Jump to content
Corsair Community

Corsair h100i rgb platinum. Extremely high temperature


Mingus81
 Share

Recommended Posts

Hello! Well, I am new at this with watercooling and I have some questions that maybe you can help me with. I just bought a new computer. Motherboard is an Asus ROG Strix Z390-F Gaming, CPU an i7 9700k. When I got the computer a couple of weeks ago and booted it up everything looked just fine but when i started playing games the CPU temp went up to 87C. I read some stuff about the watercooler

(I had an Corsair Hydro H60) and some people said it was to small for the i7 9700K. So I just bought a corsair h100i rgb platinum and installed it with the help from a friend who knows he`s way around a computer. Everything looks fine with wiring and so on, but the problem is, the radiator doesnt push any hot air at all, is this correct? I have a push configutation so I suppose I should feel some hot air flowing out from the case. It did that with the Corsair Hydro H60. It got very hot!

 

And the second and more concerning thing, accorrding to ICUE, Core Temp and MSI afterburner, the CPU in some cases goes up to 100C while gaming. The games works fine, no indication that the computer is about to crash. The first time we booted it up we checked Idle temp and it was around 36C so we thought, good to go. We started Metro Exodus and played for about 20 minutes and then we checked the temp and it was blazing hot. But no crash or any other symptoms from the computer at all. So could this be a fault in the temp reading? Or is something wrong with the cooling here? I checked the tubes and I can feel a vibration from them so I suppose the water flows correctly.

 

I hope someone have some input in this. I hardly dare to start the computer now even if it seems to work just fine :(

 

Thanks for your attention. I’m looking forward to your reply

942357355_TEMPICUE.thumb.png.2c4e2e0484ec21470410b3a4e50097f6.png

1170833467_TEMPINGAME.thumb.png.5ac7864fca8df30a19db58eae54811c6.png

DATOR.thumb.jpg.9653ff88e40d8325db340a7c8ac976d9.jpg

Link to comment
Share on other sites

That's extremely hot, especially for plain gaming, and not good for your CPU. What's your Vcore set at? Setting that too high constantly will fry your CPU.

 

Your Pump speed appears to be set to "Balanced". Try to set it to "Extreme" and see if that helps (although since the liquid temperatures aren't bad, that probably won't do much).

 

That said, the liquid temperatures seem to be okay, so perhaps the cooler's cold plate doesn't make good contact with the CPU's heat spreader. It might be a good idea to repaste it with some quality thermal compound and re-seat the cooler with enough, even pressure so it makes good contact.

Edited by Glzmo
Link to comment
Share on other sites

Thank you so much for your reply! I just tested to set pump speed to extreme and it didn't make any difference. Same temp in game, 100c on some cores. I got two LL120 for intake in the front of the case and one LL120 for push in the back. But isnt it strange that the computer doesn't show any symptoms of overheating? Like I said, im new at this stuff but I thought that a cpu at 100c would make FPS go down and maybe even turn it self of? And Im sorry, dont know if this helps but i dont know anything about Vcore settings so i send a pic :) But from what i can tell its around 1.2.

INTAKE.thumb.jpg.e8ed61550bc24e0356003de4221bc8ce.jpg

OCCT.thumb.png.67a897c2ab9001fa72823da674445c09.png

Edited by Mingus81
Link to comment
Share on other sites

Your vCore looks fine.

This is odd - if it was a pump issue, typically we'd see the pump temperature increase significantly as well. Can you feel the tubes? Is one warmer than the other?

Outside of that, it does seem to be more of a contact issue than anything.

Link to comment
Share on other sites

OK, I am trying to sort this out. Playing games with cycling 40-75% CPU usage, you get CPU temps in the 90s. Using OCCT at 100% you are in the 50s? Or is that an idle temp spike? Something is weird.

 

OCCT will occasionally run a long warm up before actually applying load. You can get around the variability by using Linpack. Another thing to try is the standard test in CPU-Z. That is linear as well and not a "stability tester", so the temps should hold solid, only going up as coolant increases. That makes it a good tool for diagnostics. (Bench Tab, "stress CPU" button at the bottom.

 

At this point, I think it would make sense to try and differentiate between actual CPU temp issues and some kind of anomaly with the Riva tuner data while playing. You should be stuttering at those temps and the coolant to CPU difference is too high for your voltage, but yet is not replicated on the stress test. Everything else about the cooler appears normal. Easiest to sort this part first, before taking things apart.

Link to comment
Share on other sites

Thank you for you time!

 

Every game i try with CPU usage above 50%, the CPU goes up to 100c in a couple of seconds. Idle stays beteween 35 and 50c. Sometimes it spikes when just opening youtube or some program. I tried the test in CPU-Z but turned it of in about 4 seconds. Every core went up to 100c very quick. The temp stats comes from both Icue, Core Temp and OCCT. I have checked the pump and when I set it to extreme in Icue i can clearly hear a huming so I suppose the pump is working correctly. But like i said, I have no clue how its suppose to sound! But the radiator is cold so maybe its a connection issue? Is it possible to use to much Thermal compound? I read somewhere that to much compound just isolates the heat. I use Arctic Silver 5.

Edited by Mingus81
Link to comment
Share on other sites

OK, that in some ways is a relief. I suspect OCCT was either in the warm up phase or downshifted itself because it detected the problem. It does both. The CPU-Z test confirms there is a problem and not a Riva tune gaming phantom. With the OCCT stuff eliminated, everything you have written points toward a contact issue. Unfortunately, this can be a very finicky thing to deal with.

 

I think the place I would start is the actual mounting post depth into the motherboard and rear support bracket. Make sure each of the 4 post threads goes all the way through and into the mounting bracket hole. I just went through this on a different brand backplate and it was enormously frustrating. I could not get this one corner in. The metal was straight. The board was straight. When doing the offending corner first, it would pop out later when the others where cross-tightened. I had to do it three times. Perhaps not coincidentally, it was on an Asus board and 8700K. In each of attempts, the TIM was spread, but there was a very tiny corner that was not quite as thin as the others. Hard to see, definite change in temps from it should have been.

 

If you had applied too much TIM, the effect would be different. You would get almost random temperature fluctuations as little air bubbles pop or it may even appear normal for long stretches of time. The instant 100C is going to be a contact issue.

 

Same thing for any pump settings. Even if the pump was off, you could watch the temps climb up 50-55-60-65.. etc. until you stop it, but it is observable in steps. When you click 'start' and it the next thing you see is 90C, it's contact/CPU related and not the cooling end.

 

The standard jumpiness when you open programs is normal for all CPUs from Skylake on. My Coffee Lake drove me nuts at first with its constant up time, but you get used to it. Unfortunately, this also make detecting contact issues more difficult since the new CPUs behave like old CPUs with problems.

 

One thing NOT to do is over-tighten the thumbscrews in an attempt to squeeze it together. The most likely result there is bent motherboard pins and you need to go shopping for a new one. Focus on the back plate first. I don't have a Platinum cooler on hand so I can not see the back plate. Most need to be turned a specific way, aligned with motherboard screw holes, etc., but thick or thin motherboards have made the back plate tricky at time.

Link to comment
Share on other sites

I just thought of something regarding the support bracket. I used the hydro h60 before and we used that support bracket when we installed the corsair h100i platinum. It looked exactly the same so we didnt remove it from the former installtion. Could this be an issue? That bracket worked and nothing seemd wrong with the h60 mounting. Edited by Mingus81
Link to comment
Share on other sites

Possibly. I am trying to remember if the H60 was CoolIT (pretty sure it was) and the Platinum series is partner manufactured with CoolIT as well, so there is a chance they are the same. However, without them in my hands, I am guessing. We do have several other Platinum owners who might be able to at least share any experience with that bracket. I will try and find some pictures.
Link to comment
Share on other sites

OK, if you were referring to the recent H60 (2018), then the Intel bracket should be identical. I don't see any clear bracket orientation , so that probably means you are looking at thread depth to the bracket and if it is even pressure. However, the older H60 (2011 - black plastic, white sails, NO lights) has some slight changes to the bracket. I do not know if that really is a factor, but something to consider. You probably will end up taking the whole thing apart to remount it, so you might as well use the current bracket if it is available, but I am not going to make the claim this is the underlying reason.
Link to comment
Share on other sites

Hello! Well, I am new at this with watercooling and I have some questions that maybe you can help me with. I just bought a new computer. Motherboard is an Asus ROG Strix Z390-F Gaming, CPU an i7 9700k. When I got the computer a couple of weeks ago and booted it up everything looked just fine but when i started playing games the CPU temp went up to 87C. I read some stuff about the watercooler

(I had an Corsair Hydro H60) and some people said it was to small for the i7 9700K. So I just bought a corsair h100i rgb platinum and installed it with the help from a friend who knows he`s way around a computer. Everything looks fine with wiring and so on, but the problem is, the radiator doesnt push any hot air at all, is this correct? I have a push configutation so I suppose I should feel some hot air flowing out from the case. It did that with the Corsair Hydro H60. It got very hot!

 

And the second and more concerning thing, accorrding to ICUE, Core Temp and MSI afterburner, the CPU in some cases goes up to 100C while gaming. The games works fine, no indication that the computer is about to crash. The first time we booted it up we checked Idle temp and it was around 36C so we thought, good to go. We started Metro Exodus and played for about 20 minutes and then we checked the temp and it was blazing hot. But no crash or any other symptoms from the computer at all. So could this be a fault in the temp reading? Or is something wrong with the cooling here? I checked the tubes and I can feel a vibration from them so I suppose the water flows correctly.

 

I hope someone have some input in this. I hardly dare to start the computer now even if it seems to work just fine :(

 

Thanks for your attention. I’m looking forward to your reply

 

Not sure if you fixed this issue, but I have the h100i and my temps are no where near that bad, I do remember that I found that the mounting bracket for the h100i wouldn't fit snug to my z370p motherboard after researching this I found out that a lot of the z370 chipset mobos are a little thinner than what the h100i was made for, no big deal if you have four extra washers that came with it like mine did, you can use those so that the mounting posts will be snug to the mounting bracket which will remove any wiggle room. Also, I don't believe in using preapplied thermal paste so I actually removed the old thermal paste (snarkily looking at you the verge) and applied a small amount of higher quality paste to the cpu using the classic pea dot method.

 

Hope this helped

Link to comment
Share on other sites

Don't use washer - the bracket is supposed to be loose before the cooler is tightened down. If it's still loose when you tighten the screws, you may have the incorrect standoff screws. I know that I've always had to triple check to make sure that I have the right ones installed.
Link to comment
Share on other sites

Ok, we just remounted the block with shorter mounting screws with no wiggling at all and now the temps are much, much better! Like night and day! So thank you all for your help!

Really appreciate it :)

Link to comment
Share on other sites

  • 6 months later...

Hi @Mingus81

 

I might have exactly the same issue as yours! I just assembled my setup two days ago and I'm not sure if I used the right screws.. I just took the longest ones (long on both sides) but I suppose as per your last message that you probably used these ones too and changed for shorter ones?

 

I have the same specs as yours, i7-9700K a ROG Strix Z390-F and the H100i Platinum..

 

This would help me a lot!

Thanks

Edited by crldeje
Link to comment
Share on other sites

I have a similar problem with my new system. I have an ASUS STRIX X299-E Gaming II motherboard, Intel Core i9-9900-X CPU, Corsair H110i RBG Platinum SE. Upon booting the BIOS initiation stops with the error comment stating that the CPU fan is not operational. Also, the CPU temp quickly ramps up to over 90 degrees Centigrade (before I turn it off). The single wire three-pin "tach" plug is attached to the CPU_header, although I have tried the CPU_Opt header and the AIO pump header. SATA connector is attached properly, and the fans are connected to the harness. With this setup I cannot get past the intial CMOS BIOS boot sequence. I have not been able to turn off the lower CPU fan threshold because the CPU temp rises so quickly and I don't know if the CPU temp is true or a false temperature. I really don't think it is a mounting problem and I have use older Corsair AIO liquid cooling systems without the tach wire.

 

Interestingly, in trying to resolve the problem, I replaced the H110i with a simple Thermaltake CPU heatsink/fan cooler and the system boots up and the CPU temperature runs between 30 to 50 degrees Centigrade, depending upon CPU usage. I would prefer to use the H110i RBG, but I am not getting anywhere. I am reluctant let the system run too long at this high CPU temp since the CPU is quite expensive to replace.

Edited by jroymd
Link to comment
Share on other sites

Here are a few things to check. I install the "tach" Connector to the CPU header to avoid any CPU Fan errors so that shouldn't cause any problems.

 

Double check the SATA power connector is connected. If it's not connected there wont be any LEDs lit up on the pump. You can also try to have the pump on its own sata power cable and see if that helps.

Double check that you are using the correct standoffs for the socket. If the cooler isn't making good contact that can cause high temps.

Double check the thumb screws are tight.

 

Something to check in iCUE is if the liquid rises when the CPU gets warmer. You can find that temp to the right of the pump's image when you first click on the cooler.

 

Hope that helps.

Link to comment
Share on other sites

 Share

×
×
  • Create New...