ASRock K7S41 issue

[alindumitru46]

I have that board for few years ago. She died suddenly and there was not a way to revive.
That evening I found it in a box by chance, looking for a replacement mobo for someone.
Thanks to the knowledge gained here on the forum, I immediately found the cause: VRM Q4 was dead.
I noticed that the two areas in figure remained united after vrm desolder. Is there a risk if the two zones are united? I tried to clean the area, but my soldering iron is pretty skinny.
VRM Q5 is 70L02H, what is best equivalent that I can use?
Thanks.

[ChaosLegionnaire]

Re: ASRock K7S41 issue

i think that could be just the body pad for the mosfet to dissipate its heat through the pcb. if in doubt, just use a good branded low wattage psu with known working short circuit protection just in case...

[alindumitru46]

Re: ASRock K7S41 issue

I would have tried anyway, but I thought to ask before doing fireworks before Christmas
There could not be a big loss: a weak Duron CPU, some RAM, in the worst case and a PSU
The only thing that keeps me on hold is the equivalent for 70L02H .

[ChaosLegionnaire]

Re: ASRock K7S41 issue

not sure how to check if its a short when the solder is bridged together. im still quite nub on component level repair.

but what u can do i think is to put the multimeter in diode test/continuity mode and put one probe on the solder pad and the other probe on the positive leg of a cap on the vrm high or low side depending on whether that mosfet is a high or low mosfet.

then put the probe that was on the positive leg onto the negative leg of the cap and see if u get continuity on one or both legs with the mosfet solder pad. if both legs get continuity, the positive and negative side of the circuit may be shorted together so u must use a wick, solder sucker or whatever to get rid of the solder bridge.

if u get continuity to only one leg of the cap, it should be fine and shouldnt cause fireworks on power on, i hope...

[alindumitru46]

Re: ASRock K7S41 issue

Bravo, young man, you're shrewd. Eventually it was logical that, but for now I have not bothered, because I do not have the component to replace.
Anyway have you spared me for an extra work
Thanks.

[alindumitru46]

Re: ASRock K7S41 issue

3 months later...
I found an AP85T03GH as replacement for 70L02H:
http://alltransistors.com/mosfet/crs...009&caps=TO252
I checked before replacement and after the origial mosfet Q3 70L02H and the replacement AP85T03GH : both has ~600mohm trough Gate and Drain but both fail the continuity test after soldering.
I really don't know...

[cpt.charlie]

Re: ASRock K7S41 issue

Maybe your problem is related to the VRM controller, remove the mosfet and test it out of circuit.

[alindumitru46]

Re: ASRock K7S41 issue

Vrm controller is the ST L6910 ADJUSTABLE STEP DOWN CONTROLLER
WITH SYNCHRONOUS RECTIFICATION with 16pins, between LPT port and 462 CPU socket?
Must verify the pinout voltages after power on?
I said already about the test of the both vrm: Q3 and Q4 out of the circuit and after solder...

[cpt.charlie]

Re: ASRock K7S41 issue

That seem to be the VRM controller, what a crappy board! socket A boards used to have 3 phase VRM modules. if you have a shorted phase your board won't start at all.

BTW, in your photo there are 2 black caps near the northbigde that looks bloated, check them.

[alindumitru46]

Re: ASRock K7S41 issue

I used the last two pictures from internet because have better resolution, so the capacitors are ok .
I'll check in the next weekend.
Thanks anyway.

[alindumitru46]

Re: ASRock K7S41 issue

My BIG mistake!
The VRM controller is not L6910 but ST L6916D!
And I cannot find datasheet.

[cpt.charlie]

Re: ASRock K7S41 issue

This is all i can find.

[alindumitru46]

Re: ASRock K7S41 issue

There is a solution for a similar issue with a different mobo - ASRock P4V88.
The solution seems to be too complex - osciloscope needing?!:
https://translate.googleusercontent....GEnOHO6N-U_ETw

[momaka]

Re: ASRock K7S41 issue

Thanks to the knowledge gained here on the forum, I immediately found the cause: VRM Q4 was dead.

Is this in circuit or out of circuit?

If the MOSFET is shorted out of circuit, then it most certainly is dead.
BUT! if the MOSFET is not shorted out of circuit and ONLY in circuit, then that's different.

It looks to me that MOSFETs Q2 and Q4 form a synchronous buck pair, with Q2 being the "upper" MOSFET and Q4 being the lower MOSFET. Thus, Q2 Drain is likely connected to a PSU power rail (3.3V/5V/12V) and Source connected to the inductor close to it (the output voltage of the buck rail... i.e. could be the CPU V_core or Northbridge Vcc). Meanwhile, Q4 Drain is connected to the same inductor (output voltage) and Source connected to ground.

With that said, in many cases where a buck converter is used for powering a CPU or Northbridge, it is typically NORMAL for the lower MOSFET to show very low resistance (less than 50 Ohms) or even a "short-circuit" between Source and Drain.

While on this topic, I shall note that Q1 and Q3 form another sync. buck rail, with Q1 likely being the upper MOSFET and Q3 the lower one. You can tell by checking which MOSFET has its Drain connected to 3.3V, 5V, or 12V rail on PSU.

I noticed that the two areas in figure remained united after vrm desolder. Is there a risk if the two zones are united? I tried to clean the area, but my soldering iron is pretty skinny.

No risk.
The two solder areas are actually united underneath with a copper trace.

3 months later...
I found an AP85T03GH as replacement for 70L02H:
http://alltransistors.com/mosfet/crs...009&caps=TO252
I checked before replacement and after the origial mosfet Q3 70L02H and the replacement AP85T03GH : both has ~600mohm trough Gate and Drain but both fail the continuity test after soldering.

I'm not sure if I am understanding this correctly...
Are you saying that you have a short-circuit between Gate and Drain on a MOSFET while it is on the board??