Damaged GTX 1070? Clevo P751tm1-G

Yesterday I bought a second hand clevo p751tm1-G with specs core i9 9900k 32GB 500gb SSD GTX 1070 8GB so when i went to collect it I tested it with heaven benchmark and GPU went up to 91c but I didn’t think much of it just thought I’ll get home and repast the GPU and CPU and I did that but now when I run heaven and benchmark it goes all the way up to 80c and shows black screen and freezes Then it resumes when the temps back at 56c then it repeats the same process like it doesn’t want hit more then 80c even sometimes it goes up to 78c and it does it again, I’ve tried playing few games and it crushes and freezes on some games but when I play resident evil 3 it doesn’t even crash or freeze it literally goes up to 90c to near 100c
I just don’t know what to do. I also have a clevo p771dm-g with i7 6700k gtx 980m but GPU doesn’t even reach up to 70c at all it stays very cool under-load makes no sense :/

 

Black screen is usually overheating VRM.

 

I could be wrong but I thought cutoff on some nvidia chips was as low as 87C? My heart might skip a beat if I saw 90-100C on the GPU even in a laptop. I would doubly check your die contact and ensure the heatsink is not warped?

 

What do you think I should do ?

 

I’ve checked the contact it’s fine so don’t know

 

I’d check that all the thermal pads are covering the modules and making proper contact with the heatsink.

 

I have seems to be all fine trying to upload a picture but it’s not working

 

View attachment 189072

 

Here’s a better picture

 

Could be a bad heat pipe.

 

How do I know if it’s the pipe

 

You know what they say; a picture is worth 1,000 words. Someone here might be able to spot something with a pic of the die, heatsink, overall, etc.

 

I saw similar behaviour in MSI GT70 with GTX870M. In that case it was short circuit in MXM connector, which was caused by liqiud metal.
However overheating VRM also gives such symptoms.

 

Put the heatsink hot water and check they all work properly.

 

I’m a beginner sorry can you be more specific

 

How do I know if it’s the vrms?

 

It will be hard to confirm without proper tools. One thing you can check is how well thermopads fill space between cooling and VRM components - you need to put thermopads from MOSFETs (those black rectangular things with ~8 soldering pads) upside down (to have clear, flat surface facing board side), then assembly heatsink, and then take it off and check if you can see traces of MOSFETs on previously flat surface.

 

Put the cold plate (the bit that touches the CPU) into hot water and check the other ends of the heatpipes all get equally hot very quickly. If one does not then it's broken and the heatsink needs replacing.

 

It sounds like your GPU die is not making any contact AT ALL with the heatsink.

Drop a .3mm SHIM on there and see if that sorts it.

 

A picture of the paste pattern left when taking it off would help too.

 

I did this with them via FB, hard to call.

The thermal pads - 1 looks extremely squashed i suspect the chip is not getting proper contact and heating rapid.

 

Can you post a picture here?