I've been using an i9-13900K in an Aorus Z790 Elite AX Rev. 1.1 for about two years with a mild undervolt: -0.075 V core, -0.050 V ring, low LLC, PL1 and PL2 at 253 W, and a 307 A current limit. With a Corsair H150 RGB, Cinebench R23 loads normally reached about 88–92°C, with the P-cores around 5.3 GHz and E-cores around 4.1 GHz.
A few days ago I installed a Thermalright LGA1700 contact frame to try to reduce temperatures. During the first stress test, temperatures initially looked normal, but after several minutes the CPU began reaching 100°C and throttling the P-cores to around 5.1 GHz or lower.
Since then I've replaced the thermal paste, switched back to the stock retention bracket, repeatedly reseated the AIO, and tried various combinations of the frame, bracket, paste, and cooler installation. I've probably reinstalled everything around 15 times. I also heard bubbling or clicking from the pump at high speed, so I moved and shook the radiator and pump to clear possible trapped air. The noise stopped, but the temperatures did not improve.
I've been careful not to overtighten anything and have tightened screws evenly in a cross pattern. The cooler appears to make firm contact, but the CPU still quickly reaches 100°C under Cinebench instead of staying near its previous temperatures.
Could the AIO be failing or partially blocked, or could the CPU's heat spreader, solder, motherboard socket, or mounting pressure have been damaged or permanently changed after years with the original retention bracket? What tests would help identify whether the problem is the cooler, the motherboard, or the processor?
4 Answers
If another known-good cooler produces the same results, then the next suspects are the socket and CPU rather than the paste or bracket. Years of pressure from the original retention mechanism can contribute to motherboard or heat-spreader deformation, although a failed solder connection beneath the heat spreader would be unusual. I’d only consider that possibility after ruling out the AIO with another cooler.
The fact that temperatures start normally and then climb suggests the cooling system may not be removing heat fast enough. Check that the AIO pump is actually running at the expected speed, verify that the radiator fans ramp up properly, and monitor coolant temperature if the cooler reports it. Also temporarily return all CPU settings to stock so the undervolt and power limits are not adding another variable.
Pay attention to the temperature behavior, not just the maximum temperature. Poor cold-plate contact often causes very sharp temperature spikes and unusually high idle temperatures because heat is not moving into the cooler efficiently. Trapped air in the pump or radiator can also cause this, especially after the AIO has been moved around. Make sure the radiator and tubes are positioned so air cannot collect in the pump.
The first thing I’d do is test with a completely different cooler, even a basic tower air cooler if you have access to one. That will separate a CPU or mounting problem from an AIO problem. Also inspect the motherboard around the socket and check that the mounting threads and screws haven’t been damaged by repeated installations. A clear, well-lit photo of the socket and CPU contact area could help.
If the H150 is a sealed, non-refillable model, it may simply be reaching the end of its life. A failing pump, partial blockage, or loss of coolant can all produce normal temperatures for a short time followed by rapid overheating.