Question H100i high package temps/low coolant temps?

[grebgonebad]

Hi all! Hoping you good folk can confirm what I think is going on here.

I've got a Ryzen 5 3600 4.05 Ghz being cooled by a Corsair H100i RGB PRO AIO. It has been running perfectly fine up until a few days ago when I started to notice some erroneous temperatures.

Prior to this it usually idled between 30-35C and never went above 70C under heavy load.

I'll preface the following figures with the fact that we recently had a pretty severe heatwave in my area. Coincidentally the day the heatwave finished my package temp shot up to 95C, at which point I immediately shut down the system (unnecessary I know, but I've never had such an anomalous change in temps before). Upon booting up again about half an hour later I have noticed that my package temps are hitting idle temps of 45C, with 60C being regularly hit when simply streaming YouTube. Running games pushes this up to the high 70's.

Confusingly, ICUE informs me that my coolant temp never exceeds 30C regardless of load.

My initial thoughts are that the recent heatwave has affected my thermal paste somehow and now is not transferring the heat as effectively to the AIO. What are your thoughts? Would I be correct in assuming that a simple re-application of thermal paste may solve this issue, or is there another issue that I am unaware of that could be causing this problem?

Any and all replies are much appreciated!

 

[lojikal]

Your thermal paste may be the culprit, however before making any recommendation, would it be possible to share your case setup including the number of fans and the configuration used.

 

[grebgonebad]

I have a Sahara Gaming P35 with 6 fans total; 3 front intakes, 1 rear out and 2 on the top mounted AIO rad also as outs.

 

[Phaaze88]

Would I be correct in assuming that a simple re-application of thermal paste may solve this issue...

Yes, it may, or may not.

is there another issue that I am unaware of that could be causing this problem?

Common points of failure with AIOs:
Mechanical: pump reaches inevitable end of life. Software can tell you if it's running or not.

Biological and Chemical: the loop gunks up with organics and galanic corrosion(mixed metal loops) over time. There's chemicals in the premixed fluid to delay this, but even that eventually expires. Software will not tell you directly about this one; you have to feel it out. Put the cpu under a load and feel both hoses. If flow is good, you will not be able to tell the difference between them, or you'll notice a very slight one. Poor flow is pretty noticeable.

Majority of AIOs are not user-serviceable, so if the issue is one of the above points of failure, there's nothing you can do about it beyond a brand new unit.

 

[grebgonebad]

Cheers for the response Phaaz88.

I'm aware of the limited life of AIO's however I feel this is unlikely to be the problem as the AIO in question is less than 2 years old, I was under the assumption that I should expect at least 5 years before any such problems should ideally occur? Please correct me if I'm wrong however.

Side note: Corsair ICUE appears to suggest that the pump is still operating, showing it as running at 2600RPM (I have it set to extreme as I read that setting it to any value shouldn't affect it's life at all).

That's a good shout about checking the hoses, I'll try this out tomorrow evening once I'm home from work.

Should it be the case that the AIO does appear to be knackered it's not a huge problem as I have another brand new identical unit that I have left over from a previous build where I ended up going for another cooling solution, so if need be I can just swap it out.

 

[Phaaze88]

I was under the assumption that I should expect at least 5 years before any such problems should ideally occur?

I believe how long each and every one of us run our PCs for plays a part in that.
Ex: someone who runs their PC 8 hours a week on average, Vs one who does 32 hours. The latter will start showing its age sooner.

Should it be the case that the AIO does appear to be knackered it's not a huge problem as I have another brand new identical unit that I have left over from a previous build where I ended up going for another cooling solution, so if need be I can just swap it out.

Nice! That's one of the biggest cons I've noticed with using AIOs. Too often, someone gets stuck, 'cause there's no other cooler nearby.

 

[grebgonebad]

I'll admit that I've been using my PC a fair bit recently, probably 5-6 hours daily for the last few months. I didn't take that into account so that may carry some weight behind it now that you mention it.

I'll check the tubes tomorrow to see if there's any glaring discrepancies that I can find, if not I'll try reapplying the thermal paste and see if that solves it. If not, as you say, at least I have a backup plan. =)

Thanks for your input! Much appreciated!

 

[grebgonebad]

Nice! That's one of the biggest cons I've noticed with using AIOs. Too often, someone gets stuck, 'cause there's no other cooler nearby.

Hey Phaaze, just thought I'd drop an update.

I checked the hoses and they felt more or less the same, plus I could feel the pump working when I touched the CPU block, so fingers crossed I'm safe for now! Ha ha.

I replaced the thermal paste and as it turns out my previous application didn't quite cover the entire heat spreader in the first place, one of the corners was a little sparse. I guess that wasn't helping any, especially considering the hot weather recently.

Anyway, everything seems A-OK now. Idle temps sit in the mid 30's and load temps don't go above 65 for the most part, so looks like problem solved (well, at least for now till the AIO packs in for real. A problem for future me I guess. Ha ha!).

Oddly enough my coolant temps still don't go above 30C regardless of load, so I'm still a little confused as to why this would be. Is it just a simple case of the system reaching equilibrium or would you say I still have a problem somewhere?

 

[Phaaze88]

Oddly enough my coolant temps still don't go above 30C regardless of load, so I'm still a little confused as to why this would be. Is it just a simple case of the system reaching equilibrium or would you say I still have a problem somewhere?

Ryzen 3600 doesn't use much power, regardless of load, and a 240mm AIO theoretically has a [up to] capacity of 250w.
There's a few layers between the cpu die and the fluid in the loop; all the heat energy from the die does not get absorbed and carried to the radiator by said fluid, only some of it.

There's no problem, but this cpu isn't going to be able to soak the cooler by itself.

 

[grebgonebad]

Ryzen 3600 doesn't use much power, regardless of load, and a 240mm AIO theoretically has a [up to] capacity of 250w.
There's a few layers between the cpu die and the fluid in the loop; all the heat energy from the die does not get absorbed and carried to the radiator by said fluid, only some of it.

There's no problem, but this cpu isn't going to be able to soak the cooler by itself.

Derp. Of course. The 3600 has a TDP of 65W. My head isn't in the game right now it seems. Thanks for putting my mind at ease!

I do plan on upgrading to a 5800X3D at some point soon, I guess with a TDP of 105W I'll see a difference then.

Thanks again for all your help mate!