Bitfenix Whisper 850W - Only 5VStby and PSON

[Francesc V.]

Hi:

I am trying to repair this modular ATX PSU. It only has the 5V Standby and PSON enabled (3.33V).


PSON is delivered by (ST9S429-PG14) in pin4. This chip is a Voltage controller but it doesn't seem to trigger any PWM signal to pass it to primary through coils and "move" the big mosfets. Also next to it there is a EST7618 that it also seems a voltage controller.

Not sure if in the right direction, but trying to find where PWM signal is generated.

Green section seems that could be the "PWM section". It's driven by a CM6901 (back side).


So far:

- I've checked the 3 opto couplers. They seem ok
- I've checked the 5 pieces (igbt, shotky......) from the khaki section and they read ok.

Any recommendation where to focus the analysis?

[Francesc V.]

I am focusing on the green section and the CM6901 chip. These are the readings I am getting when I switch on the PSU

Pin 1 - 5.35V
Pin 3 - 6.12V
Pin 4 - 4.51V
Pin 5 - 4.45V
Pin 9 - 2.09V (triangular clock signal)
Pin 15 - 14.67V
Pin 16 - 7.5V


Rest of pins: No voltage

Pin 7 (CSS) has no voltage.........maybe that's the reason the IC doesn't switch on?

Anything that might indicate that chip is faulty??? (no pwm readings in pins 11-14)

[прямо]

Pin 3 - 6.12V

Hmm, datasheet says max voltage for pin 3 is 5.5V so you should trace to where is that 6.12V coming from because 0.6V higher than what is allowed indicates an abnomal operation. It could have been the reason that killed the controller IC.

As for CSS, you can probably try to inject 1.5V to see if that will force the controller to switch on and send out some PWM signals. If it doesn't switch on, then it's probably dead.

[CapLeaker]

Check carefully how this PSU turns on. Measure the outputs and see if you see a blip of voltage and then shuts off. If it does, this sucker has TWO supervisor IC's to look at!

[ssrkh]

there must be some ic

[Francesc V.]

Hi:

First of all, thank you for your time and feedback. Thank you for the attached files. I didn't know about this web and it's amazing. So much appreciated. 😀
(There is no "miliseconds" output when I switch on the PSU). Primary doesn't do the job........

прямо Thanks for the comment: I injected 2V into pin 7 (CSS) that confirms that CM6901 is working (PWM signal on 13-14 that goes to gates at GP28S50Gs). It's pending to investigate why there is no voltage on that pin. Forcing CM6901 to work has shown that secondary output delivers 12V & -12V (rest of voltages are not present) and the PSU makes "tic-tic" noise due to the two reles that are running to prevent some damage there.

Also I have discovered that the other PFC controller is not receiving any voltage at pin 11 (CM6500UN)................maybe with current situation there is some mechanism that prevents it and it could be normal.

Last but not least, I've discovered a shorted N channel MOSFET that I've already removed and asked for replacement (LPW).

Silly question: Could I just put other N channel MOSFET (I only have A09T) to see whether there is any improvement in the PSU? Or it's better to wait? (see attachments for both items)


I see some light there..............maybe it's the train coming. 🤣

Regards,
Francesc.

[прямо]

I injected 2V into pin 7 (CSS) that confirms that CM6901 is working (PWM signal on 13-14 that goes to gates at GP28S50Gs). It's pending to investigate why there is no voltage on that pin. Forcing CM6901 to work has shown that secondary output delivers 12V & -12V (rest of voltages are not present) and the PSU makes "tic-tic" noise due to the two reles that are running to prevent some damage there.

Also I have discovered that the other PFC controller is not receiving any voltage at pin 11 (CM6500UN)................maybe with current situation there is some mechanism that prevents it and it could be normal.

Are you sure the PFC controller is not running? because when you forced the first PWM controller (CM6901) to switch on, you did get the 12V secondary output didn't you? and I'm pretty sure if you also force the 2nd PWM controller to switch on, you will get the rest of the secondary voltages (3.3V and 5V)

EDIT: Or you can check if the primary cap is charged around 400V DC when CM6901 is forced to switch on. If it is, then the PFC is definitely working.

Last but not least, I've discovered a shorted N channel MOSFET that I've already removed and asked for replacement (LPW).

Silly question: Could I just put other N channel MOSFET (I only have A09T) to see whether there is any improvement in the PSU? Or it's better to wait? (see attachments for both items)

Are you sure it's shorted? Did you check it out of circuit?
It may or may not work. 2N7002CK has twice VDS (60V) of the AO3400 VDS (30V) and higher RDS On.
2N7002 is a pretty generic N-Channel MOSFET. You can probably salvage one off a dead motherboard.

[Francesc V.]

Are you sure the PFC controller is not running? because when you forced the first PWM controller (CM6901) to switch on, you did get the 12V secondary output didn't you? and I'm pretty sure if you also force the 2nd PWM controller to switch on, you will get the rest of the secondary voltages (3.3V and 5V)

EDIT: Or you can check if the primary cap is charged around 400V DC when CM6901 is forced to switch on. If it is, then the PFC is definitely working.

Main cap is only charged to 300V.......APFC is not there, yet. We will need to wait when we demand power to the PSU whether cap goes up to 400V.

Are you sure it's shorted? Did you check it out of circuit?
It may or may not work. 2N7002CK has twice VDS (60V) of the AO3400 VDS (30V) and higher RDS On.
2N7002 is a pretty generic N-Channel MOSFET. You can probably salvage one off a dead motherboard.

Yeap......I started measuring Drain-Source on the PCB getting diode junction, moved from source to gate and back to source and measure didn't changed (this raised an alarm). I removed from the PCB with same result. It's dead. It's twin brother next to the picture is fine (Drain-Source is close to zero when you activate it's gate).

I had a surprise when I switched on again PSU without the faulty SMD..........I am now having all voltages but no PWR_OK signal. I have the feeling that when I replace with a good LPW, everything will be back to normal. Fingers crossed.


Other interesting things I've found during analysis......."some chips such us CM6500un are not powered up until you press PSON".

[CapLeaker]

Main cap is only charged to 300V.......APFC is not there, yet. We will need to wait when we demand power to the PSU whether cap goes up to 400V.

A PFC isnt coming on after the fact. It's gotta be working before current is drawn.

[Francesc V.]

I am still on it, but no progress so far.....

What could be the reason for having ALL and stable output voltages (but no PWR_OK signal) when "faulty" LPW is not installed on PCB, but when I put a new one (I've changed the whole pair) I only have the standby voltage in the power supply?

Things that might be interesting but I guess there's an explanation:

- Bottom LPW: I have a diode junction between D-S.........then I switch one probe to gate and go back to D-S having a short...........the expected behaviour for a N-Channel mosfet.
- TOP "faulty" LPW: I have a diode junction between D-S.........then I switch one probe to gate and go back to D-S having still a Diode junction. (this lead me initially to thought that mosfet was faulty).


I've checked all at component level........but i don't see anything obvious. :-(

Thanks in advance for any hint,
Francesc.

[прямо]

I thought PG (POWER GOOD) is from the supervisor IC? and PWR_OK is from the mainboard PCH, to sync up the rest of the mainboard power sequence??

[Francesc V.]

Today I had some spare time and I've tried it again. I finally fixed it. The rele wasn't making any noise so I've checked the components near to it. There was a tinny SMD capacitor shorted(C21) and the T4 diode (D5) was open circuit (not shorted but faulty). I replaced both and finally the SMPS came back to life.

What a pain in the ass this one.......🤣