Scorching hot mosfets on Asus X555Q motherboard - can't find the cause

[rddube]

Hello everyone,

I am reading temperatures above 105C (221F) on mosfets PQL8104 and PQL8103 and the board is operating fine until it goes above 105C and then it shuts down. Cpu is ok.

So can't figure it out. Attached is photo of motherboard, the schematic section showing the culprit mosfets and the whole motherboard schematic. I changed PL8101 thinking it may be bad, same results. I tested out of circuit caps PC8117 and PC8118 and they measure fine.

So I am at a loss to know what else to check to try and identify the source of the problem

Any help would be appreciated.

Many thanks!

[SMDFlea]

Post a link to the schematic please,don`t attach to your post.

Reminder: rules and organization of this section - https://www.badcaps.net/forum/troubl...f-this-section

[Sephir0th]

Hello everyone,

I am reading temperatures above 105C (221F) on mosfets PQL8104 and PQL8103 and the board is operating fine until it goes above 105C and then it shuts down. Cpu is ok.

So can't figure it out. Attached is photo of motherboard, the schematic section showing the culprit mosfets and the whole motherboard schematic. I changed PL8101 thinking it may be bad, same results. I tested out of circuit caps PC8117 and PC8118 and they measure fine.

So I am at a loss to know what else to check to try and identify the source of the problem

Any help would be appreciated.

Many thanks!

To simply say "CPU is good" is not sufficient and possibly not the correct assumption in addition.

There must be a reason why the the buck converter is getting hot. And if it doesn't have something to do with the buck converter then it originates from the load (APU).
A Inductor almost never fails end even if so, it is easy to confirm. More or less the same applies to ceramic capacitors. It is a absolute rare case that they will fail without a straight internal short of the layers. However, it doesn't sound like you have ruled out the High-Side, Low-Side and the VRM Controller for the APU so far have you?

[rddube]

[


QUOTE=Sephir0th;n3231539]

To simply say "CPU is good" is not sufficient and possibly not the correct assumption in addition.

There must be a reason why the the buck converter is getting hot. And if it doesn't have something to do with the buck converter then it originates from the load (APU).
A Inductor almost never fails end even if so, it is easy to confirm. More or less the same applies to ceramic capacitors. It is a absolute rare case that they will fail without a straight internal short of the layers. However, it doesn't sound like you have ruled out the High-Side, Low-Side and the VRM Controller for the APU so far have you?[/QUOTE]

When I say CPU is good, I meant the temps seem to be in the correct range, sorry my bad English.

Still trying to understand this (no that experienced) how would I test de Low-Side and the VRM Controller for the APU? Help please?

Posted a link to the schematices as requested. Thank you!

[mon2]

In addition to the above suggestions…This rail is for the GPU so it is a pig on the current draw. There are better mosfets that are drop ins from Texas Instruments that will run cooler but investigate the capacitors on this rail. Perhaps they are shorting and need to be replaced. Capacitor fatigue. Do review both sides of the logic board to hunt for out of specification parts on this rail.

[rddube]

In addition to the above suggestions…This rail is for the GPU so it is a pig on the current draw. There are better mosfets that are drop ins from Texas Instruments that will run cooler but investigate the capacitors on this rail. Perhaps they are shorting and need to be replaced. Capacitor fatigue. Do review both sides of the logic board to hunt for out of specification parts on this rail.

Ok, will check the caps, I guess these here could be good suspects? Just need to find them as the board itself is not numbered.

[Sephir0th]

[

To simply say "CPU is good" is not sufficient and possibly not the correct assumption in addition.

There must be a reason why the the buck converter is getting hot. And if it doesn't have something to do with the buck converter then it originates from the load (APU).
A Inductor almost never fails end even if so, it is easy to confirm. More or less the same applies to ceramic capacitors. It is a absolute rare case that they will fail without a straight internal short of the layers. However, it doesn't sound like you have ruled out the High-Side, Low-Side and the VRM Controller for the APU so far have you?

When I say CPU is good, I meant the temps seem to be in the correct range, sorry my bad English.

Still trying to understand this (no that experienced) how would I test de Low-Side and the VRM Controller for the APU? Help please?

Posted a link to the schematices as requested. Thank you!

To check a Mosfet is a simple task. You simply set the multimeter in resistance mode and measure between Source and Drain, Source and Gate & Drain and Gate. Pay attention, that the low-side Mosfet will measure low resistance between Source and Drain by design. The advise would be to remove the Mosfet from the circuit to perform a reasonable measurement. No value should be below KiloOhms.

It would be helpful to know the resistance to GND of +VDDCR_VGFX and +VDDCR_CPU.

[rddube]

To check a Mosfet is a simple task. You simply set the multimeter in resistance mode and measure between Source and Drain, Source and Gate & Drain and Gate. Pay attention, that the low-side Mosfet will measure low resistance between Source and Drain by design. The advise would be to remove the Mosfet from the circuit to perform a reasonable measurement. No value should be below KiloOhms.

It would be helpful to know the resistance to GND of +VDDCR_VGFX and +VDDCR_CPU.

Hi, Mosfets were checked out of circuit and they measure over kiloohms. I also have M3058 that I tried instead of the M3056 with the same overheating results. GND to +VDDCR_VGGX is 42.5 ohms and GND to +VDDCR_CPU is 32 ohms.

[Sephir0th]

So everything checks out as expected. This is good, and this is finally ruled out. In my opinion the next step would be to inspect PU8101 and its surrounding circuit.

However, how did you confirm where the heat spot is located? I mean with what tools?

[rddube]

So everything checks out as expected. This is good, and this is finally ruled out. In my opinion the next step would be to inspect PU8101 and its surrounding circuit.

However, how did you confirm where the heat spot is located? I mean with what tools?

Hi Sephir0th,

Can't seemto find PU8101, maybe I don't have the right schematics? Also, I am attaching a photo of the back of the MB and the circled mosfet M3054 also heats up as much as the front ones. When I turn of the motherboard, the resistance between gate and source is 2.1 ohms?

I replaced that mosfet, but get the same result.

As for heat, I have a thermal camera that I use with a macro attachment, gives me perfect readings.

Sort of lost now that I can't find PU8101??

[rddube]

Checked all the caps that seem to be on the +VDDCR_VGFX and +VDDCR_CPU rails, and didn't find any suspect. None of them seem to be shorting and I examined them with a microscope and didn't see any suspects. As for the circled mosfet on my previous post, I was mistaking because it is right under the 2 original mostfets that heat up, so this one is fine..it is the 2 on the top of the motherboard that get blazing hot.

[rddube]

Well well, I found what the issue was after going over the whole board and I thought I'd share it with others so that if anyone has this same issue, they'll have the solution. The solution was the power adapter not pumping out sufficient amps, so it would "starve" the motherboard from current, and the mosfets would try and compensate and heat up. When I found the issue, used another power adpater and bang, everything back to normal. Will now troubleshoot the power adapter. Cheers!