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

I replaced the CM6800 PWM/PFC chip, even though i sort of knew that it was probably OK
and i also replaced the 3.3V output caps (which tested good)
Same deal as before

It's not shutting down due to lack of load - when i connect a mobo and a hard drive, it shuts down a lot sooner

The FPO pin on the supervisor chip goes high when the PSU shuts down, but i'm not sure if it that's because it shutting down the PSU, or if it goes high when something else shuts it down.

Prior to this i bypassed the PFC section - and it makes no difference

The 3.3V rail starts close to 3.3V (2.29V sometimes) but quickly starts falling from the moment that it is powered up
I doubt that it is the standby chip that is causing that
Maybe the supervisor chip is faulty, but that wouldn't be causing the 3.3V rail to behave like that, unless it's somehow causing a lower and lower resistance path to ground via the 3.3V voltage/current sensing pins
But i don't see why that chip would be damaged

I think that that supervisor chip is dropping 2V across it's VCC to ground, when the PSU shuts down, which would explain why the 5VSB chip goes to 2.5V (yeah, i know that there's 0.5V missing somehere) - which is possibly normal behavior

VCC of the controller chip starts at 4.4V (from 5VSB), then goes to 11.1V (probably now supplied by the 12V rail), and then goes to 1.97V when the PSU shuts down

Something is stealing voltage from the 3.3V rail (maybe as some component heats up)
And it seems that that supervisor chip is shutting down the PSU, when it shouldn't be.
I wonder if they are separate issues, or if they are due to one fault only?

[socketa]

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

On second thoughts,
i don't think that it's the VCC pin that's on the supervisor chip that is causing the 5VSB to go to 2.5V, but rather that it's the voltage that's on FPO pin going high, which, i'm guessing, has some effect upon the 5VSB (via an optocoupler), which causes it to go to 2.5V; which might be what is supposed to happen.
And that's why i think that the 5VSB chip is probably OK

And now, i think that the falling 3.3V rail might be due to some fault in the 3.3V output circuit, or regulation (mag amp?), since the other rails are good

[CapLeaker]

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

That’s what I said early on into this threat. Check the FPO a pin going high and disable it (ground it). That FPO goes to an optocoupler. If that pin goes high, there won’t be enough voltage drop to energize the LED inside the optocoupler and the primary of the PSU goes into shut down. However a faulty supervisor or a bad main filter capacitor can do exactly the same thing too.

While you are at it... upload a straight shot, high resolution picture of the PSU.

[stj]

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

just stop for a second.
the whole psu usually runs from the standby voltage.
did you test the standby voltage on it's own BUT with a load - maybe a 1A load??
if you can run a load from the standby for atleast an hour then you know it's good.

i'v seen this type of stupid shit before and it's easy to get misled by all the other parts when it's just the standby rectifier diode or driver i.c. failing when it warms up.

[socketa]

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

Well, i spent some of the last day working on it
i was wondering how to get a 1A current, because the bulbs that i used in the past only drew 0.1A
i tried two headlight bulbs but neither of them lit up
Because of that, i eventually wound up connecting five of the other bulbs in parallel and noticed that when i connected the fifth one up that they all stopped glowing
and then the PSU didn't start up when i grounded PSON
So i grabbed another PSU and tried one of the headlight bulbs and it didn't work,
Then, i opened up the car and put it on that battery, and it lit up
i figured out that, unlike the other bulbs, that the bulb terminal that i thought was positive was actually negative (an then remembered that is how car wiring is) - so initially connected the bulb the wrong way around, and now know that these bulbs have a polarity.
So then tried it on the other PSU, and it lit up.
Then i went back to the Zalman PSU and tried it on the 5VSB, and it did not light up. Pretty sure that 5VSB voltage vanished when the fifth bulb was added, and came back when it was removed, because that's what my meter indicated.
Maybe this is progress, because it seems like the 5VSB (or part of that circuit) didn't like the fifth bulb being attached.
So currently, when powered up, the PSU fan spins weakly for a couple of seconds and then stops, with a collapsing 12V rail voltage.

[stj]

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

but unless we know the bulb whattage we cant calculate the amps!
and why would they have polarity? or did you use led bulbs?

[socketa]

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

OK, they are not polar - i wasn't giving the bulb enough time to light up from cold
It's a 12V 55W bulb that light's up in the other PSU, drawing 0.5A initially and then 0.17A
When i connect it to the 5VSB in the Zalman, it initially draws 0.5A and then 0.11A, and the bulb does not light up

When i ground PSON, the PSU still just spun for a second and stopped (no rail voltages)
Now it won't start at all when i ground PSON
Pretty sure that i could smell an electrical component burning smell when i had the bulb attached to the 5VSB

[CapLeaker]

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

C61 doesn't look too good to me.

[socketa]

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

that's a new Nichicon 50V 220uf, the reflection of the light...
it replaced the original Taicon PW (post #30) with a new Nichicon PW

[CapLeaker]

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

Did you check the main filter capacitor?

[socketa]

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

The main filter capacitor is a 330uf Teapo LH general purpose 85 degrees Celsius,
it measures 294uf and has ESR of 0.105 ohms
Anyway, i replaced it with another one from a parts PSU, that had 320uf and ESR 0.17ohms
It still won't power that bulb though, and i can see that when i connect the bulb that the 5VSB drops to zero and tries to go up every three seconds and then back to zero.
And, when i ground PSON, it shuts down several seconds after connecting
a hard drive and the fan makes a noise like it's not happy.

[CapLeaker]

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

Dang. Since the supervisor isn’t shutting it down, hit the cm6800 and the other 7pin DIP with a can of air upside down and also treat them with a hair dryer. See if you can lengthen or short the time the psu can run. You may have damaged the 5v stby circuit with all those lightbulbs. What is that 7 pin dip next to c61?

[socketa]

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

those bulbs were tail light bulbs, and only drew about 0.1A each, and i only had five of them.
so the rail's limit was about 0.5A - about the same as the headlight bulb that i used that didn't light up

that chip that's next to C61 is an STR-A6000H

[CapLeaker]

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

That STR-A6000H that is always on and produces the 5V STBY.
I am just trying to get my head around around on how this PSU a an shut down without the supervisor telling it to. It’s a more rare of an occurrence but possible that a PWM suts down by itself after like 30 seconds run time. I’ve did repair quite a few PSU’s in all kinds of variations but that has only been the case like a handful of times.
When this PSU shuts down, does the 5V STBY disappear?

[socketa]

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

Yep,
The 5VSB doesn't budge even a little bit when the PSU shuts down - it's good
What has changed now, is that the PSU shuts down after a couple of seconds, or a second if i ground PSON right away after it shutting down
So that means that i don't have to wait long now to see if it shuts down.

I just got another same model PSU (in the hope of it providing some clues), and it's got a similar problem (2.5V standby, and not powering up) as the one that this one developed; so it's probably a common failure.
So i looked back over the thread to see what brought the standby voltage back up to 5V - i see that, between posts #34 and #35, the standby voltage came up - but i didn't do anything - so that's not helpful

I am just trying to get my head around around on how this PSU a an shut down without the supervisor telling it to

maybe it's due to some faulty component heating up really fast, (and then cooling really fast, when it shuts down)
i think that sti's comment has a good chance of being along the right track

[CapLeaker]

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

Maybe replace the cm6800 with a new or equivalent.

[socketa]

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

I already did that

And i think that the standby chip is good, because i removed it, and the other other one from the other zalman, and they both tested good as far as i could see.
All pins measured the same resistance to ground pin, and no shorted internal diodes
I also checked the empty PCB holes to floating ground and they all roughly tested the same.
Then i put them back on the boards, swapping them over.
The only noticeable change was that the fan, instead of spinning for a second or two just twitches now (one blade turn at most).
So i'll swap them back over again

i'll remove the supervisor chips, test them, and swap them over.

In fact, i might see what the standby does with the supervisor chip removed.

[CapLeaker]

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

Leave the supervisor in place and ground out the FPO pin to secondary gnd. That disables it.

[socketa]

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

i was too quick, and swapped them over, but it didn't make any difference;
so i put them back as they were originally, and now the PSU doesn't even try to start when PSON is grounded (no voltages come up at all), even though the supervisor has 4.4V VCC.
And the 5V still disappears when i connect it to the headlight bulb, and pulses up to nearly 1V every few seconds
Looking at the typical application circuit, i checked Rocp resistor, which is good
The diode, resistor, and capacitor (SMD, as opposed to an electrolytic cap as the application circuit says), that are attached to VCC pin, are all good
And the capacitor that is across the feedback diode is measuring 1nf in circuit
So that section all looks good. Right?

[CapLeaker]

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

That’s one of the reasons why I said ground temporarily the FPO pin out, that disables any protection from the supervisor and the primary can’t turn off because of under voltage, over voltage, etc condition.
Problem now is that nothing works and why? Could be anything from a forgotten solder or bad solder joint, to an overheated or somehow damaged component, broken trace, etc.

Next problem is / was, that the FPO pin on the supervisor was low on some testing, then it was measured high again on other testing. Another reason to disable it because all this adds to confusion. Once this situation is set there is no variable, no matter what. You just couldn’t go Willy Nilly anymore and have to remember no to overload, put the power supply on a dim bulb tester.