Asus G53sw Power Issues

Greetings All,

I'm wondering if anyone has some insight for me into what's wrong with my laptop (Asus G53SW-XA1 which I recently received in pieces but carefully put back together):

During, or just after Windows loads, the laptop just powers off. No shutdown sequence, but (until I pulled the PMW fan pin) just before the laptop powers off, the fans sound like they're slowing down, and then 'poof' it shuts off. The funny thing is, the laptop can run FOREVER so long as it's not in Windows, so for example, the 3 hours it took to run memtest86+ the machine ran flawlessly, with the memory all passing.

Here's what I've done for testing:

- Battery in or out, it makes no difference.
- Pulled the PMW pins (blue wire) on the fans, so they always run at 100%
- The fans have been cleaned and you can feel a decent amount of airflow coming from them
- Cleaned and reapplied quality thermal paste to CPU and GPU, and ensured the the GPU's memory chips and VRMs had proper contact with the other thermal "goop"
- Cleaned the MXM 3.0b contacts on the GPU card
- (As mentioned above) tested the memory
- Tested the CPU (LINPACK)
- Tried different Windows installs on both an SSD (Windows 10 - Samsung EVO 850) and HDD (Windows 7 -Seagate 500GB)
- In the short time it's on, I have never seen any video artifacts (which would typically lead me to believe the GPU is pooched)
- It doesn't appear to be related to heat, as I have had times where one Windows startup will last 10 seconds, with the next surviving two minutes

Ultimately, it's making me think it's either the motherboard or the video card. Does anyone have any ideas of the behaviour that would point to one or the other? I'm trying to spend as little as possible to get this thing functional, and would rather not purchase a part I don't really need.

Your thoughts, experience, and wisdom would be greatly appreciated!

 

For anyone interested - assuming there is someone - here's a small update:

I replaced the motherboard, but the issue still persists (shutting down at random). With this in mind, it's likely the video card, and I am in the process of upgrading it to a GTX 770M. I will let you know how it goes when I've completed the upgrade!

 

Update - I believe I have successfully installed the GTX 770M into the laptop. Modified the vBios and can boot to windows. Modified the .inf driver file to install the drivers.

IT STILL RANDOMLY SHUTS OFF.

I have also pulled out every other component from the motherboard to ensure it's not some bad peripheral. No dice. Aside from the LCD screen, the one thing remaining is the CPU. I've gone this far. I'm going to see if I can find a cheap CPU.

 

Okay, so another update - grabbed an i7-2760QM and put it in. The result? IT STILL RANDOMLY SHUTS OFF. This thing is cursed. I'm down to the LCD, or the SATA connector assembly (which totally didn't dawn on me that it could be an issue until now). Tonight I'm going to try disconnecting both the LCD and SATA connector assembly (and plug in the drive via USB to SATA dock) to see if that makes any difference.

The only other thing I can think of is that there is some incompatibility between the motherboard and the SSD (Samsung 850 EVO) I'm using - I have tested with another drive (Seagate 500GB HDD) but I realize now that the power connector was VERY fickle on the original motherboard and perhaps I misinterpreted a loose connection with the problem I'm experiencing now when I was testing with the other drive.

 

If you're comfortable with it, try running the system outside of the chassis with the bare minimum connected (one memory module, CPU, GPU, and an external monitor). Make sure you're grounded and that you minimize any potential for ESD.

I haven't worked on this system, so I don't know the guts very well. If there are any daughter boards, don't connect them. You just want said essentials. If the power button is connected via a ribbon cable to a separate board, connect that for the sake of powering on the system and see where things take you. If it works, then you can go through the meticulous process of reconnecting other parts one by one until you can reproduce the issue.

Hopefully you can isolate the problem.

 

Thanks for the suggestion - I hope to try this tonight - I recently tried disconnecting the internal LCD, disconnected the SATA connector assembly, AND tried other drives, but to no avail, the issue still persists.

Stay tuned!

 

*Braces for impact*

It is a weird issue, for sure. Good luck!

 

Finally - a positive update! The CPU was overheating, or I should say, about half of it was - the previous owner appears to have (accidentally) bent the heat pipe of the CPU heatsink, so that when it was installed there was so much pressure coming from the fin assembly that one side of the CPU heatsink contact plate came off of the CPU, and wouldn't cool it properly! This was only discovered after I pulled the board out, and no chassis screws were forcing the heatsink into this nasty position. Thanks to Prostar for the suggestion!

I need to bend the CPU heatpipe back - it's off by about 10 or 15 degrees - does anyone have any tips on what to do, or not to do before I begin?

Thanks!

 

Ooh, that's going to be tricky. Heat will make it slightly more malleable; maybe a heat gun will help here(?).

If the pipe is damaged, hopefully the vapor inside is okay. You might be better off replacing it if that ends up being the case.

 

For the small cost of $5 plus a bit of shipping, I decided to go the new heatsink route - things are good!

The laptop has been put back together (for the most part... the new heatsink for the 770m doesn't quite fit properly, implying that I have more work to do on the chassis), and some basic stress tests have shown that the CPU isn't hitting anything higher than 75C.

I still have some concerns surrounding the cooling of the 770m - only one half of it has a heatsink touching the main GPU chip and memory chips, but the VRMs, and the memory chips on the other side of the card don't have any contact with a heatsink yet. The fan for the GPU heatsink also doesn't properly align with the fin assembly, and while the GPU has only ever been as hot as 35C (slightly more than basic use), I'm concerned with really stressing it until these issues are resolved.

This has been WAY more work (and WAY more expensive) than I was expecting, with more work to be done - but thankfully the issue that started this whole thread has been resolved!

Thanks again to Prostar for the help!

 

Wow wish I would have been around to suggest to buy a infared thermometer and check the heat transfer of the heatsink. I had a similar issue a few years ago working on a dell. The heatsink looked perfect but wasn't transferring heat to the fins. The computer was brought to 2 other computer stores before it got to me, so one of the other techs might have damaged it.

Glad you figured it out.