Please Help... Odd fan rpm issue, jump to post 11 on Dec 16

Cooling Processors quietly

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
andyb
Patron of SPCR
Posts: 3307
Joined: Wed Dec 15, 2004 12:00 pm
Location: Essex, England

Please Help... Odd fan rpm issue, jump to post 11 on Dec 16

Post by andyb » Mon Dec 03, 2012 12:32 pm

This might seem like a slightly strange question.

Will a "very good" CPU cooler be a lot better than a "good" CPU cooler.? The reason why I ask is the well documented problem with the heat transfer between the CPU and the heatspreader, this of course would make a large difference between the CPU temp and the cooling ability of an aftermarket cooler.

I wont be removing the heatspreader to manually fix the problem by replacing the TIM like some people have done because that would obviously invalidate the warranty and I cannot take that risk.

Basically I am not sure whether I should go for something like the "Coolermaster Hyper 212 Evo", or something altogether beefier like the "Thermalright Ultra-120 eXtreme Rev.C", the main point being "will there actually be a large difference in the CPU temp.?"

Whatever I will do will involve ditching the horrid little noisy cooler that came with the CPU, and will involve me overclocking the CPU as much as is possible "without over-volting", which will likely be to 4.2GHz or perhaps a little more.

By my calculations an OC to 4.2GHz is 23%, add 23% to 77W and you basically have a 95W CPU, so the cooler doesn't have to be the best of the best to deal with that level of heat...... unless the heatspreader issue changes things - that to me is an unknown quantity.


Andy
Last edited by andyb on Tue Dec 18, 2012 6:14 pm, edited 3 times in total.

CA_Steve
Moderator
Posts: 7650
Joined: Thu Oct 06, 2005 4:36 am
Location: St. Louis, MO

Re: i5 3570K OC cooler questions

Post by CA_Steve » Mon Dec 03, 2012 1:44 pm

I don't have on-hand experience with OC'ing IVB, just read a lot of experiences, as you have. Seems like IVB really heads into the weeds temp-wise when people start over-volting. That said, if I were building an i5-3470K system that I might OC and not OV with a gaming card in a quiet case, I'd lean toward the cooler that provides 5C better temps at similar SPL.

mkk
Posts: 687
Joined: Sun Sep 05, 2004 1:51 pm
Location: Gefle, Sweden
Contact:

Re: i5 3570K OC cooler questions

Post by mkk » Mon Dec 03, 2012 1:50 pm

Hyper 212 Evo or similar will be fine for the intended light overclocking. A bigger alternative like the Mugen 3 would be a little bit quieter, if that's important. Improving the factory applied paste is only really meaningful for higher overclocking or possibly when the cooling options are limited and the individual sample happens to be one with worse than average contact.

andyb
Patron of SPCR
Posts: 3307
Joined: Wed Dec 15, 2004 12:00 pm
Location: Essex, England

Re: i5 3570K OC cooler questions

Post by andyb » Mon Dec 03, 2012 2:02 pm

Thank you both, I expect that I will get the Hyper 212 Evo (hopefully before Christmas) and swap the fan with an AC F12 running via Speedfan.

I will report back with my findings.


Andy

kuzzia
Friend of SPCR
Posts: 709
Joined: Fri Jul 02, 2010 4:41 am
Location: Denmark

Re: i5 3570K OC cooler questions

Post by kuzzia » Tue Dec 04, 2012 2:29 am

andyb wrote:Thank you both, I expect that I will get the Hyper 212 Evo (hopefully before Christmas) and swap the fan with an AC F12 running via Speedfan.

I will report back with my findings.


Andy
I am very much looking forward to it. Could you do us all a favor? Many SPCR enthusiasts repel the stock Cooler Master fans based on the experiences of the stock fans found in the Sileo 500 (case), Hyper 212+ (CPU cooler) etc. But this fan looks radically different from Cooler Master's old fans so perhaps it doesn't tick annoyingly like their old fans did. So could you test the CM CPU fan prior to replacing with the AC fan so the SPCR forum would know how the CM fan performed acoustically? I assume that many in this forum would find that knowledge interesting.

CA_Steve
Moderator
Posts: 7650
Joined: Thu Oct 06, 2005 4:36 am
Location: St. Louis, MO

Re: i5 3570K OC cooler questions

Post by CA_Steve » Tue Dec 04, 2012 8:52 am

kuzzia wrote:So could you test the CM CPU fan prior to replacing with the AC fan so the SPCR forum would know how the CM fan performed acoustically? I assume that many in this forum would find that knowledge interesting.
+1

andyb
Patron of SPCR
Posts: 3307
Joined: Wed Dec 15, 2004 12:00 pm
Location: Essex, England

Re: i5 3570K OC cooler questions

Post by andyb » Tue Dec 04, 2012 3:11 pm

I am very much looking forward to it. Could you do us all a favor? Many SPCR enthusiasts repel the stock Cooler Master fans based on the experiences of the stock fans found in the Sileo 500 (case), Hyper 212+ (CPU cooler) etc. But this fan looks radically different from Cooler Master's old fans so perhaps it doesn't tick annoyingly like their old fans did. So could you test the CM CPU fan prior to replacing with the AC fan so the SPCR forum would know how the CM fan performed acoustically? I assume that many in this forum would find that knowledge interesting.
I certainly can do that.

It should be relatively simple to do as well. I have just thought up a simple plan.

The Plan:

Run the cooler with no fan attached :shock: and use speedfan to set the fan speed whilst the CPU is idling on the desktop and then simply keep on swapping the fans and changing the RPM (and speedfans reporting as a percentage) and report back my findings.

The beauty of this is that I can do my testing whilst the PC is on and type them up as I go and I wont really have to worry about the CPU overheating for a while.

I will use 10% increments in speedfan (starting at 30%) on the F12 and note the RPM, I will then run the CM fan at as close an RPM as possible and note its speedfan % and then add notes, such as didn't reliably start, was buzzy, clicky etc and relatively how each fan sounds relative to the other at the same RPM.

The main drawbacks are obvious, I have no reliable or accurate way of identifying the actual amount of air that each fan is pushing, let alone the static pressure and it would be a nightmare to even attempt to identify which is the better fan attached to a CPU cooler and at what relative noise level as I have no such equipment to test with and I don't want to be constantly swapping the fans.

I can however do a very simple test with both fans running at xxx RPM with prime95 running and check the temperature, this will also be unreliable as I am in a very small room (9x10 ft ish) so the temperature can rise very rapidly between tests which is why I think that the best that I can do is to simply listen to the fans at the same RPM's and make notes.

If anyone has any other suggestions I am willing to consider them, there might be a while before I can give any feedback as I am not sure when I will be ordering the HS/Fan.


Andy

andyb
Patron of SPCR
Posts: 3307
Joined: Wed Dec 15, 2004 12:00 pm
Location: Essex, England

Re: i5 3570K OC cooler questions

Post by andyb » Tue Dec 11, 2012 7:21 am

Hyper 212 Evo ordered.

I should be able to do testing on Friday.


Andy

kuzzia
Friend of SPCR
Posts: 709
Joined: Fri Jul 02, 2010 4:41 am
Location: Denmark

Re: i5 3570K OC cooler questions

Post by kuzzia » Tue Dec 11, 2012 7:47 am

We're very much looking forward to that. :)

andyb
Patron of SPCR
Posts: 3307
Joined: Wed Dec 15, 2004 12:00 pm
Location: Essex, England

Re: i5 3570K OC cooler questions

Post by andyb » Thu Dec 13, 2012 12:47 pm

I have the Coolermaster Hyper 212 Evo, I hope to install it and do some testing with its fan vs the AC F12 tomorrow morning, although work may interfere with my play so it might not happen until the weekend.


Andy

andyb
Patron of SPCR
Posts: 3307
Joined: Wed Dec 15, 2004 12:00 pm
Location: Essex, England

Re: Coolermaster Hyper 212 Evo on i5 3570K OC

Post by andyb » Sun Dec 16, 2012 4:36 am

Something strange has happened with speedfan or my motherboard.

I was testing the fan that came with the Hyper 212 Evo via speedfan, I had logged fan speed percentages in 5% increments and the fan speeds in rpm, I swapped over to the AC F12 and speedfan no-longer showed a fan speed. I tried all 12 combinations of setting in the advanced tab and got nothing. When I then swapped back to the included fan I had the same issue, I the tried both fans on "Case Fan 1" and checked all setting in speedfan and even tried changing the setting in the BIOS.

I cant believe that unplugging a 4-pin PWM fan and plugging in a 3-pin fan has broken something. This is pretty strange.

I have restored my system via Aronis True Image to reset Speedfan to a working state when I still had the puny Intel original fan being controlled via speedfan. Now however there is no fan control at all via SpeedFan :cry:

----

I have restored my system to a point before I meddled with either the BIOS or SpeedFan.
I Installed the latest version of SpeedFan.
I have reset the BIOS to factory defaults.
I have updated the BIOS to the latest version (1.7 vs 1.5).
I have installed MSI specific software for my motherboard to control control the BIOS from within Windows (this is weird IMO).
I have installed MSI specific software for my motherboard to control control the Fan Speed above and beyond the BIOS via software within windows, in a similar (and far easier to use) way than SpeedFan.

However nothing (BIOS included) shows FAN RPM's (or seems to work correctly in SpeedFan). Apart from this being peculiar, this is a real pain in the arse as I controlled my previous CPU fan via SpeedFan (beyond) perfectly well, and everything was going OK with this until I switched fan (for SPCR reporting reasons and questions).

Should I return the motherboard (total pain in the arse), or should I try other things first....?

FYI... I have the CPU fan plugged into "SYS Fan1" currently as I have been too lazy to re-plug it back into the CPU header. Between the "Auto" control in the BIOS and "MSI ControlCentre" (if it does anything when not running) the CPU core temps don't top ~73c and the PC has not crahed whilst playing Far cry 3 @4.2GHz (default everything else except the RAM which is rated for 2.0GHz under an XPM profile).

Please share your thoughts.


Andy

Post Reply