Question New build, problem after overheating

Jan 24, 2023
9
0
10
So I just built a mini-ITX setup with the following components.

Motherboard: ROG Strix B660-i Gaming WIFI
CPU: Intel Core i9 13900K 3.0 GHz 68MB
CPU Cooler: Be Quiet Dark Rock TF 2
Mem: Kingston 32GB (2x16GB) DDR5 6000MHz CL36 FURY
HD: Samsung 990 PRO M.2 NVMe SSD 2TB
PSU: Cooler Master V850 SFX Gold Rev.2
GPU: Asus 3070

All squeezed into a Hyte Revolt 3 case.

I built this setup and ran it over a week with no issues, then one day when I came back to computer, it had crashed and was looking at a black bios screen saying overheating. I turned the computer off (had to leave) and when I came back and tried starting it, no go, it just started rebooting itself and never showing any display output.

Nothing connnected to motherboard seems to get power (keyboard/mouse does not light up etc etc).

Things I have tried so far:
Remove GPU, still no output from motherboards own HDMI or DP output.
When GPU was still in, the fans of the GPU card started running but cut of almost right away, not sure if that is normal, probably due to the MB rebooting constantly.
The Q-leds light up CPU -> DRAM then turns off, but they turn on again shortly after, it's like the board is rebooting constantly (altho it's hard to detect, fans are running continuously).
Removed the DRAM one by one, and tested them one by one in different banks, no change.
Removed the CPU, re-applied paste, and installed again, no change.
The silly LED's on the RAM's does light up still, that doesn't say much though.
Tried clearing the CMOS, no change.

So question is, what has burned up here? I have looked around and anything from the PSU, CPU, MEM or Board can be faulty, obviously these are expensive parts and guessing by buying isn't really what I would want to end up doing.

My own theory is that since there is no display out from the Motherboard, I think the board is faulty/fried? But I have also ready a lot of PSU's can give this behavior if they do not give stable power (especially the rebooting).

Any ideas or similar experiences?
 
Last edited:
Jan 24, 2023
9
0
10
Your probably right, I will go with an AIO system, but the question now is what did I fry?

The cpu should shut itself down at 90, so hopefully that did not burn, but the motherboard or psu might not be that protected.

As I wrote, the system ran fine for over a week of tough gaming, but my guess is windows hanged the cpu and thats what cause the overheat…
 
What might have changed since all was well?

Yes, the 13900K can generate lots of heat, particularly when all threads are fully loaded.
Not so much when gaming.

I suspect that somewhere along the line the cpu was set to use full power, resulting in thermal shutdown. Possibly via some automated update.

Here is an article on how to manage 13900K with a minimally capable cooler:
https://www.tomshardware.com/features/intel-core-13900k-cooling-tested/2

It shows that if your expectations are modest, you really will not lose anything by running at lower power.

Start by resetting the bios to default.
Use the clearcmos button/short, or take out the battery for 5 minutes with the psu unplugged.

For testing, take all parts out of the case. It makes changing parts easier and eliminates the possibility of a case short.

My guess is that the psu is the most likely hardware component to fail, and the only way to test is to try with a suitable known good replacement.

If you can't borrow one, buy a replacement from a shop with a good return policy.
Expect to pay a 15% restocking return fee.
 
Jan 24, 2023
9
0
10
Thank you Geofelt, yeah testing with a different PSU seems like the quickest cheapest way to start, then the more expensive parts. Since this is really my fault I doubt I can get any refund on faulty items due to the overheating. (I really should have stress tested the CPU before assuming everything was working).

I will probably implement a watercooler system, since the case supports one, and any air cooler I checked was very much on the limit of what the CPU needs, I'm just not used to run watercooler systems that's the main reason I didn't get one, was about 10 years since I built a computer last time and these AIO systems didn't exist.
 
Overheating a cpu should not cause damage.
If the cpu detects a dangerous temperature, it will slow down or shut itself off to prevent damage.
The article I linked shows reasonable performance even with a cooler similar to yours.

I would be worried that something else caused the damage.
 
Jan 24, 2023
9
0
10
I am running the M2 hd chip on the backside, not on the heatsink frontside slot, maybe that got hot too and caused issues (if not with the M2 chip, the motherboard), cause no fan on the "backside" of the motherboard ofc. Haven't tried disconnecting that. Really guessing here, trying to make the best guess to try the cheapest stuff first. But I figure if the M2 chip is dead, motherboard would still boot.

I get no display out, and the MB seems to reboot itself constantly, as I said the Q-leds on the MB when I start the CPU lights up, then quickly goes dark and switched to DRAM (yellow) LED, then finally that one turns off too, but 3-4 seconds after that, it starts over (like if the MB restarted).
 
Jan 24, 2023
9
0
10
Update:
Tried some things today, got a hold of another PSU, tried getting it to boot with that, but no luck, same issue.
Also tried de-assembly of everything, re-apply therm paste, no luck.
I moved the M2 module to the heatsinked slot on the "upper" side of the motherboard as it should be.

My next theory is the memory got fried, it would be extremely unlucky if both modules fried, but I guess they are super close so they get about the same heat. So I will buy 1 stick that I can return and test with.
I will probably also order an AIO system that fits this case at the same time.

If that doesn't work, I think I will get a new motherboard.

Will probably take some time before next update, I am traveling tomorrow so wont be able to get back to this for a few weeks, frustrating!
 
Jan 24, 2023
9
0
10
Ok, so I bought a different motherboard and an AIO cooler, still wont boot, so I guess I burned the CPU when the computer hanged, I thought that was near impossible on new CPUs. Damn this got expensive.
 
Jan 24, 2023
9
0
10
So I got a new CPU (ugh, expensive) and with a AIO the setup is now working well again, fitting a 280 AIO into the case with the GPU was tricky though.

I also did stress testing on the CPU this time, never went above 78C temp, so if comp hangs again for some weird reason, it shouldn't burn up the new CPU.

So, note to self: STRESS TEST NEW COMP BUILDS PROPERLY!
 

TRENDING THREADS