- Jan 31, 2014
- 50
- 0
- 18,540
So, not sure if I've added this to the correct category first off - if I haven't, then my apologies.
I have a test rig setup with an Intel i7 4770K for which I've just purchased another Corsair H100i (I've been very satisfied with its performance in the past). I want to use the H100i on a new i7 6700K processor I'm going to be using as of next month so I'm testing it on the 4770K for the timebeibg, however I have concerns over it and I'm not sure whether I'll have to RMA it or if my human error factor is that bad, that I should give up building PC's and let the professionals do it for me.
The H100i's I've used in the past have never really gone past 55 to 60 degrees under full load (I had an H55 as well - same, never above 60 degrees in the summer) however this one is tapping out at nearly 70, with between 30 to 40 degrees under minimal load. Now, I'm stress testing almost as soon as I tighten up the heatsink, so maybe I'm not giving it enough time to set - also, pretty sure the first time I set the heatsink, I accidentally shifted it, so maybe air got trapped in the paste. That being said, I removed the paste with Isopropyl and replaced it, this time using Akasa paste - with the exact same results (bearing in mind, I'm offering little time for the paste to set ). Now, I have mixed views on Akasa and as I used a spreader to apply the paste, figured I must have used too much, so again I removed it and used a coolermaster paste for which I simply dropped a pea sized (maybe a small pea) drop, slightly off center I must admit and then 'squished' the paste down with the heatsink using my housemates burly, Eastern-European fingers to keep it firmly in place whilst I screwed it back on.
15 minutes later, I'm playing Payday 2 (the game I've been using to test the heatsink) and Coretemp pops up stating 'Threshold Crossed' (70 degrees) and proceeds to shutdown.
As I type this, the temperatures are jumping between 29 and 38 (core 1 is always the hottest) - it's 14 degrees here in the UK and the house can't be warmer than 21 degrees. I have the fans in a pull configuration, with the radiator on top and the fans below pulling the air through (easiest way to clean - also, LED SP fans, which look amazing) and they're set to full, turbo speed from the BIOS. The windows and doors are wide open and the case is a Corsair 450D, with two AF140 fans on the front, no HDD compartment behind it, meaning airflow is nearly uninterrupted before it reaches the AF120 fan at the rear exhaust. For reference, the two Asus Strix 980's don't go above 57 degrees, so I'm pretty sure external heat is not the issue. Lastly, no overclocking aside from Boost, but this doesn't kick in on Payday (MSG - GZ, The Witcher 3 and GTAV yes, but not Payday).
All in all, am I being impatient, overly cautious or is there something I should be concerned over here? Also, are the legends of the paste needing a minimum of 8 hours before realistic results can be taken true?
I have a test rig setup with an Intel i7 4770K for which I've just purchased another Corsair H100i (I've been very satisfied with its performance in the past). I want to use the H100i on a new i7 6700K processor I'm going to be using as of next month so I'm testing it on the 4770K for the timebeibg, however I have concerns over it and I'm not sure whether I'll have to RMA it or if my human error factor is that bad, that I should give up building PC's and let the professionals do it for me.
The H100i's I've used in the past have never really gone past 55 to 60 degrees under full load (I had an H55 as well - same, never above 60 degrees in the summer) however this one is tapping out at nearly 70, with between 30 to 40 degrees under minimal load. Now, I'm stress testing almost as soon as I tighten up the heatsink, so maybe I'm not giving it enough time to set - also, pretty sure the first time I set the heatsink, I accidentally shifted it, so maybe air got trapped in the paste. That being said, I removed the paste with Isopropyl and replaced it, this time using Akasa paste - with the exact same results (bearing in mind, I'm offering little time for the paste to set ). Now, I have mixed views on Akasa and as I used a spreader to apply the paste, figured I must have used too much, so again I removed it and used a coolermaster paste for which I simply dropped a pea sized (maybe a small pea) drop, slightly off center I must admit and then 'squished' the paste down with the heatsink using my housemates burly, Eastern-European fingers to keep it firmly in place whilst I screwed it back on.
15 minutes later, I'm playing Payday 2 (the game I've been using to test the heatsink) and Coretemp pops up stating 'Threshold Crossed' (70 degrees) and proceeds to shutdown.
As I type this, the temperatures are jumping between 29 and 38 (core 1 is always the hottest) - it's 14 degrees here in the UK and the house can't be warmer than 21 degrees. I have the fans in a pull configuration, with the radiator on top and the fans below pulling the air through (easiest way to clean - also, LED SP fans, which look amazing) and they're set to full, turbo speed from the BIOS. The windows and doors are wide open and the case is a Corsair 450D, with two AF140 fans on the front, no HDD compartment behind it, meaning airflow is nearly uninterrupted before it reaches the AF120 fan at the rear exhaust. For reference, the two Asus Strix 980's don't go above 57 degrees, so I'm pretty sure external heat is not the issue. Lastly, no overclocking aside from Boost, but this doesn't kick in on Payday (MSG - GZ, The Witcher 3 and GTAV yes, but not Payday).
All in all, am I being impatient, overly cautious or is there something I should be concerned over here? Also, are the legends of the paste needing a minimum of 8 hours before realistic results can be taken true?
Facebook
Twitter
Instagram