[SOLVED] High CPU Temps. Tried Everything. Need Some Help

[Asylth]

Hi. I have an issue I cannot seem to figure out. I just changed my case and power supply. I moved my build to a new case and cleaned it. Though after putting it all back together I noticed my CPU temps were abnormally high. On boot it was 80c and only went down to 65 while doing nothing. Launching a browser, it would spike to 88 degrees. I tried to launch games and the CPU would hit 91c and then crash after a minute. I double checked everything. Fans working, fresh thermal paste, re mounting the AIO, reset bios so on. I noticed the AIO pump temps were only at 30-40c, and I thought the AIO was broken or something even though the pump was running at max. There was no heat being output from the rad. So, I changed the cooler to a stock AMD Wraith cooler and now I’m hitting 90c when just opening a browser, so I don’t think it’s the AIO or mount that was the problem. Is it something wrong with the PSU/motherboard/power delivery or just the AIO which was working fine before?

On a quick note, when I cleaned the AIO it was attached to the motherboard and I may have lifted the motherboard by just holding the radiator when clean it. Which may have put some strain on the socket / motherboard. Not sure if this is an important factor or not.
What I’m thinking is that there might be something wrong with the CPU or the mounting, but I don’t know what it could be or what caused it. I have bought a new AIO which should come soon, but after switching to the Wraith cooler I'm fearful that it might not help at all. Would really appreciate some help and opinions.

Specs:
AMD Ryzen 5800x
Asus x470-f Gaming
Corsair GTX H100i (240mm aio)
EVGA 3080ti hybrid
Corsair vengeance 32gb 3200mhz
Corsair RM850X V2 (850W)

 

[Darkbreeze]

The wraith coolers suck anyhow, so don't use that as a metric to compare anything.

How old is your current H100i?

 

[Asylth]

The wraith coolers suck anyhow, so don't use that as a metric to compare anything.

How old is your current H100i?

I bought it in like 2016 I think. So like 5-6 years old. Which I know is pretty bad, but it cooled it fine before I changed case/psu. With "only" 55-60 idle. Now it boots up and the temp is at 80.

 

[Darkbreeze]

What case did you have originally, and what case do you have now?

Where did you have the AIO mounted, and what does the rest of the case fan configuration consist of?

5-6 years is a pretty good run for an AIO, and is generally about the time (4-5 years in most cases) that they start losing performance due to a variety of potential reasons. I'd install the new cooler, making sure that the cooler is fully seated and that you don't do this:

 

[LeviTech]

Hi.
Even with the stock amd cooler hiting 90C just by open chrome is way too much.
Well, the only thing you can do for now is wait for the new AIO and hope that there's nothing wrong with the board/cpu.
Have you checked the voltages?

 

[Asylth]

What case did you have originally, and what case do you have now?

Where did you have the AIO mounted, and what does the rest of the case fan configuration consist of?

5-6 years is a pretty good run for an AIO, and is generally about the time (4-5 years in most cases) that they start losing performance due to a variety of potential reasons. I'd install the new cooler, making sure that the cooler is fully seated and that you don't do this:

I had the AIO at the front of a phanteks eclipse p500 (pull) and now I have it at the top of an o11 dynamic (push). In both cases the AIO was not the highest point in the loop. The AIO is pulling some heat away, but not nearly as much as it was before.

 

[Asylth]

Hi.
Even with the stock amd cooler hiting 90C just by open chrome is way too much.
Well, the only thing you can do for now is wait for the new AIO and hope that there's nothing wrong with the board/cpu.
Have you checked the voltages?

Yeah, I thought the same. Hopefully a new AIO will fix it. Worst-case scenario is if there is something wrong with the cpu/motherboard/psu since I will be without a pc if I have to return one of them. HWmonitor and AI suite reports a voltage of around 1.35V but it spikes to 1.45 sometimes

 

[LeviTech]

That kind of Voltages are not to high for a turbo mode, but try switching to PBO and uninstalling the chipset drivers and check again the temps and voltage.

 

[Darkbreeze]

Problems with the motherboard would generally result in OTHER types of problems, if something is wrong with the board. Now, that's not to say it's impossible for a motherboard to have a faulty thermal sensor and be misreporting a thermal reading, or that there can't be a VRM failure, but it's pretty uncommon (As in, I've never seen it happen personally for the motherboard itself to cause the CPU to overheat, and if if it did, you can be sure it would be throwing all kinds of red flags and errors because about the only way that could happen would be from some kind of short or serious overvoltage issue, neither of which seems to be the case in your case.

CPUs CAN fail internally, but it's extremely uncommon unless something external or negligent has been done such as extended or unrealistic overclocking, dropping on the ground, running without adequate cooling for a period of time, running without ANY cooler for ANY length of time, bent pins on the CPU/motherboard (Possible, but again, likely to cause MANY other problems aside from just overheating).

The first thing I'd probably do is check to see that you have the MOST up to date stable motherboard BIOS version installed. If you don't, then I'd update to the latest stable version.

How, EXACTLY, do you have the case fans in that 011 dynamic installed and directionally configured?

I had the AIO at the front of a phanteks eclipse p500 (pull) and now I have it at the top of an o11 dynamic (push). In both cases the AIO was not the highest point in the loop.

Not sure at all what you meant to say here, as it doesn't seem to technically make sense nor does it answer the question posed in the video link I posted. Might want to take a second look at the SPECIFIC configurations shown and not recommended, in that video, or perhaps simply clarify the statement.

 

[LeviTech]

The strange thing here is that it all happen after you put all together in a new case. Something went wrong during this process.

 

[Remeca]

Don't see it mentioned, but you did clean off and properly reapply thermal paste each time right? If yes, I'd guess maybe your mount isn't letting your coolers make proper contact.

 

[Asylth]

Problems with the motherboard would generally result in OTHER types of problems, if something is wrong with the board. Now, that's not to say it's impossible for a motherboard to have a faulty thermal sensor and be misreporting a thermal reading, or that there can't be a VRM failure, but it's pretty uncommon (As in, I've never seen it happen personally for the motherboard itself to cause the CPU to overheat, and if if it did, you can be sure it would be throwing all kinds of red flags and errors because about the only way that could happen would be from some kind of short or serious overvoltage issue, neither of which seems to be the case in your case.

CPUs CAN fail internally, but it's extremely uncommon unless something external or negligent has been done such as extended or unrealistic overclocking, dropping on the ground, running without adequate cooling for a period of time, running without ANY cooler for ANY length of time, bent pins on the CPU/motherboard (Possible, but again, likely to cause MANY other problems aside from just overheating).

The first thing I'd probably do is check to see that you have the MOST up to date stable motherboard BIOS version installed. If you don't, then I'd update to the latest stable version.

How, EXACTLY, do you have the case fans in that 011 dynamic installed and directionally configured?



Not sure at all what you meant to say here, as it doesn't seem to technically make sense nor does it answer the question posed in the video link I posted. Might want to take a second look at the SPECIFIC configurations shown and not recommended, in that video, or perhaps simply clarify the statement.

I have two intake fans at the bottom of the case in the spots justified to the front. GPU AIO exhaust out the side towards the bottom two 120mm slots. And CPU AIO exhaust out the top of the case justified to the rear of the case.
In the old phanteks case I had the AIO mounted in the front (pulling fresh air in) of the case with tubes up (I know it's not ideal) but the pump was not the highest point. I also had a rear and top fan exhaust.
I updated my bios last month and I have the most up to date one (ver. 5861).

 

[Asylth]

Don't see it mentioned, but you did clean off and properly reapply thermal paste each time right? If yes, I'd guess maybe your mount isn't letting your coolers make proper contact.

Yeah, cleaned it off every time with isopropyl alcohol. And the mount should make full contact. I've tried both really firm mounts and more loose ones.

The strange thing here is that it all happen after you put all together in a new case. Something went wrong during this process.

I suspect something happened when I lifted the motherboard by the cooler. Not sure what tho since I re-mounted it so many times. Maybe the AIO just died in a weird way, idk.