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

"Nothing works" - um, no - the 5vsb sill comes up
And there is no lifted traces, and no bad solder joints
And i used a dim bulb tester before connecting directly to mains
The 5vsb was obviously getting worse over time. It now fails even if i connect a bulb that would draw less than 0.1A

[socketa]

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

Just noticed that i mistakenly replaced the Rocp resistor with the one that i was using to discharge the primary caps (they both looked the same)
The twitching fan came back, but now the 12v doesn't come up at all
i think that the chip is probably bad
And found out that the chip from the other PSU is probably faulty; as the resistance from from pin 1 to ground pin is in the Mohm range, compared to 50ohm in this one, despite all of the internal diode measurements being the same
i ordered some more chips (hopefully, not fake ones)

[CapLeaker]

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

Just noticed that i mistakenly replaced the Rocp resistor with the one that i was using to discharge the primary caps (they both looked the same)
The twitching fan came back, but now the 12v doesn't come up at all
i think that the chip is probably bad
And found out that the chip from the other PSU is probably faulty; as the resistance from from pin 1 to ground pin is in the Mohm range, compared to 50ohm in this one, despite all of the internal diode measurements being the same
i ordered some more chips (hopefully, not fake ones)

That's what I figured along the line did happen. What chip are you talking about now the supervisor, the CM6800 or the STR-A6000? The STR-A6000 is always on and does make your STBY voltage. From there it goes to the supervisor and then back to the CM6800 (FAN4800) to turn the PSU fully on.

I had the impression that the other PSU you were using was actually working? Maybe I have a wrong assumption?

Now no 12V rail? How much voltage you got in the main filter cap? Check the big diodes on the 12V rail. Maybe one is shorted?

[socketa]

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

Was still talking about the standby chip
And yes, the other PSU is faulty - only 2.5V on standby
Swapping the chips didn't make any difference to that
Back to the original PSU...
320V across main filter cap, and there is two through-hole diodes (one's connected to DC+, and the other one's connected to DC-) that both test good on diode mode.
I connected a 12V 23W bulb (that draws less than 0.1A when powered by 5V) to the standby rail, and it lit up; and noticed that the standby voltage dropped to 4.4V
Then i tried it a few minutes later and the voltage now only drops by 0.02V
The voltage goes back up when the bulb is removed
So i tried the headlight bulb again, and the voltage drops to zero, and it does it's voltage pulsing thing (briefly up to 1V, and back down for the rest of the cycle) every couple of seconds.

Thinking that 50ohms between pins 1 and 3 looked suspicious, i swapped the chips over and now the headlight bulb glows and the voltage remains at 5V.
So now the standby appears to be good, although i haven't let it power the bulb for an hour
The original chip is now definitely faulty (remembering, that it was working OK earlier on) - it looks something happened between post #65 and #67 that damaged it.

i'm weary of trying to power up the PSU now, in case it blows up the chip again
although the resistance of both secondary sides of the PSU's standby outputs are the same

[socketa]

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

All of the secondary outputs give the same measurement on diode mode, and resistance mode, with no shorts
But i'm still concerned that the chip might get damaged if i try to start the PSU up, because of that burning smell that happened before.

[CapLeaker]

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

Depending on which STR-A6000 series PWM, I am guessing STR-A6059 or STR-A6069 they only do like 11W on 120V and 17W on 230V. That’s not a lot. So don’t be obsessed by loading that STBY to the hills as it isn’t made for providing large continuous current, plus it has to handle the load connected to it already on top of the light bulb. If you wanna make reliable repeatable current measurements, you need a electronic load, not a light bulb. The light bulb is just a confirmation that the PSU can handle ‘a’ load. Cold bulbs draw more than warm bulbs etc.

If all the other voltage rails are up but no 12v, the power supply should start and immediately shut down due to the supervisor FPO pin going high. Don’t get the idea to disable the supervisor right now, figure out first why there is no 12V. Either the rectifier diodes are shorted, or something is a miss with the mosfet / gate driver on the primary.

[socketa]

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

None of the main rails come up when i ground PSON
All that happens is that the FPO pin, that's on the PS229 chip, goes from 4.4V to 61mV - which, according to the datasheet, indicates that it's not seeing any problem. (probably because the main PSU isn't even working yet)
FPO is connected to the middle optocoupler, and the primary side of that optocoupler has a voltage drop across it that goes from 9v to 127mV when i ground PSON, which is good
So, yeah, that means that there must a problem somewhere in the PWM driver circuitry

[CapLeaker]

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

Poke around the CM6800 maybe VCC is missing, thus playing dead? Or this IC is bad.

[socketa]

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

VCC, pin 13 - 14V comes up when i ground PSON
PWM voltage feedback input, pin 6 - 4V comes up when i ground PSON (has now dropped to 3.4V after rechecking a few times)
PWM driver output, pin 11 - 6.9V when i ground PSON (top side of PCB sometimes made a buzzing noise when applying the probe)
Previously, I did swap the chip over with a new one that was working with the fan spinning for a bit
Should i swap it back and see what happens?
It's from China, so maybe it went faulty since then
The two main MOSFETS test good.
The APFC part has being bypassed quite some time ago, and i don't think that it's essential for the PSU to run; as i've bypassed it in the exact same chip, that's in another PSU, and it runs fine.

[socketa]

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

Well, now the standby won't light up the headlight bulb anymore
The voltage goes to zero and pulses like before
i thought that something might go wrong when i heard that buzzing sound
i removed the chip and checked it, it seems OK (it hasn't failed like the other one), but who knows?

[socketa]

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

OK,
i just checked on a good PSU to see if the bulb wasn't faulty
The bulb lit up
And i also noticed that the voltage dropped a lot when i connected the bulb, and that there was a couple of voltage pulses until the bulb lit up and the 5V came back up
So thought that this bulb might be just on the verge of the PSU's wattage limit.
So tried it again in the Zalman, and it didn't light up.
Then thought to use a proper lead with alligator clips, and to connect to the heat sink rather than a skinny wire poked into the ATX connector, and the bulb lit up
There might have being too much voltage being dropped across the length of wire, or the connections weren't tight enough.
So, yay! , all is still good with the 5VSB.

[CapLeaker]

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

Very good! Stj’s load idea is good to, but on the simple STBY circuit not necessary at all. Is the stby ic working, yes, replace capacitor, measure voltage under small load. If PSU shuts down after this either the STBY IC is faulty or overloaded. Period. There isn’t much else to it.

A buzzing or ticking occurs if something is shorted.

[socketa]

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

4V is now back on the PWM-OUT pin, when i restated the PSU today.
There is no voltage on the gate of the heatsink-mounted MOSFETs
And i have the following voltages present on the CM6800 typical application circuit (see first photo)
From there the 13.46V goes to a reverse-biased SMD diode that's parallel with a SMD capacitor (see second photo), and that goes to the small gate driver transformer that has 3pins connected together, one isolated pin, and two pins connected together
which is different to the one here:
https://smpspowersupply.com/atx-powe...html#gatedrive

Not sure what to do now
Any suggestions?

[socketa]

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

i meant 4V is back on PWM voltage feedback input - pin 6

So all of that appears to be good
The only two paths that i can see for "PWM_DC" that's coming from the junction of those two transistor emitters is via that T4 fast switching diode; but, to me, it appears to be the wrong way around, so that "PWM-DC" can't pass through it to reach the gate driver transformer, which is connected to the other side of the diode.
Or maybe it's supposed to pass through the SMD capacitor that is in parallel with the diode?
Is that diode that's in parallel with a capacitor there to allow for pulsed DC?
I checked the voltage at the other side of the diode, and there is nothing there

[stj]

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

in theory that circuit will generate a very short pulse through the capacitor each time the transistors flip
if it strains the cap it will slowly fail - is it a substantial looking cap like a film one?
also if one of the transisters slowly gets weak it wont work well

[socketa]

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

no - it's a tiny SMD capacitor
the way that i see it, the main transformer's MOSFETs won't work at all if they don't have voltage on their gates, and it seems that that is what's happening here

[socketa]

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

i used 75W as the inline dim bulb,
and noticed that it glowed a little bit on the other PSU, and even when i removed the bridge rectifier
On the psu that has the 5VSB working ok, it doesn't glow, even after removing the bridge rectifier
So i checked the 0.68uf X2 cap that is connected directly across the socket pins of both cases, and there was a bad one (measured 28nf 64ohms ESR, rather than 680nf)
The "odd" thing is that the case with the good X2 cap (measures 623nf, 2 ohms ESR) causes the bulb to glow on both boards, whereas the other case, with the bad cap, doesn't.
The bulb doesn't glow if it's a 100W
So, in this instance, it was the glowing bulb which pointed to a good capacitor

[CapLeaker]

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

That x2 cap across Power mains is just EMI filtering. Theoretical you just can take it out and doesn’t the basic running of the PSU.

[socketa]

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

I swapped back the CM6800 chip back to the original one, and the PSU now starts
So i guess that the one that i got from China was faulty
Now, it's back to starting and then shutting down in about 15 minutes or so, when connected directly to a motherboard, with the standby going to 2.5v
Seems to shut down sooner when connected to a mobo rather than just the PSU fan and a hard drive
i traced out the standby circuits and all appears to be good
i might put the PFC diode and MOSFETs back in, and see if it makes any difference, but i doubt that it will

[CapLeaker]

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

Unfortunately a lot of electronic components from China via Ebay, Ali and others are fakes. At first they were just fakes, now they are expensive fakes. I quit ordering components from these sources.