a new monitor is on order and should be replaced within the week.

Thank you!
Poppy - Sr. Manager, Museum Services

another note from Poppy Haralson, Fri July 15

The new monitor has been installed and is working.

Please let us know if you notice any issues.

Thank you,

Poppy


from Robert Garner

One of our happy visitors to the 1401 Demo Lab this afternoon came armed with her copy of 111 Places in Silicon Valley That You Must Not Miss.
It features the CHM and our 1401 Demo Lab as Place #35 on page 78, photos attached.

First published in 2015 and revised in June, 2021, here’s what it (incongruously) says about the 1401 Demo Lab: “Among the museum displays is an IBM 1401 Demo Lab, circa 1959, which lacks only a pencil-skirted secretary to bring to life.” SHUDDER.

It also reports on the CHM’s Babbage Difference Engine #2, “a steampunk masterpiece of cogs and levers.” OUCH.

Nice to have the publicity(!), but its CHM entry is shrieking for an update.

— Robert

p.s. About the Author on Amazon: "Floriana Petersen was born in Slovenia, where she studied Art History at University of Ljubljana. In 1986 she moved to San Francisco, where she finds illumination as an interior designer, art consultant, author, and photographer. Her interior design projects were published in numerous books and magazines (www.florianainteriors.com).”
    

July 9, 2022 from Jack Ghiselli
Steve Madsen and I gave the 1401 Demo on Saturday 7/9 to 47 people. Today almost everybody was from the USA, with only one group from Canada.

We ran on CT which generally ran OK except that we got a few sporadic READ CHECK errors on the 1402. Reload corrected those problems. We powered up DE as a backup in case the READ CHECK problem persisted, but we didn't need to use it.

Same as Pat, we observed that the right-hand TV wall monitor has a bad display on the right 5 inch wide vertical column. Of course this doesn't affect the demos, since we turn off the TVs.

As usual, we had to try different BigPrint and Powers-of-Two decks to find ones that worked. I put a marker in the front of the good decks saying "Good 7/9". We still need to schedule a deck housecleaning, but I'm away for the next few days. Both Steve and I had to leave, so we didn't do any post-demo mini-demos today.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

July 6, 2022 from Pat Buder
On Wednesday Bill Worthington and I gave the demo to 60 people, which included 20 high school students who joined us after the 2 pm Revolution tour. They didn't stay for much of the demo (bus to catch?). We also had 5-10 others for whom we did mini-demos before or after the main event.

We ran on DE, which performed well. After the demo when we were processing BigPrints, Bill switched to CT, which also ran well.

Stan came into the lab during the demo, so we had the audience give him a hand on behalf of the entire restoration team. He bowed graciously.

Problem (for CHM A/V):
The right wall monitor has an approximately 5 inch wide vertical area on the right side where the video appears hazy.

Note: I have been present for all Wednesday demos since we restarted them. I will NOT be available for next Wednesday, 7/13, so we will need both lead and assistant for that day.

Pat

July 6, 2022 from Ken Shirriff - CT #4 tape drive - suggestion from Iggy
Marc and I worked on the CT #4 tape drive today.
Summary: The drive still doesn't work and we found new issues. More investigation is needed.

The main question from last week was why did relay R109 burn out, and why is the replacement relay very hot. We determined that R109 is very hot on all the drives, so it's presumably working as intended and IBM just likes to keep things toasty. Marc checked that the voltage was correct, -48V.

We checked the circuit breakers and found that the -7.5V circuit breaker doesn't trip and reset. It should probably be repaired at some point. While looking at this, Marc noticed that there were disconnected wires on the circuit breakers, determined that they were for the fuse light, and fixed that.

We noticed that the right capstan was stuck halfway out. Fixing this didn't help overall.

Marc made a relay extender cable so he could probe the relays during operation. He found that the relays R107 and R109 appear to be operating properly so the problem must be elsewhere.

Current behavior:
When doing a load-rewind, the tape drive motors start, but then nothing happens. The next step is supposed to be the tape take-up motor lowers the tape into the columns, followed by the head descending and the capstans coming out, but these steps don't happen.

Hitting reset, unload causes relays to change state, but the drive ends up stuck in a semi-unloaded state where it will neither load nor unload. Setting R109 manually gets it out of this state, so another unsuccessful load-rewind can be performed.

Ken


suggestion from Iggy

Perhaps it may be a good idea to measure the current through that relays 109, 107, and the replaced resistor that was burned out…
Then compare against a drive that does not have the problem.
There’s a possibility of a wiring error, or a short in that rat’s nest area, with the close proximity of the wires inserted in the edge connectors.

Just my $0.02 worth.

Iggy

July 2, 2022 - Demo, from Stephen Madsen & Jack Ghiselli, comment by Robert Garner
from Stephen Madsen

Jack Ghiselli and I gave a demo to 49 visitors. In addition to the local people, we had visitors from Japan, Australia, and Germany. We used CT, which ran fine. The sorter and key punches also worked fine. Thanks to the restoration team.
Regards,

Stephen H. Madsen


from Jack Ghiselli

Hi Steve,
A pleasure working with you yesterday, as always. Again, you got through the main presentation in about 35 minutes flat. You are really efficient, and I'm going to try to learn from your presentation.
Just for your information,I stayed behind to continue my project to verify object decks. As usual, we kept getting late-arrival visitors continually coming in in small groups, so I had to break off and do mini-demos for them. It was a BUSY afternoon. I counted 58 additional people during the afternoon (after our main demo).

Wow. Anyway, in between I worked on BIGPRINT decks, running VOBJ to compare them, and repairing what errors I could. I ended up with three (3) good BIGPRINT Version 16.8 decks in the demo tray. Several others I discarded because of punch misaligned cards and other non-repairable errors. I hope to start on any other BIGPRINT decks are lying around, and on Powers-Of-2, and maybe LINCOLN.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051


from Robert Garner, Sunday 7/3/2022

Jack,

Thanks for your steely quality control campaign!

— Robert

June 29, 2022 - from Stan Paddock, and others about Liebert air conditioner
Received from Stan Paddock Fri 7/1/2022. 9:06 PM

Status report Tuesday June 29
People who attended the CHM were:
X Stan Paddock 4.0
X Tom Szolyga 6.0
X Ken Shirriff 6.0
X Marc Verdiell 6.0
Ken Shirriff and Marc Verdiell went back to work on the number 4 tape drive on the Connecticut machione.
When they replaced the relay that was determine to be found, the tape worked for less than 50 seconds and failed.
In addition, the new relay became hotter than a two dollar pistol. (What ever that means.)

Suspect now switches is the circuitry that drives the relay.
Marc took home parts from the IBM 024 parts machine from the workshop in order to build a jumper to be able to measure the voltage and current of the relay that is failing

Next week we will again attack tape drive four.
Report from Tom Szolyga
Tom has now been able to communicate between the Arduino and an IBM PC over a single USB cable.
That is a big step forward tower communication between the German machine and an IBM PC.
................................................................................................
Liebert heat problem
It was discovered that the Liebert air conditioning system had failed in 1/2 of it's capacity.
The question was is 1/2 of the Leibert is enough for our continued use?
Stan brought in a digital thermometer and turned on the German machine and the Connecticut at the same time.
Within a few minutes, the temperature dropped two degrees and stayed low the rest of the day.
This means CHM does not have to spend $12,000 to fix the failed part of the Leibert machine.
We will continue with the standard practice of turning on the Leibert when the German machine is powered up.

Stan Paddock


from Robert Garner to Stan Paddock - Wed 6/29/2022 11:32 PM

Stan,

Thanks for your sensible insights and temp test today with our Liebert problem. I hadn’t realized/appreciated that it has two compressors (and evaporators).

From the Liebert manual..
http://ibm-1401.info/SL-18100_REV1_08-06.pdf
… assuming our unit is the DH/VH199A model, it’s rated at 50kW (or 50/3.5 = 14 tons).

The manual notes: "The Deluxe System/3 is designed for maximum energy efficiency. Beginning with semi-hermetic compressors, the most reliable and energy efficient compressor available, Liebert adds an A-Frame coil that can provide 60% of total capacity with only one of the two compressors operating. Highly efficient fan motors pull air through the coil, providing not only better coil coverage, but reducing fan motor horsepower requirements."

Our two running 1401s (and the 60-to-50-Hz power converter) may dissipate about 20 kW(?).
so there’s not a whole lot of margin for a demo crowd if our Liebert can only handle 30kW with one compressor disabled.
(and it might be marginal on a hot day).

Perhaps we should ask the Liebert technician for his opinion?

Nevertheless, it does seem possible that we could permanently disable the failed compressor (to disable related error messages.
However, if the 2nd compressor starts acting up, then no demo’s until fixed (and a higher bill?)

Cheers,

— Robert


On Jun 29, 2022, at 10:02 PM, Stan Paddock wrote:

Gary,

When I came in today, the temperature in the 1401 demo room was 76 degrees with the Connecticut 1401 on and the German 1401 and the Liebert turned.

I turned on the Liebert and the German 1401.

Within ten minutes, the temperature dropped two degrees.

For the rest of the afternoon including the 3:00 demonstration, the temperature stayed under the initial temperature.

It appears that the Libert single compressor test was a success.

Stan


>> ---------- Forwarded message ---------
>> From: Gary Matsushita
>> Date: Tue, Jun 28, 2022 at 8:27 AM
>> Subject: RE: Liebert Machine
>> To: Stan Paddock
>>
>> Hi Stan
>>
>> Thanks for the input. If your 1401 team reviews and feels this is the move that is best for the next step, they can let Collections/Exhibitions (Dag/Kirsten) that this might be best use for the exhibit on-going. What was set up with the two Lieberts dates back to the room build-out, and the placement & use of the Lieberts was from the 1401 team. Changing if your solution is better makes sense. Re funds, there was very little still remaining for the 1401 project (a couple thousand max), so Exhibitions was going to have to pull from its general funds to do the repair. I’m sure they will be happy with a less costly fix!

From: Stan Paddock
Sent: Saturday, June 25, 2022 8:54 PM
To: Gary Matsushita
Subject: Liebert Machine

Gary,

The Computer History Museum gained a bunch when they purchased 1401 Shoreline Blvd.

As the corporate headquarters for SGI, the building came with a large computer room with a raised floor, a large UPS (Uninterpretable Power Supply), a large generator in the parking lot and a large Liebert air conditioner.

We had no need for the UPS or the generator in the parking lot and have been removed.

The Connecticut IBM 1401 consumes 12 kilowatts when in operation without requiring any external air conditioning.

The German IBM 1401 also consumes 12 kilowatts when operational and an additional 12 kilowatts with the 60 cycle to 50 cycle power converter operational.

Our operation guideline is that if the German machine is turned on, turn on the Liebert air condition,

Without doing this, the room gets warm within 15 minutes.

The Liebert is a monster with two systems inside.

It was designed to handle a room full of computers when SGI occupied the building.

The current problem is that one of the systems inside of the box has leaked its entire supply of R-22 refrigerant.

According to the Liebert service technician, it is ok to run the Liebert on one system.

Might I suggest we run the Connecticut IBM 1401 and the German IBM 1401 at the same time with the Liebert running 50% capability to see if that is sufficient air conditioning for operation.

If this works, you could have the Liebert service people wire the Liebert to run as a single system.

Then you will only get an error message if the single system develops an error.
This will cost far less than the $12,000 it will cost to fix both systems.

Stan Paddock

June 29, 2022 - demos, Wednesday, from Jack Ghiselli
On Wednesday 6/29, Pat Buder and I started verifying demo decks.

1. BigPrint. We found the current version is 16.8, generated 3/18/2016. This identifier can be found in object cards 7-8. In the ROPE compiled object deck, there is a one-byte error in character maps. The result is that the character "D" has undesired spaces when rendered as middle- or large-size. The correction is to patch object card #98, column 28, from a "4" to an "8". Here is the printout symptom:

In the patched card #98, we added the comment "PATCH" in an unused area (column 30 ff), to remind ourselves in the future. We also ran VOBJ to verify one other copy of BigPrint. Another copy of BigPrint had so many errors we trashed it.

2. Powers-Of-2. We found our demo deck had one card out of sequence. Although this was not a fatal error, we moved the out-of-place card into correct sequence.

Tomorrow (Thursday 6/30) afternoon Poppy has scheduled 100 high school students to visit, and we will give them a 1401 demo. If we have extra time, we will try to verify some additional demo decks.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051


from Robert Garner - Wednesday, June 29, 2022 10:59 PM

Jack,

You’re genuinely disciplined in the IBM way. :)

We need to THINK about a bonus/salary raise! :))

— Robert

June 25, 2022 - demos, Saturday, from Jack Ghiselli, June 26 added comment from Frank King
Update,
026 Keypunch #2 unjammed
Bill Worthington and I gave the 1401 Demo today to 35 visitors. We used CT which worked perfectly. Only the 4th Tape Drive is still unavailable, and that is no problem for the demos. Yay, Restoration Team!

026 Keypunch #2 is still unusable. (See update :--)) When powered on, the drum continuously rocks back and forth. Something is jammed. We tried moving the cover around (which helped in the past), but no joy. When we have large groups, three keypunches really helps.

Everybody should take a turn working with Bill Worthington. That dude has been squared away by 10000 years working for IBM. Bill says at end-of-day in an IBM office you had to clean everything off your deck top and leave it pristine! So, today first thing he did was find the horseshoe key-ring I'd misplaced last Wednesday. Next, he whips out a NEW CARDBOARD BOX he's brought to hold all the demo props. Since we're a museum, we probably should have had Dag Spicer there to curate the century-old previous cardboard, but what we did was THROW IT AWAY, and clean out the heap of trash at the bottom and reorganize. We discovered the two demo SMS boards which I thought had been stolen, so we're back to seven. And the scissors, which we returned to the tool chest "knives" drawer. And a whole bunch of card weights, which we returned to the rack in the back room (the demo box now has exactly 5 card weights: 083 Sorter, CT reader/punch, DE reader/punch). Much easier to find things now. Bill says the new box was originally used by his grandfather, Reverend Worthington, so the contents are now blessed.

Now that the CT 1402 punch works, our next objective is to clean up the various decks in the Demo Tray. I ran VOBJ on the BigPrint and Powers-Of-2 decks we've been using, and found a couple of innocuous errors, but the decks still run OK. I wrapped printouts of the problems around the BigPrint and Powers-Of-2 in the Demo Tray, if you're interested. I started on some of the other decks and found they were missing cards, and/or shuffled. There's also a copy of VOBJ and operating instructions in the Demo Tray.

But, we had a constant stream of late visitors after the main demo had completed at noon, lasting almost until 5:00 pm. When visitors come from far away, and are here only for the day, you feel you need to stop and talk to them. We gave mini-demos after the main demo to small groups totaling 52 people. Maybe we should schedule a Monday or Tuesday to spend time reviewing demo decks when there are no visitors?

We had overseas visitors from Australia (Sydney), Bangladesh, Belarus, Brazil ( São Paulo Salvador), Canada (Waterloo), China (Chengdu), Peru, and Russia (Moscow). Also an interesting group of aeronautical engineers from Joby Aviation, a local developer of autonomous electric VTOL craft.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051


From Frank King, Sunday June 26

Hi, I always love these reports. Thanks Jack
I think I have a fix for Key Punch # 2. The back space key is jammed down, I think.
Remove the cover and pull it back up. If the cap comes off you may need to pull it back up with a pair of pliers. It is probably bent and needs a gentle touch.
If it is still bad Wednesday, I will fix it. Good Luck. :-)

Frank King


(See update :--)) dated 11:19 AM Wed 6//29/2022

From Ken Shirriff

I unjammed the backspace key following Frank's advice and keypunch #2 now works.

Ken

- - - - - - -

From Jack Ghiselli - June 29, 2022 11:46 AM

Thanks, Ken. That will help the demo today. Sorry we couldn't fix it ourselves.

June 22, 2022 - demos, Wednesday, from Jack Ghiselli
Hooray to the restoration team! Yesterday (Tuesday) Dale Jelsema, Frank King, Mason King, and Stan Paddock fixed the CT 1402 card punch, which has been down for a couple months.

I came in yesterday and punched some ROPE object decks which had been waiting on the 1402. One is the VOBJ verifier to help check if our demo decks have any problems. I punched its object deck and rant it, and sure enough, it showed some minor errors in both our demo-copy BigPrint and Powers-Of-2 decks (but they run).

The Wednesday 3:00 demo was so popular it ran until almost 5:00, so we didn't have much time. When we get more time, we'll check all the copies of all the demo decks.

Thanks again to the Restoration Team.

--Jack Ghiselli - jghiselli@sbcglobal.net, Cell 408-839-1051

June 21 & 22, 2022
from Stan Paddock, June 22th
DaleJelsema6.0
KenShirriff6.0
TomSzolyga6.0
MarcVerdiell6.0
Report from Tom Szolyga
Today I worked on the SW to communicate between the PC and the Arduino.
Making progress.

From: Ken Shirriff
Marc and I looked at the CT tape drive problems today.
Last week we found an exploded resistor-capacitor module RC108 on tape drive #4 and Marc removed it with great difficulty:
Marc tested it on his fancy HP RLC machine and surprisingly it turned out to work just fine! So we took a closer look at latching relay R109 next to it:
As you can see, the relay coil shorted out. Apparently the heat from this damaged the case of the RC module next to it.

I found three replacement latching relays in the workshop. One looked like it had been underwater (extensive rust), one had a bent contact that wasn't closing (not the contact wire that moves, but the U-shaped fixed part was bent), and one had some dirty contacts. Marc is cleaning up the contacts and we'll replace the relay next time.

Second tape drive issue.

A few weeks ago, Frank and Stan noticed that the CT tape drives gave errors almost continuously when operated from the TAU control panel.
I looked at the bit signals on the CT machine and compared them with the DE machine.
On the DE machine, the signals from the tape drive are 8 volts peak-to-peak as they should be.
On the CT machine, the levels were all over the place.
Here's a look at tape signals on the DE machine:
The yellow line is the raw signal from the tape drive, 8 volts peak-to-peak.
The cyan line shows the signal after rectification, so it has twice as many peaks.
The pink line shows the signal after differentiation, producing spikes.
Finally, the blue line shows the data as stored in the TAU bit latch.

On CT#1, when all bits were supposed to be set, bits 2 and C were missing from the A register.
That is, the TAU didn't read these bits.
Here's how bits 1 2 4 and 8 look as received from CT drive 1.
These should all look like the yellow line above.
Note that bit 2 (cyan) is about 3 volts peak-to-peak, much lower than the others, which would explain why it doesn't get read.

The others are 6 volts instead of 8, so they are significantly lower than they should be, but the TAU still mostly reads them. I didn't look at bit C, but it's probably low like bit 2.
Probably the amplifiers need to be adjusted, although it could be a problem with writing instead.

Drive #2 is even worse, with all the signals around 2 volts peak-to-peak.

Drive #3 has signals around 5 volts peak-to-peak.

I didn't test drive #4 because of the relay problems.

Since the drives all have different issues, I think the problem is with the drives, not with the termination.
So it looks like the CT tape drives all need some adjustment before they will read data reliably.

Ken
--------

From Marc Verdiell

And while Ken figured out the bad relay and the bad tape reads, I replaced the cracked RC module with one that looked good (although, as Ken pointed out, the cracked one probably still works, it appears to be just cosmetic damage).

I think we should reverse engineer the cracked module construction for the benefit of science, it looks like a highly interesting and advanced construction, maybe using thin film ceramics.

I then proceeded to reconnect all the backplane wires at the right places, those that I removed on purpose and had marked, and those that had decided to remove themselves on their own will without being marked.

It’s very hard to access and see through the thick mess of entangled wires in the backplane, and I was proceeding quite carefully to avoid introducing wiring mistakes, so that took me all day.

All the wires are back in, hopefully on the correct pins.
We should be able to just plug in a new cleaned-up relay in location 109 next week and see if we corrected the problem.
Or if we created a new exciting one…

Marc


from Stan Paddock, June 21th
DaleJelsema5.0
FrankKing6.0
StanPaddock5.0
Visitors

Frank King has a grandson Mason King who came in today to see what his grand father does at the CHM.
Mason was a great help with today's work on the punch.

Connecticut IBM 1402 Card Punch

For several weeks, we have tried to fix the card registration of the cards punched on the Connecticut machine.
We have tried cleaning and adjusting this and that.
We have three books on maintaining the 1402 card punch.
None of these books seem to match our punch.
One of the books seemed to be close to our punch.
It described an adjustment that has a range of +- one degree.
When the guys checked, it seemed the adjustment was out of range.
Frank, Dale and Mason adjusted the setting and it fixed the problem!!
With the knowledge from fixing the Connecticut card punch, we think it is time to fix the card punch on the German card punch next punch.

Leibert room PC to IBM 029 keypunch

Stan is continuing to change the software to support the new hardware configuration.
Several weeks to finish the job.

Stan

------------ This is an addition from Frank King in reference today's work. ------------

We cleaned and adjusted the timing on the grip and release, of the secondary alignment for the card to be punched.
We reproduced a new big print card deck and left it on the reader.
I hope this fixed the problems with the punch unit.
You guys can try it out and let me know your results.

Thanks for all you do.
Frank

Present were:

Dale Jelsema
Frank King
Mason King
Stan Paddock

June 14 & 15, 2022
from Stan Paddock, likely June 14th
KenSherriff6
Ken Shirriff is well known for reverse engineering printed circuits.

He found that there is a relay in the third IBM 729 on the Connecticut machine that was not playing along with the band.
Ken took the SMS card that drove the relay home and checked the driver circuitry that drove the relay and found the transistor that drove the relay was blown.

Ken replaced the SMS card at the CHM and tried it again.
he relay stilled did not work.
Not only that, the same transistor as before was blown.

This narrowed the problem down to the relay or the arc suppression diode across the relay coil.

Failure of either one of these is very rare.
The arc suppression diode had shorted so that when the transistor wanted to activated the relay, the shorted diode blew the transistor.

In the shop, we have multiple copies of the transistor so both blown transistors were replaced.
Marc found a generic diode that had a higher PIV and current rating than the original diode.
With a little soldering, the tape drive was put back together and tested.
Everything on drive three is AOK.

Stan


Reports from Tom Szolyga and Ken Shirriff via Stan Paddock, June 15th
KenSherriff6
TomSzolyga6

Today I {Tom} worked on the SW to communicate between the PC and the Arduino.
Making progress.

Stan, here's Ken's Wednesday status explaining the RC module problem.


I investigated the tape drive CT#4 more today and it needs a new RC 108 module. This resistor-capacitor module snubs the relay coil. As you can see in the photo, it has self-destructed. (This is the module to the right of relay 107.)
Click to enlarge

To summarize what I've found:
Latching relay 109 closes to indicate an unloaded state. (It latches to store this state across power off.)
When this relay opens, it turns on the blower relay DP4.
When the blower vacuum switch closes, it turns on the main run relay R1.

The observed problem is that when the drive is powered on, the blower turns on and then the drive starts running, even though it is unloaded. So the problem is in the R109, DP4 path. I figured one of these relays might be faulty. However, the bad resistor-capacitor module R108 is across the coil of R109, so if it's messed up, it would explain the problem.

I need help to replace the RC module though:
  1. Do we have any of these 330 ohm .1 mfd modules?
  2. The module is hard to replace; it has taper pins plugged into it, and of course they are behind a bunch of wiring cables. So I'll need a hand with the wiring.

Click to enlarge
There's nobody else in today, so we can work on this next week.

Also, about 30 people came through the lab and I explained the computers.
The computers weren't powered up, so I couldn't do demos apart from the manual punch


June 11, 2022, from Jack Ghiselli
Bill Worthington and I gave the 1401 demo to 38 visitors, plus 11 more afterwards for mini-demos. Special thanks to Jesse Nichols who went through Revolution and urged visitors to come to the 1401 demo. We had foreign visitors from China, Ireland, UK (London), and Italy (Calabria). Our timing ran a little long (45 minutes). We ran on CT, and all hardware worked perfectly. CT 729 Tape Drive #1 seems not to do high-speed rewind, but stays in low-speed. That doesn't affect the demo, though.

We had apparently used up almost all the non-rechargeable AA batteries for the microphones, so I brought in a couple additional packs. We should have enough for several months now.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

June 7, 2022, from Stan Paddock and Robert Garner
from Stan Paddock
FrankKing
StanPaddock
KenShirriff6.0
When we came into the 1401 demo lab, we noticed 4 things that need our attention:
  1. The third tape on the Connecticut machine was partially loaded and the tape was cut.
       Ken took the reel down, cut the damage portion of the tape off and added a new beginning-of-tape on the tape.
  2. We configured the first tape drives as 1,2,3 and ran them from the TAU and got errors from each tape.
       Ken said Marc was working with these tapes and the configuration with the terminator and may know what is going on.
  3. Tape 4 was know to be out of operation.
       What Ken discovered is that when the tape is powered up, it goes into a status of 50% up and 50% down.
       It is not obvious what is going on.
        Ken took home books on the tape drive to learn what make the tape work.
  4. Frank and Stan checked out the card punch on the Connecticut system.
       Last week we pulled the punch unit, Cleaned every thing and put it back together.
       We punched some cards and discovered we had not fixed the problem.
        We also noticed that a pair of lower bearing are very worn.
       We will try to remove the roller shaft and get to the bearings.

e-mail from Robert Garner to CHM facility folks, June 8

Gary, Dennis, Poppy, et al,

Per Jack Ghiselli's May 28th email below, the Liebert in the 1401 demo lab is in need of service.
I understand that a Liebert field engineer checked it out a couple of weeks ago and may have concluded that one of the (two) compressors was faulty.

Any news on the repair status? Or plan for getting it repaired?

Thanks(!),

— Robert

p.s. Someone had tuned Off the main On/Off lever switch. I turned it back on for this afternoon’s demo.
When both on, the 1401 units and the 60-to-50-Hz power converter combined can dissipate up to 25kW of heat.
If not removed, it can get too warm during the 1401 demos and maintenance sessions.


From Marc Verdiell

If I remember correctly, a technician came last week. One of the two systems was out of refrigerant. It must have a leak. The refrigerant was replenished and the system was repressurized, but that is a temporary repair. The technician recommended that a proper leak detection should be made in a subsequent visit and the leaky part replaced. Not sure who is in charge of that at the Museum (facilities?).

Marc


from Luis Valencia, Facilities Manager, Computer History Museum

Hello All
As of right now Facilities is waiting on repair proposal from ACCO.
We’ll keep everyone posted on this.

Thank you,
Luis,

June 4, 2022, from Jack Ghiselli
I gave the 1401 Demo today, Saturday 6/4/2022. Steve Madsen had kindly signed up to help, but at the last minute had a conflict.

I suspect the nefarious Amazing Restoration Team is having fun teasing us demo docents. When I powered up CT, at first the 1402 wouldn't read cards, although I tried several decks. Finally opened up the covers and lo to my surprise found a greasy paper towel in the read rollers. I bet you guys have a camera somewhere and were just laughing at us. OK, I have put the greasy paper towel SOMEWHERE in the maintenance area -- you find it! After that, all hardware worked perfectly -- the CT, its first two 079 tape drives, all the 026 Keypunches, and the 083 Sorter. Anyway, if you guys fix the CT 1402 punch, all will be forgiven.

We had a lot of visitors -- 42 for the main demo, and 36 others who wandered by later in small groups for mini-demos. We continue to get quite a few international visitors, today from Australia (Melbourne), India, Kazakhstan (Almaty), Switzerland (Zurich), and Uruguay. With only one docent, the after-demo BigPrints and especially the mini-demos for later groups went slowly. We try to shut down after the main demo, but it's really hard to turn away a visitor who's traveled a long way, with only today to see things. So mini-demos went on until 2:00 PM.

I had to remind a few visitors that the CHM is requiring masks for 1401 demos. The front desk has extras.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051


Addendum -

Well, Ed, there was GREASE on that towel. We demo docents mostly just stand around looking beautiful. I had to go wash my hands. Yuck!

May 31 & June 1, 2022
from Stan Paddock
Status report Tuesday May 31
from Stan Paddock
KenShirriff6.0

Ken Shirriff is well known for reverse engineering printed circuits.
He found that there is a relay in the third IBM 729 on the Connecticut machine that was not playing along with the band.
Ken took the SMS card that drove the relay home and checked the driver circuitry that drove the relay and found the transistor that drove the relay was blown.

Ken replaced the SMS card at the CHM and tried it again.
The relay stilled did not work.
Not only that, the same transistor as before was blown.

This narrowed the problem down to the relay or the arc suppression diode across the relay coil.

Failure of either one of these is very rare.
The arc suppression diode had shorted so that when the transistor wanted to activated the relay, the shorted diode blew the transistor.

In the shop, we have multiple copies of the transistor so both blown transistors were replaced.
Marc found a generic diode that had a higher PIV and current rating than the original diode.
With a little soldering, the tape drive was put back together and tested.
Everything on drive three is AOK.

Having three operation IBM 729 tape drives on one 1401 computer is pretty good.

Next week we will attack tape drive four.



Status report Wednesday June 1
from Stan Paddock

People who attended the CHM were:
FrankKing6.0
StanPaddock6.0
TomSzoolyga6.0
KenShirriff6.0
MarcVerdiell5.0
RobertGarner2.0

Report from Tom Szolyga
Today I worked on the SW to communicate between the PC and the Arduino.
Making progress.



The 1402 card punch on the Connecticut machine had decided to start punching cards out of registration.
Today, Frank King and Stan Paddock pulled the punch out of the Connecticut 1402.
This allowed us to clean and lubricate the internals.
The timing was checked.
We will put everything back together next week and test the operation.
Next Tuesday, Frank and I will complete the re-installation and testing of the punch.

The German machine runs on 50 cycle power.
This requires an 60 cycle to 50 cycle power converter in order to operate.
The power of the 60 cycle to 50 cycle power converter and the power of the German machine generates more heat than the standard air conditioning can handle.
When running the German machine, the Liebert Air conditioner has to be on to dissipate the heat.

The docents noticed an error displayed on the Leibert air conditioner in the back room.
They passed the error on the CHM staff and a Liebert serviceman came out today.
The serviceman told us that the machine can be used as is until they can come out to fix the problem.

Robert Garner invited guests to the CHM today.
His guests today were Geoff Thompson, Ron Rider, John Shoch, and Dan Swinehart, from the early days of Xerox PARC, 1972-74. 1
I believe they liked the IBM 1401 exhibit more than the rest of the CHM.

Stan

1 from: Robert Garner
To those interested/involved in Ethernet's history,

Here's the YouTube recording of the April 13th “Ethernet’s Emergence from Xerox PARC” IEEE webinar:
https://www.youtube.com/watch?v=Z_cgixHivqw. (about 2.5 hours)
Click Here for time sequence of speaker and topic

May 28, 2022, from Jack Ghiselli
NOTE: The Liebert air-conditioner unit in the back room is sounding an audible beep, and its tiny screen is displaying the error message
"01 ALARM PRESENT LOW SUCTION, COMP. 2".
Turning the Liebert off and then back on makes this message go away briefly, but it soon reappears. I hope somebody more competent than me can look at the Liebert and see if it needs fixing.

Today, Steve Madsen and I gave the 1401 demo to 43 people, and mini demos to 10 additional people who wandered in later. There were visitors from Connecticut, whom we told about the history of the CT machine, and two separate families from Germany (Bremen and Hamburg) whom we told about the DE machine (they liked the internal 1403 labels in German). Another group came from India (Bangalore), and we showed them the photo of the (non-functional) IBM 1401 on display in Bangalore, which they now intend to go visit. There were also international visitors from Russia (two different people) and Switzerland (Zurich). The museum seems to be attracting lots of overseas visitors.

A BIG thanks to Steve, who signed up to assist at the last minute when he noticed nobody else had. Yay, Steve! We finished the demo in 40 minutes time, and got all the after-demo BigPrints done and handed out to this large group by noon. We used CT. All the hardware worked well, including all 029 Keypunches and the 083 Sorter. On the 083, we are using Pat Buder's suggestion to alternate between columns 74 and 75 on the Sort Deck, which seems to result in a pleasing Sort demo. Thanks to the Restoration Team's work last Wednesday, two 729 Tape Drives on CT were working so we could demo tapes using the TAU.

As a reminder, the CHM is asking 1401 demonstrators to require masks on all visitors and demonstrators in the 1401 lab.

Stan: I forgetfully left my USB thumb drive in your PC hooked to CT. Please don't throw it out, and I'll get it next time I come in. PCWRITE function 2 works great for printing ROPE listings on 14x11 paper.

May 24 & 25, 2022, from Ken Shirriff, Stan Paddock, Robert Garner, and Antoine Bercovici
from Ken Shiriff via Stan Paddock - May 24

Ken worked 6 hours
I was the only one at the lab today, so I studied the documentation for the CT drive that isn't loading properly. I didn't power up the system since I figured that poking around in the tape drive by myself was potentially hazardous.

Summary: I think the problem is that the capstan motor isn't starting. I hope to probe the capstan motor logic (TU.9.40.1) tomorrow.

Details (mostly for my reference, but in case anyone wants more):

To review, the symptoms when loading a tape on this drive:

  1. The tape down motor goes on for a few seconds and stops. This turns the worm gear that rotates the brake clutches, moving the tape slightly into the columns. This behavior is correct.
  2. The head up motor moves the head down (down, not up, because R7 motor reverse is active). The head up switch opens and the head down switch closes as expected. However, the motor doesn't stop, which is wrong.
  3. The capstans don't come out, and the right clutch doesn't spin the tape into the column. This is also wrong.
I think that figuring out why the capstan motor doesn't start is the key, since the clutch is activated by the capstans. The capstan logic is TU.9.20.1:
The capstans depend on a bunch of signals: R7 motor reverse, LD rewind stat, tape in L col, tape in R col, unload status, and head down. The tape in column signals are supposed to be the ones that cause it to activate. They looked okay when I looked on the oscilloscope a couple of weeks ago, as did the head down switch. So maybe another signal is messed up, or a board is bad here, or there's something wrong with relay R3 or the capstan motor. I hope to probe these signals tomorrow which with luck will make the problem obvious.
As far as why the head down motor doesn't stop, I've puzzled over the logic for a long time. My conclusion is that the logic doesn't match the written description in the manual:
Supposedly the head down switch stops the motor, but as far as I can tell, this is not how the logic is implemented and not how this 729 should behave. My guess is that this 729 is a bit unusual (since it's the only one with this circuitry in the lab). Specifically, on TU.09.35.1, unload gate goes low, which blocks the signal from head down and then ld gate, so the motor stop signal is blocked. All the signals here appear "correct". As far as I can tell, the circuit just doesn't match the description, and the motor is not stopped by the head down switch.
My theory is that this drive is designed so the head down motor will be stopped when the capstans move, rather than when the head down switch closes. (The logic supports this.) Since the capstans aren't activated (due to the problem), the head down motor will keep running, which we see.

Thus, fixing the capstans should make everything else work.

Ken


Status report Wednesday May 25 from Stan Paddock
People who attended the CHM were:
FrankKing6.0
TomSzolyga6.0
Ken Shirriff6.0
MarcVerdiell6.0
RobertGarner2.0
Report from Tom Szolyga

Project Report: Today I worked on the SW to communicate between the PC and the Arduino. I have converted the PCWriter app to Visual C++ and Visual Studio 22. It compiles and runs on Windows 10 or 11. So far it takes commands from the user and opens/reads and writes files. Next step is sending the files to the Arduino. I am investigating two ways to do that. One way is by doing IN or OUT instructions to the I/O port. The second way is by using Windows I/O functions calls. Doing I/O instructions in a C user application program to a Serial port mapped through a USB port is dodgy because Windows is not a real time system. I am thinking that using Windows function calls is a more definite, easier to read approach. Unfortunately it takes more time and effort to create.

Report from Frank King

Ken and I worked on CT tape drive # 3. The head down switch seems to be operating properly but the circuit that starts the capstan motors doesn’t seem to be working. If we manually pick relay 3, the capstan motors start but the circuit doesn’t pick R 3 or R to 8 like it is suppose to. Ken is going to take the card home to see if there are bad transistors.

and Stan

I duplicated Big Print deck using CT and it ran ok except for one reader check. So, we should make the alignment of the punch a priority next Tuesday.

Mark and a buddy from Paris (Antoin Bercuvri), I think, continued body work on the covers for the tape drives. Looks like they knew what they were doing.
They fit for the first time. :-)

Robert showed up for a few minutes at lunch and then came by the 1401 area for a short time.

We had a visit from Kate McGregor.
She stopped by on her way from Alaska.
What a wonderful person.
The Museum was lucky to find her.
Too bad she has to work remotely.

Stan


From Robert Garner

Attached are four photos of Marc & Ken working on the ailing CT 729s with Marc's guest Antoine Bercovici visiting from Paris yesterday!
It was fun showing off/sharing the 1401, RAMAC, and Revolution exhibits with Antoine.

In his spare time, Antoine has beautifully micro-photographed hundreds of chips (from his collection of thousands):
https://zeptobars.com/en/
https://twitter.com/Siliconinsid
His "day job" is researching paleolandscapes, paleobotany, and sedimentology:
http://www.paleoserver.com/aberco/site/homepage.html
(I dare say that integrated circuits could be viewed as an instance of small-scale sedimentology! :))


From Antoine Bercovici

Adding my own shots of the day, this working 1401 display is some of the most unique vintage computer exhibit, having all the sound and the smell of the vintage equipment filling the room.

I must say that Zeptobars is not me, but https://twitter.com/Siliconinsid as well as https://www.instagram.com/siliconinsider/ is!

And a bit of shameless promotion for our work in Colorado, elucidating how life recovered in the 1 million year following the dinosaur killing asteroid impact: https://www.pbs.org/wgbh/nova/video/rise-of-the-mammals/

Cheers!

Dr. Antoine D. Bercovici
Research Associate (Paleobotany and sedimentology)
Department of Paleobiology, NMNH Smithsonian Institution
10th Street and Constitution Ave. NW, Washington, DC 20560-0121
http://www.paleoserver.com/aberco/

May 21, 2022
from Jack Ghiselli

I gave the 1401 demo on Saturday 5/21. Nobody signed up as Assistant. There were 32 visitors for the main demo, and 19 people who wandered in later for mini-demos. There were international visitors from Australia (Sydney), Canada (Vancouver), Germany (Hamburg), Poland (Warsaw), Russia, and Switzerland (Zurich). A very lively group, with lots of questions. I used CT, which worked perfectly. The Amazing Restoration Team had even got one CT tape drive working, so there was no need to use DE for the tape demo. All 029 Keypunches and the 083 Sorter worked perfectly.

I put a good BigPrint deck and a good Powers-Of-Two deck at the front of the metal Demo Cart tray, and recommend you use these. There are several other copies which don't work, and a quick inspection showed at least one was missing cards. I'm looking forward to a working 1402 punch so we can punch a new VOBJ object deck, run it to see what the problems with the various Demo decks are, and make all copies good.

The main demo, including 083 Sorter, took only 37 minutes, which is about the desired duration. Without an Assistant, the multiple after-demo BigPrints took longer than usual. One minor good news item: I discovered our missing third demo vial of cores at the bottom of the cardboard carton on the Demo Cart, so it didn't get stolen after all. Thank you, Esteemed Honest Visitors. "Trust but Verify", as Ronald Reagan used to say, quoting the old Russian proverb, “doveryai no proveryai”.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051


May 18, 2022
from Bill Worthington

On Wednesday, May 18, Pat Buder and I conducted the 1401 demo for 14 museum visitors. They were joined about 5 minutes after the demo started by a dozen highschoolers who were enjoying a field trip. along with the 14 visitors was a couple from Italy.

We had a problem finding a working BIG PRINT deck(I thought there were about 3 copies in the program tray.and borrowed Jack Ghidelli's Thanks Jack!

May 17 & 18, 2022, from Stan Paddock
May 17
People who attended the CHM were:
Stan Paddock 6.0
Marc Verdiell 6.0
Stan worked on the CT 1402 punch. I cleaned the feed rollers and lubricated the feed rollers bearings. While everything looked better, it did not fix the registration problem. Marc and I discussed how to put the CTE tape drive back together.
May 18
People who attended the CHM were:
Stan Paddock 6.0
Tom Szolyga 6.0
Marc Verdiell 6.0
Tom was continuing his work on the new German PC connection and said he was making progress.
When I came in, Marc was attempting to mount the front door of the first tape drive with the help of Dale Jelesma.
Dale is a friend of Frank King and worked for IBM for over 40 years.
We are glad to have Dale join us when he can.

Marc and Dale worked on the tape drive getting all the sheet metal reattached to the drive.
It took the skills of a body shop to get the covers to fit back on the machine.
That tape drive is now fully operational.

Stan worked on the punch for the CTE machine.
The punch produces cards that are out of registration.
The work I did not fix the punch.
Where is Frank when you need him?

May 14, 2022 - from Jack Ghiselli
Paul Laughton, Ken Ross and I gave the 1401 Demo on Saturday 5/14. I was Lead, and Paul and Ken decided Ken would be the Beautiful Assistant. I have to say I couldn't decide who was more Beautiful, but Paul was about to travel out of town so decided he'd not expose himself to COVID. We had two visitors early, 18 people for the main demo, and 20 wandering in afterwards for mini-demos. Today it was a smaller group of visitors, but quite varied -- we had visitors from Australia (Perth, no less), Bulgaria, Chile, Italy, The Netherlands, and Thailand. We tried to remember to use metric measures during the presentation, since real-world visitors get annoyed at inches, yards, hogsheads, etc.

The visitor from Bulgaria had worked back in 1968 on punch card computers built in the Soviet Union, which he said were copies of IBM 360s.

When we came in, there was a multi-card jam in the 083 Sorter. I had missed seeing Pat Buder's report from Wednesday about this, but luckily we were early, so had time to clear the jam and get the 083 running again. While testing the 083, I noticed that cards were sorting into the correct 0 to 9 pockets, not "one off" as in the past. The 083 directed some cards to the REJECT stacker, although the cards seemed to have good punches. I suspect some of these had been punched sightly mis-aligned, so I'm not sure whether our past sorting problem is actually an 083 problem or a 1402 punch alignment problem. I agree with Pat Buder's suggestion that after the 1402 Punch is fixed we should punch a new demo Sort Deck.

We ran the demo on CT, but since none of its tapes are working yet, we demonstrated 729 Tapes on DE, using the TAU. Everything ran beautifully. Thank you again, Restoration Team. We got the demo finished in 45 minutes, which is slightly long. Still working to shorten it up.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

May 11, 2022
Apparently Stan is having medical treatment for his back
- Ah, he is back, this report received May 15, 2022
People who attended the CHM were:
X Frank King 6.0
Stan Paddock
X Tom Szolyga 6.0
X Marc Verdiell 6.0
Tom was doing his thing and said he was making progress.
Mark got some parts rearranged looking for a wire clamp and when I left was getting ready to install the shaft.
Frank reproduced a big print deck a few times and got a punch check each time. Not sure why. Nothing accomplished.

Others are "filling in the blank"
from Marc
Likely present, Marc Verdiell, Frank King, Ken Shirriff, Tom Szolyga, others??

I put the shaft back in the machine [CT Tape], re-pinned the hub, and re-mounted the brushes (always a challenging operation, this last one). The tape drive should be good to test next week after I re-tension the main belts.

May 4, 2022
Apparently Stan is having medical treatment for his back
- Ah, he is back, this report received May 15, 2022
People who attended the CHM were:
X Frank King 6.0
Stan Paddock
X Tom Szolyga 6.0
X Ken Shirriff 6.0
X Marc Verdiell 6.0

The DE tape drives fail to write bits without hanging after a few seconds. The Disc light is on but I think it should be if it hangs

I had three groups of young adults about 12-14 years old that punched cards. Once they got the hang of it, they punched like pros. All that wanted a Big Print so I ran it for them. First group was about 10 -12, the second group about 20 - 25 the third group was 10 - 12.

The 1401 group had lunch together.

Tom sounded like he was making progress on his German project.

Mark found a lot of new things about the tape clutches and seems about ready to install the rebuilt braking system into the tape drive.

Ken and I worked on the CT # 3 729 that had the head down problem. It still will not dump tape in the right side when attempting to load.


Others are "filling in the blank"
from Marc
Likely present, Marc Verdiell, Frank King, Ken Shirriff, Tom Szolyga, others??

I rebuilt the CT tape brake clutch: pressed new bearings in, put new felt seals in (hopefully correct size and correct place this time), refilled with 29g of special magic iron powder, tested that it worked at the required current, then reassembled the main shaft with all 3 clutches.

I know that Tom continued to work on his serial interface over these last two weeks. He is making it quite user friendly by adding a nice color LCD screen.

Frank tested the machines and started working on the CT punch, which Jack Ghiselli had reported defective.

Ken worked on trying to figure out why the CT tape next to mine gets stuck in its state machine logic when trying to load a tape, and made some diagnostic progress. It is the only Model II tape with NOR logic that we have, which means we can’t compare its devilish load logic it with another working one, which is very annoying. Ken tried to explain it to me, and it sounded very complicated.

In addition, Google Gobble, the local wild turkey, was waiting for me at the side door when I went to get my lunch. He looked annoyed and suspicious, and let out a loud gobble. Incidentally, lunch was a turkey sandwich, so he had good reasons to be annoyed and suspicious.

May 7, 2022
from Jack Ghiselli
Steve Madsen and I did the 1401 demo today to 24 people. In addition to USA, we had visitors from Dubai, Delhi India, and Bratislava Slovakia.

We used the CT system for most of the demo, but since its tapes don't work we used the DE TAU and the one working DE tape drive. All three keypunches and the 083 sorter worked fine.

Still hoping for a working 1402 punch when the Restoration Team gets time.

Weekly 1401 demos on Wednesdays begin 5/11.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

Apr 30, 2022 - demo
from
Jack Ghiselli and Pat Bruder
from Jack Ghiselli
Pat Buder and I gave the 1401 Demo on Saturday 4/30/2022. We had 28 visitors for the main demo and about 10 who wandered in later. In addition to the USA, we had visitors from Rio de Janeiro Brazil, Mexico City Mexico, and Warsaw Poland. We're trying to shorten the demo, and we got this one completed in under 45 minutes. Yay.

We used CT for the demo, except we demonstrated tapes on the one working tape drive on DE, which was fine. During demo prep, while using the TAU to space the tape forward, dummy me let it run off the end of the tape. Pat Buder was kind and did NOT laugh and take pictures of me clumsily performing the exercise of re-threading the tape back on to the feed reel. We had a card jam inside the CT 1402 Reader, but after we removed the brush assembly and cleared it the reader worked fine. All three keypunches and the 083 Sorter worked fine. The extra weight bar had come off one of the 1402 reader weights, so we cleaned off the goo and re-attached it with some new double-stick tape.

Our demo cart seems to be down to 5 SMS cards (there used to be 7) and 2 vials of cores (there used to be 3). That's still enough for the demo, but we'll try to keep track better.

We had a little extra time, so we took some Windows screen shots on the PC attached to the CT system, and I will fit these into the PCWriter operator's instructions I'm working on for Stan.q

Thanks again to the Restoration Team for all your hard work. When your busy time permits, we're still hoping for a correctly aligned 1402 Punch so we can replace up some of our bad demo decks.

We're hoping to add an additional weekly 1401 Demo on Wednesdays at 3:00, starting 5/11. We need to encourage some more 1401 demo volunteers to cover two a week.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051
from Pat Bruder
Last Saturday I was asked to do a special Revolution tour for a group of three blind visitors. It was one woman and two men, all in their late 20's. At least one of them had done some coding. As I often do, I tacked on a 1401 mini-demo at the end.

I did a similar demo several years ago, but that was for a single person and his attendant. Also, another docent did the tour and he and I did the demo. We both found it very rewarding.

Before this tour I did some research on leading tours for blind people and got some great tips.

Revolution Tour

The CHM staff gave me gloves for everyone so that we could touch selected items. Unfortunately Revolution isn't much of a hands-on experience, but I planned some places that they could touch.

One challenge was moving through some of our narrow, twisted pathways. They suggested we form a line with each person touching the one in front of them on the shoulder or arm. Sometimes they also used their blind canes to feel the way.

Because of the additional time it takes to describe everything I drastically reduced the number of stops we made.

IBM 1401 Demo

It turns out that there are many things to touch and hear in the 1401 lab:
   Cards
   Key Punch
   Sorter
   1401 Cabinet and Panel
   Card reader
   Tape and tape drive
   Large cable
   Circuit cards

All of these were guided touching, and none were in motion or had anything exposed that would be dangerous.

All three of the guests can type, so keypunching was feasible. However, our vintage 026's don't have the raised dots on the "F" and "J" home keys found on modern keyboards. I had to help them position their hands, but they managed.

Lessons Learned / Opinions

Group size is important. One or two is best, and never more than three per docent.

It's best when the museum is closed, but we worked around the public.

Practice this special docent activity in advance. Carefully select each stop you will make. Think about how you will describe things for this audience.

Do research, especially seeking articles about museum tours written by blind people. An example of something I discovered before the tour is that you need not go out of your way to avoid terms that refer to seeing. These are prevalent in everyday conversation, and blind people are used to hearing them. Avoiding them is unnecessary.

Ask them if they want some particular assistance, and they will tell you how to help them if needed.

Conclusion

At the end of the day, they asked me to guide them to the entrance walkway by the parking lot so they could wait for the Uber they ordered. As before, I had a very pleasant and rewarding experience.

Pat


Apr 27, 2022
from
Stan Paddock and Marc Verdiell
from Stan Paddock,
Robert Garner 1.5
Frank King 6.0
Stan Paddock 6.0
Tom Szolyga 6.0
Ken Shirriff 6.0
Marc Verdiell 6.0
Allen Palmer 6.0


German System:
All three tape drives fail to work.
Allen and Frank checked the TAU,
Other than the tape system does not respond, no errors show on the TAU control panel.
(To be continued)

Connecticut System:

#1: the right brake clutch locks up.
Marc has pulled the brake unit.
Somebody, in the past, has worked on the brake and did not put it back correct.
A felt washer was left out allowing the break power to get into a bearing.
He says the felt seals are worn and he will replace them.

#2: Continuous errors for a few weeks.
Before bit 4 seemed to be bad.
Now I see problems with bits A, B, C. I.e. compare errors when writing from the TAU.
Frank and Ken measured the signals at the preamps and they looked fine.
Frank thinks it could be a cabling/termination problem since we've been moving cables around while working on other drives.
Ken and Frank still working on the problem.

#3: It still as problems loading, so Marc took the head assembly and fixed some broken parts. Frank put it back in and the heads move down and the tape starts to load, but the right reel doesn't work right and the load doesn't work.
Frank and Marc determined the head down switch is working as designed.
But the tape will still not load.

#4: The motors keep running and it doesn't load right.

We have a staffing problem so this problem has not been addressed.

Unit Card Equipment:

#2 keypunch ingested a punch card and it was difficult to recover.
Skilled unit record technician Frank King was able to fix the problem.

Stan Paddock
from Marc Verdiell

And big thanks to Allen that told me about the new felt seals that had been remade by Grant, how to remove the bearing by pressing it out, how to replace said unobtainium bearing with two obtainium ones, how many grams of magnetic powder to put back, etc, etc…
Marc

Apr 20, 2022 etc.
from
Stan Paddock, Jack Ghiselli, Marc Verdiell & Saturday Docent Report
from Stan Paddock,
Robert Garner 3.0
Frank King 6.0
Stan Paddock 6.0
Ken Shirriff 5.0
Tom Szolyga 6.0
Marc Verdiell 5.0


German System:
Did not address the whole German system on the belief that a week of rest may make it work better.

Connecticut System:

The Connecticut system has four tape drives: #1,#2,#3, and #4.
(Hi Stan, you wanted an update on the CT tape drives)

#1: the right brake clutch locks up.
Marc removed it.
He says the felt seals are worn and he will replace them.

#2: Continuous errors for a few weeks.
Before bit 4 seemed to be bad.
Now I see problems with bits A, B, C. I.e. compare errors when writing from the TAU.
Frank and I measured the signals at the preamps and they looked fine.
Frank thinks it could be a cabling/termination problem since we've been moving cables around while working on other drives.
I noticed bulb A was burnt out on the TAU panel, so I replaced it.

#3: It had problems loading, so Marc took the head assembly and fixed some broken parts. Frank put it back in and the heads move down and the tape starts to load, but the right reel doesn't work right and the load doesn't work.
The head load motor keeps running, so we suspect the head down micro switch isn't closing.
We suspect the mechanism that drives the head up and down is not synchronized when it was installed.
Next week.

#4: The motors keep running and it doesn't load right.
We are gaining experience on the tape drives but making progress every week.

Stan and Tom worked on the German to PC system.
The Connecticut to PC system communicated via the printer port on the PC.
The German to PC system will be communicating via USB.
The software in the PC in the German PC will be 70% of the Connecticut PC.

Stan Paddock
from Jack Ghiselli Fri 4/22/2022 11:40 AM

Re: Status of CT1402 Card Punch
I came in last Saturday to try again to punch an object deck from ROPE. Apparently, the CT 1402 Card Punch is punching cards out of alignment. Sometimes we blame this on poor quality blank card stock, so I tried several different types of blank cards, including old (probably good) stock pre-printed with various company information. Same problems.

Attachment #1

Attachment #2

Some of the mis-punching was really wild, and usually got 1402 Punch Checks. See Attachment 1. But most mis-punches were much more subtle and you had to use the punch alignment fixture to see the errors (see Attachment #2).

Interestingly, some of the mis-punched cards read OK sometimes on the 1402 reader, and sometimes not. Also, the 1402 reader seems much more forgiving of slight vertical errors than the 029 keypunch, which drops vertically misaligned zone punches when you try to duplicate a card. Probably this is because the 1402 reads cards from bottom to top on the long edge, while the 029 reads them side to side on the short edge.

Anyway, this mis-punching might be part of the cause of the flaky loading of demo decks we've recently punched. In the words of Native American actor Chief Dan George, "Sometimes the magic works and sometimes it doesn't."

Since the DE 1402 Punch is still down, card punching seems to be unreliable. If our overworked Restoration Team gets some free time, it would be great if you could look at this.

Incidentally, this does NOT affect the 1401 demo, since we don't use the 1402 Punch.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051
from Marc Verdiell Fri 4/22/2022 7:43 PM
Re: Status of CT1402 Card Punch

Good thing is, while working on the German punch, I got pretty good at doing the mechanical alignment. When I’m done with the 729 clutch I can realign the punch.

Marc
Saturday Docent Report, from Jack Ghiselli

Bill Worthington and I gave the 1401 demo today to 45 visitors. Foreign countries represented were Australia, France (Paris), and India.

We ran the demo on CT. Couldn't get any of the CT 729's to work with the TAU, so we used the DE TAU and 729's to demonstrate tapes.

CT tapes are still being worked on by Restoration.

1402 punches on DE is still down and the one on CT is misaligning.

026 Keypunch #2 is unusable. The program wheel is stuck oscillating and it won't feed or punch.

--Jack Ghiselli

Apr 13, 2022
from Stan Paddock and Marc Verdiell (and later from Frank King)
from Stan Paddock
Robert Garner 1.0
Frank King 5.0
Stan Paddock 5.0
Ken Shirriff 5.0
Tom Szolyga 5.0
Marc Verdiell 5.0

German System:
Determined the punch is still out of order.
We will have to determine when it will have to move it up on the priority list or it will not get repaired.

Also tested the tape drives.
Some worked some of the times.
Not available for usage for demos.

Connecticut System:

Marc continued his work on the first Connecticut tape drive.
He finished assembling the tape drive and tested it.
It started out running as design.
Then the right spindle started to jam even with the power off.
This is the same problem as before.
With all of the parts on the table in the workshop two weeks ago, we could not find any error and we resembled the spindle.
Next week we will remove the spindle and break in down again.
Then we will take the brake clutch and see what is wrong with it.
If we can not be sure what is causing the problem, we have a spare brake unit that we can install in place of the failed unit.

Frank noticed that the read/write head on the third tape was stuck in the down position.
Frank removed the belt that drives the head up and down.
With Frank, Marc and Ken working the problem, it was discovered that there was a slice of metal from the erase head that had followed on the read/write head.
This kept the head from going complete to the down position.
It was also noticed, there were two small screws that had broken off in the main block.
This was a further problem as both of them were broken off below the surface.

Marc has a well equipped shop at home.
He took the read/write head assembly home to see what he could do with it.
I will send along another e-mail from Marc specifying how he fixed the problems.

With luck, we might have both drives operational next week.

Stan Paddock
from Marc Verdiell

Subject: IBM 729 tape head repaired

I think the tape head is repaired. I extracted the two screws. I made a locating punch in the center of the screw stub, drilled a small hole in the middle, then used the screw extractor, both came (almost) right out.

The erase head plate did fit very snugly in its former footprint, so I just put fresh liquid cyanoacrylate and it locked back right in there. Also the erase head itself has a locating peg, so when I screwed it back on, it went right back to where it was. It should have retained the original alignment, which seems non critical anyhow, it’s pretty far from the tape. Ken believes it’s not used in our tape models anyhow.

I think we should be all good to go and can reuse this head if we wanted to. Plus it’s all cleaned up now.


Screws extracted:

Holes nice and clear

Erase head with the missing plate

Erase head with the plate back on:

Went back into the perfect place:

It’s all fixed (I think):
Marc


from Frank King in response to question by Jack Ghiselli

We ran programs from both systems. Everything worked with the exception of the tapes and the German punch. Only 1 of the tapes works on CT so, I would suggest for demoing the tapes, use the TAU panel. All 3 work on the German machine from the TAU using go forward and go backward.

Apr 9, 2022
from Jack Ghiselli
We demonstrated the CHM IBM 1401 on Saturday 4/19/2022 to 27 people, which was a more manageable size, and close to the size suggested by CHM staff. After the main demo, another 28 people came by and we gave mini-demos to them. I was Lead and Bill Worthington was Assistant. Our demo ran 50 minutes, which is a little long, and I'm working to shorten it up.

One of our visitors worked in 1957 with Honeywell on early magnetic tape drives. In addition to visitors from the USA, we had visitors from Germany and Italy, pushing our language skills.

The CT 1402 Card Reader/Punch was inoperative. There were cards left in the reader throat, the READER STOP light was on, and we were unable to perform a Non-Process Run-out on either the Reader or the Punch. So, we used DE. On DE, one 729 Tape Drive was marked as Out of Operation, so we used the right-most two drives. Using the Tape Exerciser (Ron 2.1) program, we were able to run the right-most tape, but the next to the left only did one write cycle and then stopped. So, we ran the demo using the TAU, and both drives worked OK. We didn't have time to diagnose the tape issue further. We ran BigPrint and Powers of Two for the demo, and both ran OK. All three 026 Keypunches and the 083 Sorter worked OK. Our sort deck is getting a little ratty, but with both 1402 Punch units down, we couldn't punch a new one. We found the Tape Exerciser (Ron 2.1) deck in the Demo Tray was missing its last 3 cards, so we duplicated new ones from another deck on the Keypunch.

Mar 30, 2022
from Stan Paddock
Attending were
X Robert Garner 1.0
X Frank King 5.0
X Stan Paddock 5.0
X Ken Shirriff 3.0
X Tom Szolyga 5.0
X Ed Thelen 3.0
X Marc Verdiell 5.0

German System:
(Paid no attention to this system today)

Photo Ops:
It has been a long time since we had seven 1401 people together at the CHM.
I thought it was time for some video evidence. Here

Connecticut System:

Marc continued his work on the first Connecticut tape drive.
Marc had rebuilt one of the brush block at home and installed it in the shop today.
He is now completing the installing the other parts.
The tape drive should be back on line in the next couple of weeks.

Frank notice the right take-up tape reel was not working right.

Using his many years of experience, he suspected is was a brush problem as we found on the first drive.
Frank adjusted the brush block and the problem went away.
Next week when all four of the Connecticut tape drives should be operational, we will check for errors on all drives.

The new keypunch ribbon being tested on keypunch #1 failed as we expected.
I am going to send them back.
I am going try and build a device that will re-ink our dry existing ribbons.

Docent demo cards:

On the disk of the pc that talks to the Connecticut system, Stan found a punch card file for BigPrint and powers of two.
Stan punched the two decks.
While these two decks are not the current revision, they both work fine.
They are on the punch joggle plate on the1402.

I expect to be able to be able to punch all of the current demo decks next week.

Visitor

We had a visitor to the CHM last week.
His name is:
Daniel Van Dalsem
(669) 264-7553
DanDalsem@UCSB.edu

Dan is a college student from Santa Barbra.
Depending on what he thought of our meeting last week,
he might join us during summer break.
His parents live in Santa Clara.

Tom spent the day working in the shop developing his PC to the German system.
This will be functionally the system we currently use on the Connecticut system.

Stan Paddock

Some relaxed pictures taken Wednesday, March 30, 2022
     - Music, stage play, and photos by Sandra Shu-Lee of CHM
     - Lyrics by Ed Thelen
     -
Rap sheets of below suspects
     - "Is there anyone here that I haven't offended yet?"

The Line-Up. left to right, Robert Garner, Marc Verdiell, Tom Szolyga, Frank King, Stan Paddock, Ed Thelen, Ken Shirriff

Another angle

Ed with screw driver. "I might break something !!"

Ah, Ken will keep Ed out of trouble! a chaperone !

OK, somebody do something.

Frank at work. What more can I say??

This is Stan. Back on the farm, we would call him the "straw boss".

Guilty as charged !! Lined up for sentencing..
     - more pictures taken a few years ago

Mar 26, 2022
Demo Report and VOBJ Instructions, from Jack Ghiselli
We gave the 1401 demo today. I was Lead and Pat Buder was Assistant.

Excellent Handoff:       I want to compliment the Restoration Team for an excellent handoff after your work last Wednesday. Stan and whoever else was there (sorry, I don't know who) made brand-new crisp copies of BigPrint and Powers-Of-Two, on good-quality card stock. They also left a supply of good-quality blank cards in the CT 1402. They also left copies of single-card 80/80 list and 80/80 punch programs, plus a brand-new copy of "Server" which we needed to use for punching ROPE object decks. Everything was well labelled and placed right on top of the CT 1402. You guys rock!

Hardware Issues:       We used CT. It looked like DE would also have worked, but we wanted to punch some cards and DE's 1402 Punch is still INOP. Immediately after power-up, decks wouldn't load on the CT 1402 Reader. However, after a short 15-minute warmup, it worked perfectly. We had an initial problem getting the 1403 printer to work, until we noticed that the cover over the Carriage-Control Tape was open. We closed it and the 1403 worked perfectly. The 083 Sorter and all three 026 Keypunches worked fine. Again, many thanks to the Restoration Team.

Safety Issues:       It appears that the fire extinguisher is missing from the wall cabinet next to the A/V closet. Is this OK with CHM management?

Demo:       We had lots of visitors. We had 66 people for our main demo, plus 7 more who wandered in during the demo. This exceeds the COVID guidelines Poppy had given us, but we were too busy to guard the door. If we do limit attendance, we should consider running multiple demo sessions. We also had about 20 people who wandered in after the demo, so we gave snap mini-demos to them. We had visitors from Michigan and New York, and foreign visitors from Poland and South Korea. One visitor, Duane Sand, was an ex-Burroughs guy who was possibly interested in joining the CHM volunteers. I think he talked to Poppy also, so I hope she follows up. Due to poor planning on my part the demo ran a little long -- about 45 minutes, so I'll try to tighten it up in the future.

Follow-up on Bad Decks:       For the past several weeks, we've had problems with demo decks that won't run. Today we used the new Demo decks built by Restoration last Wednesday, and they go a long way to fixing the problem. To help further, I wrote a program VOBJ (Verify Object Decks), intended to help us in the future verify that our Demo object decks haven't been shuffled by poor card-handing. I developed it under ROPE, and brought a thumb drive of the object deck. Then we used the PC and Stan's "Server" program on the 1401, to punch VOBJ into cards on the CT 1402. Thanks, Stan, for excellent documentation. We put the VOBJ program deck in the Demo tray. We assumed the new BigPrint, Powers-Of-Two, Server, and 80/80 decks were intended for us, so we put them too into the Demo card tray. If they weren't intended for us, you can find them there. Operating instructions for VOBJ are attached and I'll work with Ed Thelen to get VOBJ put into his data base.
Data Base with instructions

Staffing:       Pat Buder and I have been doing the 1401 demos the past few weeks. However, both of us will be out of town next Saturday (4/2), and as of now nobody has signed up. Perhaps we won't have a 1401 demo next Saturday?

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

Mar 23, 2022
from Stan Paddock
Attending were
X Robert Garner 1.0
X Frank King 5.0
X Stan Paddock 5.0
X Ken Shirriff 3.0
X Tom Szolyga ??
Marc Verdiell 5.0

German System:
(Paid no attention to this system today)

Connecticut System:
While installing the brushes in the first tape, Marc noticed that one of the brushes was worn down to nothing.
In the past, Marc had built new brushes for the 729 tape drives.
Marc took the brush bracket home to install new brushes.
When Marc is able to come back to the CHM, he will complete the install of parts to the tape spool.

Frank powered up the number 2 tape drive.
A slight squeal had become anything but slight.
After searching for an hour, Frank found the source.
On the right reel shaft, there is a brake that was squealing.
What was interesting, not all of out tapes have this brake.
Frank checked the book as what to do to fix the problem.
Using his vast experience, Frank attacked the problem and the squeal is gone.

Frank said "I remember how to install new ribbons in a keypunch."
He installed the tape rather quickly.
He also removed the new ribbon and turned it over rather quickly.

We are running one of the new ribbon to insure it does not come apart like the last ones.
These are the same ones we ordered before.
We will try to find another way to make our old keypunch work again.

Docent demo cards:
On the disk of the pc that talks to the Connecticut system, Stan found a punch card file for BigPrint and powers of two.
Stan punched the two decks.
While these two decks are not the current revision, they both work fine.
They are on the punch joggle plate on the1402.

I expect to be able to be able to punch all of the current demo decks next week.

Visitor
We had a visitor to the CHM today.
I have misplaced the punch card that I had him to write his full name and other ID.
So all I can tell you is his name is Daniel [Van Dalsem] and he is a college student from Santa Barbra.
Depending on what he thought of today, He might join us during summer break.
His parents live in Santa Clara.

Tom spent the day working in the shop developing his PC to the German system.
This will be functionally the system we currently use on the Connecticut system.

Stan Paddock

Mar 20, 2022, & 22nd
from Jack Ghiselli
I came in today (Sunday 3/20/2022) to investigate why some of our BigPrint decks are failing to run.

In the tray on the demo cart, we have three copies of BigPrint. I've labeled these "Good", "Bad #1", and "Bad #2". As you might guess from these clever names, the first one works and the other two fail, but with different symptoms. By comparing 80/80 lists of the object decks, I found the following problems.

  1. In BigPrint "Bad #1", card #0041 (object decks from ROPE have sequence numbers in columns 72-75) has several zone punches missing in columns 1-39. In particular, column 1 is supposed to have an "M", but due to the dropped zone it erroneously has a "4". This changes the "move" instruction (M) at memory location 1513 into a "punch" instruction (4). During execution, the deck successfully reads the date and first name card, but then hangs at this instruction waiting for the 1402 Card Punch to be made ready. If you did make the punch ready, the program would get into further trouble attempting to execute following damaged instructions. The fix for this deck would be to replace card #0041 with a correct card copied from the good BigPrint deck. I speculate that this problem occurred when originally punching the deck on the 1402 Punch or when re-punching a damaged card on the 029 Keypunch

    Recommendation: Whenever you use the keypunch to replace a damaged object-deck card, always hold the old and new cards together up to a light and verify that all holes got copied correctly.

  2. In BigPrint "Bad #2", cards #0041 and #0042 are missing from the deck. During execution, after reading the date and first name card, the program branches to the missing code and fails with a CPU Check Reset error light at Instruction Address 1525 and Op-Code of Blank (Invalid op-code). The fix would be to insert new cards #0041 and #0042 copied from the good BigPrint deck. I speculate that this error was introduced when a card load failed part way through, and the operator failed to do a Non Process Runout on the 1402 Reader and probably the two cards were left in the rollers and mistakenly discarded.
    Recommendation: Whenever a 1402 Card Reader load fails, be sure you use Non Process Runout to clear any cards remaining in the 1402. If you're unsure, check sequence numbers in object deck columns 72-75 to be sure you get the entire deck back together correctly.
  3. All three BigPrint decks in our demo tray are revision "BIGPRINT REV 16.8 GENERATED ON 03/18/2016". You can find this in columns 1-39 of Card #0007 and columns 1-2 of Card #0008. Unfortunately, this latest and greatest assembly listing has not yet been put into Ed Thelan's database of demo programs. Doing that would make chasing this type of problem easier.

    This investigation suggests that these particular BigPrint problems we've been having are NOT hardware problems. Instead they are probably card-handling mistakes that damaged the decks. However, we have seen some other errors. One example was sporadically we saw a BigPrint load fail part way through with a CPU Check Reset light. This might be related to hardware, or might be other deck errors that I haven't noticed.

    I have NOT FIXED these problems, because I've been investigating at home. I am scheduled as one of the Demo team for next Saturday 3/26, and will try to fix the decks then.

  4. All three BigPrint decks in our demo tray contain three copies of Card #0098. Due to the way the loading works, this does not cause any execution problems, other than reading two unnecessary cards. I speculate that when punching the original deck from which these copies were cloned, a 1402 Punch error occurred and restart procedures punched the duplicate cards. While not strictly necessary, the fix would be to remove the second and third copies of Card #0098.
Recommendation: If you are punching object decks on the 1402 Punch and a punch error occurs, be sure you check sequence numbers to verify that no output cards are lost or duplicated.

We demo folks seem to have deck problems fairly often. When I get time, I'll try to write a diagnostic program to check demo object decks for missing cards, and/or to compare a suspected deck to a known good copy. That might help.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051


Robert Garner says:
"Jack,
"Your perseverance in sorting out BigPrint deck faults deserves a raise! :) "
from Jack Ghiselli, March 22, 2022

I came in today (Tuesday 3/22/2022), and thought I'd try to repair our two bad BigPrint demo decks in the 1401 Demo Tray.

Progress: The problem with BigPrint "Bad #1" was that card sequence # 0041 was missing some zone punches in columns 1-20. I thought the fix would simply be to duplicate a replacement card from the good (working) BigPrint deck. Ah, that life were so simple. I took Card #0041 from the good BigPrint and duplicated it in the 026 keypunch. To my surprise, the duplicated card also dropped the zone punches. I tried slowing down the process by duplicating one column at a time, but still bad. Tried a different keypunch; still bad. I put Card #0041 from the "good" BigPrint deck on the card-hole-alignment jig, and the hole alignment looks bad, especially in the first few columns, which are the ones the keypunch can't duplicate correctly. Interestingly, the "good" BigPrint deck does load and run successfully (I tried it on CT), so apparently the 1402 Reader is more forgiving of bad hole alignment than are the 026 keypunches. Perhaps some of our other sporadic loading problems were cases when the 1402 was less forgiving? I considered hand-punching a replacement card from the 80/80 listing, but several other cards also looked poorly aligned. I speculate that when the original BigPrint deck was punched, the CT 1402 Punch was slightly out of alignment. Then later someone duplicated a damaged card in the deck, but failed to notice the missing zones in the duplicated card.

Status: We seem to have one "good" BigPrint deck that does load and run successfully (at least on CT). So, we probably can run the 1401 Demo this coming Saturday. We also have two "bad" BigPrint decks which might be difficult to repair, due to poor hole alignment on the "good" deck.

Recommendation: Perhaps on Wednesday (3/23), if the Restoration Team has time, they could get the latest and greatest BigPrint ROPE object deck on a thumb drive, punch a new "master" copy of BigPrint, and verify that the 1402 is punching correct alignment. The demo folks can make extra copies from the "master". I talked to Stan Paddock on Sunday, and as usual he was way ahead of me on this approach. Hooray for the Restoration Team, who actually know how to fix things.

--Jack Ghiselli jghiselli@sbcglobal.net, Cell 408-839-1051

Mar 16, 2022
People who attended the CHM were:
Frank King 6.0
Stan Paddock 6.0
Ken Shirriff 5.0
Tom Szolyga 5.0
Marc Verdiell 5.0

Warning: Stan says he is tired after these maintenance sessions
and has given me (Ed Thelen) permission to "edit" his reports.
This is chancy !!
I'm too tired to even show up for work :--((

German System:
We used the German system to try and isolate the problems we had with were having with program decks.

Connecticut System:
The last demonstration had a problem with the decks.
We had some of the same problems on both systems.
Ken was looking into the problem.
He noticed that with one BigPrint deck failed at 20% of the deck.
The second deck failed at 60% of the deck.
By taking the first 50% of the second deck and the last 50% of the first deck, BigPrint worked!
Way to go Ken.

In discussion, we determined the problem.
A couple of weeks ago, the docents helped making sure we had three copies of all of the demo programs. After testing a demo program, they made copies of the program on the Connecticut system.

We missed it this morning but the I/O check switch was off on both machines.
This will allow the system to read bad cards into memory.
Since there is bad data into memory, the system will crash when the bad section is hit by the processes.

Suspecting the problem, we checked cards punched by the Connecticut and found all of them were off registration.
Ed Thelen popping off:
Off registration as determined by what ??
We should have a "card gauge" - an aluminum plate with guide projections and 80 x 12 black rectangles indicating where possible holes should be.
If you can see aluminum in the card holes, you know which way, and how much, the holes are misaligned.
This is by far the best method of determining mis-registration ...

We were using the cards we paid to have made by a company who had never made punched cards before.
Comment from Robert Garner
Ah, so it sounds like the TimeCardsUSA stock may be slipping in the CT 1402 punch.
Bummer, but not surprising— we know that their card stock is a tad slippery.
(Last stock paper fab shut down 20 years ago now..)

I concur that our demo (and other program) decks should only be punched in our authentic/period IBM 5082 stock.
(Nevertheless, the TimeCardsUSA cards seem to work as date/name cards for BigPrint.)

- Robert

p.s. From our punched card info page: http://ibm-1401.info/index.html#CardSource
"According to Tim, the last supplier of IBM and "US Gov't" spec card stock, which he called 100-lb tabulating index, was an International Paper mill in Louisiana, shutdown ~2008. He had supplied 20 - 40k punched cards at a shot to Bob Swartz of Cardamation, Pennsylvania, our last source of high quality/IBM-spec 5081 cards, whose business closed in 2009 after his death.

We have decided to take all of those cards and put them in the Audio room to be used for the IBM 001 punch.
Next week we will be punching all vintage cards for the docents.


Two weeks ago, Marc took the right hand spool and break assembly for the first drive of the Connecticut system. It was suspected that there was a bad bearing binding the shaft.
After Marc and Ken got the assembly all apart, they could not [ find? ] a bad bearing.
They oiled the bearings, today the re-assembled the whole shaft.
Marc was heard to say "After you have taken the whole shaft apart, it goes back together much faster.
He also said the 4 stove bolts and heavy tie wraps made it much [ easier? ] while installing the assembly.

While installing a brush block, Marc noticed one of the brushes was way too small. [ short? ]
Marc took the brushes home to install with brushes [ he ] had built in the past.
This short brush, more than likely, caused the bad performance on this drive.
Ed Thelen popping off:
Brushes just plain wear!! Are we doing "periodic maintenance"?
Then again, what is "periodic" in our usage ??

Tom Szolyga is continuing his work on his German machine to PC interface.

Stan Paddock

Mar 9, 2022 - removed -

Feb 16, 2022
From Stan Paddock - received March 13, 2022
Attending:
Frank King 5.0
Stan Paddock 5.0
Ken Shirriff 5.0
Marc Verdiell 5.0
German System:
(Paid no attention to this system today)

Connecticut System:
Marc and Ken found two books that tells you how to take apart and put back all of the parts of an 729 IBM tape drive.
So they took out the shaft from the first tape drive.
The took the shaft with the clutches into the work shop.
In the shop, they took all the items off the shaft.
Now came the hard part.
When the shaft, the shaft behaved as if one of the bearings did not like turning.

With the bearings (3) on the table top, all bearings seem to turn with no problem.
I am sure when Marc and Ken get together, they will figure it out.

Frank said I remember how to install new ribbons in a keypunch.
He installed the tape rather quickly.
He also removed the new ribbon and turned it over rather quickly.

We are running one of the new ribbon to insure it does not come apart like the last ones.

Stan worked on the PC-029 transcription system.

Stan Paddock
from Ken Shirriff - tape drive
Update from last Wednesday's tape drive effort. The pin holding the front hub onto the shaft won't come out, so we're kind of stuck. Any advice from the experts?

Marc and Frank tried to drive the pin out with a hammer and punch, but it wouldn't budge. (This should be the easy part of the disassembly...) Apparently whoever put the pin in used a lot of force and it doesn't want to come out. We're considering penetrating oil or heating it. Drilling it out would probably be difficult because it's hardened. Any other ideas?

Here's a photo. The pin is behind the hub, indicated by the arrow.

from Allen Palmer - tape drive
THE PIN IS TAPERED it should have a slight inset on the small end. You may need to look very close to see which end it is. Some time in the past someone may have driven it in so hard as to extend the pin out on both ends.

I would not think of either heating or drilling it. Can you see the recessed end of the pin ?

from Grant Saviers - tape drive
If the ends of the pin are mashed, a light grinding might help sort the large end from the small end. If you need a replacement pin, I have many, but some caliper quality diameter dimensions will be needed at a defined separation.

Usually, pins are extra long. A trial fit determines the proper length and position, marked and then cut off with extra on the large end. A dremel cut off wheel is perfect for the job.

from Paul Laughton - 1401 Microphone Issues
Mic 1 volume is very low. The master volume control must be turned up high to make Mic 1 heard. As a consequence, the input volume control on Mic 2 must be turned down. This works; however, the demonstrator currently not speaking needs to mute their mic to avoid feedback.

I have observed that there is some cross channel activity between Mic 1 and the video. What I see is that when the video is playing but its output volume is turned off, the input level lights on Mic 1 mimic the input level lights for the video.

Note: I have also sent this message to Jon Plutte

from Ignacio Menendez
As last resort, take your high speed cutter rotary saw, and come in front to back on the aluminum hub… VERY CAREFULLY, and take it down to the point that all you have left is an UNDAMAGED reel shaft…
The pin should be then easy to remove by cutting it shorter, again with the cutter, and tapping it out. NOTE: all the area must be well protected with rags, to prevent the filings to get into the rest of the tape drive head, columns, capstans, etc.
AS LAST RESORT, I believe we have plenty of hubs, to sacrifice that one.

Feb 9, 2022 - from Stan Paddock and Ken Shirriff, with comment from Grant Saviers, and Allen Palmer and Ignacio Menendez
from Stan Paddock

Attending were:
     Frank King 5
     Stan Paddock 5
     Ken Shirriff 4
     Pat Buder 5

German System:

The German system was not powered today. It still needs adjustment of the punch unit.

Connecticut System:

Pat Buder is continuing on making sure we have three copies of the demo programs.

Stan worked on the stacks of cards in the Leibert room. Found in excess of 30,000 un-punched cards.

There are several boxes of cards to be put away next week.

Frank and Ken worked of the first tape drive. It appears that the brake in the right take up reel is putting on the brakes even with the power turned off. Work on this tape drive will continue next week.

Last week two of the Connecticut tape drives were flakey. This week, three of the Connecticut tape drives were flakey.

Liebert Room:

In the Liebert room there is an PC tied to an 029 keypunch. This has worked to punch cards from files on the PC for 10 years.. It broke down several months ago.

Frank showed me a simpler way to connect the units together. Today, I made the final connection between the PC and the 029. Next week we may have the system up and running again.

Stan Paddock


from Ken Shirriff

Frank and I took a look at tape drive 1 on the CT machine today, with some help from Stan and Pat. It's failing to load the tape; the mechanisms move but the tape doesn't go down into the right column.

After some investigation, we found that the right reel was kind of stuck. With the release button held, it would move only a bit, rather than rotating freely. I turned off the power and it was still stuck, so it's not an electrical issue. Frank and I investigated more and it started rotating a bit, and then rotating more, and then eventually rotating freely. We hoped that it was fixed, but when we tried the tape drive again, the reel immediately wedged up again.

Our conclusion is that the brake clutch is at fault. Each shaft has three clutches: one to rotate forward, one for backward. , and one to stop motion. The forward and backward clutches work fine, but the brake clutch seems to be getting stuck. The clutches are magnetic, with iron/graphite powder that binds the clutch plates when the coil is energized. So it's unclear what would cause the clutch to stay engaged with no power applied. Maybe Iggy has some ideas?
Frank and I considered removing the clutch to examine it, but it is a complex process (see below) and we didn't have time today. This would be a good job for someone with more mechanical skills than me, e.g. Marc.

Ken

One thing I forgot to mention. Frank noticed a lot of debris on the bottom of tape drive #1. This appeared to be magnetic tape shredded into 1/16" bits; there was one larger fragment about 1/16" by 1/2" that was clearly tape. It's hard to figure out how this would get created. At first we were concerned that it was bits of metal or clutch powder, but that doesn't seem to be the case. I vacuumed it up.


Comment from Grant Saviers

The iron powder in the clutch has a proprietary additive (it's not graphite) that keeps the iron powder from sticking together. We sourced the replacement powder from a company in VT or NH, I'd have to search a bit to find the reference. If moisture got in or if it wore out then then replacing the powder might be the solution. Look at my report for the amount.

The other causes could be a failed original bearing (not available) or the make do replacement ball bearing pair that I found adequate for the application. Then there are the wool felt seals that hold the powder in the clutches, these do wear out. I made the punch to make these and left it for the shop when I moved.

Grant


from Allen Palmer (a day later)

Just some thoughts

Loss of powder would not cause the problem, just the opposite - the cultch would turn freely on the shaft but fail to engage to the shaft. Hard to picture the ‘powder’ getting so solid as the design of the internals are made to break up the powder.

From the symptoms you list I would suspect are physical like grant listed, most likely the bearing or the shaft has ‘rusted’ tight inside the reel.

You will need to pull the whole clutch assembly out. If it is the break clutch 9 if memory serves me still) you can pull off the break reel by itself and check it.

There are some ’tricks’ to supporting the rear bar & the pins that align it. Also replacing the brush pads can be a ‘pain’ getting them on & adjusted. Just takes time & patiences. Be very careful about hammering on the end of the shaft - you do not want to do any ‘mushrooming’ to the shaft.

Allen


comment from Ignacio Menendez

Just to complement Allens comments, there are 4 long ‘stove bolts, that you insert in place of the 4 bolts securing the rear heavy casting, and as you are pulling the rear casting back, don’t forget to hold the shafts up (I use old coat hanger wire), to keep the weight of all the components from bending the shafts.
Just my $0.02 worth.
Iggy
P.S. these 4 stove bolts are inside one of the two large red tool boxes, in one of the drawers.

Feb 2, 2022 - from Stan Paddock
Attending were:
Frank King - 5.5
Stan Paddock - 5.5
Pat Buder -5.5

German System:
We powered up the German machine today because it was 50 degrees when we came in today

Connecticut System:

Pat Buder took over the work on making sure we have three copies of the demo programs.

Stan worked on the stacks of cards in the Leibert room.
Found in excess of 30,000 un-punched cards.

I think there are several boxes of cards to be put away next week.

Frank has worked on the IBM 083 sorter.
It is still a little flaky.

Last week two of the Connecticut tape drives were flakey.
This week, three of the Connecticut tape drives were flakey.

Stan Paddock

Jan 26, 2022 from Stan Paddock
Attending were:
Frank King - 5.5
Stan Paddock - 7.5

German System:
Did not power it up today.

Connecticut System:
We brought out a new ribbon to the printer and installed it.

Stan Worked on the demo programs.
The demo team was in and Pat Butler asked for three copies of each be made.


One copy for a master deck accessed by the hardware team.
One copy for the docents to use their master during demonstrations
One copy for the docents to use as a backup of their master deck.

Next week, the plans are to punch all the demo programs (primary and back up copies.)

We could use some help from people who can read IBM 729 technical books.
Is not a complicated as Apollo devices but still fun to work with.

Stan Paddock

Jan 19, 2022 from Stan Paddock
Attending were:
Frank King - 7.5
Stan Paddock - 7.5

German System:
Tried to test the tape drives.
Tried the tape test software.
Did not work.
Tried to test the drives with the TAU and all three drives will not respond.
We marked the tapes as down and went to lunch.

After lunch, we thought we would give it another try.
Frank re-loaded the tape test program and all three tapes came to life..

Connecticut System:
Frank used the tape test on all four tape drives.
The first tape has a problem that needs to be addressed.
The second has a prolay stutter that needs to be addressed.
The third and fourth drives seem to be fine.
We brought out a new ribbon to the printer.
We will replace it next week.
The knob the select the number of lines per inch needs to be tighten.
This will be fixed next week.

Stan found several copies of demonstration programs for the docents to use.
Next week, the plans are to punch all the demo programs (primary and back up copies.)

We could use some help from people who can read IBM 729 technical books.
Is not a complicated as Apollo devices but still fun to work with.

George Ahearn's son-in-law Alex stopped by today.
He dropped off a box of blank cards, some pictures of a 1401 work shop and a computer book.
Stan Paddock

Jan 12, 2022 from Stan Paddock
Attending were:
Stan Paddock

I knew Frank and Tom were not coming in today but I did not know if anybody else would be coming in.
Nobody else came in.

Following our safety rules, The only machine I turned on was a PC.

Last week I got the old PC <> 1401 operational.
I can punch cards from the PC!!

I am now assembling all of the demo programs and will punch two decks each.
One deck will given to the docents to use for demos and the other deck will be stored away as master.
If one of the docent decks fails, the docents will let one of 1401 team know and we will punch a new docent deck for them.

In searching for something in the Leibert room, I found we have at least two years NOS blank punch cards in there.

German Printer:
Last week Frank King, loaded the German printer with a lot of hydraulic oil.
I placed three Pig Blankets under the hydraulic of the printer.
I checked for oil leakage today. None!

It must only when the printer is running.

Jan 5, 2021 from Stan Paddock
Attending were:
Frank King - 7.5 hr.
Stan Paddock - 7 hr.

German System
We have known that the IBM 1403 printer was leaking oil. We have noticed that the leaking has stopped.
(The lack of an oil leak is because the oil level had dropped below where it leak spot)
Frank brought out the big can of oil and filled the reservoir.

Connecticut System:
Frank checked the oil level in the IBM 1403 printer and found the oil level is correct. Frank started going through the demo programs for both systems.
We found two copies of BigPrint.
One works fine and the other does not.

Connecticut to PC connection:
Stan re-assembled the cable from the PC to 1401 Serial Port, and attached them together.
Since it has been so long since the PC was used, it forgot the date and time.
Had to run a Setup program to reset all of the BIOS parameters.

Demo Card Decks:
Stan found an old control program copy for the 1401.
Stan loaded both programs into the PC and 1401 and ran some tests.
The ability to punch cards works fine.

Next week, the plans are to punch all the demo programs (primary and back up copies.)
Ed Thelen here, we should draw a diagonal line on the tops of all card decks,
to help identify when a deck not complete or shuffled.
I also used to label the first and last card to help assure a complete deck.
IBM 083 Card Sorter:
The sorter showed an capability to jam the cards and put the cards in the wrong cards in the wrong bit. Frank talked to the sorter every time it failed.
Frank made adjusted every time it failed.
When we left the CHM today. the sorted had learned its lessons.

Stan Paddock

Send comments etc. to Ed@Ed-Thelen.org