Running hot. SOS

[BoinkTV]

Hi all,

 

Upgraded over Christmas to an i7 9700k. I have a Corsair H115i AIO in an IBP Snowblind S Case. Prior to my upgrade, from an i5 8400, I would sit at 70deg under load. I have noticed even if I leave my PC on during the day it ends up in the high 80s, low 90s even at times. Mostly when it creeps into the 90s, it will be under load, but I have caught it in the high 80s just idling. If I sleep my PC during the day, when it runs for a few hours under load when I am home, it sits in the 70s.

 

I have dusted it a few times since, even reseated and pasted my CPU.

 

Do I need a new AIO? Do I need to switch to a CPU fan? I have debated hardline water cooling, but now with the Snowblind S, I would hardly see it. My GPU rarely creeps over 50.

 

 

                   Side Question- PC Blue screens whenever I move my Focusrite Scarlet 2i2 audio interface. Not that concerned with this as I can just you know, not touch it, but if anyone has encountered this before it would be greatly appreciated!

 

 

TLDR: Specs

 

CPU: i7 9700k

GPU: 2060 8GB OC

PSU: Thermaltake GOLD 750W

RAM: 2x 8GB Corsair Vengeance 3200mhz cl16

AIO: Corsair H115i (2x 120mm rad)

Case: IBP Snowblind S

 

 

Thanks in advance for your help!

[Bombastinator]

Depends on how you define too hot.  Is the liquid inside you AIO that hot too?  Probably no good way to tell.  Are the fans on the rad having any problems? Pump issues are usually more severe than this.  Might try a repaste.  The thing is that’s a 240 rad so it should have enough horsepower. 

[-rascal-]

The iBuyPower Snowbird S has pretty poor airflow.

Your ONLY form of air intake is the RESTRICTED side panel.

The front is COMPLETELY closed off....and no ventilation at the top of the case.

Only other form of ventilation is the single rear fan spot. 

 

The CPU temperature slowly climbing higher the longer the system is on would indicator airflow issue.

Hot air / heat is not being removed out of the system fast enough, and "heatsoaking."

 

Temporarily remove the glass side panel, and see if there is much change in temperature (i.e. 5*C or more).

If you see a significant change/drop, it would confirm airflow issue.

[Mark Kaine]

4 hours ago, -rascal- said:

Your ONLY form of air intake is the RESTRICTED side panel.

The front is COMPLETELY closed off....and no ventilation at the top of the case.

Only other form of ventilation is the single rear fan spot. 

to be really fair, my inwin has very similar "front intakes".

 

(its getting ~70C during C23 multi)

 

although the openings are a bit more sensible i guess… (and its also completely open at the bottom with just a dust filter basically, tbf)

 

 

idk you're right tho, it does sound like airflow issue, so i wonder

about the fan setup, also what it usually is, "bad mounting / pressure" perhaps?

 

[Bombastinator]

Wow.  That is exceptionally bad.  One can sometimes power past restrictive airflow by using high static pressure 120mm fans but if there really is zero airflow that won’t help.  A lot of older cases with blocked fronts like that had an air inlet in the bottom front.  It wasn’t great but it was better than nothing with a high static pressure fan on it.   Move for super restrictive cases like that might be go heavy negative pressure.  Point all your fans out.  Try to maximize intake from things like cracks betweeen the case and the door and stuff like that. Heavy negative pressure can have problems only one of which is dust bunnies in weird places. But it might be needed just to get enough airflow.

 

one quick test to see how bad a hot case problem might be is to run the thing without the side panel for a while and see if it gets a lot better.  If it does you have an airflow limitation. 

Edited June 10, 2021 by Bombastinator

[-rascal-]

1 hour ago, Mark Kaine said:

to be really fair, my inwin has very similar "front intakes".

 

(its getting ~70C during C23 multi)

 

although the openings are a bit more sensible i guess… (and its also completely open at the bottom with just a dust filter basically, tbf)

 

 

idk you're right tho, it does sound like airflow issue, so i wonder

about the fan setup, also what it usually is, "bad mounting / pressure" perhaps?

 

 

Good point.

It looks like the iBuyPower version does not have a ventilated bottom -- appears to be completely closed panel, minus the  vents for the PSU.

The side vents is helluva lot more restricted than the InWin 301 that you have -- they look like just tiny 1 or 2mm cuts...

 

Fan set-up and CPU block mounting is a good-to-know as well.

[BoinkTV]

16 hours ago, -rascal- said:

 

Good point.

It looks like the iBuyPower version does not have a ventilated bottom -- appears to be completely closed panel, minus the  vents for the PSU.

The side vents is helluva lot more restricted than the InWin 301 that you have -- they look like just tiny 1 or 2mm cuts...

 

Fan set-up and CPU block mounting is a good-to-know as well.

I have my AIO Rad at the front of the case mounted to pull cold air from those tiny slots on the side, and an exhaust fan. there is only space for 3 120mm fans in the case, would it be better to exhaust them all?

 

[BoinkTV]

17 hours ago, Bombastinator said:

Wow.  That is exceptionally bad.  One can sometimes power past restrictive airflow by using high static pressure 120mm fans but if there really is zero airflow that won’t help.  A lot of older cases with blocked fronts like that had an air inlet in the bottom front.  It wasn’t great but it was better than nothing with a high static pressure fan on it.   Move for super restrictive cases like that might be go heavy negative pressure.  Point all your fans out.  Try to maximize intake from things like cracks betweeen the case and the door and stuff like that. Heavy negative pressure can have problems only one of which is dust bunnies in weird places. But it might be needed just to get enough airflow.

 

one quick test to see how bad a hot case problem might be is to run the thing without the side panel for a while and see if it gets a lot better.  If it does you have an airflow limitation. 

Ran it for an hour with the side panel off. Temp was mid 70s under 50% load and low 80s under load. say about an average of 5deg drop

[BoinkTV]

22 hours ago, -rascal- said:

The iBuyPower Snowbird S has pretty poor airflow.

Your ONLY form of air intake is the RESTRICTED side panel.

The front is COMPLETELY closed off....and no ventilation at the top of the case.

Only other form of ventilation is the single rear fan spot. 

 

The CPU temperature slowly climbing higher the longer the system is on would indicator airflow issue.

Hot air / heat is not being removed out of the system fast enough, and "heatsoaking."

 

Temporarily remove the glass side panel, and see if there is much change in temperature (i.e. 5*C or more).

If you see a significant change/drop, it would confirm airflow issue.

The side airflow slots are actually bigger than they seem in the photo. it actually does unload a fair bit of air, but I am currently using it as intake, as I have an AC unit on that side of the PC.

[Bombastinator]

4 minutes ago, BoinkTV said:

Ran it for an hour with the side panel off. Temp was mid 70s under 50% load and low 80s under load. say about an average of 5deg drop

That implies not too big a problem with the case.  The numbers are still higher than expected though.  Dependent somewhat on what cpu and cooler you have of course.

[-rascal-]

13 minutes ago, BoinkTV said:

Ran it for an hour with the side panel off. Temp was mid 70s under 50% load and low 80s under load. say about an average of 5deg drop

 

Okay, so with the side panel removed had some difference....so there is room for improvement.

However, I was expecting even more temperature drop.

 

The i7-9700K will dump out more heat than a i5-8400, without a doubt.

 

7 minutes ago, Bombastinator said:

That implies not too big a problem with the case.  The numbers are still higher than expected though.  Dependent somewhat on what cpu and cooler you have of course.

 

Corsiar H115i.

 

My next thing to target would be the AIO itself.

Like @Mark Kaine mentioned, AIO mounting pressure, and if the thermal paste was applied properly.

 

Speaking of thermal paste, did you RE-USE the thermal paste, or did you clean off the old, and put new thermal paste on, when you did the i5-8400 to i7-9700K swap?

You should NOT be re-using thermal paste, as remove the CPU cooler off the CPU will "tamper" with the thermal paste layer, making it less effective.

[Bombastinator]

Just now, -rascal- said:

 

Okay, so with the side panel removed had some difference....so there is room for improvement.

However, I was expecting even more temperature drop.

 

The i7-9700K will dump out more heat than a i5-8400, without a doubt.

 

 

Corsiar H115i.

 

My next thing to target would be the AIO itself.

Like @Mark Kaine mentioned, AIO mounting pressure, and if the thermal paste was applied properly.

 

Speaking of thermal paste, did you RE-USE the thermal paste, or did you clean off the old, and put new thermal paste on, when you did the i5-8400 to i7-9700K swap?

Cold plate stuff.  Yeah that could be checked.  There are also some higher quality than stock  pastes that can make a real difference.  I used some and got 8° out of it.  Even several years later the stuff is still fine. 

[BoinkTV]

1 hour ago, -rascal- said:

 

Okay, so with the side panel removed had some difference....so there is room for improvement.

However, I was expecting even more temperature drop.

 

The i7-9700K will dump out more heat than a i5-8400, without a doubt.

 

 

Corsiar H115i.

 

My next thing to target would be the AIO itself.

Like @Mark Kaine mentioned, AIO mounting pressure, and if the thermal paste was applied properly.

 

Speaking of thermal paste, did you RE-USE the thermal paste, or did you clean off the old, and put new thermal paste on, when you did the i5-8400 to i7-9700K swap?

You should NOT be re-using thermal paste, as remove the CPU cooler off the CPU will "tamper" with the thermal paste layer, making it less effective.

Yes, there was a difference, but like you said it is still worrisome. I did expect the 9700k to put out more heat, but not more than 5-10deg.

 

Obviously I can't know if the mounting pressure is 100% correct, however the paste was new both times I did it. I used one that I bought on amazon as a recommendation from one of the tech youtubers.

 

Would it be worth swapping to like a Noctua CPU fan? custom looping for only my CPU? (if paste ends up not being the issue)

[Bombastinator]

3 minutes ago, BoinkTV said:

Yes, there was a difference, but like you said it is still worrisome. I did expect the 9700k to put out more heat, but not more than 5-10deg.

 

Obviously I can't know if the mounting pressure is 100% correct, however the paste was new both times I did it. I used one that I bought on amazon as a recommendation from one of the tech youtubers.

 

Would it be worth swapping to like a Noctua CPU fan? custom looping for only my CPU? (if paste ends up not being the issue)

Nah.  Fans are fans.  As long as they spin the difference between one and another isn’t huge for these purposes.  One may blow more air or make less noise that about it.
 

It could be that either your cpu ihs or your cold plate is dented and far from flat I suppose.  It’s a somewhat rare thing but it can keep a cooler from performing well even if everything else is good.  The move to check it would be take something you know is very flat like the edge of a metal ruler or something, place it against the cleaned part in question and see if you can see light coming out.  If it turns out there is a problem there is repair by lapping which is an immensely involved process. Paste is meant to take up slack in that though so unless the problem is really bad a decent paste should make it not a very big deal. 
 

GN had a big Corsair cooler with a not flat heat plate and showed what it does in one of their videos. It’s something that would either be that way from the start or would require serious bang to change it though.  The things don spontaneously warp.

[-rascal-]

1 hour ago, BoinkTV said:

Yes, there was a difference, but like you said it is still worrisome. I did expect the 9700k to put out more heat, but not more than 5-10deg.

 

Obviously I can't know if the mounting pressure is 100% correct, however the paste was new both times I did it. I used one that I bought on amazon as a recommendation from one of the tech youtubers.

 

Would it be worth swapping to like a Noctua CPU fan? custom looping for only my CPU? (if paste ends up not being the issue)

 

Also, check if Multi-Core Enhancement (MCE) is enabled in the BIOS.

90%+ of the time, it will be on 'Auto' or 'Enabled.'

Just curious if it was 'Disabled' when you had the i5-8400, or running at a less "aggressive" mode.

 

Keep in mind, the i5-8400 is a "65W" chip, while the i7-9700K is "95W."

But if MEC is enabled, and/or running at the more aggressive/performance mode, that i7-9700K is NOT "95W" anymore.

More like 100W+.

 

https://www.gamersnexus.net/hwreviews/3421-intel-i7-9700k-review-benchmark-vs-8700k-and-more

Quote

Keep in mind that this is with MCE disabled, so it’s following the boost duration specification and sticking to a power consumption of 90W. Liquid temperature is at around 6 degrees over ambient during this test pass.

 

MCE enabled violates Intel's TDP specs...for higher CPU frequency, and more heat.

(If you care to read / watch a video about it)

https://www.gamersnexus.net/guides/3389-intel-tdp-investigation-9900k-violating-turbo-duration-z390

[BoinkTV]

22 hours ago, -rascal- said:

 

Also, check if Multi-Core Enhancement (MCE) is enabled in the BIOS.

90%+ of the time, it will be on 'Auto' or 'Enabled.'

Just curious if it was 'Disabled' when you had the i5-8400, or running at a less "aggressive" mode.

 

Keep in mind, the i5-8400 is a "65W" chip, while the i7-9700K is "95W."

But if MEC is enabled, and/or running at the more aggressive/performance mode, that i7-9700K is NOT "95W" anymore.

More like 100W+.

 

https://www.gamersnexus.net/hwreviews/3421-intel-i7-9700k-review-benchmark-vs-8700k-and-more

 

MCE enabled violates Intel's TDP specs...for higher CPU frequency, and more heat.

(If you care to read / watch a video about it)

https://www.gamersnexus.net/guides/3389-intel-tdp-investigation-9900k-violating-turbo-duration-z390

Have some time to tinker today- do I want MCE on? or off.

[-rascal-]

15 hours ago, BoinkTV said:

Have some time to tinker today- do I want MCE on? or off.

 

I am just curious how much MCE affects temperatures, and IF it's the cause.

If you want the CPU to run within Intel specs, have MCE off, or set to a less "aggressive" mode (some motherboards have different levels).

 

Also, what motherboard are you using?

Make and model?

[BoinkTV]

8 hours ago, -rascal- said:

 

I am just curious how much MCE affects temperatures, and IF it's the cause.

If you want the CPU to run within Intel specs, have MCE off, or set to a less "aggressive" mode (some motherboards have different levels).

 

Also, what motherboard are you using?

Make and model?

I have the ASUS Prime Z370-A