Has my pump failed? Advice please on my Corsair H115i.

[The-Hoff]

I am getting terrible cooling all of a sudden on my i7 8700k. The temps are really high when stressing. The fans work fine. I looked at the Corsair Link software and there are two speed settings for the pump, quiet and Performance. I can't hear or feel a thing (if i touch the unit) when I go between these two settings. The onscreen display shows the speed changes but I cannot hear a thing from the pump on either speed.
Surely there a noticable noise increase between the two speeds, or is the pump super silent anyway? I am starting to think the pump could be jammed, regardless of the speed readings. Does this seem plausable?
Thanks in advance.

 

[Lordgrayone]

How old is it? Check the pipes/rad temperature, are they all the same heat? If one pipe is hot and the other cold chances are the pump has failed / broken

 

[The-Hoff]

How old is it? Check the pipes/rad temperature, are they all the same heat? If one pipe is hot and the other cold chances are the pump has failed / broken

Thanks, it is about 12 months old. I have had my system sat for a couple of months while upgrading a few parts and I have only just started using it again. I haven't checked the rad temperature but I did hold both pipes, mainly to feel any kind of vibration whilst changing the pump speeds from silent to performance, and I can't feel any temp difference, they are neither hot nor cold. The thing that makes me feel that it is the pump is the fact that I cannot hear/feel a thing from the main pump housing when changing speeds. Surely it should ramp up? Corsair Link still reports the changes in the pump speed rpm's but I am guessing the motherboard is still reporting the switching but mechanically, the pump is not working. Or would a failed pump give 0 rpm? I just wish I could hear some pump noise.

 

[Lordgrayone]

If the pump was not rotating I would assume 0rpm reading however if the impeller had snapped the pump could still be running but not pumping any water around, I would assume you would hear/feel vibration yes, what sort of temps are you getting? If it's been sat for a while it could just be the paste on the cpu has dried up, I guess it depends on how different /extreme the temp difference is from how it was before. Might also be worth giving the pc a shake as it could just be an airlock

 

[The-Hoff]

Thanks again for the advice. I am getting about 30-32 degrees when idle and 75-80 degrees when stress testing with Realbench. My i7 8700k has been delidded by 8pack. I have reapplied paste a couple of times. I know if I have had my system down for some time, I hear a little 'sloshing around' during start up until it fades out. Maybe that is the water settling and the pump is fine after all.

 

[britain4]

I accidentally started stress testing once with the pump on my H60 unplugged (!) and it shot up to nearly 100 degrees almost instantly before I realised and shut the test off (delidded 4770K). I think if your pump wasn’t working your temps would be much higher. My pump is 100% silent and with no vibration so no way to tell it’s even running other than temps!

So I think your pump is working to some degree at least but only way to be certain is to unplug it and see what happens! Is your CPU overclocked at all?

That said those temps do sound rather high, first thing I would do is try replacing the thermal paste (I am very much doubtful it would have dried up unless you used toothpaste but it may have been a bad application), second thing is which way up is your rad because my H60 isn’t completely full of fluid and if I used it “pipes up” it would be sucking in air. Are your fans getting enough cool air to them? should be easy enough to crack this by process of elimination.

 

[The-Hoff]

I accidentally started stress testing once with the pump on my H60 unplugged (!) and it shot up to nearly 100 degrees almost instantly before I realised and shut the test off (delidded 4770K). I think if your pump wasn’t working your temps would be much higher. My pump is 100% silent and with no vibration so no way to tell it’s even running other than temps!

So I think your pump is working to some degree at least but only way to be certain is to unplug it and see what happens! Is your CPU overclocked at all?

That said those temps do sound rather high, first thing I would do is try replacing the thermal paste (I am very much doubtful it would have dried up unless you used toothpaste but it may have been a bad application), second thing is which way up is your rad because my H60 isn’t completely full of fluid and if I used it “pipes up” it would be sucking in air. Are your fans getting enough cool air to them? should be easy enough to crack this by process of elimination.

I must apologise, I missed out an important bit of information. I am running at 5.0 at 1.35v delidded. The paste is fine, it is new (thermal Grizzly).
I will try unplug the sata power cable to the pump and see what happens to the temps.

 

[britain4]

I must apologise, I missed out an important bit of information. I am running at 5.0 at 1.35v delidded. The paste is fine, it is new (thermal Grizzly).
I will try unplug the sata power cable to the pump and see what happens to the temps.

I’m running at 1.35V delidded on my 4770K. With a H60 (push pull fans) mine stay around 80 ish under full load so you would expect a H115i to be quite a bit better. (That said I have no idea if mine is cooling 100% but I hope to change it soon)

Yes it’s a good way to tell but naturally there is a real possibility of temps shooting up without warning so do be careful.

Are your fans ramping up with the CPU temps - I know you said they “work fine” but I had an issue a few days back where the fans worked but wouldn’t react to CPU temps

 

[tyler_jrb]

Thanks again for the advice. I am getting about 30-32 degrees when idle and 75-80 degrees when stress testing with Realbench. My i7 8700k has been delidded by 8pack. I have reapplied paste a couple of times. I know if I have had my system down for some time, I hear a little 'sloshing around' during start up until it fades out. Maybe that is the water settling and the pump is fine after all.

If the pump had failed completely you would know, it probably would last a minute or two at most at idle until the chip overheated.

My older h115i I couldn't feel or hear the pump until link was set to perf mode where it ramped up to a much higher speed. And even then it wasn't overly loud.

What are things like your airflow, rad mounting. Is your rad mounted in the roof with an AIB Ti as that can heat up its intake temps quite a fair bit.

What are the stress temps like at first cold boot?

 

[Beastahoythere]

Hi my 8600k at 5100 on a h110 was idle at 29 and about 87 on full load with all working fine but i think pump is failing / getting noisy so go back to custum loop after 5 years of AIO

 

[The-Hoff]

If the pump had failed completely you would know, it probably would last a minute or two at most at idle until the chip overheated.

My older h115i I couldn't feel or hear the pump until link was set to perf mode where it ramped up to a much higher speed. And even then it wasn't overly loud.

What are things like your airflow, rad mounting. Is your rad mounted in the roof with an AIB Ti as that can heat up its intake temps quite a fair bit.

What are the stress temps like at first cold boot?

Well I finally resolved the issue, although I still don't know what caused it. My rad is in the roof and my 1080ti is the Sea Hawk with the fan in the rear. You were correct about knowing if the pump had failed. I removed the power from the pump when in the desktop, and within 3 minutes the temp hit 60c and my pc froze. My pump must have been working after all, perhaps I was just lucky having one that was ultra silent?
My PC still failed to respond so I turned it off, reconnected the power to the pump, and rebooted. The crash must have done something to the registry, as my PC was constantly crashing after that. Sometimes it wouldn't even boot, it would hang before posting. Any XMP profile would crash. Clearing the bios, resetting Cmos, uninstalling/reinstalling the Corsair Link drivers didn't help. I spent a couple of hours diagnosing everything and reseating/testing the ram, my cpu, my ssd's and I unplugging everything. Nothing made a difference.
Enough was enough, I formatted Windows.......and low and behold, everything was perfect from there on. When I eventually came to testing my delidded i7 8700k it was a completely different story this time round, my OC now had lower temps. This enabled me to drop the voltage bit by bit from 1.35v to 1.27v. Even now, I am still testing and dropping my volts. I am so far getting a stable overclock using Realbench and Prime95. My max temps now are 65c which is great.
Who knows where my final stable OC will end up. Maybe I can start jumping from 5.0 to 5.1/5.2. Thanks for all the advice guys. It looks like my chip and cooling are now doing okay

 

[Lordgrayone]

Sounds like you went on a right mission there but worth it in the end and glad you have sorted it, sounds like it could have been something running in the background using the cpu and pushing up your idle temps.

 

[The-Hoff]

Sounds like you went on a right mission there but worth it in the end and glad you have sorted it, sounds like it could have been something running in the background using the cpu and pushing up your idle temps.

Thanks, much appreciated. I must admit, I was getting quite concerned. First it was high temps from a delidded cpu, then a major system issue caused just by yanking the power from the pump. Having built PC's for many years I still couldn't pin point what that had done to my PC. I was lucky a simple format solved both problems. I finally have the excitment of pushing my cpu to its max over the weekend.

 

[OCnOoB88]

hi all,
I know im a little late to the party,
I was having issues overclocking my setup. instant blue screen or stuck on black spinning load then crash,
Specs are:
Mobo: Asus z270e Gaming
CPU: i7700k kabylake
RAM: Corsair Vengeance 3600Mhz DDR4 (2 x 16gb) 32gb
GPU: Asus GTX 1080 ti OC
SSD: Samsung evo pro (several)
AOI: Corsair H115i
PSU: Corsair 650w (i think- cant check without dismantling my case, as it has a tricky cover on the PSU)

I initially tried to OC using the Eztuner in BIOS, which boosts 17% = BSOD
Then i tried to put the 5ghz profile (without modifying voltage)= BSOD

Then i came to my senses and decided to stress test using Prime95.
it shoots up to 100 degrees C within 2 seconds. So i suspect my AOI has lost it?
the fans and pump are all running at high/max speeds (verified in iCue corsair software)

safe to assume my PC is tripping on the instant high temp when booting the new settings from BIOS?
Worth to change thermal paste on it, or just buy a new one.. as it doesnt even lower the temp at all when doing the test
No errors or warnings on any of the cores

could it be:

- thermal paste issue?
- AOI is dead inside?
- Faulty sensor somewhere?
- im too noob to know?

any help is appreciated