Running a Megahalems passive in a high airflow case

[faxon]

So has anyone tried this before? right now i have a thermaltake armor (original, the one without the side fan), and i am really unhappy with its feature set and wire management options vs what newer cases have, and it's airflow isnt really that great. i have always had better temperatures across my system with the door off vs with it on, and i have stopped using the 92mm fans in the top and by the PSU/top HDD cage since they werent helping my HDD temps any and my cpu vents out the back.

Now, right now i have 2 of these fans mounted on my megahalems with fan silencers, one mounted a foot from the cooler horizontally where my HDDs are sitting in the 3 in 3 bracket, and a 4th in the back vent on my case about an inch from the cooler. all 4 fans are blowing in the same direction, creating a sort of wind tunnel across my CPU (basic specs in sig). As you can probably tell by now, im big on the silent factor, and im looking to get a case that has excellent hard drive mounting options, wire management, is alluminium so it's lighter, and has high airflow. I have decided on the ATCS 840 as my case of choice, and i was wondering if anyone has tried running a passively cooled configuration with this case.

I have a Megahalems cooler on my CPU, and under linpack my cpu doesnt really go passed 70c (at 68c as i write this under linpack), so i was thinking that if i had enough airflow (2x230mm fans and a 120 all in close proximity pulling air out) i might be able to get away with not even having any fans on the cooler at all.

so has anyone tried this configuration before? i dont really need a higher OC than what i have, and making my system even quieter would be great if i can pull it off. given, i probably wont notice the difference, but it would be a fun experiment anyway.

 

[yh125d]

Could probably be done, but not with that OC. You're already at 68c in a well ventilated case and push-pull config, no way you can go passive at that voltage. A more realistic solution would be to take the 4 fans in the setup you have now and change it to two. One on the COU, and another at the back doing exhaust.

 

[faxon]

yea but the one on the front of the case is also moving air across my 3 really hot barracuda drives which i use to serve files to the network until i move them to a dedicated file server. when i get my 840 all i will have in it to start is my Vraptor and 2 Spinpoint F1s spaced out for airflow, and im probably going to pick up a new PSU as well, since i want to move my file server into my armor case, and it would be easier to not have to rewire the PSU wiring i have set up now.

mostly just curious if someone tried this already. my chips temps are usually a lot lower (low 50s in FC2 and crysis), and turning off my pull fan only raises my temps 1.5-2c on average. if anyone has tried this, it would be great to know how it went.

 

[Rubycon]

Your temps will go significantly higher without solid airflow through the fins. Unless you build a duct to direct as much air as possible in and out of the heatsink itself, don't expect this to work unless you keep loads light! Linpack? :laugh:

 

[OILFIELDTRASH]

I have a 840 case and the top 2x230 fans create loads of negative pressure. I turned them upside down so now they pull air in and with those 2 fans pulling cool air in you could have a decent overclock with no fans. No way with 4ghz. You dont need such a high oc anyways. I went from a Q9400 to a i7 build and can't tell the difference.

 

[Zap]

Don't do it.

Heatsinks that are able to run passive have fins that are spaced wider apart.

A fan on the heatsink will be quieter than the same fan (at same voltage) attached to the case as an intake or exhaust.

I think the best bet for trading a bit of temperature for less noise is to reduce the number of fans you have, and undervolt the rest a bit more. How many fans do you have anyways?

 

[BonzaiDuck]

I'm pretty sure that the Megahalem is about the best heatpipe cooler out there -- unless you want to choke up your computer case with an IFX-14. From what I gather from several cross-referenced reviews, it's worth maybe 4 or 5C reduction in load temperatures for a thermal wattage around 100+W. It's also about the same size as the TRUE, which brings me to my . . . . ah . . . . "pet topic."

Someone spoke of "negative pressure" from too many exhaust fans in a particular case, noting they reversed the airflow to make things all wonderfulness and happy. Good point there.

Last time I posted results and pictures for my 2007 motherboard-ducting project -- and in response to a request, no less -- there were no comments.

In principle, whether Thermaltake ARmor or HAF 920 (which has got to be one of the best of the best for airflow) -- you can improve cooling by channeling airflow through narrow, restricted apertures around the hot components, coolers and heatsinks. At best, this means forcing the highest percentage of intake air through those apertures (which means a ducting mod) before exhausting the air from the case immediately. Conversely, this means that cool intake air in the case doesn't mix with air already-warmed by those hot components.

If you want to do that sort of thing, it's cheap. You can use foam art-board and the proper "foam-safe" glue from a hobby shop. Art-board is cheap; the glue is pricey, but you shouldn't need a lot of it. And you have to commit yourself to an afternoon of measuring, planning and design, and application of an Xacto knife.

So -- here's my story. I'd built my "ducting mod" in 2007, to accommodate standard-sized memory sticks. I had to replace my memory recently in that system, counting on a spare-a-pair of Corsair DHX's. And the DHX modules have these heatspreaders which sit really high above the motherboard.

My original ducting mod didn't fit. I had to modify it -- with much trepidation. But I only had to cut slots in the foam-board to accommodate the DHX heatspreaders, and I think that cooling -- the distribution of airflow across the memory and other components -- has actually improved.

IF you don't want to take the time and tedium to "duct" -- then it becomes a matter of brute force. Example: I have another machine in a CM Stacker 830 with five intake fans (3x140mm and 2x120mm), a 120x38mm rear-exhaust and the now not-so-popular CM barrel-fan as exhaust -- its airflow reversed from what was intended.

That works . . . . too . . . . . But the ducting-mods will allow you to reduce the overall number of fans, therefore reduce the amount of power-draw, and get the same or better load temperatures. It's just a big P-I-T-A, though . . . .

PS and EDIT: The ducting allows me to avoid hanging fans on the cooler (TRUE or Megahalem -- doesn't matter) altogether. Rubycon mentioned having the direct exhaust-pressure from a fan hung on the cooler itself, and she also mentioned the exception with ducting. You may lose a little "direct pressure" for an adequately ducted cooler with either a single exhaust ("puller") fan or fans fore and aft, but the cooling effectiveness doesn't seem different from a configuration with fans hung directly on the cooler.

 

[Tullphan]

Bonzai, wasn't you tempted (seduced?) by the HAF 920?
If you get it & work your magic, don't forget to post pics!