Fan Direction Advice Needed

[SCTony]

I am finishing up my first build and need some help deciding which way to route my airflow. I realize airflow is not too critical in a HAF-932 case but I want to set it up correctly. Here is how I currently have it-
109078

Not pictured is another 230mm 110cfm fan on the side cover currently blowing IN. Should I have the front 230 blowing IN and the side 230 blowing OUT? Or-
shoud I turn the Noctua to blow IN and have the three bottom fans blowing OUT? Thanks.

<side note> My computer has a flat tire :) The extra weight with the LC gear was too much for the stock casters. One broke and one popped off. Ha.

[Creekin]

ha! flat tire :)

i've always gone with the "hot air rises" theory, bearing in mind that is only really relevant to convection not fan assisted.
but why try and pump air down thats trying to rise naturally?
go with the flow! :D
id have the front and side going IN and the top and rear OUT

also depends if you want to avoid dust in the case and whether you use filters, i aim for positive pressure (more in than out) and then filter the inlets.

edit: i assume those 2 castors next to you kb are replacements?
they look good

[SCTony]

Hi Creekin-the reason I started with the rad fans IN was to bring fresh air over the rad. I am thinking there won't be a great deal of heat as I am only cooling a Q9550 quad. I would like to get as much flow as possible over the NB/SB/Mosfet heatsinks. It seems like the airflow kind of loses it's "oomph" with the 932 case being so open.
-Ya, 4 new wheels (2" ball-bearing swivels) to be installed soon, I hope. They will require a little tinkering to fit. :)

[Creekin]

i'm sure some will say never put warm air through the rad, but i really don't think theres a huge diff in air temp,
plus, with a well wc'd case there shouldn't be any heat in it, it should all be in the rad. especially with ur radless front fans
so you wanna exauhst it out the top, not back in to the case, back into the blocks and sinks and back into the loop!

alot of ppl seem to forget, the goal is to REMOVE HEAT, not introduce cold.

[SCTony]

Ok- so if I push OUT through the rad- have both 230s IN- then should I turn the Noctua around so it blows IN to try for positive pressure? Thanks.

[Creekin]

the noctua is kinda 50/50 standard would be for out,
but as IN it would cool the mofs and feed the 360 nicely AND increase positive pressure,
sounds like the G.O. :up:

[Conumdrum]

SCTony, the best temps will be rad fans pulling in and everything else out. Best CFM for the rad that way. If your temps are good like that and you can spare some temps, then reversing everything the other way would be better for all your other hardware. It's up to you of course.

[aerial]

Imo it is best to blow cool air from outside of the case on radiators. All other laws, such as convection are inferior to power of fans. So all other holes in our case (even these without fans) will be used as ventilation, while fans are taking as cool air as possible. Note that for full watercooling stup, it isn't that important to have very cool air inside the case, since all crucial components are cooled by water independently.

[Creekin]

Note that for full watercooling stup, it isn't that important to have very cool air inside the case, since all crucial components are cooled by water independently.

I think that's the key point in this topic.

BUT hot air coming into the case will get re-absorbed by the water blocks and "loop" back into the loop.

In a perfect world rads should be external or mounted across a section of case sealed off from the main chamber, like often seen in the bottom of a tj07.
But the challenge here is to get it all internal. http://forums.bit-tech.net/images/smilies/bit-dremel.gif

In my rig i am using the airflow from the rads to also cool my HDDs.
front intake (cool fresh air),top and rear exhaust (warmed by front rad/mobo ambient/hdds).

The other consideration is filtering and dust.
I think it's difficult to make a general rule for this, every case is a different....case. :)

[Church]

Imho rising rad efficiency by using colder outside air more then offsets impact of warm air blown inside case. As others noted, it's LC components cooled where it counts (cpu/gpu+maybe motherboard), and rest of hardware doesn't dissipate enough heat nor are that much low temperature dependent to suffer from that noticeably. And even moving warm air cools better then colder but still air. After all, for each LC build there are more then 10K common PCs where warmer air inside case does no harm.

[antiacid]

the reason you use case fans is to move the heat generated from the board / components that aren't being LC. Therefore, you want to use natural convection and work with it to get the best results. Top should be out, noctua should be out, side hould be in, front should be in. You'll get noticeably less than 110cfm on those 700rpm fans I believe but you should be able to keep the inside of the case very close to the outside temperature that way.

[defect9]

if you turned all the non-radiator fans to be intake fans, would you need the radiator fans or would the positive case pressure be enough to shove all that heat out through the top?

[jim0ne]

if you turned all the non-radiator fans to be intake fans, would you need the radiator fans or would the positive case pressure be enough to shove all that heat out through the top?

I think you still need the fans. For the static pressure, it should help with removing the heat from the water.

[Creekin]

good q defect9,
you would still need rad fans i reckon,
perhaps if all the other fans were really high cfm, and you sealed the case up airtight besides the rads, MAYBE.. lol

ppl often run rads fanless for low heat apps. a decent size rad mounted horizontally would naturally shift alot of heat by convection, just how much would be an interesting experiment..

[CrazyNutz]

1).Your rad needs good airflow, so If you do build positive pressure in your case and your fans on your rad are blowing in, then you will slow your air flow through your rad.

2). The fans on rads are more effective in push config than pull.

3). I just did a build with that case, you have ample amount of fresh cool air comming in, so I say front & side = blow in, and top & back = blow out, also your PSU blows out.

[MrBean]

Mmm....what works really well for me, is rad fans in, front in, and rear fans out - you will also need to put rad-fans push-through mode, they will be more efficient that way. All depends on max practical airflow of your fans though, and best to run system at ful-load while trying different combinations, while minimising effect of ambient temp.

Particularly look at things like MOSFETs et al, monitor them closely - CPU (watercooled) temps won't be to indicative. Also, using front as an intake should improve Graphics'card overal temps.....

[avddreamr]

Either will work, you aren't anywhere near saturated.

My setup, similar to yours....
ATCS 840 /XSPC RX360 , 5 GT AP-15, 3 on the radiator, one in a scythe kama bay, the other on the rear. Both are intakes, with the radiator functioning as exhaust.
I have no problems being stable while I keep my fans under 1000 rpm at load.

Cooling a q6600, and a 5870.

So you'll be fine.

[penguins]

1).Your rad needs good airflow, so If you do build positive pressure in your case and your fans on your rad are blowing in, then you will slow your air flow through your rad.

2). The fans on rads are more effective in push config than pull.

3). I just did a build with that case, you have ample amount of fresh cool air comming in, so I say front & side = blow in, and top & back = blow out, also your PSU blows out.


Mmm....what works really well for me, is rad fans in, front in, and rear fans out - you will also need to put rad-fans push-through mode, they will be more efficient that way. All depends on max practical airflow of your fans though, and best to run system at ful-load while trying different combinations, while minimising effect of ambient temp.

Particularly look at things like MOSFETs et al, monitor them closely - CPU (watercooled) temps won't be to indicative. Also, using front as an intake should improve Graphics'card overal temps.....


the push is always better is not a factual representation.

here are a couple links for you to read through.
showing
slow fans = pull
fast fans = push

http://www.xtremesystems.org/forums/showthread.php?t=260161&highlight=push+pull+martin

http://www.xtremesystems.org/forums/showpost.php?p=3026299&postcount=18

[Creekin]

those links seem to be broken,
first time ive heard someone mention a diff between slow and fast fans,
interesting :) thanks

[Church]

Creekin: Martin's test about shrouds and fan orientation efficiency (http://martin.skinneelabs.com/Radiator-Fan-Orientation-And-Shroud-Testing-Review.html).

Push Vs Pull - This depends on fan speed/power. The high speed fans at 2000RPM with a 38mm fan thickness provided the best performance in a push condition. The slow speed fans with 1350RPM with a 25mm fan thickness provided the best performance in a pull condition. I would estimate that performance line is likely to cross in the 1500-1700RPM range where they are equal. So.... slow speed = pull, high speed = push, medium speed = it doesn't really matter.

[SCTony]

Thanks for all the input everyone. I spent yesterday making adapters for the new wheels so I haven't had time to work with the fans much yet. With the rad fans as is - pulling IN- there is a lot of nice cool air being dumped onto the mainboard :) The front 230 is turning normally but I don't feel much flow. Maybe the hard-drive bays being full is hampering it? I imagine it has very little static pressure power. I am thinking about replacing that one with a GT-15 one day.
Is there a way to check for dead zones visually- ie fog or smoke trails? Just curious. Thanks :)

[Creekin]

Thanks churchy, that's making me rethink my setup, again :D as i have 3xSR1 rads with 750rpm enermax fans
my personal theory from years of working with smoke and fans was push is better, for various reasons gone into ad infinitum elsewhere....
but who can argue with martin? :shrug:

SCTony, you could use smoke, but cigarette smoke will leave a tar residue on everything in ur case, miniscule but still...
and it is hard to tell whats happening anyway as it just gets dispersed.
i would use a flame from a lighter, and move it around and watch it move :up:

do u know the rpm of that 230? i imagine its pretty slow.

[Church]

Creekin: it's just a matter of simple putting fan on other side of rad anyway. Also, if there is enough space, imho best way to go - push-pull. Not that much thicker resulting fan-rad sandwitch then single row of fans, more effective then single row+shrouds, and not as monstrously thick as push-pull + shrouds on all of fans (25*2+50-60 or rad+25*2 - ~ 15(!)cm).

[Creekin]

ive always thought that push/pull was unnecessary overkill, in a sport where we spend thousands to cool pcs to surf pron on :D

not to mention the added $$ of in my case 7 more fans @$22ea, not that $$s is an issue, just seems like a very minor benefit for $150 odd bucks. i will have a read of that link and see what he has to say. i have read some of his work b4 but not this page. :up:

adding 4 more fans to my front rad would be a pain but easy on my top and rear. i might play around with my 280 and 4 fans and see whats what. :)

size isnt much of an issue for me, ive got an x2000 which is HUGE! plus im happy to mount rads to the outside rather than inside.
i was a little worried about the pissy 750rpms thats coming out of my enermax....

been reading that link while typing this reply...hmm looks like hes onto something, damn it! ahh fark it...sorry bank acct... :shrug:

[Church]

Creekin: it's just that in all-internal builds sometimes it's simplier and/or cheaper too to add another row of fans with rising cooling efficiency by some 25%(?), then another rad section one often simply has no more place for inside case. Depends on specifics of each build and wishes of builder of course.

[penguins]

ive always thought that push/pull was unnecessary overkill, in a sport where we spend thousands to cool pcs to surf pron on :D

not to mention the added $$ of in my case 7 more fans @$22ea, not that $$s is an issue, just seems like a very minor benefit for $150 odd bucks. i will have a read of that link and see what he has to say. i have read some of his work b4 but not this page. :up:

adding 4 more fans to my front rad would be a pain but easy on my top and rear. i might play around with my 280 and 4 fans and see whats what. :)

size isnt much of an issue for me, ive got an x2000 which is HUGE! plus im happy to mount rads to the outside rather than inside.
i was a little worried about the pissy 750rpms thats coming out of my enermax....

been reading that link while typing this reply...hmm looks like hes onto something, damn it! ahh fark it...sorry bank acct... :shrug:

read the link first before getting this kind of idea :) it has a lot of info on single fans push or pull also.