3930K cooling

Uncle Petey

Uncle Petey

Uncle Petey

Soldato
Joined
22 Jul 2012
Posts
16,160
Location
London
I recently saw a post here that claimed a more than 20c drop in temps (load and idle) on a 3930K @4.5Ghz after switching from a Silver Arrow SB-E to an H100i.

Naturally, I called BS, but I hear a few people say that water (including CLC's) have a significant advantage over air on SB-E chips. That's fine with me, but 20c difference?

What say you gentlemen? Any 3930K owners that have tried high end air vs CLC?
Btw, this isn't about discrediting another poster. I'm genuinely curious to know what the deal is here and whether it's just a case of a bad install or something.
 
Only way I could see that much of a difference is the H100i was getting good separation of intake and exhaust air and possibly mounted to cool by pulling outside air though it into case as well. and the SA SB-E setup was was not.. The heated SA-SB-E exhaust air was mixing with air in case and causing SA SB-E intake air to be pre-heated by 15c so CPU was 15c hotter.
 
I got a 5c difference moving from a Phanteks PH-TC14PE with the 2 standard fans to the Corsair H100i with 2x SP120 Quiet fans as push and exhausting out the top of the C70.
Half of that is probably the TIM, now using IC Diamond 24.
Hell of a lot quieter though, that was the main objective.
 
Vortez review says H100i 7.8c cooler than SA SB-E Extreme mounted in top of Corsair 500R venting out and SA SB-E Extreme dumping it's exhaust inside of case. Stock fan setup in case is Includes 200mm side panel fan (92cfm), 2x front-mounted 120mm fans and rear 120mm fan rated at 43cfm. SA SB-E Extreme is rated 130cfm..3 times as much as exhaust fan and 2 times as much as intake fans.. so pretty obvious that the hot air coming out of cooler just in front of fan was not being removed from case. Maybe rear fan was getting a third of the hot air. ;)

At stock settings the H100i was only 2.7c better with same heated air issue.
 
Rossi~ said:
I got a 5c difference moving from a Phanteks PH-TC14PE with the 2 standard fans to the Corsair H100i with 2x SP120 Quiet fans as push and exhausting out the top of the C70.
Half of that is probably the TIM, now using IC Diamond 24.
Hell of a lot quieter though, that was the main objective.
Click to expand...

Agree TIM and intake/exhaust separation are the difference. Did stock fans cool H100i better? I'm sure they weren't as quiet.
 
I have got a h100i with push/pull with 1900rpm scythe fans on my 3930k @ 4.78GHz/1.46VCore and it idles at 31degC on cores. This is in the Cosmos 2 using a fan controller when maxed
 
BaDBoY_uK said:
I have got a h100i with push/pull with 1900rpm scythe fans on my 3930k @ 4.78GHz/1.46VCore and it idles at 31degC on cores. This is in the Cosmos 2 using a fan controller when maxed
Click to expand...
But what are temperature maximums when running full load tests?
 
doyll said:
But what are temperature maximums when running full load tests?
Click to expand...

I purposely didn't put that in because I haven't checked since I raised it to 4.7GHz. I have only played games like ARMA2 and BF3. I will check for you tonight mate. I'd be surprised if it goes above 65 but I could be wrong
 
BaDBoY_uK said:
Me too. They looked... errrr... a tad weak

Have you got the OS software working? I am probably doing something wrong. But it doesn't matter anyway
Click to expand...

Nope, haven't touched the software either.

I put the SP120 Quiet fans on and plugged them into the CPUFAN headers on the motherboard. I don't need the software.
 
BaDBoY_uK said:
I purposely didn't put that in because I haven't checked since I raised it to 4.7GHz. I have only played games like ARMA2 and BF3. I will check for you tonight mate. I'd be surprised if it goes above 65 but I could be wrong
Click to expand...
Was just curious to see how the Scythe compare to stock fans.
 
doyll said:
No problem. Doesn't really make any difference unless you know exactly what the Phanteks max was to compare it to.. as Phanteks and SA SB-E are similar at cooling.
Click to expand...

The last IBT Max test before installing it to compare, the Phanteks hottest core was 82c and with the H100i the hottest was 77c. Hence where i got the 5c from. This is 4.6Ghz @ 1.385v

I bet the temperatures would be a lot better with some faster fans on a fan controller. Tempted to get the SP120 Performance fans and a simple controller, may do before summer.
 
I would be curious to see how well H100i would do with both on level playing field... Phanteks having a duct on exhaust going out of case so none of the heated exhaust was able to mix with case air. There is a review doing this with H100, SA SB-E and SA SB-E Extreme on i7 3820 @4.75GHz.

Temperature is delta.
H100 . . . . . . . . 41c 2500rpm. . . . . . . . 55dBA 2
SA SB-E . . . . . . 42c 1100rpm & 1300rpm 38dBA TY-150 & TY-141
SA SB-E Extreme 34c 2500rpm. . . . . . . . 56dBA 2x TY-143
http://www.youtube.com/watch?v=q6xrsuPwDbo
 
^Only because of the fans it uses.

Rossi, you've settled it for me :) Nice looking build btw.

Your findings and doyll's take on the airflow situation make me think either his airflow was suspect and/or the heatsink was mounted poorly. He claims it was mounted fine and perhaps it was. The massive temp difference tells another story imo.

Btw doyll, was that Vortez review the same one we were all wishing had been done by Helios instead?
 
You must log in or register to reply here.
Back
Top Bottom