E4438C SMPS Repair

[fourierpwn]

Hi all,

I am working on a dead power supply for an E4438C. It's a multi layered design and no schematics seem to be available so I am having some trouble with reverse engineering the board. The unit was completely dead, though when inspecting the SMPS, it seems as though the fault is within the PFC circuitry. The main PFC workhorse is the L4981AD. Troubleshooting revealed no main PWM clock, which the L4981 is responsible for. Though, the VCC of the L4981 is missing so I began reverse engineering to see where this was coming from. Again, due to the complex nature of the SMPS, I am struggling. I found a 15V zener which is tied to the VCC pin of the L4981 to be dead though upon replacing the unit is still dead (zener didn't blow again).

While probing around, I noticed that there is +300VDC (I'm in Australia so 240VAC rectified -> ~320VDC) across a 10R resistor which is awfully close to a ground pin. Regardless, the 10R resistor is open circuit. I am reluctant to blindly replace this resistor before confirming whether it is meant to have such a large voltage across it.

If anyone has this unit or access to a similar SMPS, would it be at all possible to check if this pin is tied to the main DC line. Here are some pictures to help illustrate things for anyone willing to be of assistance.

A. B. C. D.

Looking at image C. the main filter cap after rectification is near the top, labelled by 300VDC. There's a resistor divider from the filter cap, which is typically where you can get a good idea of the usage of the supply as the tracks here run quite hot and degrade the board (FYI for looking at second hand parts). The pin labelled 1. is the 10R resistor and pin 2. is the ground pin. If anybody would be kind enough to test on a working unit, a simple continuity test between the positive pin of main filter cap and pin 1 from image C. will tell me what I need to know.

Any other suggestions/ideas are welcomed.

Also, the red, blue and black wires are for debugging purposes as the under side of the SMPS is not accessible when inside the unit so they can be ignored.

Merry Christmas!

[George Edmonds]

Hi

You have 300VDC plus across a 10R resistor, I don't think so.

George

[asis]

Hi,
You are faced with a difficult task, but doable.

Merry Christmas!

[fourierpwn]

Duly noted!

What is the example file which you have linked asis? R35 is interesting...

[fourierpwn]

Also, I see the E4438C-0BV option (discontinued) is for a hard copy of the component level service manual. As this supply is a third party option, does anyone know if the CLIP includes the SMPS?

[asis]

This is a server  PSU  diagram just for example.
It was given to the customer long ago.
Apparently,  you have something with an STB .
I do not see the type of microchip - the blue wire is in the way.
Indicate which MOSFET the STB has on the finned heatsink (TP57 bottom) and check the resistor in the "source" circuit and all ECAP's.
You draw a STB fragment, there are few components - it will be easier to understand.

[fourierpwn]

I do not see the type of microchip - the blue wire is in the way.

This is a 2845B PWM controller.

Indicate which MOSFET the STB has on the finned heatsink (TP57 bottom) and check the resistor in the "source" circuit and all ECAP's.

IRFBG30 power FET. I've checked this and all seems fine.

You draw a STB fragment, there are few components - it will be easier to understand.

This was going to be my next step.

[Tony_G]

Apologies for resurrecting an old post but I managed to dumpster dive an E4438C and it has a blown power supply.

You can see the blown part here:


Zooming it:


Zoomed in a bit more:


fourierpwn - From your photos, it looks like that part of the supply was perfectly fine - Any chance you might have a photo that shows the resistors in better detail than I have? All of mine will need replacing but due to the cap bursting all over them I can't read the rings (or at least the rings when converted make no sense in terms of values).

If anyone else might have one that could help I'd really appreciate it as well.

Thanks,

TonyG

[fourierpwn]

PM sent

[eyesofthunder]

I just swapped my E4428C power supply into a E4438C that was down.  The E4438C is more valuable to me than the E4428C.

Tony, looks like mine has failed very similar to yours.  How successful was your repair?