Repairing a Sorensen XPL 30-2D

[suicidaleggroll]

Hopefully some of the more knowledgeable people here can help me.  I do have a BSEE, but my experience is mostly in programming (high level and embedded) and digital circuit design, I'm of pretty much no use when it comes to analog.

So we have had this Sorensen XPL 30-2D at work for about 4.5 years now, and it's been great.  A few weeks ago though it started acting up.  Given the sensitivity of the equipment it powers and the method in which it's failing, I managed to convince the higher-ups to purchase a replacement supply (Keithley 2220-30-1) and write the Sorensen off (aka: give it to me so I can try to fix it and use it for personal projects at home).

For those who don't know, it's a dual channel, 0-30V, 0-2A, analog control, digital display power supply.  Note: the digital display shows the voltage set point, not the actual output voltage (when in CV mode).

Channel 1 is fine, channel 2 is the one acting up.  The coarse voltage control knob works correctly, the problem is the fine voltage control knob.  When powered on, the coarse voltage control works normally, while the fine voltage control knob has no effect from min to about 80% of max.  Beyond 80% of max the set voltage according to the display skyrockets to beyond 60V (it's only a 30V supply), BUT the ACTUAL output voltage drops down to 5-10v.  So there's a major problem with the set voltage (at least according to the display) and there's a major disconnect between the set voltage and the actual output voltage when the fine voltage control knob is above ~80%.

I initially thought it might be a problem with the pot used on the fine control, but unfortunately that's not the case.  The label on the pot show "Alpha 10k".  In-circuit measurements of the channel 1 (the one that's working) fine voltage control pot show it at ~6 ohms at min, rising to about 3k around 60%, and dropping to about 1.5k at 100%.  In-circuit measurements of the channel 2 fine voltage control pot show it at ~3 ohms at min, rising to about 60 ohms almost immediately, and staying within about 58-62 ohms all the way to 100%.  I then pulled the channel 2 fine voltage control pot off the board and measured it, and I get a nice linear rise from 0-10k.  So the problem is not the pot itself, it's the circuitry surrounding the pot, and this is where I'm at a loss.

Pics are as follows.  First, the main display/interface board in its entirety.  You can see the six control pots in the middle (the 5th from the left is missing, that's the fine voltage control pot for channel #2).  The order of the pots goes: 1) coarse voltage ch1, 2) fine voltage ch1, 3) current ch1, 4) coarse voltage ch2, 5) fine voltage ch2, 6) current ch2
http://thesuicidaleggroll.com/sorensen_xpl_30-2d/IMG_1635.JPG

Next, a closeup of channel #2, with the fine voltage control pot missing since I had already pulled it off.
http://thesuicidaleggroll.com/sorensen_xpl_30-2d/IMG_1636.JPG

Finally, a closeup of the back side of channel #2 to see the traces.
http://thesuicidaleggroll.com/sorensen_xpl_30-2d/IMG_1637.JPG

[SeanB]

Seeing as the connectors for each side are the same, and they likely are isolated have you tried swapping them, to see if the problem is the board itself or an external component, like a failing capacitor or a pass transistor that is faulty?

[suicidaleggroll]

So I tacked the pot back onto the board, and then...

I tried the ch1 controls on the ch2...rest of the circuit, and everything was fine, loaded or unloaded
I tried the ch2 controls on the ch1 rest of the circuit, and everything was fine, loaded or unloaded
I tried the ch2 controls on the ch2 rest of the circuit, and everything was fine, loaded or unloaded
I reassembled the entire unit and tried both ch1 and ch2 simultaneously and everything was fine, loaded or unloaded
I finished soldering the pot (since it was just tacked before), and again everything was fine

So, I can't reproduce the problem now, and the only thing I've done is disassemble the supply, remove the ch2 fine voltage control pot, solder it back into place, and reassemble the supply.

Maybe it was a bad solder job on the pot?  It looked good to me, nice clean joints with good flow.  That's the only thing I can think of though, since it's the only thing I changed.

[megajocke]

Is the back side of the pot made of metal? Could it have been shorting to those vias under it on the board?