Question New CPU overheating, any ideas?

[Daniel545]

Hello, basically as the title says. After replacing the CPU from Ryzen 5 5600 for Ryzen 9 5900X the ide temps are between 80-90 degrees. The PC was working great before I replaced the CPU.
Things I have tried

• reinstalling the AIO cooler multiple times, with different thermal paste patterns. No difference in temperature overall.
• tried powering on the computer without the cooler connected to see if it could be a pump problem. Temps immediately went to 90 after couple of seconds, so I assume the pump works (maybe). I can feel some vibrations on hoses and on the pump itself, but I dont know if it is the pump spinning or vibrations are produced by other components
• There are no protective foils left anywhere

I have no idea what I am doing wrong. Is there a way to see if the pump is going bad? After examining the AIO I found out the pump is really hot, also one hose gets hot and the other remains cold, also one side of the radiator gets hot, while the other side is cold.

It is hard to believe I broke the AIO just by replacing the CPU, but I dont know what else to look for. Any ideas what this could be? Is Corsair H100X enough for cooling down a Ryzen 9 5900X or should I look for a different cooler?

PC specs:
CPU: Ryzen 9 5900x (Ryzen 5 5600 before)
Cooler: Corsair H100X High Performance
Mobo: ASUS ROG B550 F gaming
GPU: AORUS RTX 3070
PSU: Corsair RM750x
RAM: 32gb Corsair vengeance DDR4

 

[Lutfij]

BIOS version for your motherboard? How is the AIO mounted in your case? Make and model of your case? If the AIO's pump block has an air lock, you can remove the entire AIO from the case, have the pump dangling by it's own weight and then shake the block as well as the entire radiator assembly to dislodge the air lock in the pump.

 

[danny009]

I would just advice if ur running on Windows 11, disable Windows Search from Services.msc then head on to Windows Power Settings on Control Panel, choose an plan then hit advanced settings of that plan, scroll down Processor Management, reduce to %99 max and min to %5. This will turn off unneeded turbo mode of your CPU as it is pretty useless these days since CPU are fast enough.

Get HWinfo64 and track temps. This is the software I recommend tracking temps. Set up alarms in hwinfo to get to know when CPU reaches certain temp. It is an awesome tool really.

 

[Paperdoc]

Your details of the feel of hoses at the pump and at the rad says that there IS fluid flowing around the loop. However, normally the temp difference is not large. You describe HOT and COLD. That suggest to me that the RATE of fluid flow is too low, indicating possibly that the pump is running too slow.

Tell us exactly which mobo headers you have used for this cooling system. Ideally the PUMP speed cable should be plugged into the CPU_FAN header. Its is also vital that the pump's other cable is plugged into a SATA power output from the PSU. Then the two RAD FANS (using the splitter supplied) should be connected to the CPU_OPT header. If connections are made this way, the CPU_FAN header will report a "fan speed" which really is the PUMP speed. So, what does that show? It should be full speed all the time, and that is probably about 3000 RPM. The CPU_OPT header will show the speed of one of the rad fans (it cannot tell you about the other), and that can range from 500 to 1700 RPM.

There is a fine detail that MAY be important to you, but may not be - the power source to your pump should be from the SATA connection and NOT depend on a detail in the CPU_FAN header configuration. But just in case, check for this on that header. The header may have an item to set the Fan Control MODE with options of Voltage (or DC), PWM, or AUTO. Make sure this is set to PWM, NOT to Voltage or Auto. Now check the CPU_OPT header IF it is available separately, and make the SAME settings. (On some mobos the CPU_OPT header is just a mirror of CPU_FAN, so no options are offered for CPU_OPT separately.)

Now, it also is possible that there is a bubble of air trapped inside the pump and this is slowing pump action, resulting in slow fluid flow. Lutfij (above) has outlined one way to get any such air bubble out of the pump so that (hopefully) it will move to some part of the Rad and stay there.