AMD FX-9590 causeing shutdown with a corsair H110i cooler

maximum_1 · Dec 11, 2016

Hello, I have recently been having trouble with my rig, as it seems to randomly shutdown. I've checked the error logs and they all tell me that the CPU is causing the shutdown. The problem seems to have started after some Windows updates, so I reinstalled Windows, but the problem persists. I thought it could be that the CPU is overheating, since it runs around 60-70°C. However, I have not made any changes to the CPU cooler and it used to run fine. I have already reapplied Thermal Paste, cleaned the radiator and fans, and have even tried running it without the case on.
Any suggestions would be greatly appreciated, thank you.

My rig:
AMD FX-9590
AsRock 970 performance motherboard
Corsair H110i GT
MSI GTX 1080
and I've tried the ThermalTake TR2 700W and Corsair AX760i as the PSU.

The Paladin · Dec 11, 2016

I've checked the error logs and they all tell me that the CPU is causing the shutdown.

would you have a copy of such a log for us to see. ?

synphul · Dec 11, 2016

Have you checked the air blowing out of the radiator, is it warm or cool? It should be warm, if it's cool air and the cpu is overheating then it may not be working properly. Try feeling the cooling hoses to see if they feel warm. Both should be warm to the touch, if not the pump may not be working. Try checking the temps in amd overdrive by looking at the thermal margin, the closer the temps get to 0 the less thermal headroom you have before overheating. Thermal protections will cause that. It could be a problem with the pump on the cooler or the cooler not receiving power.

Karadjgne · Dec 11, 2016

Those cpus use incredibly high voltages, especially in vcore. Normally this isnt so much of an issue, but you are using an AIO. This leaves the voltage regulatory circuitry surrounding the cpu at the mercy of any airflow, or lack thereof. Try zip ties on a fan to the hdd cage, aimed directly at the AIO pump. It's probably the VRM's overheating and supplying unstable voltages to the cpu, resulting in instability at such high OC, and shutdowns.

Deleted member 217926 · Dec 11, 2016

Exact motherboard model? The 9xxx series CPUs have extensive issues and aren't even compatible with many 970 chipset boards. Even on officially supported 970 chipset boards they don't do well unless underclocked.

Dunlop0078 · Dec 11, 2016

That motherboard is woefully inadequate for a 9590 at its stock clocks, a 4 phase VRM is not enough for that cpu. TJ max for that cpu is around 62c much past that and you will throttle. If I had to guess your shutdown is caused by the VRMs on the motherboard overheating.

maximum_1 · Dec 11, 2016

ASRock 970 Performance ATX AM3+ Motherboard

maximum_1 · Dec 11, 2016

Dunlop0078 :

any suggestions for alternative motherboards? I've had this one for roughly a year and a half, and this has been the first time it has had these problems.

maximum_1 · Dec 11, 2016

synphul :

The air is quite warm coming out of the radiator, but one of the tubes is quite cold (roughly 80°F with IR thermometer), and the other tube is roughly 114°F. Right now the computer isn't staying on long enough to check using on board software.

Do you know of a way to test if the pump is working?

synphul · Dec 11, 2016

When I looked that board up I get linked to the fatal1ty 970 performance which if the specs on asrock's page are accurate should do ok. It says it handles cpu's up to 220w and is an 8+2 phase vrm board, not 4 phase. Vrm overheating could still be an issue without acting cooling like top down airflow but if the cpu is actually reaching 60-70c then that's basic cpu overheating regardless of the vrm. It's likely throttling or shutting down to prevent overheating. If it used to run fine and now isn't running fine with 60-70c temps it sounds like a cpu cooling issue.

Given the difference in temps on the tubes I'm thinking there's a coolant flow problem. Maybe a weak pump, failing pump, pump speed too low, partial blockage in the radiator. A 'hot' and 'cold' side or tube would make more sense if it were like a car with a thermostat which only opens and allows flow once a specific temp is reached. Aio coolers don't have a thermostat though so the coolant is constantly flowing. It should be similar in temp along both hoses.

Something else you can try, wiggle the hoses a bit. Not enough to pull on the water block or radiator or anything. Also try tilting the case toward its side and back upright a few times to try and work any air bubbles loose if it got trapped in the pump somehow. If nothing works you may need to rma the cooler if it's still under warranty. I've seen people remove the radiator and move their cooling hoses a bit suddenly improve temps. If the cooler pump is powered by the cpu_fan header try going into the bios and making sure the cpu fan is set to full speed (so the pump is running full speed).

maximum_1 · Dec 11, 2016

synphul :

When I looked that board up I get linked to the fatal1ty 970 performance which if the specs on asrock's page are accurate should do ok. It says it handles cpu's up to 220w and is an 8+2 phase vrm board, not 4 phase. Vrm overheating could still be an issue without acting cooling like top down airflow but if the cpu is actually reaching 60-70c then that's basic cpu overheating regardless of the vrm. It's likely throttling or shutting down to prevent overheating. If it used to run fine and now isn't running fine with 60-70c temps it sounds like a cpu cooling issue.

Given the difference in temps on the tubes I'm thinking there's a coolant flow problem. Maybe a weak pump, failing pump, pump speed too low, partial blockage in the radiator. A 'hot' and 'cold' side or tube would make more sense if it were like a car with a thermostat which only opens and allows flow once a specific temp is reached. Aio coolers don't have a thermostat though so the coolant is constantly flowing. It should be similar in temp along both hoses.

Something else you can try, wiggle the hoses a bit. Not enough to pull on the water block or radiator or anything. Also try tilting the case toward its side and back upright a few times to try and work any air bubbles loose if it got trapped in the pump somehow. If nothing works you may need to rma the cooler if it's still under warranty. I've seen people remove the radiator and move their cooling hoses a bit suddenly improve temps. If the cooler pump is powered by the cpu_fan header try going into the bios and making sure the cpu fan is set to full speed (so the pump is running full speed).

I was hoping that it would be the CPU cooler since that is easier to replace than the motherboard. This one is still under warranty, so I was going to contact Corsair about that tomorrow when their chat line is open. In the meantime we ordered the H115i as a possible replacement. I have gone into bios and the cpu_fan1(the radiator) is set for full but their is still no change.

Dunlop0078 · Dec 11, 2016

synphul :

When I looked that board up I get linked to the fatal1ty 970 performance which if the specs on asrock's page are accurate should do ok. It says it handles cpu's up to 220w and is an 8+2 phase vrm board, not 4 phase. Vrm overheating could still be an issue without acting cooling like top down airflow but if the cpu is actually reaching 60-70c then that's basic cpu overheating regardless of the vrm. It's likely throttling or shutting down to prevent overheating. If it used to run fine and now isn't running fine with 60-70c temps it sounds like a cpu cooling issue.

Given the difference in temps on the tubes I'm thinking there's a coolant flow problem. Maybe a weak pump, failing pump, pump speed too low, partial blockage in the radiator. A 'hot' and 'cold' side or tube would make more sense if it were like a car with a thermostat which only opens and allows flow once a specific temp is reached. Aio coolers don't have a thermostat though so the coolant is constantly flowing. It should be similar in temp along both hoses.

Something else you can try, wiggle the hoses a bit. Not enough to pull on the water block or radiator or anything. Also try tilting the case toward its side and back upright a few times to try and work any air bubbles loose if it got trapped in the pump somehow. If nothing works you may need to rma the cooler if it's still under warranty. I've seen people remove the radiator and move their cooling hoses a bit suddenly improve temps. If the cooler pump is powered by the cpu_fan header try going into the bios and making sure the cpu fan is set to full speed (so the pump is running full speed).

Oh yes my mistake. I was going by this VRM database I linked below, they dont have the 970 fatality listed, I was assuming it would be like all of asrocks other 970 chipset boards with a 4 phase VRM. If you have a asrock 970 fatilty that should be fine, I agree your cpu may just be overheating.

http://www.overclock.net/t/946407/amd-motherboards-vrm-info-database

Deleted member 217926 · Dec 11, 2016

^ I agree, if it ran fine for over a year and is now overheating it's most likely the cooler. You wouldn't believe how many threads we see about the 9xxx series chips not being stable at stock clocks on boards that are advertised to support it though. 62C is max for that chip.

maximum_1 · Dec 14, 2016

I just installed my new H115i and it works great, the CPU has been about 40-50C and it is much quieter. Thank you for you help.

Search

AMD FX-9590 causeing shutdown with a corsair H110i cooler

maximum_1

Commendable

Dunlop0078

The Paladin

Polypheme

synphul

Champion

Karadjgne

Titan

Deleted member 217926

Guest

Dunlop0078

Titan

maximum_1

Commendable

maximum_1

Commendable

maximum_1

Commendable

synphul

Champion

maximum_1

Commendable

Dunlop0078

Titan

Deleted member 217926

Guest

maximum_1

Commendable

TRENDING THREADS

Latest posts

Moderators online

Share this page