Author Topic: Sorensen XPF 60-20DP repair (Read 6941 times)

rmacintosh · « **on:** October 26, 2017, 03:52:25 pm »

I have recently acquired a Sorensen XPF 60-20DP bench power supply. It is untested so I got it dirt cheap. In anticipation of having to possibly troubleshoot or repair this supply so I can put it into service, I was searching/googling for a service manual. All I was able to find was the product brochure and a operation manual for this model, which is only half useful as it just gives the specs and the calibration procedure beyond the usual specs and feature rundown. There is a maintenance section that only says this:

Quote

The Manufacturers or their agents overseas will provide repair for any unit developing a fault.
Where owner wish to undertake their own maintenance work, this should only be done by skilled
personnel in conjunction with the service manual which may be purchased directly from the
Manufacturers or their agents overseas.

So I know one exists. Anyone have this manual or know of a better source than google has been for me?

glarsson · « **Reply #1 on:** October 26, 2017, 04:47:59 pm »

This unit is the same as the Thurlby Thandar CPX400D/CPX400DP

https://www.aimtti.com/product-category/dc-power-supplies/aim-cpxseries
https://www.aimtti.com/cpx-series-tour

I asked TTI for service manual for my power supplies in the TSX series (2x TSX3510P and 2x TSX1820P) and they sent some PDF files. Perhaps they can be of some help.

rmacintosh · « **Reply #2 on:** October 26, 2017, 05:48:09 pm »

Excellent. I'll try emailing both TTI and Sorensen/Ametek, see if I can get a service manual from any one of them.

rmacintosh · « **Reply #3 on:** November 06, 2017, 04:42:36 pm »

TTI was able to get back to me, they sent me a CPX400A service manual and a CPX400D/DP 'service guide' which is marginally useful. It at least has a parts list, although it only goes down to the assembly level not component level.

Also sent the CPX400 main board schematics.

They said there is currently no service manual for the CPX400D/DP units.

rmacintosh · « **Reply #4 on:** November 06, 2017, 04:47:22 pm »

1 more
too big to attach

Updated link!!!!!!
https://drive.google.com/file/d/0B9olCHx94L2ZZE43UF9nUTBNUkN1SFI4U3RNeUJMNEJQQVBZ/view?usp=drivesdk&resourcekey=0-BoOeCj9GlUnXV2CIyRzY0Q

glarsson · « **Reply #5 on:** November 06, 2017, 05:44:57 pm »

The CPX series seems to have a parallell history to the TSX series. The circuit diagram has changed several times and both have non-PFC and PFC versions. The non-PFC is the older models and the PFC models are the later models. PFC here means active power factor control. The design of the enclose says late design, i.e. look at the PFC version circuit diagram (the separate diagram) not the one in the service manual.

rmacintosh · « **Reply #6 on:** August 26, 2018, 01:19:54 am »

Well, it has been 9 months since this unit arrived. All I have managed to do so far is tear it down quickly and do a initial scan of the PCB's and some simple DMM continuity and diode tests. Nothing glaring wrong so far.

The 9 months to only get this far along is on account of having a new baby arrive at the beginning of the year and as a result any sort of free time I used to enjoy for projects such as this has been erased. I'm slowly trying to get back to a normal schedule and this is one thing I am really itching to dig into again.

So I decided to start with taking some pictures. since this has now been sitting on one of my benches in pieces for the past 9 months.

The first to are the standard front and back pictures. MFG date - OCT 2014

rmacintosh · « **Reply #7 on:** August 26, 2018, 01:24:41 am »

Next three are of the main power board, top and bottom view. Not much going on here - some heat stress on the board by the large pair of resistors. The components themselves look aand measure fine, although those MOV's at the tail end of the resistors have seen happier days.

I think my troubleshooting will begin here, I quickly tried turning on the unit and .... nothing no life, as if the fuse was blown or something which doesn't seem to be the case after a quick DMM test. More probing necessary.

rmacintosh · « **Reply #8 on:** August 26, 2018, 03:05:24 am »

Now we have the left and right channel boards. These boards are basically mirror images of each other as they sit upright perpendicular to the main power board attached via a couple dual row header connectors.

Note the board indicates the part CPX400 power - L (left)
This psu is identical the the TTi CPX400DP unit.

This appears to be the original board for this unit. I have noticed some of the resistors at the edge of the board to the left of T1 seem discolored and have been running hot but their values seem to be good.

I could be wrong, but I don't think any of these boards have been repaired before. No signs of rework.
I believe the flux residue on the transistors/diodes attached to the heatsinks are from original assembly and those heatsinks and parts were hand installed.

rmacintosh · « **Reply #9 on:** August 26, 2018, 04:02:10 am »

This is the right channel board. Noticeably different. I don't think there is any difference, just has a slightly different layout.
That and the part number being different CPX400S Power (35555-5540-2A)

Nothing too concerning, some noticeable heat discoloration around R9. Everything reads normal with a DMM in that area. I even compared with the left board.

coromonadalix · « **Reply #10 on:** August 26, 2018, 12:10:57 pm »

There's lot of scratchs on one of the pcb, the solders on the mosfets, transistors diodes doesn't seems clean to my eyes, solder or flux residues

I think the boards where probed and had some rework, check where all the wire connect ...

I would change r11 r15 r16 on the main board and lift them higher to help them dissipate and save the pcb from burning / darkening more

Same goes for R9 and maybe R52 on the power boards, its cooking the board

rmacintosh · « **Reply #11 on:** August 26, 2018, 10:52:16 pm »

Totally agree. Great idea raising the resistors from the PCB.

I am currently reassembling the unit so I can power it up and do some more troubleshooting to try to isolate the problem to one of the boards. (With the hope there is only one problem )

Seekonk · « **Reply #12 on:** August 26, 2018, 11:52:38 pm »

I've repaired many power supplies. If the unit is totally dead, I suspect it is the control power supply that powers all the electronics. Look for a small transformer with a FET and switching chip. Find the nearby input and output electrolytics and look for voltage. That is where I would start. I got real lucky once with a really nice Xantrex for dirt cheap that wouldn't power up. It had the remote power on option. Just had to remove the relay board.

rmacintosh · « **Reply #13 on:** August 27, 2018, 01:43:26 am »

Well, after reassembling the unit fully minus the top cover i was ready to do some digging and troubleshooting for symptoms of a completely dead PSU.
Then I got this curveball.

https://youtu.be/1aBJ_t9QagI

it lives! but not quite.

Right channel seems operational, as long as the front panel has power. The left channel is just dead, there is a brief flash on the right display but that's it.
I'll have to take the front panel off and disassemble. Take a few pictures and will post those once I get that apart.
It appears to be 2 PCBs stack on top of each other.

Quite strange the way the right channel display slowly fades in and after a bit slowly dies out. Must be a power issue, maybe simply a bad regulator or cap.
Will see....

bitseeker · « **Reply #14 on:** August 27, 2018, 06:36:45 am »

I usually start from the rear of the supply and work forward. If you don't have any or too little power coming in, nothing after that will work properly.

Make sure you have high-voltage DC being output by the main power board. It looks like all it's doing is rectifying and smoothing the AC into unregulated DC somewhere in the neighborhood of 300V. If you get significantly less, then see what's wrong on that board.

If the main board is outputting OK, then follow the low-voltage DC power rails from the right channel's display board back to the right channel's SMPS board to see where its power is being generated, whether the voltage looks reasonable (probably ~5V), and if it's stable. Since the display on that side sometimes comes on, but doesn't stay on, it could have leaky electrolytic caps (for kick starting the switching IC or filtering the power rails) or a bad opto-coupler.

The left channel looks like it just isn't starting up. The startup cap for it might be bad, the switching IC could be dead, the power MOSFET that the IC drives could be fried, or some power resistors might be bad.

You could also swap the display boards to see if the problem follows them (i.e., there might be a short or heavy load on one of the display boards).

A UPS I was working on had a shorted MOSFET, burned up current sense and gate resistors, dead switching IC and blown opto-coupler. Having the datasheet for the switching IC is useful for systematically checking the critical components that keep the SMPS going.

rmacintosh · « **Reply #15 on:** August 28, 2018, 08:06:37 pm »

I got a bit of time to do some poking around on the main power board today.
I measured the output voltage after the bridge rectifier, 160v seems ok

Then i measured the output of the boost converter, HT. That was also 160v
hmmm, seems like the switching circuit is toast.
I have to say, the first measurement I took was good. 392v, but this was the only time I got this.
Also, when this rail was present the right display was working. Mostly, like in the video I posted earlier.
Once the displays died so did the HT rail. I haven't been able to get that back.

I also measured the other rails, not expecting much as they are derived from the HT rail.
+18vprim -- measured 11.5v
+12vprim -- measured 10v

Interestingly the two 3W resistors on the main board that are charring the PCB are located in the boost converter circuit acting as snubbers for the switching node.

At the moment, I am thinking a few things.

A. There is something gone bad in the boost converter, find that...replace part(s)....fixed

B. There is something wrong with the PFC section (IC4) If thats the case it looks like replacing IC4 would fix the problem.

C. There could be a problem with the control circuitry around IC1. The relay is working but the SS signal to startup the PFC and drive the boost converter may not be present.

A seems most likely to me, but I could be wrong. Considering the gate drive circuit for Q5,6,7 is powered by the +18vprim rail which is reading in at 11.5v I am still kind of stumped. To top that off the +18vprim rail is derived from the HT rail by IC2

Seekonk · « **Reply #16 on:** August 28, 2018, 08:46:27 pm »

Something is dragging down your 18V. The 10V can be expected from that thru the regulator. I hate to say capacitor, but the one on that 18V could be open. Try bridging one. Can the rest of the circuit be disconnected with that connector? Switchers like that either work, don't work or pulse. It is either tht capacitor or further down that 18V line.

rmacintosh · « **Reply #17 on:** August 29, 2018, 04:39:43 pm »

I was able to check all the electrolytics with an ESR meter in-circuit them took them out and tested them on a Sencore LC analyzer - they all seem to be good.
Im going to try reassembling the unit again and taking some more measurements. Even try using the IR thermometer to see if I can pinpoint any parts getting hot.

rmacintosh · « **Reply #18 on:** August 31, 2018, 03:59:55 am »

After combing over the main board schematic a bit more I started tracing the V+ rail. On the main board all the V+ rail does is setup the inhibit signal for the left/right power boards but V+ is also routed to the L/R power boards.

V+ is used in a number of areas on the power board including generating a +15v rail. Across the V+ rail I am measuring a low resistance - 220 $\$\Omega\$$
Measured on the right power board, the left power board has roughly 14k $\$\Omega\$$ across the rail.

This puts a few things a culprits.
The 7815 regulator, which takes V+ as input.
The fan switch circuit
Some filter/bypass caps on this rail

I'll do some removing or parts this weekend to determine what exactly is the bad part and will need to place a parts order.

Just for fun, i reassebled the unit leaving the right power board out and it works! sort of...
Not quite out of the water yet though.
FAIL 4

Unfortunately there are no error code references in the documentation i have (posted in this thread).
This error code flashes right at power on, and if i push the output on button (whose LED blinks constantly. should be solid on or off) the FAIL 4 code is displayed. No dc output despite what the meters show.

I am also not sure if the fail 4 is due to the missing power board and the second PSU channel not being installed. Ill have to fix the right power board before I can test that theory.

bitseeker · « **Reply #19 on:** August 31, 2018, 05:47:06 am »

Well that's some good progress. Get the main rails functioning correctly and then see what's what.

rmacintosh · « **Reply #20 on:** August 31, 2018, 10:08:19 pm »

Placed an order for parts.

Here is what I found.
D9 is on the T1-B primary winding of the left channel power board which appears to be generating V+. The service manual describes it as "A power-back winding is rectified by D9 and filtered by C6, C7."

Also Q11 is completely shorted. This is the switch that turns the case fan on. I better take a look at the fan as well, make sure its not the fan that caused all this.

I'll know more by mid week next week when I get parts.

bitseeker · « **Reply #21 on:** August 31, 2018, 10:53:41 pm »

Sounds good. Looking forward to see how it turns out.

rmacintosh · « **Reply #22 on:** September 06, 2018, 05:32:51 pm »

Update.

Repaired!
Probably. Replaced the dead parts.
Still have to do the mods where I replaced the hot resistors and install new ones raised from the board.
...but i was too impatient to do that before trying out the power supply.

So impatient in fact that I had forgotten to check out the case fan as I had mentioned in the previous post.

So I turned on the PSU, and it worked!
Turned out the FAIL 4 message I was getting before was a result of not having the comm board installed that handles the GPIB/USB/LAN/RS232 communications. Once I assembled the fully PSU it worked a treat, for about 10 mins........

Then I got this:
https://youtu.be/gjXewvMHMCM

Shortly after I took this video it dawned on me that I didn't check the fan yet. After all, the failed components on this supply were generating the rail and doing the switching for the case fan. It turns out the fan was just TOAST!!!

I quickly tested it on another PSU and the current limit kicked in immediately. Got as high as 4v and about 1.75A before I gave up trying to feed it the nominal 24v it wants. I installed a new fan and it has been chugging along nicely for about an hour. I will try a performance test next with a load as well as check how accurate and in-spec this unit is.

Not bad for $200 considering this guy will retail for around $1800 new right now (Newark seems to be selling it for $2600

).

bitseeker · « **Reply #23 on:** September 07, 2018, 01:14:15 am »

Thanks for the update. Good progress. Hope the burn-in goes well so she becomes a keeper.

Zucca · « **Reply #24 on:** September 07, 2018, 07:44:56 am »

When I see those fan pics, I kind of smell that brown plastic burned crap...


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Sorensen XPF 60-20DP repair (Read 6941 times)

Share me