Compatibility of Asrock Extreme 6 AC and Thermalright HR 22

All about them.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
Atlantis 666
Posts: 13
Joined: Sun Mar 23, 2014 3:37 am

Compatibility of Asrock Extreme 6 AC and Thermalright HR 22

Post by Atlantis 666 » Sun Mar 23, 2014 5:07 am

Hi all: i'm new writing here, but it's years and years i follow this site (mostly the articles), so after reading a lot of interesting posts usefull to decide the configuration of my new silent pc, i decided to join the forum to ask some questions.

I would like to know, in particular, if does anyone know if the huge Thermalrigth HR 22 will fit in an Asrock Z87 extreme 6 AC (after reading many advice about true PWM fan connector and Bios fan control managing, i understood it is the best choice to build a fanless/semifanless pc). In particular i'm afraid by the 2 VRM heatsink size. But i saw some pictures were it was placed over a Z87 Extreme 4 mobo (http://www.overclock.net/t/1437284/ther ... t-hr-22/20).
I wasn't able to find heatsink size and socket area measures.

I want to give it a try for a fanless semifanless solution (until 55° are reached, from bios i want to set 0 rpm fans) with i5 4570!

Let me know if i'm OT.

Thanks in advance.

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

Re: Compatibility of Asrock Extreme 6 AC and Thermalright HR

Post by Abula » Sun Mar 23, 2014 3:06 pm

Welcome to SPCR!

Looking at pictures of the motherboard, i see it has as all motherboards the cpu socket (with the square and semi layout of the cpu mouting) clean, so i don't think you should have any issue mounting the HR22.
I want to give it a try for a fanless semifanless solution (until 55° are reached, from bios i want to set 0 rpm fans) with i5 4570!
Here i dont want to discourage you, but im going to address two important things.

1) PWM fans.
Not all PWM fans are created equal, some can drop more than others, and more importantly in you case, some PWM fans can drop to 0rpms, while others don't, it really depends on how the PWM of the fan was design. Im going to leave you a couple of examples where even at 0% it spins and other that the fan should stop if the %PWM is reached. So chose careful the fan you will use, and search if it can drop to 0rpms or not.

Image

Image

Also the motherboard PWM fan control might or not affect this, some have failsafe or restriction into going into 0rpms, or alarms, etc.

Another thing worth mentioning, is that going fanless certain portion might or not be good, depending on the fan you chose, like for example the NF-A15 PWM, i personally cant hear anything at 200rpms, and most of time, the very little air movement does impact a lot the temperatures, so lets say a 200rpm could help you not to reach 55C as easy, as with it being fully fanless for that portion, and still not netting any perceivable noise, this you have to test on your own when the PC is built.

2) CPU TEMP
Haswell run hot, specially hyperthreaded CPUs, i seen better temps on quads that don't have hyperthreading so your might be different than mine. I usually idle around 30C with the fan at minimum rpms, around 530rpms, but with very light load i can reach 50C, like browsing the internet, this is the reason i assign not increse the rpm below 50C, where it starts to gradually ramping up, if i set for example 40C, then the fan would have been ramp up and down a lot, after 50C i do need a middle load for it to go up, so it plays good like this on my setup. Now on a fanless startup, idk this is something you will have to test with your hardware, as i would think that you will pass that 50C much easier than me, my guess is you will be in the zone that i tried to avoid, but then again my CPU is hyperthreaded and runs much hotter than an i5, so you might not have the issue, this is why i say its more important to test it once you have it up and running, as there is no way of predicting it.

Now one thing that you do have realize is that the fan that has the biggest impact on the CPU temp is the CPU fan, so this was the main reason i left mine unrestricted, my case fans can only reach 37.5% PWM, which tends to be 650rpms, while the TY150 on the HR22 was left unrestricted so it can reach the rated 1100rpm if its needed, but no game will take it there, only benchs like prime95 can, again this will all depend on all your setup, and this has to be setup like you want to, but until you build it.

What i would do is also play with the case fans, when i set up mine, i ran probably more than 40 times prime95 with different fan settings, restricting CPU, restricting case fans, both on minimum and maximum rpms, etc to feel the noise the setup could reach and temps. In my experience (this might vary on yours), the case fans dont have a huge of an impact as the cpu fan, also the cpu fan could spin higher and i wouldn't hear it as easily as the case fans, so for my setup, it worked out better to have higher rpms on the CPU FAN and lower on the case fans. Again all of this is very dependant on each setup, so test it once you build it.

If you don't understand anything i posted let me know, ill try to detail it more.

Atlantis 666
Posts: 13
Joined: Sun Mar 23, 2014 3:37 am

Re: Compatibility of Asrock Extreme 6 AC and Thermalright HR

Post by Atlantis 666 » Sun Mar 23, 2014 5:30 pm

Hi Abula, thanks for this and many other usefull post i have read here and over other forums (but expecially here on SPCR) in last weeks. These were very usefull to decide my new pc config.
Looking at pictures of the motherboard, i see it has as all motherboards the cpu socket (with the square and semi layout of the cpu mouting) clean, so i don't think you should have any issue mounting the HR22.
Also i noticed the lines drawed on the pcb, and i found a review (on Anandtech) that was speaking about some "Intel xy specifications with the VRM heatsinks and the memory slots", but i didn't find anything more about: does exist any standard about?
This is exactly what says Anand review about this:
Starting with the socket area, and we can see ASRock are right up to the edge of the Intel xy specifications with the VRM heatsinks and the memory slots, meaning big air coolers might have some trouble with high profile memory here.
I also wasn't able to find the height of VRM heatsink anywere.
What scares me more is the heatpipe low height against VRM heatsink than the heatsink body itself (that should be ok) over the few capacitors.
Do you know if HR 22 can be mounted with the asymmetric larger side over the ram instead than toward the rear fan? It could be a work around in case of incompatibility with mobo.


I totally agree when you say that i will have to do many test by my self because the specific scenario can't be predicted ex ante. Infact in my mind i decided to leave some room for improvements to try to reach my target (fanless/semifanless pc at low loads): for example undervolt, or buy a discrete graphic card (this should help staying low with cpu temp), use an horizontal case, stay with side panel open in the actual case, etc.... And i will have to test a lot of settings from bios (this is one of the reason why i choose Asrock). And also with the actual hardware i'm planning to buy i think i will have much to play to reach the optimal settings for a silent pc.

I decided to choose HR 22 over Macho or Zalman Fx100 (this was the second choice, 54€ aginst 69 of HR 22) because also if some reviews doesn't give very good rating about HR 22 (i read also some your interesting post somewhere over the internet :) ), i thought it's ability seems to be that allow slower cpu temp rise than other heatsinks do in passive mode; in my scenario it means that i will be able to stay in passive for more time than other heatsink will allow. Aside this, it's true that CR95 perform this scenario better than HR 22 (but it would cost much more to me, and has no possibility to use a fan), and is also true that with a fan, the other heatsinks i spoke above, have a levelled performance. But i'm interest (and for this i'm paying more) in the first part of the work: the ability to hold low temps in fanless mode for much time as it is possible; in this scenario the 8°/10° less than the macho (for eg) for me will mean more fanless computing.

But is worth to take in mind also your strategy: allow a little (and inaudible) spin of the fan to help more the heatsink staying under 55° threshold (expecialy with hotter haswell cpu). I will test which method will give the best result for my needs!

One thing i haven't still understood, is which daily temp is safe for haswell, and which pick temp can i reach before thermal throttle: this will affect a lot my decisions about fan rpm settings. But it is something i will check after ordering the pc.

About PWM fans: again i totally agree with you (i read some other post you wrote about it, so it is something i already considered). Infact i opened another specific topic to collect advice for my specific needs (and i see you had already answered! :!: ). I will write more about this arguments on the other topic. :wink: It's a crucial point of all my way to (semi)fanless strategy.

About the cpu fan affecting more the cpu temp than case fan: ok, but i ask my self: with a little boy like HR 22 is still possible to speak about cpu fan and rear fan? I think that this traditional way to categorize fans is ovveride: am'i wrong? Also in tests/reviews comments is disputed if using Hr 22 with fanduct can be classifed as passive or semipassive mode or whatelse.... What i'm trying to say is that when i will place hr 22 in my case (and maybe the final case will be horizontal and with upper side opened) i could decide to place (or maybe it could fit) only 1 or 2 fans: one placed on the rear that will be also the cpu fan (expecially if i will have room to use fan duct), and maybe another, that will start spinning on high load just to provide little airflow to ram, VRM heatsink (because with a passive cpu heatsink they will suffer more of high temp), and also again to HR 22 (that so would take advantage of 2 fans no necessary directly attached to it), if i will be sure that SSD won't need to be cooled with front fan to stay in safe temp range. But again: i will have to test in my specific scenario. For now i just take in mind this second advice you gave and take it ready for the future tests.

Post Reply