On 26 Feb 2006 13:39:18 -0800, "no_go" put
finger to keyboard and composed:

Hi,

I am looking for help to fix a dead ATX SMPS.

Background;
I bought a new NorthQ 4001AT 400Watt SMPS. The unit was DOA, and the
seller replaced it immidiately and didn't want the defect one in
return. So I want to fix it - as all components in it are new it may
just be a minor fault and it may serve me for a long time, besides I
would be very satisfied if I could manage to fix it.
The SMPS is dead - when PS_ON (green wire) is connected to earth I hear
a small 'tick' from the SMPS and the fan blades moves half an inch.
There is no voltages on any of the outputs. When doing these tests I
load the +3,3 +5 and +12 Volt outputs with minimum loads (as read on
the box) in the form of power resistors.
When testing there is over 300 V across the two HV capacitors, and half
the voltage between the two. There is 3 power transistors in the
primary circuit 2 pcs J13009 and 1 pcs C5027-R. When testing these with
an DMM all junctions seems fine. There is +5V between pin 5 and 9 on
the ATX power connector.

That's the +5VSB standby voltage.

I would like to ask for help to continue troubleshooting from here. At
least parts of the primary seems to be OK. I have read somewhere that a
possible fault can be the startup resistors of the primary transistors,
and that these resistors usually have quite a high value. But I cannot
find any resistors with large values around these transistors.

In an ATX PSU, there would be one startup resistor for the chopper
transistor in the +5VSB circuit. Yours appears OK.

I have
some experience with electronics, but little with SMPSs. I have
experience with tube amplifiers so I am used to watch myself from B+.

The semiconductor components I have located in the SMPS are;

Primary;
2 x J13009 NPN Power BJT

MJE13009 ???

1 x C5027-R Hi-Volt BJT

Probably a 2SC5027

diodes

Secondary;
2 x SB3040PT Schottky Barrier Rectifier
4 x component TO-220 3-pin
1 x LM 339N Quad comparator

Overvoltage and power good sensing ?

1 x KA7500B 5V reference

PWM controller

1 x 4-pin DIP component

Optocoupler for +5VSB voltage sensing?

8 x small 3-pin BJT or similar
lot of diodes

Thank you for helping me.
marty

This is a typical ATX PSU circuit:
http://www.pavouk.comp.cz/hw/en_atxps.html

The TL494 is equivalent to the KA7500B PWM controller. If pin 4 of the
KA7500B is at +5V, then I would suspect the PSU has shut down due to
an overvoltage fault.

I'd measure the resistance looking into each of the secondary outputs.
A short would point to a diode.

Here is a good repair guide for switchmode PSUs:
http://www.repairfaq.org/REPAIR/F_smpsfaq.html

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.