Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

[socketa]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

Hmm. So between these 2 pins there is only a 1.1v difference when on?

Yes, that's correct

[socketa]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

PC123
Type Long Creepage Distance Photocoupler
Forward voltage typical: 1.2V

[CapLeaker]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

Boy, that’s right on the threshold. And that’s the opto that goes to the cm6800, I expected more voltage difference than that.

[socketa]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

OK thanks,
I'll swap it with another one from another random parts PSU, and will measure the voltage again, and see if it makes any difference.
Will post results here

[CapLeaker]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

i don't think it is the opto. Could be, but me thinking its not.

[socketa]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

i swapped it with an Everlight EL187 opto, which, according to the datasheet, has a typical forward voltage of 1.2V.
the difference between the pins on the secondary (when PS-ON is grounded) is now 1.03V (9.82 - 8.78)
when i turned it on for the fist time with the replaced opto, it wouldn't start, because the standby was at 2.5V already
after a few more attempts, standby is now 5V for at least 30 seconds, but it was dropping to 2.5V, like before (initially), without me doing anything
and like before, after about 30 seconds of powering on the PSU, it shuts down, and standby drops to 2.5V

The fact that it didn't behave normally when i first plugged it in (standby was already 2.5V) would mean that it's not the APFC chip?
Or would a faulty APFC/PWM chip/circuit affect the standby supply even when the PSU is not powered up?
It does seem like the chip is doing something in standby mode, since (as you previously mentioned) that's what the CM6800 OVP protection part does (makes standby go to 2.5V)

I notice that there is no MOVs installed on this PSU - maybe there was a voltage spike that screwed up the chip?

[CapLeaker]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

Usually the stdby on these cm6800 or FAN4800 have a separate 5VSB circuit, likely using an TNY277 IC or something similar. When the power is off the cm6800 shouldn't do anything at all. Post a straight shot high resolution picture.

[socketa]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

A6069H PWM 7-Pin chip is on the primary side - frontside of board
PS229 4-Channel secondary monitoring IC is on the secondary side - frontside of board
CM6800 is on the backside of the board
and it does has a MOV, but not an NTC thermistor

[CapLeaker]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

Check replace those 5 caps near the heat sinks.

[socketa]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

i checked the four caps on the primary;
the bigger one (Taicon PW 220uf 50V), that's next to the 7-pin chip, capacitance is good, but the ESR is very low when i first connect it to the ESR Meter: 0.000 ohms
i guess that that means that it's not a good test
i'm having a fossick to see if i can find a replacement.

[socketa]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

It appears to be connected to the middle-sized transformer (is that the standby transformer?) via a resistor and what looks like a diode.
Going to have to order some
Probably take around a month

[CapLeaker]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

A low ESR is good, a high ESR is bad. Google something like ESR chart. As long as the cap doesn’t show a constant short (bad), a little short for a second, then charging, that’s fine. Probably you measured it in circuit. Measuring caps in circuit, you have to understand the circuit and what you are measuring. Caps are best measured out of circuit at room temperature. A hot cap measures lower ESR as a cold cap. Keep that in mind too.

[socketa]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

i measured all of them out of circuit with a dedicated ESR meter at about 25 degrees Celsius.
The other three caps all had around expected normal ESR. But this cap is testing lower ESR than Nichicon HN VRM caps
and ,yes, i zero the meter before measuring

[CapLeaker]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

just stick something else in there that's at least 50v 220uF rated. For example a 50V 330uf or 470uF. It's just for a test...

[socketa]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

Well, it doesn't look like it was a good idea to remove those four caps all at once, because now that i put them all back in the stby doesn't even get to 5v but falls to 2.5V straight away.
Couldn't find a 50V replacement cap to test with - i tried using two 25V 470uf caps in series, despite the resultant ESR measuring too high, then i tried a couple of 25V 220uf caps separately (voltage didn't go higher than 25V), but in both instances it doesn't get to 5v now - 4.something volts, and then drops back to 2.5V.
i'll try again early tomorrow when it's cooler and see what happens

[socketa]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

This morning i tried again,
and the 5VSB now stays on, and, with a rubycon YXF 25V and PSON grounded, the PSU shuts down after about 3 minutes - same story as before.
(noted that it's a bit high though: 5.07V)
So maybe the original Taicon PW is good, despite it having extremely low ESR.
There is also around 340V on the primary cap.

[CapLeaker]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

Now you’re back to square one. The PFC isn’t boosting voltage as it should be closer to 400V dc on the primary cap. I guess I would try a different cm 6800 or equivalent if you got one handy. If not it’s time to reverse engineer that thing and draw a schematic. Other than that it isn’t looking good, since you don’t have proper equipment to deal with this kind of stuff.

[socketa]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

Ok
Next step is to disconnect the PFC mosfets and bridge the boost diode.
I did that with a Silverstone that's got the same CM6800 chip, and it's being running fine.
Looking through the posts on this chip, it looks like it's PFC is likely going to result in stressing out the primary cap anyway.

[CapLeaker]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

I suppose you got nothing much to lose at this point, since the PSU isn't worth much to start with

[socketa]

Re: Zalman ZM500-GV stops after 40s and 5VSB drops to 2.5V

why is it worthless?
anyway...
so, now i'm checking the pins of cm6800 and looking for movement that might indicate something going wrong prior to the PSU shutting down, seems all good up to pin 10 so far.
on the datasheet pin 11 drives two transistors, which is the same on my board, but in between that pin and the transistors is what looks like a diode, you can see it in the top right quadrant of the PCB trace photo. It has a marking of T4 C4 on it and a bar at one end (markings not visible in photo).
Is that a diode?
it tests 10ohms in both directions, in circuit
I'm a little bit suspicious of it.