High temps with TRUE rev.C

Cooling Processors quietly

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
doveman
Posts: 870
Joined: Mon Jan 29, 2007 10:30 am
Location: London

High temps with TRUE rev.C

Post by doveman » Fri Jan 13, 2012 5:53 am

I've got a Phenom II X4 955 running at stock 3.2Ghz on a MSI 990FXA-GD80 board. It's on the bench at the moment, so the only fan is the Nexus 120mm on the True rev.c.

I've been trying to overclock the NB to get better RAM performance and did a Prime95 test with it at 2600Mhz last night (Custom FFT 1024K, 8000MB RAM) and after a while the AMD 10h+ CPU Thermal Sensor was showing 71c in HwInfo and the other temps were in the 60s, although I was more concerned with stopping the test than noting them down at that point so they may not all have been that high.

The Nexus 120mm runs at a constant 1055RPM, which is it's full speed so it can't go any faster when the CPU heats up, so I'm thinking maybe I need to replace it with something with some headroom, albeit I wouldn't normally be doing Prime95 Torture Tests. Is there anything reasonably cheap that will be about as quiet as the Nexus most of the time, but able to spinup when needed?

I'm running the test again now after updating the BIOS to v11.8 which reset everything, so my NB is at 2000Mhz, the CPU and CPU-NB voltages are on Auto (were 1.388v and 1.3v) and my RAM at 1333Mhz (was 1600Mhz) and the CPU Thermal Sensor is at 62c, CPU 51c, Motherboard 37c and T1/D1+ 52c. I'll set my overclocks and test at those again in a bit.

EDIT: At idle, I get CPU Thermal Sensor 33c, CPU 23c, Motherboard 35c, T1/D1+ 24c

Arbutus
Posts: 234
Joined: Wed Jun 02, 2010 9:15 pm
Location: Calgary, Alberta, Canada

Re: High temps with TRUE rev.C

Post by Arbutus » Fri Jan 13, 2012 7:53 am

doveman wrote:Is there anything reasonably cheap that will be about as quiet as the Nexus most of the time, but able to spinup when needed?
GELID Silent 12 PWM 750-1500RPM

doveman
Posts: 870
Joined: Mon Jan 29, 2007 10:30 am
Location: London

Re: High temps with TRUE rev.C

Post by doveman » Fri Jan 13, 2012 8:07 am

Thanks, looks OK. I'll see if I can find one locally.

EDIT: Unfortunately doesn't seem to be available in the UK. All I could find was the non-PWM 120mm, which only runs at 1000RPM.

I just remembered my CPU is overclocked to 3.9Ghz, which I did with ATI CCC so it doesn't show in the BIOS :oops: But I want to be able to run it overclocked, so I need to sort out the cooling to ensure it can cope when necessary.

lodestar
Posts: 1683
Joined: Fri Aug 05, 2005 3:29 am
Location: UK

Re: High temps with TRUE rev.C

Post by lodestar » Fri Jan 13, 2012 8:46 am

There's alway the Scythe Slip Stream 120mm PWM Fan SY1225SL12LM-P which is one of the quietest PWM fans around. See the SPCR Mugen 2 review for details. It only costs around £9 from sources such as
http://www.quietpc.com/gb-en-gbp/produc ... am-120-pwm

Abula
Posts: 3662
Joined: Tue Nov 02, 2004 12:22 pm
Location: Guatemala

Re: High temps with TRUE rev.C

Post by Abula » Fri Jan 13, 2012 10:07 am

Another good PWM but hard to get is Scythe Kama FLEX PWM 120mm SA1225FDB12H-P (300-1600rpm).

Now if you just want another 3pin fan with a little more rpm for heatsink, consider Scythe Gentle Typhoon 120mm x 25mm Fan - 1450 RPM (D1225C12B4AP-14), but dont expect the same noise level as you 1000rpm nexus, it will be louder.

doveman
Posts: 870
Joined: Mon Jan 29, 2007 10:30 am
Location: London

Re: High temps with TRUE rev.C

Post by doveman » Fri Jan 13, 2012 10:17 am

lodestar wrote:There's alway the Scythe Slip Stream 120mm PWM Fan SY1225SL12LM-P which is one of the quietest PWM fans around. See the SPCR Mugen 2 review for details. It only costs around £9 from sources such as
http://www.quietpc.com/gb-en-gbp/produc ... am-120-pwm
Thanks, I'll probably get that. I've already got a Scythe fan lying around somewhere but I don't think it's a PWM one.

doveman
Posts: 870
Joined: Mon Jan 29, 2007 10:30 am
Location: London

Re: High temps with TRUE rev.C

Post by doveman » Fri Jan 13, 2012 10:19 am

Abula wrote:Another good PWM but hard to get is Scythe Kama FLEX PWM 120mm SA1225FDB12H-P (300-1600rpm).

Now if you just want another 3pin fan with a little more rpm for heatsink, consider Scythe Gentle Typhoon 120mm x 25mm Fan - 1450 RPM (D1225C12B4AP-14), but dont expect the same noise level as you 1000rpm nexus, it will be louder.
Thanks. I've spent quite a bit of time and money to make this system as quiet as possible though, so don't want to ruin it for the sake of a few quid ;)

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: High temps with TRUE rev.C

Post by ces » Fri Jan 13, 2012 3:35 pm

There are devices that can convert your nexus fan into a PWM fan.

doveman
Posts: 870
Joined: Mon Jan 29, 2007 10:30 am
Location: London

Re: High temps with TRUE rev.C

Post by doveman » Fri Jan 13, 2012 4:19 pm

ces wrote:There are devices that can convert your nexus fan into a PWM fan.
They're not going to make it run any faster though are they?

doveman
Posts: 870
Joined: Mon Jan 29, 2007 10:30 am
Location: London

Re: High temps with TRUE rev.C

Post by doveman » Fri Jan 13, 2012 5:55 pm

lodestar wrote:There's alway the Scythe Slip Stream 120mm PWM Fan SY1225SL12LM-P which is one of the quietest PWM fans around. See the SPCR Mugen 2 review for details. It only costs around £9 from sources such as
http://www.quietpc.com/gb-en-gbp/produc ... am-120-pwm
Hmm, looking at that review I note the Intel Pentium D 950 with Mugen 2 and Nexus fan at 12v only hit 36c under load. I need to test my PhII X4 955 at stock 3.2Ghz but even then I'm sure it's going to go up a lot more than 3c (it idles around 33c, the review doesn't seem to give any idle temps). Obviously different processors, but still I find that surprising.

Abula
Posts: 3662
Joined: Tue Nov 02, 2004 12:22 pm
Location: Guatemala

Re: High temps with TRUE rev.C

Post by Abula » Fri Jan 13, 2012 6:10 pm

doveman wrote:
ces wrote:There are devices that can convert your nexus fan into a PWM fan.
They're not going to make it run any faster though are they?
No, its more to use the PWM signal of the mobo and control 3pin fans with it, check this video, Sunbeamtech Rheosmart 6 PWM Fan Controller BIOS Setting Tutorial and Product Showcase SUNBEAMTECH Rheosmart 6 Fan Controller (PL-RS-6).
doveman wrote:
Abula wrote:Another good PWM but hard to get is Scythe Kama FLEX PWM 120mm SA1225FDB12H-P (300-1600rpm).

Now if you just want another 3pin fan with a little more rpm for heatsink, consider Scythe Gentle Typhoon 120mm x 25mm Fan - 1450 RPM (D1225C12B4AP-14), but dont expect the same noise level as you 1000rpm nexus, it will be louder.
Thanks. I've spent quite a bit of time and money to make this system as quiet as possible though, so don't want to ruin it for the sake of a few quid ;)
Here is the thing, i tried multiple fans and brands and in my experience i yet to see a quiet fan above 800rpm, specially above 1000rpm is noticible, but there is where its independant to each, you might find what good enough what i think its noisy or even noticible, etc. Scythe Gentle Typhoons imo are among the best fans for rads and coolers, they have good static pressure, you might be able to run them at lower rpms and get better results.... there is way to many varibles to be able to predict the outcome. There are other versions, i own 6x 1850rpm Gentle Typhoons that i used to use on my antec 1200, and at 1850 imo they are noisy, the are still noticeble until 800rpm where they are good, around 550rpm i cant tell they are on even on a quiet room, but they dont move much air at this rpms.

D1225C12B1AP-11 (500 rpm)
D1225C12B2AP-12 (800 rpm)
D1225C12B3AP-13 (1,150 rpm)
D1225C12B4AP-14 (1,450 rpm)
D1225C12B5AP-15 (1,850 rpm)

If you want similar rpm you can go for the AP-13 which is 1150rpm, i still think it will be a little louder than nexus, but my guess is that it will also preform a little better. One thing worth mentioning is that Gentle Typhoons use real ball bearings, some people like them others dont.... here is very subjective.

Good luck with your choice,

doveman
Posts: 870
Joined: Mon Jan 29, 2007 10:30 am
Location: London

Re: High temps with TRUE rev.C

Post by doveman » Sat Jan 14, 2012 8:36 am

doveman wrote:
lodestar wrote:There's alway the Scythe Slip Stream 120mm PWM Fan SY1225SL12LM-P which is one of the quietest PWM fans around. See the SPCR Mugen 2 review for details. It only costs around £9 from sources such as
http://www.quietpc.com/gb-en-gbp/produc ... am-120-pwm
Thanks, I'll probably get that. I've already got a Scythe fan lying around somewhere but I don't think it's a PWM one.
I found the fan I've got and it's the non-PWM SY1225SL12LM, the specs for which are 1,200 rpm, 24.00 dBA, 68.54 CFM (for the PWM one they're 1,300 rpm, 26.5 dBA, 74.25 CFM). The Nexus is 1,000 rpm, 18 dBA, 36.87 CFM, so the Scythe fan I've got can push a lot more air at full speed and at 10v should be as quiet (or slightly more so) than the Nexus whilst cooling slightly better/pushing more air. http://www.silentpcreview.com/article961-page5.html

Other than running slightly faster/pushing a bit more air, are there any advantages of the PWM fan over the one I've already got?

Abula
Posts: 3662
Joined: Tue Nov 02, 2004 12:22 pm
Location: Guatemala

Re: High temps with TRUE rev.C

Post by Abula » Sat Jan 14, 2012 9:27 am

doveman wrote:
doveman wrote:
lodestar wrote:Other than running slightly faster/pushing a bit more air, are there any advantages of the PWM fan over the one I've already got?
Depending on the mobo, you can control it dynamically, meaning its speed can vary depending on the sensor on the mobo, most of the cases cpu temperature, it could be running at 500rpm at idle and ramp up to 800rpm below at 50C, and maybe 1300rpm above 70C.. etc (some mobos can do this with 3pins via voltage reg), this is all dependant on the mobo and how it handles fan control via bios/software, now if you are running them at a steady speed or dont mind tweaking it manually (fan controller), then there should be no gain, personally i love PWM fans, specially with the way i can have virtually silent computers at idle and ramp up when im gaming/encoding. Now more CFM not always means better cooling, some fans do better with heatsinks due to its design, look for high static preassure for better performance behind a rad/heatsink. Either way since you have the slipstream laying around just test it see what temps you get with it and how much noise it has at different rpms, no need to spend unless you need to.

Good luck,

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: High temps with TRUE rev.C

Post by ces » Sat Jan 14, 2012 7:35 pm

Abula wrote:Here is the thing, i tried multiple fans and brands and in my experience i yet to see a quiet fan above 800rpm
I think that says it all. If you want quiet... keep it under 800 rpms.
Abula wrote:Either way since you have the slipstream laying around just test it see what temps you get with it and how much noise it has at different rpms, no need to spend unless you need to.
More good advice.
Last edited by ces on Sat Jan 14, 2012 7:40 pm, edited 1 time in total.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: High temps with TRUE rev.C

Post by ces » Sat Jan 14, 2012 7:39 pm

doveman wrote:Other than running slightly faster/pushing a bit more air, are there any advantages of the PWM fan over the one I've already got?
Most of the time computers are running not much above idle. A PWM fan permits you to run the fan slowly (low rpms = quiet) most of the time.

With non-PWM fans you have to have them running at all times at the speed you need them for at peak performance (which is very little of the time)

doveman
Posts: 870
Joined: Mon Jan 29, 2007 10:30 am
Location: London

Re: High temps with TRUE rev.C

Post by doveman » Sun Jan 15, 2012 4:27 am

ces wrote:
doveman wrote:Other than running slightly faster/pushing a bit more air, are there any advantages of the PWM fan over the one I've already got?
Most of the time computers are running not much above idle. A PWM fan permits you to run the fan slowly (low rpms = quiet) most of the time.

With non-PWM fans you have to have them running at all times at the speed you need them for at peak performance (which is very little of the time)
Except (as Abula mentioned) with motherboards that can control the fanspeed by adjusting the voltage. My motherboard has SmartFan in the BIOS so I'll try that, although even at fullspeed (1250 RPM) I can't say I find the Scythe noticeable, and it's not even in a case at the moment.

Abula
Posts: 3662
Joined: Tue Nov 02, 2004 12:22 pm
Location: Guatemala

Re: High temps with TRUE rev.C

Post by Abula » Sun Jan 15, 2012 6:20 am

doveman wrote:although even at fullspeed (1250 RPM) I can't say I find the Scythe noticeable, and it's not even in a case at the moment.
MikeC wrote an article where he says the following,
There are two rules of thumb here:
1) A fan should be slowed down to the point where it provides just enough airflow to prevent overheating.
2) A fan should be slowed down to the point where there is no acoustic benefit to reducing the speed further — i.e. the fan has become inaudible.
Both of these rules are system-specific; it is impossible for us to know what the ideal speed for your system will be.
If you cant hear the slipstream at full rpms then keep it like that, all systems are different so are people hearing and preferences. Check if its better though, in temps i mean.

doveman
Posts: 870
Joined: Mon Jan 29, 2007 10:30 am
Location: London

Re: High temps with TRUE rev.C

Post by doveman » Mon Jan 16, 2012 6:12 am

Abula wrote:MikeC wrote an article where he says the following,
There are two rules of thumb here:
1) A fan should be slowed down to the point where it provides just enough airflow to prevent overheating.
Overheating suggests to me the point at which the system shuts itself down, but I wouldn't like to run that hot or close to it.
If you cant hear the slipstream at full rpms then keep it like that, all systems are different so are people hearing and preferences. Check if its better though, in temps i mean.
Actually, I have realised the Slipstream is louder than the Nexus. It's still very unoffensive and not really intrusive, but I will want to run it slower when idle so that it only spins up when necessary. It hasn't made any difference to my idle temps, which is strange as it's pushing twice as much air, but it has brought my load temps down a bit (except the MB which is the same or slightly higher) although I didn't record the ambient temps when testing the Nexus so it's not very scientific (if I recall correctly the Nexus tests were done with the heating on and the window shut in the evening, whilst the Slipstream tests were done with the heating off and the window open in the afternoon with an ambient of around 19c).

With Nexus:
Core 34c, CPU 24c, Motherboard 32c, T1/D1+ 24c

Prime95 Blend: Core 66c, CPU 55c, Motherboard 36c, T1/D1+ 55c

With Slipstream:
Idle: Core 33c, CPU 24c, MB 34c, T1/D1+ 24c

Prime95 Blend: Core 62c, CPU 51c, MB 38c, T1/D1+ 51c

doveman
Posts: 870
Joined: Mon Jan 29, 2007 10:30 am
Location: London

Re: High temps with TRUE rev.C

Post by doveman » Sun Feb 19, 2012 12:56 pm

Actually I've changed my mind. I find the Slipstream quite annoying and offensive now, running at afull speed round 1300 RPM:)

I haven't actually put the Nexus back on yet as it's one of the annoying ones which I had to cut the corners in order to fit the TRUE mounting clips and I struggled to get it back on when I tried due to my not very good modding of it, so I might just buy one of the Nexus' with the corners already cut. I can't actually slow this Slipstream down at all with my motherboard, so if I had the PWM-version and could run it slower most of the time it would probably be OK and then able to spin up when needed, but as it hasn't significantly reduced the temps compared to my Nexus running silently, I'm not sure it's worth getting a PWM-Slipstream. I could also reduce the speed of the Slipstream with one of the 7v/5v Zalman resistors I've got, but then it obviously can't run faster when needed, so there's no point or benefit over the Nexus.

I get the impression that there's less air coming through my TRUE with the Slipstream as well, which is wierd as it's CFM suggests it should be pushing a lot more air than the Nexus. I might just be mis-remembering what the airflow felt like with the Nexus though, so I'll have to compare it properly when I get round to putting it back on again.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: High temps with TRUE rev.C

Post by ces » Sun Feb 19, 2012 5:59 pm

doveman wrote:
ces wrote:There are devices that can convert your nexus fan into a PWM fan.
They're not going to make it run any faster though are they?
Use two fans then.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

thermal interface material (TIM)

Post by ces » Sun Feb 19, 2012 6:07 pm

You might try the new Maingear EPIC T1000 thermal interface material (TIM)
http://www.legitreviews.com/article/1852/1/

All things equal, if you improve the performance of your thermal coupling you should be able to get along with a bit less fan. Replace your one fan with two slower and quieter fans, you have a win/win.

Post Reply