return to main page

Activity Report - October 2, 2004,

by Robert Garner

On Saturday, 10/2, Bill Flora & I compared these schematics against the Deutsche 1402 (DE 1402) and the Visible Storage 1402 (VS 1402). It was nice having them sitting side-by-side!

On first inspection, the DE 1402 wiring appears to match these schematics, with the following six caveats:

  1. The DE 1402's original remoteControl+manual "Circuit Breaker I" was replaced with a combined 3-phase line filter + remoteControl "Contactor I," while eliminating "Circuit Breaker II" for the 729 tape drives, according to the schematic below.
    (I wonder if there are circuit breakers in the 729's?)
    The contactor's remote control logically would be from the on/off button on the 1401 main control panel. See pics below of both 1402 AC power input setups. The AC low-pass line filter must have been a field mod for noisy power environments. It's certainly not in the VS 1402. (The VS 1402 circuit breakers seem to be manual on/off only - I don't see their remote control connection.)

  2. The DE 1402 schematics specify 10-gauge power lines. The undersized, non-standard power cord that came with it seems to be something like 14 gauge (although I didn't measure it).

    A 10-gauge wire is good for 33A (enclosed air space), so three, in-phase lines could safely carry 3*33A*220V = 22 kVA, more than adequate. 3-phase 14-gauge wiring would be marginal at 3*17A*220V = 11 kVA, esp. with power factors, etc..

  3. A worrisome issue, as 1st noted by Don, is that the DE 1402's ceramic AC wiring assembly is cracked at its upper left and upper right corners, at both ends of the neutral return bus bar, causing the metal posts on the strip to peel inward; and it's plastic cover is melted, most certainly due to high temps.
    The picture is below.

    (The VS 1402 does not have a neutral bus bar, as it is 4-wire Delta connected, without a return neutral.)

    I've been postulating as to what may have caused this over heating. (The neutral bus bar metal strip itself was clearly oversized by design.

    And they must have expected high currents there to, or else why the protective plastic cover? Perhaps the bus bar/ceramic may have been over heated by one or more of the neutral wires connecting to it?) Here are the possibilities I've come up with (most-to-least likely):

    1. One of the 3 phases in one (or more) motors is disconnected or has a poor connection? This unbalances the 3-phases, resulting in the other two, non-canceling return current phases flowing into the neutral line.

    2. The neutral post screws came loose over time, causing overheating?

    3. A misbalance in the 3-phases, causing too much return current on the neutral line? This might be the case if most of the power were on one of the phases. (From the wiring schematics, I noticed that the 1403 DC power, 729 bulk, & 1401 blowers get their power from Phases B & C, whereas Phase A supplies connector PWB-13/14 as "system power".

      All the motors are all 3-phase.)

    4. The 3-phase, undersized, 14-gauge wire is resistive, forcing additional current to return via the neutral?

    5. 3rd harmonics on the neutral due to non-linear supplies, causing overheating?

    I'm not sure how to best approach figuring out the problem. In any case, I'll purchase a 50-500V "Megger" to test for wire-to-wire and wire-to-ground leakage faults. We'll test paths not only in the 1402, but also in the 1401 and 729's.

  4. It was interesting to compare the VS 1402 to the DE 1402 power cords. (We do not seem to have schematics for the VS1402 - I'd like to find some.). The VS 1402 is a 3-phase, 4-wire Delta power connection (3 current phases plus earth ground) with a large gauge wire (perhaps 8), and the DE 1402 is a 3-phase, 5-wire Wye connection (i.e., 3 current phases plus neutral current return line and earth ground wires) with a little smaller 10-gauge wire. (I've read Wye is/was favored in Europe and Delta is/was favored in US.)

    I do not think we can use the VS 1402's big, 4-wire delta cable, as we should not tie neutral and ground together at the 1402. That would unsafely place neutral return currents on the chassis of the entire system, and is not to code.

    Jim Yee can get us a new 5-wire, 10-gauge cable for connection to the Elgar.

    And we'll need the proper 5-wire Russellstoll receptacle/female plug.

    (I would prefer a Russellstoll screw-on to a twist-type, given that the connector is situated out in the room where someone can easily bump it.)

    Standard charts show 30A, 5-wire, 277Y/480V, which would handle our 220Y/380V config.

    The Elgar output bus enclosure has a neutral wire inside it (i.e., it's 5-wire).

  5. The VS 1402 also has a hefty copper bar as an earth ground point, and the DE 1402 had a poor, loose ground screw connection to the frame.

    (Pictures in next msg.)

  6. We figured out what the large missing item that the fan cools in the left front of the DE 1402 is: It's the +-3V @ 5A power supply for margin checking.

    (I wonder if it was packaged separately in another box, or is gone? Arnold?)

In summary, I think we next need to "Megg-out" the wires, understand the cause of the over heating, and get a 5-wire cable and receptacle for the Elgar.

- Robert

Below are 3 scanned pages of the DE 1402 power wiring, which powers the entire 1401 system.

Here are the schematics:

1402 Power Line Filter

1402 Power Connections

1402 Power Wiring
DE1402ACWiring - Here the picture of the DE 1402's AC power input to line filter block and cracked ceramic block. The upper 5 posts are connected together as the neutral return bar and the 3 lower left posts are the 3 phases. (the right crack is hard to see because of bright flash. The left neutral post is clearly bent inward from cracking). The undersized ~14-gauge white neutral wire is on the right.
Clearer images are at ActivityReport-Sept-22.html
VS1402ACWiring -