The Corsair User Forums  

Go Back   The Corsair User Forums > Corsair Product Discussion > Cooling

Reply
 
Thread Tools Search this Thread Rate Thread Display Modes
  #1  
Old 11-07-2017, 12:45 PM
DrJester DrJester is offline
Registered User
DrJester's PC Specs
 
Join Date: Nov 2017
Posts: 7
POST ID # = 925639
DrJester Reputation: 10
Default Are these temps ok with the H60?

I'm scratching my head here, after replacing a defective watercooler with a Corsair H60, I'm getting these temperatures. Somehow I expected them to be a bit lower, but I'm not so sure, can someone elucidate me if these are fine? I'm afraid I may have botched the installation of the H60 and its thermal paste.

I'm using a push fan system, the CM Sniper case(so plenty of fans), clean cable management and a very old Intel i7 960 running at the default clock of 3.2GHz. (I can't afford to upgrade)

I'm running the Furmark CPU Burner for 3 or so minutes in the screenshot.
Attached Images
File Type: png tempscpu.png (345.3 KB, 85 views)
Reply With Quote


  #2  
Old 11-07-2017, 05:58 PM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 10,319
POST ID # = 925676
c-attack c-attack Reputation: 107
Default

I don't use Furmark for anything and can't really compare. Either way, if you get an unexpected result, it's always good to run an alternate test to see how it compares. Run something CPU specific and moderate, like Intel XTU. It is a fairly easy stress test with a sine wave like loading pattern. Before you begin, click the wrench on the line graph and configure it to show your 4 cores (0-3).

I am thinking back to my old 930 and voltatges, but I don't see anything particularly out of the ordinary at 1.22v. That might be more than necessary for 3.2, but if that is what you have been using there is no need to change it until this is resolved.

When you load into windows, are your CPU temps normal, but then slowly slide upwards the longer you are running? Or are they somewhat erratic? Look OK one moment then high and all over the next?

Make sure the header with the pump lead is set to 100%/PWM or whatever supplies a constant 12v. Single fan should not be an issue nor the radiator size for this wattage.
Reply With Quote


  #3  
Old 11-07-2017, 08:24 PM
DrJester DrJester is offline
Registered User
DrJester's PC Specs
 
Join Date: Nov 2017
Posts: 7
POST ID # = 925680
DrJester Reputation: 10
Default

Quote:
Originally Posted by c-attack View Post
I don't use Furmark for anything and can't really compare. Either way, if you get an unexpected result, it's always good to run an alternate test to see how it compares. Run something CPU specific and moderate, like Intel XTU. It is a fairly easy stress test with a sine wave like loading pattern. Before you begin, click the wrench on the line graph and configure it to show your 4 cores (0-3).

I am thinking back to my old 930 and voltatges, but I don't see anything particularly out of the ordinary at 1.22v. That might be more than necessary for 3.2, but if that is what you have been using there is no need to change it until this is resolved.

When you load into windows, are your CPU temps normal, but then slowly slide upwards the longer you are running? Or are they somewhat erratic? Look OK one moment then high and all over the next?

Make sure the header with the pump lead is set to 100%/PWM or whatever supplies a constant 12v. Single fan should not be an issue nor the radiator size for this wattage.

Thanks for the reply! I've downloaded the Intel XTU, and I could do the stress test, and read the CPU load, but it can not read the temperature nor the "Max Core Frequency". But while doing the 5 minute stress test, it reached 82C in one of the cores.
Reply With Quote


  #4  
Old 11-07-2017, 09:39 PM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 10,319
POST ID # = 925685
c-attack c-attack Reputation: 107
Default

Do CPU temps seem higher than expected for normal operations (programs, browsing, anything but stress tests)? If you launch an application (non-game) like your web browser, do you see a large temp spike? Do the CPU temps seem abnormally jumpy?

Odds are the cold plate and CPU are not in perfect contact, but only if the answer to the above questions are yes. 82C is pretty high for any CPU on XTU. Most CPUs will be under 65 on that. My 5930K will stay under 50C at the stock 3.6GHz, so something does seem off. Gently press down on the pump block and see if temps drop. You may also want to take the block off and see how the TIM spread. If it is still clumpy in places, the block is not all the way down.
Reply With Quote


  #5  
Old 11-08-2017, 10:13 AM
DrJester DrJester is offline
Registered User
DrJester's PC Specs
 
Join Date: Nov 2017
Posts: 7
POST ID # = 925713
DrJester Reputation: 10
Default

Quote:
Originally Posted by c-attack View Post
Do CPU temps seem higher than expected for normal operations (programs, browsing, anything but stress tests)? If you launch an application (non-game) like your web browser, do you see a large temp spike? Do the CPU temps seem abnormally jumpy?

Odds are the cold plate and CPU are not in perfect contact, but only if the answer to the above questions are yes. 82C is pretty high for any CPU on XTU. Most CPUs will be under 65 on that. My 5930K will stay under 50C at the stock 3.6GHz, so something does seem off. Gently press down on the pump block and see if temps drop. You may also want to take the block off and see how the TIM spread. If it is still clumpy in places, the block is not all the way down.
When idling, or browsing, it stays around 50C to 60C, but mostly 50C and 55C(when opening the popup window to upload this screenshot, still holding at 55C and 56C). Thanks for the suggestion, I'll try that as soon as I get some assistance, since I'm still recovering from surgery and I can't exert myself much yet.

I'll get back as soon as I have someone who can help me with heavy lifting.
Attached Images
File Type: png temp2.png (97.5 KB, 48 views)
Reply With Quote


  #6  
Old 11-08-2017, 10:44 AM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 10,319
POST ID # = 925718
c-attack c-attack Reputation: 107
Default

OK, it stays relatively smooth and steady in 50's at idle, I think we are looking at a pump or flow problem. 50-80C and really jumpy might be a contact issue, but a constant (more or less) 50-60 at idle most likely means cooler issues.

The AUXFAN value in HWmonitor likely corresponds to the pump RPM for the H60 (you are connected AUXFAN on the board). That ~4300 speed is what's expected, so I don't think it is undervolted or a power failure. That leaves a physical flow restriction of some type as the problem. This is not overly common on the H60 and certainly not brand new, but perhaps this exact one had been sitting in inventory for a long time.

If you are still in your return period with the vendor, returning it to them as defective is likely the fastest resolution. If not or if they are unfriendly, contact Corsair directly through the Ticket Support system in the link at the top of the page. You will need a copy of the receipt to establish ownership and warranty start date (purchase date).

In the meantime, you can probably get away with browsing and light desktop work, but NO stress testing and gaming or any other sustained load is going to be too hot. Another classic sign to look for is the slow and steady rise of CPU temperature. If you boot up in the morning and it is steady at 40C, it may slowly drift toward 50 and then possibly 60C. The longer you are up and running, the warmer it gets. Another classic sign is if you run the stress test the baseline CPU temperature will be higher after the test and take seemingly forever to cool back down. So, it was at 50C, you run the stress test for 60 seconds and then it is at 60C and won't go back to 50C. If these symptoms do not match what you see, let me know so we can take a hard look at something else. All of the above require replacement. A simple offset backplate would not.
Reply With Quote


  #7  
Old 11-08-2017, 11:42 AM
DrJester DrJester is offline
Registered User
DrJester's PC Specs
 
Join Date: Nov 2017
Posts: 7
POST ID # = 925724
DrJester Reputation: 10
Default

since doing quick tests doesn't require physical exertion. I did another one with the speedfan graph.

Here's the result using Intel XTU as CPU stress test.

EDIT: You are right, they were also the only 3 and 4 pin slots I had free on my motherboard that were close enough for the power cable
Attached Images
File Type: png temp3.png (329.8 KB, 54 views)
Reply With Quote


  #8  
Old 11-08-2017, 12:23 PM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 10,319
POST ID # = 925726
c-attack c-attack Reputation: 107
Default

Well, I see the general temp rising relatively quickly after the initial jump - maybe 1C every few seconds. That does suggest a cooler limitation or mild flow issue. The after effect is not as clear. The general 100% load temp goes up 8C or so in that minute. That is the coolant temperature going up 8C inside the cooler. When you stop, the baseline CPU temp should be +8C as well, fort example 40C before and 48C after. However, it doesn't look like that much in the graph. Maybe my perspective is off. It should also take the cooler a good 10-20 minutes to come back down, where as a working cooler would dissipate the 8C within 2-4 minutes and likely would not have a *C rise in the first place.

This may be a mild restriction, but the other variable is fan speed. I don't want to make assumptions from the HWMon data. The first one made sense, but your second screen shot show a 17000 rpm speed. Not sure what that is. Is CPU fan the radiator fan? How fast do you have it going?
Reply With Quote


  #9  
Old 11-08-2017, 12:52 PM
DrJester DrJester is offline
Registered User
DrJester's PC Specs
 
Join Date: Nov 2017
Posts: 7
POST ID # = 925730
DrJester Reputation: 10
Default

Quote:
Originally Posted by c-attack View Post
Well, I see the general temp rising relatively quickly after the initial jump - maybe 1C every few seconds. That does suggest a cooler limitation or mild flow issue. The after effect is not as clear. The general 100% load temp goes up 8C or so in that minute. That is the coolant temperature going up 8C inside the cooler. When you stop, the baseline CPU temp should be +8C as well, fort example 40C before and 48C after. However, it doesn't look like that much in the graph. Maybe my perspective is off. It should also take the cooler a good 10-20 minutes to come back down, where as a working cooler would dissipate the 8C within 2-4 minutes and likely would not have a *C rise in the first place.

This may be a mild restriction, but the other variable is fan speed. I don't want to make assumptions from the HWMon data. The first one made sense, but your second screen shot show a 17000 rpm speed. Not sure what that is. Is CPU fan the radiator fan? How fast do you have it going?

CPUFANIN0 is the radiator fan, whilst the AUXFAN is the pump.

Why it reached 17000rpm I have no idea!

When I first booted the PC with the new watercooler on, I noticed that it took a while to start. I was in the bios, and it read 0RPM for quite some time.

Thanks for all the help so far.
I'll do a further tests later, but I'm afraid of having to take it out since I used the already pre-applied thermal paste that came with the H60 pump.
Reply With Quote


  #10  
Old 11-08-2017, 01:31 PM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 10,319
POST ID # = 925737
c-attack c-attack Reputation: 107
Default

There is no need to take the H60 or the cold plate off right now, except of course to replace it. That tight little wavy pattern in XTU indicates you put on right and made good contact. If you had not, those little waves would have been jagged, tall, and somewhat erratic. That combined with the immediate slow escalation in load temps suggests the problem is with the cooler itself. 2000 rpm is plenty of fan speed. Your results should have been OK with a lot less.

The odd pump speed readings are curious and may be related to the problem. On newer variable speed pumps with their own software control systems, it is not uncommon to see wacky readings. However, the H60 is a simple design and should read more or less the same all the time in the 4000-4400 range. The zero and 17000 readings are unusual and it may be a stuttering or erratic pump is partially responsible. Either way, you do not need to distinguish between a partial blockage and a failing pump. The H60 is not working as it is supposed to and you should return in first chance you get. With new coolers that have coolant temp and CPU temp data, separating factors is a lot easier. It takes more steps to work it out on something like a H60 when there is no obvious cause.
Reply With Quote


  #11  
Old 11-08-2017, 02:22 PM
DrJester DrJester is offline
Registered User
DrJester's PC Specs
 
Join Date: Nov 2017
Posts: 7
POST ID # = 925741
DrJester Reputation: 10
Default

Quote:
Originally Posted by c-attack View Post
There is no need to take the H60 or the cold plate off right now, except of course to replace it. That tight little wavy pattern in XTU indicates you put on right and made good contact. If you had not, those little waves would have been jagged, tall, and somewhat erratic. That combined with the immediate slow escalation in load temps suggests the problem is with the cooler itself. 2000 rpm is plenty of fan speed. Your results should have been OK with a lot less.

The odd pump speed readings are curious and may be related to the problem. On newer variable speed pumps with their own software control systems, it is not uncommon to see wacky readings. However, the H60 is a simple design and should read more or less the same all the time in the 4000-4400 range. The zero and 17000 readings are unusual and it may be a stuttering or erratic pump is partially responsible. Either way, you do not need to distinguish between a partial blockage and a failing pump. The H60 is not working as it is supposed to and you should return in first chance you get. With new coolers that have coolant temp and CPU temp data, separating factors is a lot easier. It takes more steps to work it out on something like a H60 when there is no obvious cause.
Thank you so very much for the help!
I'll do one last test tonight, because it is when I can get help, and if it is the same thing(which I will post the results here) I'll contact the store. But considering how difficult they are, I have a feeling I may have to contact Corsair later.
Bought it new, factory sealed, arrived on monday by mail, so, it should still be within the return period.

Once again, thank you so much!
Reply With Quote


  #12  
Old 11-10-2017, 10:09 AM
DrJester DrJester is offline
Registered User
DrJester's PC Specs
 
Join Date: Nov 2017
Posts: 7
POST ID # = 925914
DrJester Reputation: 10
Default

Quick Update:
It seems you are, indeed right, no change at all, even by pressing the pump with my fingers.

Contacted the store and they have, miraculously, accepted to exchange the product for another H60. Hell froze over! hahaha Lets hope it is not one of those that stayed on their stock for too long.

Once again, thank you for the help!
Reply With Quote


Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -4. The time now is 02:00 AM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2020, vBulletin Solutions, Inc.