Help with power supply repair

[jerryk]

I'm trying to repair an old power computer power supply (AcBel PC6001).  It died in a recent mains power brown out.  The defective parts that I can find so far is a Toshiba K3878 (E 128) Mosfet and two diodes in series to the gate.  I'm trying to find a replacement for the mosfet since it's obsolete and trying identify the dead diodes so I can replace them.  I have attached photos of the two diodes in question.  The smd diode has no markings other than a color coded band and the axial diode has the letters/numbers printed around it that are TZX 24 B.

Any help identifying a replacement mosfet and an id of the diodes would be greatly appreciated.  Also if there is a schematic of this PS floating around that would be great to have too.

Thanks - Jerry

[Andy Watson]

The TZX is probably a 24V zener.
http://datasheet.elcodis.com/pdf/22/16/221675/tzx6v2b-tr.pdf
Can you doodle-out the circuit surrounding the other diode?

[jerryk]

I'm just learning kicad so I hope this helps.  There is a current controller ic(UC3843b) that is controlling the mosfet if I I'm seeing this correctly. D2 is the smd diode.

[Andy Watson]

Going to need a bit more circuit than that

The diode looks like a relatively low power signal diode - but without the schematic ...

Can you doodle a bit more? I doubt that you will find schematic/service manual online. Sometimes it is worth contacting the manufacturers but I suspect it is a bit of a long-shot in this case.

[jerryk]

I will try to draw something with more detail.  Here are pictures ( if it helps) of the main circuit board showing the mosfet/mosfet heatsink removed and the zenier (as ZD1) diode removed.  The smd diode was on the back side in series with the axial zener on the front by the heatsink.

[jerryk]

Here is a little more detail on how the mosfet fits into the circuit of the power supply.  The varistors are actually mov's, L1 is a passive PFC as I understand it, and Q1 is a 2SK3878 mosfet.  I do appreciate the help on what the axial zener diode (D1)is but I am still looking to ID that smd diode (D2) and find a compatible substitute for the mosfet (Q1)which is a 2SK3878.  These are the only defective parts that I can find.  I hope I have drawn the schematic close enough to work with.

 

[Andy Watson]

Have you got the source and drain correct on the diagram?
My best guess is that the zener and diode are there to clamp the gate voltage on the Mosfet. The mosfet datasheet gives a gate-source breakdown voltage of +/- 30V, this would tally with the zener being 24V. The diode (D2) is there to prevent the zener from clamping negative gate voltages (why ? I don't know - need more of the circuit), so any common diode would do - 1N4148 or whatever the SMD equivalent is.

[jerryk]

Hi Andy.  Thanks for the help

I have revised my attempt at a schematic of the area of the PS that I'm finding issues.  The damage seems to be contained to the input side but the parts body count is increasing.  I have attached a picture of the other blown parts.  I'm stumped on the resistor because it doesn't fit any 5 band scheme that I can find.  It's definitely fried.  Any help Identifying it would be appreciated.  I also attached a picture of a transformer that I cannot find any data on.  I did my best to approximate how it is wound but have no idea what pin 2 on the input side does since it is not connected to the windings.

I have ordered a couple of replacement mosfets that were as close as I could find.  For the 2SK3878 I am going to try a FQA10N80C_F109 and a STW11NK90Z and will see if they work. 

You are right that I had the drain and source incorrect.  Thanks for putting up with my newness to this and I am trying to get the schematic close enough to ease your efforts to help.

The good new is I have replaced my friends power supply with a new one but just want to try to make this one work again as an exercise in learning.  I have a long way to go.

Thanks - Jerry   

[KuroZ]

I've seen silver being used as gray before in some resistors and green is 0.5% tolerance when used as a tolerance band.

Maybe it's a 15.8ohm resistor. Where was it in the circuit you attached?

[jerryk]

It is depicted as R8 on the schematic above.  It's between the source and the negative dc rail from the bridge rectifier.

Jerry

[wedgetail]

jerryk
 A suggestion slightly different, I recognise the IC it is a very commonly used, download the datasheet and see how a typical interface to other board components, this will give you an idea of what can be used.

[Andy Watson]

Silver multiplies the value by 0.01, so the resistor value is 0.15 ohms. 5% tolerance (gold). I believe the extra green band is a temperature co-efficient.

[KuroZ]

Silver multiplies the value by 0.01, so the resistor value is 0.15 ohms. 5% tolerance (gold). I believe the extra green band is a temperature co-efficient.

Yes, this makes more sense with the updated schematics, as it looks like a current sense resistor.

[jerryk]

Thanks guys for all the help so far.  I am learning as I go on this attempt at fixing this PS.  I am concerned though about the main transformers viability.  I have searched every iteration of the markings on the transformer itself and come up empty handed.  Perhaps there is a standard in the labeling that dictates how these things are wound etc.  The transformer label reads API 2 25D40 -0001I R HI-POT B-8 1146 AE (picture attached).  The reason for my concern is that on the primary side there are 3 pins that have wires soldered to them.  I have continuity between the outside pins which makes sense as primary winding however the center pin, even though you can see the wire attached and that it goes up under the tape in to the windings, tests open with all other pins on the transformer.  In the schematic I left it disconnected since that is how it test out.  Yet on the board itself it is clearly soldered to the negative output from the bridge rectifier.  If it did tap the primary windings, there would be a direct short to the positive side of the bridge rectifier which would not work either.  Maybe it's just a wire to nowhere and all is ok.  If there is an issue with this transformer my guess is I'm sunk since it appears these transformers are made specific to each PS and finding a match may be difficult.  Hopefully I'm out in left field in my thinking as I have become quite attached to this little guy and don't want to cash in on it's organ donor card just yet.

Jerry

[Andy Watson]

The unconnected TX pin could be an electrostatic screen, although I would have expected it to be connected to a ground/earth rather than the negative of the supply.

[jerryk]

Here is the back side of the circuit showing the center pin of the input side of the transformer, the neg output pin of the bridge rectifier, and the earth ground.  Earth and the neg have a cd102m cap connecting them.  I hope I have drawn it correctly.

Jerry