Question AIO cooler making a grinding noise when it receives more power

[Zhao_Zilong]

So I recently upgraded from a FX 8370 and a gigabyte mobo, to a 9900k n a z390 gigabyte mobo.
Using the same, coolermaster nepton 240m on both builds.

having to cool a lower TDP CPU (even without an OC on the 8370) theres problems with the new CPU/mobo

the pump turns off after a few mins of the pc being online. leads the cpu to hit 100C (210F) on all cores in 5-10 seconds

when that happens it doesnt cut out, and i can manually swap the 4 pin power connector for the pump, i get the GRRR pump noise and the temps instantly drop back to 35ish (C)
I've gotten it to the point where my system is stable, I've forced the bios to use the pump on a manual fan curve
on this curve it works perfectly, havent seen a max temp over 70C which is insane. but it still makes the same grinding noise once it goes over the 45C (aka the threshold I set it for the 'fans' to rev up)

So is my AIO pump failing or is it getting the wrong power?
any insight would help

 

[Lutfij]

Have you isolated the source of the grinding noise from the pump/block or the fans on the system? The grinding noise might come to be with the pump on it's way out, with the impeller brushing against the housing or you've ended up with an air bubble in the pump chamber which is what the impeller is hitting. You might want to see if you can remove the AIO from the system, have the pump dangling and then see if shaking the radiator and the pump dislodges the air bubble, assuming that it's an air bubble that's causing the issue.

Pertinaing to the pump header/fan curve, what is the model for your Z390 chipset gigabyte board? Are you on the latest BIOS update for the board?

 

[Zhao_Zilong]

It's the z390 gaming sli board
https://www.gigabyte.com/Motherboard/Z390-GAMING-SLI-rev-10#support-dl-bios

updated it to the F7 bios, had the same pump drop before going from F4-7 so dont believe that is the issue.

It is most definitely a pump issue though (or pump control) as fans can be set to 100% (and in fact instantly jump to 100% when the pump stops) but it doesn't help. stops the system from dying outright, but it sits at 100C and I can touch both pipes coming from the CPU, both are HOT for 2cm, n supercold after that, so the radiator is cooling... but the liquid isnt flowing.

as for dislodging an air bubble, removing the pump is easy enough... the radiator is a bit harder. custom drilled the holes , easy installation... while nothing else was in the case so now access to screws are almost impossible without removing everything.
how could a closed system get an air bubble though? thats the part that confuses me, and made me instantly jump on the broken pump wagon