Jump to content
Corsair Community

High Liquid Temp on H115i Pro


puritos
 Share

Recommended Posts

Corsair send me a new h115i pro because my old h110 stopped working and i think this unit is not working fine, in Idle the water temp is 29-31º degrees but when im playing or similar the water temps go to 49-50º (pump at 2100rpm and fans at 1000 rpm) the cpu temp never pass 55º gpu its at 65º and motherboards temps are between 35-40.

 

It is normal 50 degrees on liquid?¿ i think my h110 usually stay at 35-38 at max.

Link to comment
Share on other sites

We need a bit more information to analyze this. Specifically, what GPU do you have and how is your case airflow configured?

 

Most of the time when we see this, it's either a bad cooler - which seems unlikely as your idle temps are fine - or it's an airflow issue with GPU waste heat running through the radiator.

Link to comment
Share on other sites

We need a bit more information to analyze this. Specifically, what GPU do you have and how is your case airflow configured?

 

Most of the time when we see this, it's either a bad cooler - which seems unlikely as your idle temps are fine - or it's an airflow issue with GPU waste heat running through the radiator.

 

Thanks for the fast reply.

 

-I have a 2080TI (asus strix).

 

- My case is Cougar Panzer Evo

-Front: 3 intakes fans

-Back : 1 exhaust fan

-Top : H115i Pro Fans at exhaust in pull configuration (this case doesnt allow Push config in the top)

 

The computer works fine and i never have any problem but i was worried about that Water temp because i remember it to be under 40º with my old unit and my 5820k @4.6 that have higher temps compared to my new CPU (now i have an r7 2700 withc stock clocks).

 

I

Edited by puritos
Link to comment
Share on other sites

A 20C water temp change is well above average and questions should be asked. It is possible the GPU is playing a contributing part, but you can cut that out by running a CPU only stress test. Use something mild, like CPU-Z or OCCT Linpack. Both of those should have steady loads and make for a good coolant temp check. Depending on fan speed, I would expect around +6C rise in coolant temp.

 

The idle coolant temp seems about right for a 22-24C room temperature, although I am not as familiar with your case and solid top panels always add some complexity. However, since this appears to have been a non-issue on the prior 280mm in the same place, it seems less likely case structure is responsible. Run the CPU only test and see what happens to coolant. It should go up the +5-8C (only need to run it 10-15 minutes at most), then it should drop 4-5C in 1-2 minutes after. The last couple of degrees often takes slightly longer.

Link to comment
Share on other sites

A 20C water temp change is well above average and questions should be asked. It is possible the GPU is playing a contributing part, but you can cut that out by running a CPU only stress test. Use something mild, like CPU-Z or OCCT Linpack. Both of those should have steady loads and make for a good coolant temp check. Depending on fan speed, I would expect around +6C rise in coolant temp.

 

The idle coolant temp seems about right for a 22-24C room temperature, although I am not as familiar with your case and solid top panels always add some complexity. However, since this appears to have been a non-issue on the prior 280mm in the same place, it seems less likely case structure is responsible. Run the CPU only test and see what happens to coolant. It should go up the +5-8C (only need to run it 10-15 minutes at most), then it should drop 4-5C in 1-2 minutes after. The last couple of degrees often takes slightly longer.

 

Thank you so much i tested what u suggested (with cpu-z) and those are the results with 24º ambient temp and 30 minutes stress test.

 

 

Test Start:

 

- Coolant temp: 30º

- Cpu temp : 30º

- Speeds : Fans 600 rpm Pump 1100 Rpm

 

 

Test "peak":

 

- Coolant temp: 38º

- Cpu temp : 57º

- Speeds : Fans 1000 rpm Pump 2800 Rpm

 

 

It goes back to 32 coolant temp in 3-5 minutes.

 

Conclusions: Looks like only heavy cpu "works" give me a normal and expected temperatures and it only goes to 50º when im doing cpu+gpu intensive works.

 

Can the gpu be culprit? (Aio tubes are really close to it) or can be something related to airflow ? (as i said i got 3 intakes + 1 exhaust at 900 RPM + h115 fans at exhaust).

 

I dont care too much about lower my cpu+gpu temps because they are not really bad and my pc its silent with this profile but im worried because i think that maybe 50º Water temp can be dangerous for the AIO.

Edited by puritos
Link to comment
Share on other sites

It does appear another source of heat is impacting the coolant temps -- surely the GPU. The coolant is not going to be spot heated from tubes close to the GPU. The usual issue is the entire case gets heated up, including pump block, tubes, and radiator. The next step would to take a look at your case temps when gaming. The motherboard temp sensor will show in most programs. Any drive temps if they are internal may give a clue. Obviously any temp probe or other means of measuring air temperature will help as well, including measuring exhaust temp from out the back fan.

 

50C is not specifically dangerous (rubber does not melt, weaken, etc. anywhere near that temp), but it does present a noticeable CPU temp penalty. Even on my delidded 8700K that would get me into some high numbers. If the cause is as suspected, then everything in the case is going to be 45C+ which may not be ideal. Make sure you are getting a decent clip from the rear exhaust fan to encourage air to go out that path and not necessarily though the top fans. Blasting the top fans at the expense of the rear fan may draw more waste heat through the radiator rather than out the back like you would prefer.

Link to comment
Share on other sites

It does appear another source of heat is impacting the coolant temps -- surely the GPU. The coolant is not going to be spot heated from tubes close to the GPU. The usual issue is the entire case gets heated up, including pump block, tubes, and radiator. The next step would to take a look at your case temps when gaming. The motherboard temp sensor will show in most programs. Any drive temps if they are internal may give a clue. Obviously any temp probe or other means of measuring air temperature will help as well, including measuring exhaust temp from out the back fan.

 

50C is not specifically dangerous (rubber does not melt, weaken, etc. anywhere near that temp), but it does present a noticeable CPU temp penalty. Even on my delidded 8700K that would get me into some high numbers. If the cause is as suspected, then everything in the case is going to be 45C+ which may not be ideal. Make sure you are getting a decent clip from the rear exhaust fan to encourage air to go out that path and not necessarily though the top fans. Blasting the top fans at the expense of the rear fan may draw more waste heat through the radiator rather than out the back like you would prefer.

 

 

Thank you so much again.

 

When im running cpu+gpu and the coolant go to 50º my motherboard temps go to 40-47º (it have 2-3 sensors) and my drives go to 51º (Nvme 970 evo) and 38º (ssd Crucial).

 

I will try to tweak and try different profiles for my exhaust/intakes fans and see what happens.

 

Again thank you so much for your time.

Link to comment
Share on other sites

OK, that's seems like a strong correlation and we are looking at internal temp issues. My Ti is full water, so I don't have a clear idea of how much heat the card produces in terms of air temp. The wattage is certainly big and even on full water I was amazed at the heat going from a Titan X(P) to a 2080 Ti.

 

The only other thing that immediately comes to mind is the top glass (?) panel is restricting the air flow out of the radiator. It looks like it has diversionary side channels to let the air out. That's never going to be as efficient as a straight blow through, but that does not mean it is the problem or it cannot be effective. The typical test for that is take the top panel off, but I don't know how feasible that is on your case.

Link to comment
Share on other sites

One thing we've seen with other cases with restricted tops is that reversing the flow from exhaust to intake leads to a massive improvement. You would need to make sure that you have filters in place for dust management but it may be worth at least considering.
Link to comment
Share on other sites

You are basically pulling in an extra 200W+ through the radiator with the GPU under load, that's probably double or triple the CPU output.

 

Run the rear as intake, run the top as intake, front as exhaust etc.

Link to comment
Share on other sites

You are basically pulling in an extra 200W+ through the radiator with the GPU under load, that's probably double or triple the CPU output.

 

Run the rear as intake, run the top as intake, front as exhaust etc.

 

That is a rather grand supposition without any actual measured thermal data. Power draw in watts does not equal radiant heat dissipation and that is ignoring whatever air flow is going through the case. Changing some directional flow may be in order if the current set-up cannot be managed, but a fairly radical reverse flow does not work for most people and the there are some definite obstacles to overcome.

Link to comment
Share on other sites

Today i tested one thing i reverted my h115i fans and put them as intake (in the top) and my h115 temps are clearly better ( 41º maximun on load) but my gpu temps are a bit higher. Looks like my case doesnt work pretty well with the radiator exhausting in the top so i ordered 2 extra fans to test it as a intake in the front (with push+pull config) and i will post the results when they arrive.

 

Again thank you all for your time and your suggestions.

Link to comment
Share on other sites

 Share

×
×
  • Create New...