Alienware missing 3v 5v

[jagiro84]

Hi there, after cleaning my Alienware it lost "almost" all signs of life.

Before I did anything on it while it was still working I took some notes and voltages of the main inductors and resistance measurements.

I noticed it's missing the main 3v and 5v,

Here is the current situation in easy to follow order.

- power button has 3v
- Bios chip has no voltage on its pins
- present voltages 20v , 12v, 3.9v , 3 and 5 volts in certain areas only
- first 2 MOSFETs near the 3v and 5v inductor have voltage BUT not on the gates
- BIOS chip reprogrammed (no change )

my main concern was the GPU since its the only place it has extremely low resistance to ground 0,5 to 0,6 ( it had those same values before while working ) , but in any case

- took out all 5 gpu MOSFETs just to disable it in case that was the issue, but to no avail.

- The GPU is 3070ti with i7 12th gen.

- On thermal camera the only thing showing as generating most heat is the PUS005 (marking code 3w=d0E 5 pins ) , some small heat from the EC chip (ene KB9542gf) and also small heat on the chip(marking 1225c Ti 198 a50j ) directly connected to the the 4 MOSFETs Gate pins)

Even though I have some small experience with repairs I feel pretty lost at the moment as I am still learning and would appreciate any path clearance forward , a point to the right direction

Notes, I couldn't find any schematic / boardview for this system x15 r2. The closest I could find at the moment is schematics / boardview of the Alienware M15 r3

[mon2]

Attached is the datasheet for the TPS1225C (PU301 on the pcb silkscreen) which will be supplying your always on rails for the 3v3 and 5v0 power.

Confirm, carefully the voltage to ground of VREG5 and VREG3 on this component. Post each measurement.



If you do not see these always-on rails, next confirm if the VIN pin 12 has a valid input voltage to create these LDO rails.

If yes, then check the voltage to ground of the respective ENABLE pin voltages @ pin # 20 and pin # 6. Each of these enable pins must be a logic '1' to enable the respective always-on power rails.

[jagiro84]

I have done as you instructed, in the attachment are the results of the measurements

[mon2]

My bad...on the TPS1225C, the VREG5 and VREG3 are always-on as documented in the datasheet. Next of concern is that the EN1 and EN2 lines are at 0 volts so the inductors that are linked to this regulator will be OFF. That is, the low current VREG5 and VREG3 are fine at the measured values but the heavier current rails are OFF. These ENABLE signals are most likely being driven the local EC that is onboard.

Perhaps the EC is defective? Not familiar with EC programming but check if there is a local flash device nearby this controller (or it may be embedded inside this part).

[jagiro84]

..

This is the area of the EC with 2 chips near it, I couldn't find any datasheet for the EC, maybe it's new type.

[mon2]

Looks too small to be the EC chip. Keep searching for a larger square chip on the board.

[jagiro84]

Looks too small to be the EC chip. Keep searching for a larger square chip on the board.

There seems to be no larger square one chip as can be seen on front and back of the motherboard

The chip I was referring to as EC chip is the UE5 ENE KB9542GF

[jagiro84]

Okay I think I made some progress, before connecting the PSU I pushed the BGA Ene kb9542 towards the board while measuring the voltages of alwon 3v and 5v, when I push the chip inside the voltages appear and when I let go they disappear. So I think bad connection underneath the chip, I will reflow the chip and post the results.

[jagiro84]

Hi there.

I have update

- after reflowing the EC chip KB9542gf alwon 3v 5v has returned

- GPU MOSFETs resoldered back to its original position

- system can be turned on (no display) ,

- power button light comes on with flashing error code " no memory found"

- when turned ON, under thermal camera CPU and GPU both getting hot , receiving their voltage(alive)

- under thermal camera system memory not getting online (no visible heat as when working normally)

Included in pictures voltage measurement around gpu, cpu , sys mem.


Question is, if the mem coil , (and I assume it's the one with 1v near the memory chips) is getting it's voltage, why is the voltage not being passed on to the memory chips themselves, I assume if the voltage was reaching the chips it would be visible hot on thermal camera as it was visible when the system was functioning.

Is there a similar chip system that is responsible for passing the voltage?

[mcplslg123]

Check the markings of controller connected to the coil and find out which controller is responsible for memory voltage(not gpu vram supply). It should be 1.2V for DDR4.

[jagiro84]

Check the markings of controller connected to the coil and find out which controller is responsible for memory voltage(not gpu vram supply). It should be 1.2V for DDR4.

It's ddr5 memory, it's giving 1,055 volts through the coil