HP Probook 640 G5 Short circuit on the 19 volt line

[Rofix]

Hi everyone, I wanted to ask you for some advice
on this board: HP ProBook 640 G5
6050A3028601-MB-A01(A1)
I have a short circuit on the 19-volt line.
I initially found and replaced this faulty MOSFET, Q60C10.
Then, when I powered up the board, I had the same problem. The MOSFET failed and is giving me a short circuit.
I wanted to ask if it's possible to try to find the faulty component by injecting current, and at what point, so I can check the damaged component with a thermal imaging camera.
Thanks everyone for the advice.

[Rofix]

I did further checks and found 0 Ohm on several inductors.


.

[Rofix]

There's something wrong with the diodes underneath too.

[mon2]

Is your meter auto scaling or a manual type? If manual type, select the lowest resistance scale and then touch the meter probes together. This is the 0R ohm resistance or baseline for your meter. Subtract this value from every resistance measurement for the true resistance of the points under test.

Place one meter probe on the inductor showing above; other meter probe to the 19v rail. What is the resistance? If a low resistance, then the 19V rail has leaked onto the CPU vcore rail through a leaky high side mosfet. A low resistance to ground of the region being measured (multi-phase power supply used by the CPU and GPU rails) is often bad news. The CPU is shorted but you can test with a voltage injection and the CPU die will most likely heat up and draw the peak current set by your power supply. Always use the lowest voltage for injection = 0v8 is recommended.

[Rofix]

Is your meter auto scaling or a manual type? If manual type, select the lowest resistance scale and then touch the meter probes together. This is the 0R ohm resistance or baseline for your meter. Subtract this value from every resistance measurement for the true resistance of the points under test.

Place one meter probe on the inductor showing above; other meter probe to the 19v rail. What is the resistance? If a low resistance, then the 19V rail has leaked onto the CPU vcore rail through a leaky high side mosfet. A low resistance to ground of the region being measured (multi-phase power supply used by the CPU and GPU rails) is often bad news. The CPU is shorted but you can test with a voltage injection and the CPU die will most likely heat up and draw the peak current set by your power supply. Always use the lowest voltage for injection = 0v8 is recommended.

Hi, I use the multimeter in manual mode. The resistance between the two leads is 0.1 Ohm.
Now I noticed something strange on the 19V line and the GND. I have 71 Kohm.
If I always check the inductor shown above on the 19V rail, I get 71 Kohm.

[Rofix]

Is your meter auto scaling or a manual type? If manual type, select the lowest resistance scale and then touch the meter probes together. This is the 0R ohm resistance or baseline for your meter. Subtract this value from every resistance measurement for the true resistance of the points under test.

Place one meter probe on the inductor showing above; other meter probe to the 19v rail. What is the resistance? If a low resistance, then the 19V rail has leaked onto the CPU vcore rail through a leaky high side mosfet. A low resistance to ground of the region being measured (multi-phase power supply used by the CPU and GPU rails) is often bad news. The CPU is shorted but you can test with a voltage injection and the CPU die will most likely heat up and draw the peak current set by your power supply. Always use the lowest voltage for injection = 0v8 is recommended.

Where can I inject power to see if the problem is with the CPU or GPU?

[mon2]

You will inject 0v8 at the inductor showing the 0R ohms to ground. Select 1A for now. The 1A should peak and heat up the part that is shorted. Then increase the current to 2A-3A and confirm the same part continues to heat up. With some luck, perhaps a filter cap on the same power path is shorted and it will heat up on the thermal camera / IPA alcohol testing.

[Rofix]

You will inject 0v8 at the inductor showing the 0R ohms to ground. Select 1A for now. The 1A should peak and heat up the part that is shorted. Then increase the current to 2A-3A and confirm the same part continues to heat up. With some luck, perhaps a filter cap on the same power path is shorted and it will heat up on the thermal camera / IPA alcohol testing.

Hi, thank you for the advice, unfortunately the part that heats up is the CPU, there is nothing more to do for this card, I don't know if it's worth getting a new one