Aio fails to maintain 5900x at elevated temperatures
Aio fails to maintain 5900x at elevated temperatures
This seems like a typical issue, though it's quite unusual. I can't grasp what's happening with my computer. My details: r 5900x Pure loop 360 Tuf gaming x570 plus gtx 970 32 gb 3200mgz cl 14 Win 10 on SSD Crucial Bx100. I've had this machine for about six months, and initially temperatures were around 83°C. Now they hover near 85°C, but occasionally spike to 90°C and linger. After that, the BIOS restarts to protect the CPU from overheating. Recently, my CPU hit 90°C twice. I connected the AIO pump directly to the power supply as recommended, but after overheating it was switched to the motherboard. That didn't resolve anything. I ran a render in Blender using the CPU; at full load temps were about 78°C, dropping to 50% and then jumping back up to 83-85°C. When it slowed down to 50%, it would hit 83 or 85°C again. After some time, it went over 90°C and stopped. I even changed the CPU speed from 4.00 to 0.5 GHz and disabled TDP auto, but it didn't help. I'm hearing the cooling fan run nonstop, yet temps stay above 90°C. It might be related to software issues, but I'm unsure what's causing it. I've been using compressed air monthly and used Thermal Grizzly paste as a temporary fix.
It's warm but safe. I recommend swapping out the fan and changing the grease.
Check if the pump is still functioning properly. It might be stuck or there could be an obstruction preventing water flow. Temporarily disconnect the PWM pin (the fourth pin in the header) to run the fans at full speed. This should help maintain cooling via airflow. If the CPU temperature remains low while the radiator stays cool or only slightly warm, the issue likely lies in heat transfer from the CPU to the radiator. Possible causes include a faulty pump, insufficient water movement, or loss of coolant.
This action has eliminated power caps, meaning the CPU now draws more than 140W and gets even hotter. Your temperatures are high—have you turned on PBO? Also, check your pump performance in HWinfo and consider undervolting for better cooling.
I recently applied heavy undervolting which actually improved performance, but unfortunately the problem persists and remains unusual. The maximum temperatures reached 55°C. I reconnected my AIO pump to the power supply again, so the system didn’t detect it and ran at full speed continuously. However, suddenly temperatures spiked and stopped without any warning. I can’t confirm if the pump is functioning properly since it isn’t connected to the motherboard, but I tried to feel for movement and found nothing. After resetting the system, it worked again. It seems software isn’t the cause. I’ll send the AIO for repair, but I’m puzzled by this strange behavior. Sometimes I can run for up to 8 hours without issues—my power supply is solid (11A Platinum, 850W), and the pump isn’t exposed to air. The placement of the pump is noted on the attached file.