Question CPU Cooler Options

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

advancedmixedgaming

Distinguished
Nov 5, 2015
468
4
18,865
I just installed my entire PC into a new case and I saw some pretty major temperature differences on my CPU due to more fans fitted into my new case. Now I am in the market for a CPU cooler that can fit and isn't much more than $60. I used to be up to date on coolers, but now I'm not too sure what to throw in. Any suggestions?
 

advancedmixedgaming

Distinguished
Nov 5, 2015
468
4
18,865
You are right. I am sorry. I gave you the wrong test procedure. I appologize.

I gave you my Prime test for memory testing. This is the correct procedure for verifying thermal compliance.


Run Prime95 version 26.6 or the latest version, in which case you will want to disable AVX and AVX 2 in the options at the bottom of the torture test options screen and choose the "Small FFT test option". Do not choose "Smallest FFT", just "Small FFT". Run this for 15 minutes while monitoring your core/package temperatures to verify that you do not exceed the thermal specifications of your CPU.

(This should be considered to be 80°C for most generations of Intel processor and for current Ryzen CPUs. For older AMD FX and Phenom series, you should use a thermal monitor that has options for "Distance to TJmax" (Like AMD Overdrive, (which is NOT compatible with Ryzen platforms) and you want to NOT see distance to TJmax drop below 10°C distance to TJmax. Anything that is MORE than 10°C distance to TJmax is within the allowed thermal envelope.)

Take a screenshot of the Ryzen master window about five minutes into the test as that is long enough to saturate the package and heatsink in most cases when running a stock configuration. If you are overclocking, I would wait until between 10-15 minutes, because I've seen thermal peaks happen occasionally after ten minutes but never beyond 15.
Test with case
Test with panel off
 

advancedmixedgaming

Distinguished
Nov 5, 2015
468
4
18,865
I've tested multiple times with and without the side panel on and I am lost at this point. Any input at all would be greatly appreciated.

If it helps, I tried reseating and reapplying paste twice now. I used Thermaltake TG-7 in an "X" pattern across the CPU. If this could be a problem, please let me know whomever that may read this.
 
Have you configured the case fan and CPU cooler fan profiles in the BIOS? You might want to try a slightly more aggressive fan profile or custom curve for your CPU cooler fan.

You are 100% certain you got a good mount of the heatsink? How much paste and what method did you apply it?

-Some methods "Recommended" around the web-


1zfhjl2.jpg





-The likely, but undesirable results of those methods-


2m5le92.jpg




I don't recommend any of those, and to clarify, I'm not claiming to be an authority or trying to indicate there is no other way but mine, but it is one that has worked on literally hundreds of builds over the years without issue.

I personally prescribe to the large cooked rice grain (For high mounting pressure type heatsinks) or a dab about half the size (Of the part that sticks past the metal) of a #2 pencil eraser (Low mounting pressure type heatsinks) when using any of the commonly available pastes that come in a tube applicator like the Arctic Silver 5 or most of the included pastes that come with premium coolers. If you are using a stock, low pressure mounting system type cooler like those which use the stock mounting hardware and backplate, you will want to probably use double the amount of thermal interface material (paste) because the low pressure will not effectively force the paste outward from the center point nearly as much as aftermarket high mounting pressure coolers do.

I've never had excess paste squeeze from between the CPU lid and heatsink using this method nor had an issue with inadequate cooling due to any lack of sufficient TIM. You can make your own decision on what works best for you. Depending on the cooler, a very small snow pea sized dab will work well too, so long as you don't get carried away or misconstrue the definition of snow pea. For our purposes, perhaps a dab about 1/3 the size of a number 2 pencil eraser is a better description.



-More promising methods-

o086c9.jpg





-Probable results-

33d9k6u.jpg

27ya137.jpg





Notice that even with this method there is substantial filling of the heat pipe seams. It is NOT necessary to lay multiple lines along seams.
 

advancedmixedgaming

Distinguished
Nov 5, 2015
468
4
18,865
Have you configured the case fan and CPU cooler fan profiles in the BIOS? You might want to try a slightly more aggressive fan profile or custom curve for your CPU cooler fan.

You are 100% certain you got a good mount of the heatsink? How much paste and what method did you apply it?

-Some methods "Recommended" around the web-


1zfhjl2.jpg





-The likely, but undesirable results of those methods-


2m5le92.jpg




I don't recommend any of those, and to clarify, I'm not claiming to be an authority or trying to indicate there is no other way but mine, but it is one that has worked on literally hundreds of builds over the years without issue.

I personally prescribe to the large cooked rice grain (For high mounting pressure type heatsinks) or a dab about half the size (Of the part that sticks past the metal) of a #2 pencil eraser (Low mounting pressure type heatsinks) when using any of the commonly available pastes that come in a tube applicator like the Arctic Silver 5 or most of the included pastes that come with premium coolers. If you are using a stock, low pressure mounting system type cooler like those which use the stock mounting hardware and backplate, you will want to probably use double the amount of thermal interface material (paste) because the low pressure will not effectively force the paste outward from the center point nearly as much as aftermarket high mounting pressure coolers do.

I've never had excess paste squeeze from between the CPU lid and heatsink using this method nor had an issue with inadequate cooling due to any lack of sufficient TIM. You can make your own decision on what works best for you. Depending on the cooler, a very small snow pea sized dab will work well too, so long as you don't get carried away or misconstrue the definition of snow pea. For our purposes, perhaps a dab about 1/3 the size of a number 2 pencil eraser is a better description.



-More promising methods-

o086c9.jpg





-Probable results-

33d9k6u.jpg

27ya137.jpg





Notice that even with this method there is substantial filling of the heat pipe seams. It is NOT necessary to lay multiple lines along seams.
I did try the X pattern, so maybe there's too much thermal paste? I also tightened until the cooler gave a definitive stop, which the Scythe Ninja 5 does. I'll reapply the paste and see what happens.
I also believe I have the CPU cooler running at %100 at all times because it is so damn quiet.
Should I use the Scythe thermal paste it came with, or my TG-7?
 
Well, that's definitely an improvement. Honestly I think you should be even lower than that though. When you are running Prime95, you are running the latest version AND, and this is VERY important, you are putting a check in the box next to "disable AVX2" and then it will un-gray the option for AVX, and then you put a check next to "disable AVX" as well, right?

If not, you need to do that. Disable both AVX2 and AVX instructions in the main window that pops up when you run Prime. Also, make sure you are running "Small FFT", NOT "Smallest FFT".

After figuring out that you had one of the fans backwards, I would double check that your case fans are all pointing the right direction as well. Not to poke fun, but seriously, because mistakes can happen. Make sure that your front fans have the fan blade side facing out towards the fresh air, and that your exhaust fans in the top and rear have the fan blades facing in. The side of the fan with the supporting arms for the fan motor are always supposed to face towards the direction you want air to move towards. Also, there are usually arrows indicating the direction of airflow on any quality fan.
 

advancedmixedgaming

Distinguished
Nov 5, 2015
468
4
18,865
Well, that's definitely an improvement. Honestly I think you should be even lower than that though. When you are running Prime95, you are running the latest version AND, and this is VERY important, you are putting a check in the box next to "disable AVX2" and then it will un-gray the option for AVX, and then you put a check next to "disable AVX" as well, right?

If not, you need to do that. Disable both AVX2 and AVX instructions in the main window that pops up when you run Prime. Also, make sure you are running "Small FFT", NOT "Smallest FFT".

After figuring out that you had one of the fans backwards, I would double check that your case fans are all pointing the right direction as well. Not to poke fun, but seriously, because mistakes can happen. Make sure that your front fans have the fan blade side facing out towards the fresh air, and that your exhaust fans in the top and rear have the fan blades facing in. The side of the fan with the supporting arms for the fan motor are always supposed to face towards the direction you want air to move towards. Also, there are usually arrows indicating the direction of airflow on any quality fan.
AVX was all disabled, I am sure of that.
I just took a look at all of the fans, and they are all %100 facing the right way. The way I have my case configured at the moment is having 2 intake fans on the front panel, and one exhaust fan on the top of the rear panel. I don't have any chassis fan headers left, but I have CPU fan #2 left that I ASSUME I could attach a case fan to. I have room for 2 fans on the top, 2 on the bottom, and I believe one more in the rear. Would putting a fan in any of these places critically improve airflow or temps at all?
 
You don't want to use the CPU fan #2 header. If you want to add another fan to the TOP rear fan location, which is the only place I would recommend adding one at, you can just use a fan splitter cable. It would cost you maybe five bucks max on Amazon or Newegg. If you are using PWM four pin fans, then you need a PWM four pin splitter. If you are using three pin DC controlled fans, then you need a three pin fan splitter cable.
 

advancedmixedgaming

Distinguished
Nov 5, 2015
468
4
18,865
You don't want to use the CPU fan #2 header. If you want to add another fan to the TOP rear fan location, which is the only place I would recommend adding one at, you can just use a fan splitter cable. It would cost you maybe five bucks max on Amazon or Newegg. If you are using PWM four pin fans, then you need a PWM four pin splitter. If you are using three pin DC controlled fans, then you need a three pin fan splitter cable.
Do you think the temps would noticeably improve if I added an exhaust there?
 
Even a 2 degree improvement is a score. Some of us work very hard at gaining one degree here, and one degree there. People spend upwards of 20-30 dollars on thermal pastes, specialized metallics etc., while from top to bottom the difference in temps from the best pastes to the worst rarely achieve more than a couple of degrees difference.

Yes, I think it is worth adding. It might not offer a drop on the max core temps, because the cores become heat saturated long before it can be conducted away into the heatsink, which is why a high quality motherboard that has the ability to handle some of that itself can be important, but it will certainly help with average temps and may help to INCREASE the amount of time it takes for the CPU to reach it's max temperature. In some cases, that could mean that it doesn't reach it, which would in effect lower the max temp for normal usage but might not for fully saturated thermal testing. And then again, it might. Every configuration is somewhat different.

I personally wouldn't run any standard ATX system without at least two intake and two exhaust fans, unless it was a fairly low TDP processor and graphics card. Could help considerably with GPU temps as well.

Did you pull the heatsink off and redo the paste? If not, I'd probably do that anyhow even though you found that fan configuration problem. Be sure to clean the bottom of the heatsink AND the top of the CPU heat spreader using 91% Isopropyl alcohol. You can get it very cheap at any Walmart or pharmacy, and most home centers. Usually like less than two bucks. Then apply a dot dead center on the CPU heat spreader (Lid) about the same size as if you cut a #2 pencil eraser off right at where the metal on the pencil starts, and then cut the loose portion of the eraser into quarters. Or about the size of two cooked rice grains. I'm saying to use slightly more than I recommend on some other heatsinks since that one has a dead stop tightening protocol while others can be variably tightened and may have higher mounting pressure that doesn't require as much paste to get full coverage.

Likely, a single line of paste the width that it comes out of the tube and about 1/8" long would also work fine and be plenty of paste. It does not take a lot of paste. Those other methods I showed you, show why.

In truth, the paste is ONLY there to fill the micro pores in the metal surface AND to bridge the gap if there is any convexity or concavity in the surface of the heat spreader or heatsink base. If both were 100% flat surfaces and there was no porosity to the metal, it wouldn't be necessary at all.
 

advancedmixedgaming

Distinguished
Nov 5, 2015
468
4
18,865
Even a 2 degree improvement is a score. Some of us work very hard at gaining one degree here, and one degree there. People spend upwards of 20-30 dollars on thermal pastes, specialized metallics etc., while from top to bottom the difference in temps from the best pastes to the worst rarely achieve more than a couple of degrees difference.

Yes, I think it is worth adding. It might not offer a drop on the max core temps, because the cores become heat saturated long before it can be conducted away into the heatsink, which is why a high quality motherboard that has the ability to handle some of that itself can be important, but it will certainly help with average temps and may help to INCREASE the amount of time it takes for the CPU to reach it's max temperature. In some cases, that could mean that it doesn't reach it, which would in effect lower the max temp for normal usage but might not for fully saturated thermal testing. And then again, it might. Every configuration is somewhat different.

I personally wouldn't run any standard ATX system without at least two intake and two exhaust fans, unless it was a fairly low TDP processor and graphics card. Could help considerably with GPU temps as well.

Did you pull the heatsink off and redo the paste? If not, I'd probably do that anyhow even though you found that fan configuration problem. Be sure to clean the bottom of the heatsink AND the top of the CPU heat spreader using 91% Isopropyl alcohol. You can get it very cheap at any Walmart or pharmacy, and most home centers. Usually like less than two bucks. Then apply a dot dead center on the CPU heat spreader (Lid) about the same size as if you cut a #2 pencil eraser off right at where the metal on the pencil starts, and then cut the loose portion of the eraser into quarters. Or about the size of two cooked rice grains. I'm saying to use slightly more than I recommend on some other heatsinks since that one has a dead stop tightening protocol while others can be variably tightened and may have higher mounting pressure that doesn't require as much paste to get full coverage.

Likely, a single line of paste the width that it comes out of the tube and about 1/8" long would also work fine and be plenty of paste. It does not take a lot of paste. Those other methods I showed you, show why.

In truth, the paste is ONLY there to fill the micro pores in the metal surface AND to bridge the gap if there is any convexity or concavity in the surface of the heat spreader or heatsink base. If both were 100% flat surfaces and there was no porosity to the metal, it wouldn't be necessary at all.
Yep, I cleaned it off and reapplied the paste as the size of rice.
Would it be necessarily BAD to plug in the chassis fan into the second CPU header, or is it just advised against? If its just advised against, I'll probably hook one in to test the temps and just buy a splitter in the mean time.
 
It's not "bad", but it's VERY annoying, because that fan is going to increase and decrease very rapidly in speed, just like the CPU cooler, based on CPU temp rather than based on the temp of the motherboard thermal diode which is what you normally want case fans running off of. You need to look in the BIOS in your fan controls. Many boards these days allow you to change WHICH thermal sensor each fan header uses for rate control. You might be able to change that CPU 2 header to run off the same sensors as the rest of the case fans, but you may not as well. Usually the CPU headers are not changeable as they assume it will always either be a CPU fan or a pump that is connected to those.

But if the noise doesn't bother you, you can certainly use it so long as your fan is a four pin PWM model.
 

advancedmixedgaming

Distinguished
Nov 5, 2015
468
4
18,865
It's not "bad", but it's VERY annoying, because that fan is going to increase and decrease very rapidly in speed, just like the CPU cooler, based on CPU temp rather than based on the temp of the motherboard thermal diode which is what you normally want case fans running off of. You need to look in the BIOS in your fan controls. Many boards these days allow you to change WHICH thermal sensor each fan header uses for rate control. You might be able to change that CPU 2 header to run off the same sensors as the rest of the case fans, but you may not as well. Usually the CPU headers are not changeable as they assume it will always either be a CPU fan or a pump that is connected to those.

But if the noise doesn't bother you, you can certainly use it so long as your fan is a four pin PWM model.
I'll install it tomorrow if I remember and let you know how the temps are affected .