HELP! MSI GT72 power issues

Recently, the PSU for my GT72 6QD turns off while i'm gaming (the LED on the charger turns off). It works flawlessly on desktop, internet browsing etc. but when gaming the PSU will eventually turn off (sometimes after 5 mins, sometimes an hour).

My laptop battery has been dead for a while now, so when the PSU turns off, the laptop has a minute or so before it also dies. If i'm able to unplug the PSU from the mains and reconnect it before the laptop dies, then i'm able to resume gaming without the game or laptop crashing.

I suspected the issue was related to my battery, so i disconnected my battery, to try the laptop on AC alone. Again, the laptop boots and works fine under light load (desktop, browsing etc.). However, the laptop now turns off immediately when a game is launched. Unlike when the battery was connected, this is immediate and does not allow for any gameplay whatsoever. Also unlike when the battery was connected, when the laptop dies the PSU now stays on.

From what I've read here and on other forums, the problem is likely either with the battery, the PSU, the GPU or the power IC unit. I've ordered a third party PSU from amazon to test (awaiting delivery) but i'm unable to find a trustworthy battery replacement. It seems unlikely to me that it would be a GPU problem, as games continue to run after the PSU dies (when battery is also connected). My technical knowledge is very limited, so i know almost nothing about power IC units and how to diagnose problems with these.

@Kevin@GenTechPC @Khenglish @hacktrix2006 @Falkentyne

I've tagged you guys as I've seen you help in similar posts. Thanks in advance for any responses! Any insight or help with diagnosis & repair will be greatly appreciated!

I'm happy to provide further information and to perform tests if instructed how to do so.

hacktrix - i noticed that you had a similar issue and discovered a fix related to an intel ME firmware update. I'm not sure how to try this fix, any advice?

Background info:

Battery - As mentioned the laptop battery has been dead for some time now. When running on battery alone, the laptop will turn off without warning. Sometimes this happens within a few minutes, other times within an hour. This does not happen at a consistent battery level (according to windows), for example the battery may state 35% or 80% when the laptop dies. The laptop will not turn back on until the PSU is connected.

Power jack - Recently my power jack failed and a friend and i were able to replace it. This appears to have been working properly since the repair. I'm not sure if this is related.

PSU - There have been a few instances where my PSU has shut down in the past. This has been very infrequent (once or twice within the last year) until now and predates the power jack repair.

GPU - over the past 6 months or so, i've been overclocking my GPU. A custom VBIOS was provided by a member of the MSI forums to allow me to do this. Since finding a stable overclock, i've not had any problems with artifacting. Again, not sure if this is related.

New monitor - I recently purchased a new monitor which has a resolution of 3440x1440, a significant upgrade to the 1920x1080 monitor that i currently game on. I tried playing a game on the new monitor with my laptop and was able to achieve around 100FPS (relatively low graphics game). However, after a short while of playing, both the monitor and laptop screen turned off and required me to do a hard reset of my laptop. After restarting and trying again, this happened a second time, except this time the PSU also turned off. This was actually the start of my recent issues with the laptop/PSU, which now occur even when gaming on my 1080p monitor.

 

Right so as you have already pulled the battery out, that rules that out which leaves GPU or Power IC,. So remove the GPU and test load after if the issue is then solved try placing the GPU back in to see if the issue returns if it does it's the GPU.

As for the Intel ME firmware unless you updated to the very latest it will not be that causing the issue.

Sent from my SNE-LX1 using Tapatalk

 

Thanks for the swift reponse @hacktrix2006.

I'll pull the GPU out and give that a try ASAP. Quick question on this though - is there a good way to test load without a GPU in? I can try to run a game but will it even run without a GPU?

The new PSU literally just arrived, so will also test that now. Its aftermarket but hopefully will be good enough to test.

On Intel ME firmware - its possible that i updated it but i'm honestly not sure. How do i check the firmware version? and which firmware version is problematic?

 

update - I just tried running a game with the new PSU, battery still disconnected. The laptop died pretty much immediately, this time the PSU also turned off (unlike with the original PSU with battery disconnected).

 

So it's the GPU

Sent from my SNE-LX1 using Tapatalk

 

Ok, will check the GPU next. Any recommendation on how to test?

Just to confirm - the intel ME firmware that i've got installed seems to be 11.8.50.3425, am i right in thinking that this isn't the latest version and, therefore, isn't the cause of the problem?

 

IME is fine. GPU check is required

Sent from my SNE-LX1 using Tapatalk

 

I've not yet physically removed the GPU but the laptop has a button that disables the GPU and switches to integrated graphics. I just tried to run a game using integrated graphics and it runs without the laptop crashing (albeit at 6 FPS).

Does this confirm that its a GPU problem? Or could it still be the power IC unit (or something else)? I'm guessing that disabling the GPU reduces the power requirement, so could also reduce the symptoms of other, power related, issues. Is there anyway to confirm what is causing the problem?

 

What GPU do you have.

You also mention that you have a modded vbios have you tried to see if a stock vbios resolves the issue?


Sent from my SNE-LX1 using Tapatalk

 

I've got the stock gtx 970m. I'll try flashing the original vbios today, to see if that resolves the issue.

I've been using the laptop in integrated graphics mode for a couple of days now with no problems. However, this morning my laptop began doing something strange. When i turned it on, I could hear (what i presume to be) the fans starting, then a clicking noise and the laptop fails to start. It does this over and over again, without me pressing the power button again, until the MSI logo eventually appears on the screen. The laptop then boots as normal.

 

Physically remove the GPU from your system.

Sent from my SNE-LX1 using Tapatalk

 

Just tried the stock VBIOS but that didn't help. Will physically remove the GPU when i get a chance.

I've noticed something else strange over the last couple of days. For the past 6 months or so, i've been using a thunderbolt 3 hub with this laptop. The hub behaves strangely when connected to this laptop but works absolutely fine with other laptops. Since disconnecting my battery a few days ago, the thunderbolt 3 hub has been working flawlessly with this laptop. Not sure if that's related to the issue that i'm having but thought it might be worth reporting, just incase.

 

Might be time to get another laptop TBH.

Are you using the original PSU btw?

Sent from my SNE-LX1 using Tapatalk

 

Unfortunately i don't have the budget for another laptop at the moment. If i'm able to identify the problem then i'll either repair or replace it, if possible.

Yes, I'm using the original PSU. I tried an aftermarket PSU yesterday but had the same issues, so went back to using the original.

 

Gdgd, as you remove the GPU look at the SMD's around that area and the CPU (Easier to remove both heatsink just to check VRM area on the CPU side as well.



Sent from my SNE-LX1 using Tapatalk

 

I'm removing the GPU card now. I wanted to share some pictures while I've got the laptop open, incase I need to further investigate/photograph/test any components.

I can't see any visible damage to SMDs around the CPU or GPU. Attaching pictures.

 

Will look later.

Sent from my SNE-LX1 using Tapatalk

 

Ok thanks. I'll hold off on reassembling the laptop until you reply.

What will I learn from removing the GPU card by the way? I already know that the laptop works when the GPU is disabled. Maybe it's worth reseating the GPU card instead?

 

The images are showing broken, did you upload directly to this site?

 

No I didn't, the files are too large. Is there an alternate site I can host on? Or I can email them to you?

 

Place them on something like mega for me and then link me the folder that the files are in.

 

Try these links:

https://ibb.co/51nT0gF

https://ibb.co/4mQg12D

https://ibb.co/nr8FVTW

https://ibb.co/GpwvJL2

https://ibb.co/cYK4h72

https://ibb.co/LxGwqM3

https://ibb.co/txv7Lg3

https://ibb.co/xCtDcfH

https://ibb.co/nmKbq4h

 

Right looking now.

 

GPU: Section c17 looks like something is going on there, looks like some kind of corrosion give it clean, D504 on the back of the card doesn't look good, there is some browning on the PCB see if you can clean that off or if its a burn mark.

Mainboard looks ok, a bit of dirt here and there but power rail seems fine as far as i can see.


Other then that it needs a good clean in alcohol.

For now keep the GPU out the mainboard for now and see if it stabilizes on that.

Replace the pads if you can as well as clean the heatsinks.

 

Ok, will try and clean those parts and will send you some more pics when they're clean

 

Thank you.

 

I see what you mean about the browning around D504. I'm assuming you mean section C7, as I couldn't find a C17.

I've tried to clean up both of those areas and have taken some closer pictures (some before cleaning, some after). I've also included pictures of the SMD on the other side of D504, which seems to have written on it:
87350D
TI553
ZCRT E3

Pictures here:

https://ibb.co/Vghn5zV

https://ibb.co/L1Pj37S

https://ibb.co/DLyYFy5

https://ibb.co/YNjRYpd

https://ibb.co/y6ZG8Hn

https://ibb.co/nj8sQ5Z

https://ibb.co/Mpbv23n

https://ibb.co/kGqhMjK

 

C17 is top near vrm.

Sent from my SNE-LX1 using Tapatalk

 

Do you mean c77, shown in the picture below?

https://ibb.co/pv16m5y

If not, I really can't find C17 anywhere! Any chance you can mark it in the picture I sent you and send it back?

 

Yes, however it looks ok. Your pads on GPU need replacing as they look busted.

Sent from my SNE-LX1 using Tapatalk

 

What do you mean by pads? Thermal pads?

Do the close up pics all look ok then? C7, C77 and D504?

 

Yes thermal pads, C77 looks fine.

D504 area seems either to be burnt dust which should clean nicely or burnt PCB.

Also I noted on some of the contacts there is a blown/yellow gloop on solder joints for the MOSFETs/inductors the R solve squares not even sure what that is all about.

Give it a good clean with some alcohol isopropyl.

Sent from my SNE-LX1 using Tapatalk

 

I've reassembled the laptop without the GPU card. On first boot, it seemed to initially fail, then boot as normal (the same as i have previously reported, before removing the GPU). Other than that, the system seems stable. Though it was stable with the GPU card installed anyway, as long as it wasn't enabled.

I'll clean the GPU card tomorrow and reinstall it.

As for B504 - the close up pictures were taken after I attempted to clean it. As the brown wasn't removed, does this mean it's burnt PCB? If so, what does this mean?

 

There is a member in here more suited to PCBs and electronics more then me.

@Khenglish are you able assist here bud.

Also you said it initially failed, you mean rebooted 3 times right if so that's normal.

Also I am solidus1983 on Reddit btw I noticed your post on there.

Sent from my SNE-LX1 using Tapatalk

 

In the past few days, sometimes it has rebooted three times before starting, yes. However, now it only reboots once before starting. I think this started after I disconnected the battery. Just like in this video, except mine eventually boots:

Haha sorry for bothering you on there then!

 

Reconnect the battery, also check your CMOS battery as well normally the 3 reboots is the system flipping from dGPU mode to iGPU mode. This might also be why it does it once.

You can also go into the bios and load defaults save and quit then go back in and set it how you had it, makes sure that the nvram is clean from corruption.

Sent from my SNE-LX1 using Tapatalk

 

@Khenglish - I've been following the diagnostic process that you instructed @Darker01 to take on this thread.

I firstly tested resistance between the two giant pins on the front of the mxm slot. I got a reading of roughly 110, when testing at 200k Ohms. What does this mean? Until i tested that I was convinced i had a mosfet issue, but does this reading rule that out?

 

I decided to physically remove the battery yesterday, as its been dead for a while, regardless of whether its causing this issue or not. Some of the surrounding tape ripped, so probably best not to reconnect it. I've ordered a new battery from Amazon. Unfortunately, i could only find chinese aftermarket batteries, but i'm hoping it will be good enough to test with, until i can get hold of an OEM.

How do i check the CMOS battery?

I'll try loading the default BIOS settings when i next get a chance.

 

Check the CMOS battery with a volt meter, just to make sure it near the rated voltage on the battery cell.

As for getting a battery of amazon you might have the same issue as I did where it would just continually charge even after it was at 100% I ended up getting battery from MSI Directly.

Sent from my SNE-LX1 using Tapatalk

 

I have just reset BIOS settings to default. I'm unable to test whether this fixed anything until i reinstall the GPU, as the laptop runs smoothly without the GPU anyway.

I will try to test the CMOS battery when i open the laptop up again to reinstall the GPU. I'm hoping to hear from Khenglish or Darker01 before doing so, so that i can run further tests on the GPU while its out of the laptop.

On the aftermarket battery - yeah, i don't have particularly high hopes for the Amazon battery. Hopefully it will allow me to confirm whether the battery was the problem, or eliminate it as the cause of the problem. If the issue turns out to be battery related, i'll return the Amazon one and buy one from MSI directly.

 

The new battery has arrived. I'll be reinstalling the GPU card and installing the new battery later on today.

Before reinstalling the GPU, i wanted to carry out some tests on it. I've tried to carry out the tests suggested by Khenglish on this thread. The tests i carried out were as follows:

• as mentioned above, when testing resistance between the two large pins on the right hand side of the mxm slot (supply voltage) - 110k ohms
• when testing resistance across each of the caps in the row of 6 caps - 3.5 ohms
• when testing resistance across each of the caps in the row of 2 caps - 11.3 ohms
• when testing the resistance between the bottom side of the 6 caps (core voltage) and the furthermost right pin on the mxm slot - 110k ohms
• when testing resistance across each of the power FETs - seems to vary, depending on the pins that are tested, but all readings are in the thousands of ohms.

It seems to me that none of these readings suggest a problem (but my interpretation could easily be incorrect). Does this mean the issue isn't with the GPU?

 

Glad to hear battery has arrived as for the electronics side of things I don't know what you just said. I'll leave that response to the experts.

Sent from my SNE-LX1 using Tapatalk

 

Yes those all sound correct.

 

Update: Reinstalled the GPU card, applied new thermal paste and changed the battery.

I tried gaming before connecting the battery, to see if cleaning/reseating the GPU had any effect. To my surprise, I was able to run a game for 2 hours on AC alone. However, after 2 hours the laptop turned off mid-game. Prior to this, the laptop died immediately after a game was booted, whenever the battery was disconnected.

I connected the new battery and was able to run a game for 20 mins or so without the laptop/PSU dying. I've not had a chance to stress test though (i.e. several hours of gaming). I'll try that tomorrow.

Given the GPU reseat has had some effect, it seems unlikely at this point that the issue is related to the battery.

@hacktrix2006 @Khenglish - what would you recommend testing next? To recap - so far i have:

1. reinstalled the stock VBIOS - no effect
2. removed the battery - instead of PSU dying mid game (and laptop/game remaining on), laptop began dying immediately after booting game (and PSU remains on)
3. tried alternate PSU (with same specs as original) - no effect
4. disabled/physically removed GPU card - laptop is stable (but its always stable outside of gaming)
5. tested the resistance across various parts of the GPU card - all seems normal
6. restored default settings in BIOS - no effect
7. cleaned GPU with alcohol, reinstalled GPU card and applied new thermal paste - laptop is now able to run a game for some time while on AC power, however, it still turns off eventually
8. tried alternate battery (aftermarket) - laptop is able to run a game but have not tested over a long time period

How about testing the power management on the motherboard? Is this worth doing and, if so, can you instruct me on how to test it please?

 

Are you monitoring temps etc?

Use HWinfo and make it save to a csv file. There has to be a reason why the laptop is shutting down.

 

I've been keeping an eye out on temps using Dragon Gaming Center and GPU-Z and I haven't seen the GPU reach even 70 Celsius. I'll install HWinfo, as you've suggested, and report my findings when stress-testing with the new battery tomorrow.

 

Update - I managed to use/stress-test the new battery today (with AC also connected). I found that:

• The laptop still cuts off after a few minutes on battery power (just on desktop, not gaming) - This happens with no warning, just like with the original battery. Perhaps my original battery isn't dead after-all. Could this be a manifestation of the wider issue?
• The laptop is unable to complete battery calibration due to above issue. The laptop dies as soon as it enters the "discharging" phase of the calibration process.
• the laptop and PSU turned off when running the heaven benchmark. This happened about 20 mins in.

This shows that, despite my ability to game for longer since reseating the GPU, the problem still remains.

I've attached HWinfo reports that were taken during each of the above tests and one which was recorded while i ran a game for an hour without a crash (can be used as a baseline). These can be found here:

• baseline (no crash while gaming)- https://mega.nz/file/gtQFUAgB#M0e5vY2_sxml68Ry40m4JgFBjx0wK7xL2wrjkGjrpDU
• battery prematurely dying - https://mega.nz/file/8pZTEC6L#SsiBqR3ElPPEH8pwsGK8krEqz43KjnfE1AbWhv4lKBI
• battery calibration crash - https://mega.nz/file/ooY3kYBB#5OzWYPdhBMDxpoc7A0Y0tdcJsvROu5sSTbX4N3211AE
• gaming crash (while on AC & Battery) - https://mega.nz/file/soIVTAaB#-bvC6rQa3NfYeCVcyobt1i-9T9BldUZ1Ra2YffTCYoM

 

Will have a look when on my laptop again.

Sent from my SNE-LX1 using Tapatalk

 

Right looking at your battery ones i noticed that your laptop when from 12.58v to 12.18v within 5 minutes which is not good "report_7 (desktop on battery only -laptop crashes in less than 5 mins)", however everything else looks fine PCH, CPU and GPU.

"report_8 (heaven benchmark on AC&Battery - laptop and PSU crash within 20 mins)" looks like you ran this before battery was fully charged, however PCH, CPU, GPU looks fine.


"report_5_(battery calibration - laptop crashed on discharge, PSU remained on)" yet again you battery voltage is all over the shot here but PCH, GPU and CPU are fine

"report_4(gaming with AC& battery - no crash)" the battery voltage is showing something completely different to the others however temps are fine with PCH, GPU and CPU.

So what i would suggest is to first run in a CMD (Admin Mode)

Code:

powercfg.exe /batteryreport /output "c:\battery_report.htm"

zip that file then upload that here whilst your on AC and fully charged, then fully discharge the battery and then do an EC Reset (After laptop power off unplug the AC and then with a paper clip press in the little hole which has a Battery ICON for 30 seconds), then fully charge the battery again whilst the laptop is switched off still then re run the above command again and reupload the battery report again.

The fact that the Battery Voltage dropped so quick would more then likely cause the EC firmware to panic and shutdown your laptop.

 

Thanks for all the help hacktrix, i really appreciate it.

• here is a battery report taken just now, on AC power with the battery at 99% (doesn't seem to be increasing) - https://mega.nz/file/VowjCaYR#0zGMHUWbT9aOjr3oNvet903gp1OGJADDEoCxCnUjXWU
• here is a battery report that i created last week, with the original battery (unsure what % the battery was on at the time) - https://mega.nz/file/1thhjSRJ#HEep2V1aB-PVX-8jZYcO1v4TJ-Tu_fdbaxv9bdWnDmE

I did an EC reset when i first installed the new battery, before booting. I'm not able to fully discharge the battery, as the laptop switches off within a few minutes of disconnecting the AC. I can try doing an EC reset at that point though.