Troubleshooting a Bench Power Supply Stuck in Current Limiting Mode

[jbenward]

Hi all,

Long time lurker, first time poster.  I've been flirting with the idea of building a little hobby electronics lab for a while now and I've finally got the space for it after moving into a new place.  What spurred me to finally register and post on here is purchasing my first bench supply.  I got what seemed (at the time) like a good deal- $70 (+ $30 shipping) for a LIP 40V/1.5A dual linear tracking power supply (LIP Elec. Eng LTD Model LPS-401D).

So it arrived today.  I got it hooked up and turned it on. Channel "B" works as it should... channel "A" is a different story.  It immediately current limits, even with no load.  Here is a video of me doing the unthinkable... turning it on before taking it apart.

https://youtu.be/mmNxnKIk9w4

What I'm looking to you, wonderful EEVbloggers, for is for a bit of assistance/advice on troubleshooting/fixing this power supply.  I'm no EE, but I've designed and built a few synthesizer modules over the years, so I'm not a complete newbie.  So let me clue you into what I've found out so far...

After powering up the unit, it starts up in standby, I set the voltage to 18V and the current to 1.0A.  Turning it off of standby- channel "A" goes into current limit mode immediately.  The terminals are outputting SOME voltage- on the order of 32mV when current limiting at 1A.  Adjusting the current up or down adjusts the voltage.  So the current/voltage adjustments are working, just not properly.  The (+) & (-) terminals (on channel "A") read approx. 3Ω.  Open from GND to either terminal.  So there is a short between (+) & (-).  The terminals on the rear are not shorted.  (FYI, Channel "B" reads about 3kΩ from (+) to (-) terminals & open to from either terminals to GND- so this is, I'm assuming, the proper reading).

I opened the unit up and everything looks good.  No black marks, no bulging caps, no burned ICs.  Nothing obvious.  The unit is laid out very well and almost everything on the power supply board is socketed.  Diodes, resistors- I have never seen this before.  Hopefully will make it easier to fix.  Below are some photos of the unit and boards.

Tomorrow, I'm going to start actual troubleshooting, testing the capacitors first (with in-circuit ESR tester), then the diodes, then the pass-transistors.

If you guys have ever run into a similar issue or know the ins/outs of linear power supplies, then I'm all ears.  I searched FAR and WIDE for any info on this unit.  Besides a Reddit post and a couple of eBay listings- there isn't anything that I could find (about the company or about the model).  Without a schematic (or even a manual) it's going to be on the _difficult_ side for me.  Any help would be greatly appreciated.

-Ben

Unit Photos:
Front of Unit - https://photos.app.goo.gl/RYPK2PMaNUyUiLgQ9
Rear of Unit - https://photos.app.goo.gl/ApHbUWGhwzdsvKfs9
Top-Down of Unit with Cover Off - https://photos.app.goo.gl/uykMQBzW32nLdxck9
Top Board (Linear Power Supply Board) - https://photos.app.goo.gl/F8QBrqUo1BXg6qgq8
Bottom Board (Controller/Logic Board) - https://photos.app.goo.gl/srUpnaSEUQCfYZYx9

[BrokenYugo]

Check for a shorted parts across the output.

[jbenward]

That was my first thought about there being a physical bridge across the output somewhere. I could not find one.  I checked all over- outside, inside the case, front and rear jacks, PCB traces- everywhere I could think of.  I'm hoping that the apparent shorting of the outputs is indicative of a certain failure in linear power supplies that I'm just unaware of.

Basically- I've checked all of the mechanical (ie. bridging) possibilities.  Got to be something a bit harder to find.

-Ben

[Ground_Loop]

Have you checked for shorts on the power transistors?

[jbenward]

I've checked the power transistors and they check out properly.  It's something else that's failing short.

I think I'm going to return the unit- this is above my pay-grade and it could be just about anything inside.  I was hoping there was some obvious culprit here- but as I said I checked those and it's likely a diode or small cap that's failed.  My little Chinese ESR tester absolutely crapped the bed trying to test anything in-circuit.  Everything I tested had to be tested out of the circuit.

It's unfortunate- it was a really cool little power supply.

-Ben

[Ground_Loop]

I had a similar situation with a power supply where the negative jack was shorted to the ground jack.  I found a shorted decoupling cap across the two jacks.  I see in the last photo what looks like a similar setup where your output jacks have disc caps across the outputs to ground.  I assume you checked those good as well.  I really don't see that this problem could be very far up the circuit, given the indications.

[macboy]

Some linear power supplies may have down-programming and/or a crowbar circuit. Either feature means a power transistor (or maybe SCR) across the output to sink current. If one of those power devices goes short-circuit, the symptom can be the exact problem that you are seeing. On the analog board, I at least two TO-220 devices with a heatsink, and one without a heatsink, per channel. Check those devices... what are they connected to, and are they shorted.

[jbenward]

Thanks for the advice on this.  Turns out the return process involved the seller not wanting the unit back.  So I guess it is more troubleshooting since I'd really like to find the issue.

I checked the caps across the outputs.  They are both good.  Plus he's shorting between positive and negative, no continuity between ground and neg or pos.

I also checked all of the TO-220 on Channel A and they all seem to be working properly.  I even pulled the front panel wiring harness and isolated the board and there is still continuity between pos and neg on Channel A.  It has to be something on the power board.  It's not in the micro (continuity is  there both on and off).  I some of the protection diodes and tested them and they were good as well.

I'm going to look into the crowbar circuit.  That might be a good lead...

Also, there is a TO-220 package here with a number that I can't find a datasheet for- anywhere.  It's a TAG8442.  Any ideas?

-Ben

[macboy]

Could be a second source (equivalent) of Fairchild FDP8442 N Channel MOSFET. It could conceivably be used in a crowbar. Check if it is shorted pin 2 to 3, and if those pins are across the output.

[jbenward]

So about the TAG8442.  I pulled the board, desoldered it, and bingo, no more continuity.  I tested the suspect TA8442 on my component tester and it reads as 2 resistors (0.25ohm and 150ohm).  Since this isn't a part that is made anymore (even the FDP8442), I replaced it with an equivalent N Channel MOSFET (FQP30N06L) I had on hand.  After resoldering the new MOSFET, I put the board back, powered it up and yep- that was it.  I ran a host of load tests on it to be sure there wasn't another issue.

Thanks Macboy- you had the right instincts on this.

Well thanks everyone for all the help on this.  I'm hoping that this is the end of this and I've got a dependable power supply now.

-Ben

[BrokenYugo]

Pretty typical with linear supplies that somebody did something ill advised with it, blew some protection device, and then it gets replaced and sold as industrial surplus rather than repaired.