CPU cooling on a GA-EP45

[Chike]

Re: CPU cooling on a GA-EP45

I have seen a review where they tested both exhust and intake configurations of the H50 and results were pretty much the same.

[wallypiper]

Re: CPU cooling on a GA-EP45

Hmm.
I finally noticed that the intel cpu fan DOES eventually start. The little herky jerky thing I was seeing was some sort of pre starting behavior. So maybe that's not the problem with my system. So I downloaded and installed new drivers for the on board LAN, the on board HD audio and the EVGA graphics card. A couple of the BSOD's I've gotten are, according to various google hits, possibly related to the LAN drivers. And the audio drivers have been troublesome before. And of course video drivers are notorious trouble makers. None of my drivers were very old but that's Windows. MS changes something and it ripples through the system I guess. We'll see. I'm still interested in upgrading the cooling but I don't see it as a major thing at the moment. Big thanks for all the input, though.

At this point, my next "project" will be to go through the BIOS settings and make sure I have all of the recommended settings for a non OC mobo and see what happens from there.

Thanks again. I'll be back, I'm sure.

[profJim]

Re: CPU cooling on a GA-EP45

If needed, it looks like you can reverse the air flow of both 80mm fans so that they will exhaust air from your case if you install the H50's fan so that it draws cool air into your case.

It looks like you might be able to replace the front 80mm fan with a 120mm fan.

Let us know how you make out or if you need more help.

[wallypiper]

Re: CPU cooling on a GA-EP45

That's an idea but it would mean using the hot air coming off the radiator as the heat sink for the rest of the system. Not really sure how that would play out. It would probably be more effective to add more air inlet to the case with a larger supply fan at the front and maybe replace some of that way cool clear plastic on the side with filtered openings. or even try adding another small fan to that area. Flooding the case with cool ambient air would probably be better overall and reduce the loss of efficiency for the radiator. I do heat transfer for a living and know that a critical part of the equation is the approach. That means the temperature difference between the fluid circulating through the radiator and the air passing through it. The fluid must be cooler than the CPU itself or it wouldn't absorb heat from it. I don't know what fluid is used but assume it's water based with possibly some biocide and corrosion inhibitors added. If the CPU surface is 50 C and the air is 30 C, you have 20 C (36 F) of room between the cpu and the air. The mystery is the heat transfer efficiency of the two interfaces.

Let's say your cpu is running at 200 watts. It could be that in order to transfer the 200 watts to the air, the fluid has to enter the radiator at a temperature 10 higher than the air temp. As it passes through the radiator, it will cool to some temp closer to the air temp but obviously never below it. So maybe that number is 40 C entering and 35 C leaving. Then the fluid returns to the cpu at 35 C where it absorbs heat from the surface of the cpu and is heated back to 40. Those numbers could be higher though. Maybe 50 entering the radiator and 40 leaving. But in any case, the fluid will always be warmer than the air by some amount so using warm air from inside the case has to make a difference. Any test that shows blow in vs blow out not making much difference has to have been performed on a system that had very good airflow INTO the case so that the air temp inside the case was not much higher than the ambient air temp. Fluid cooling in and of itself is not guaranteed to be more effective than air cooling especially when the "air" cooling utilizes heat pipes which use change of phase in the recirculating heat transfer fluid. Selecting the right fluid for that could result in much more effective cooling than a system which uses a medium that is always liquid.

Of course, the proof is in the pudding. There's a pretty good article over at Tom's where they try to control all the variables and make an apples to apples comparison of the H50 and some air cooling options, including case temperature effects.

[Ray422]

Re: CPU cooling on a GA-EP45

The fan for the H50 replaces the normal exhaust fan in the case and Corsair recommends installing the fan in a "blow in" position, to blow cool outside air through the radiator. My case doesn't have a place for a replacement case exhaust fan. I'd be a little concerned about the amount of heat inside the case with the cpu heat being blown INTO the case and the case exhaust fan removed. I suppose that, again, I'm being somewhat hamstrung by the Raidmax case design. Cheaped out and got what I paid for I guess. I could possibly find a place to put an exhaust fan but there's really not any empty space that I see at the moment.

Well.........of course, when you switch the rear exhaust to an intake, then you switch the front intake to an exhaust...If you have no front intake for switching to exhaust, then use a side fan for exhaust. I also have a raidmax, and I merely flipped the front intake.

You go into all this detail about cooling and air flow? Do you not have a front intake you can reverse?
If ya know so much about airflow, the starting point should have been a better case with more fans, especially a good front fan :) No cooling works well when you start with poor circulation.

[wallypiper]

Re: CPU cooling on a GA-EP45

It's not just airflow. It's temperature. The blow in arrangement means the air being supplied to the case to cool everything BUT the CPU has already been heated up by the CPU radiator. How much?

My Q9550 has a max TDP of 95 watts, according to Intel. The fan on the Corsair moves a max of 59 CFM. That's about a 5 degree F rise in the air temperature. (CFM * 1.08 * delta T(F) = BTU/h, BTUH/3.414=watts)

Probably not a big deal but if you're overclocking and changing the airflow in and out of the case, its worth knowing. The temperature rise would change in direct proportion to the wattage or the airflow.