A question on double rads in the bottom compartment

[shazza]

Hey - I've got a question for you gurus. I've seen tons of work logs with two rads in the bottom compartment of cases like the TJ-07, V2000, etc. I'm working with the Black Pearl, and have lots of room.

My question has to do with performance. While my case can easily fit two rads in the bottom - even two quads if I want - what's the performance hit from one of the rads taking in warm air from the other? While it's possible to have fresh air coming into both rads, it's going to heat up the center, with no where to go.

My initial thought was one rad - maybe push/pull with GT1450s, and then GT1850s on the other side exhausting the air from the bottom compartment. Just wondering if I'd be better off with an additional rad on the exit side? I'm some time from being able to test it myself, so just wondering what the consensus is.

Thx.

[LostInRehab]

Let me give you another thing to think about. Your case has a lot of vents at the bottom of the case what about having both rads vent outside the case while the vent openings passively pull air in.

[SNiiPE_DoGG]

In my opinion, you have two distinct options here shazza:

1) use high CFM on the rads (push-pull-push-pull if you can ) and have the fans blowing unilaterally. The more air volume flowing through the case the less heat saturated air flowing across the fins of the second rad in the airflow series.

2) (a cool option think ) Put 4 1850rpm gt's mounted to the floor of the case blowing in then use 1450rpm gt's on each rad in push blowing out both sides. Simply put a powered air intake in the bottom and an air exhaust out each side.

[shazza]

Hmmnnn - interesting ideas. @ Lost - that's a thought, just not sure how much air I can get into the case with the current setup - only openings on the bottom of the PSU plate. Maybe if I added a fan in the front.

@ SNiiPE = Option 1 would probably work. Option 2 is difficult because of the PSU layout.

Keep the ideas coming. I'm guessing I'll just have to wait and test to know for sure what the impact is of different rad combos.

[iandh]

If the cold air rad was looped to the CPU, and the warm air rad was looped to the GPU(s), there would probably be only minimal change in final GPU temps.

[alacheesu]

I see Skinnee has posted air out temps on most (or all) his rad testing. It seems most rads use around 90% of available air capacity in a typical scenario, meaning the air/water delta on the second rad would be 1/10th of what the first rad has to work with. Consequently, it would contribute only 10% extra heat dissipation provided they're in the same loop. That's less than half of what you would get from adding a single rad taking in cool air. It doesn't seem worth it to me.

I'm assuming the air goes directly from one rad to another without mixing with cool air from somewhere. The 90% number is just what I found most reasonable going by the numbers on Skinnee's site, but it seems to differ a bit from rad to rad, heat load and fan speed.

[LostInRehab]

Hmmnnn - interesting ideas. @ Lost - that's a thought, just not sure how much air I can get into the case with the current setup - only openings on the bottom of the PSU plate. Maybe if I added a fan in the front.

I was looking at the floor here.



Seems like a lot of openings for air unless you're going to cover them up but thats just a thought.

[Serpentarius]

ever thought of having some air duct?

instead letting the hot air goes wherever they want (physics), why not direct them through proper channels?
besides, air is highly compressible, you can have small air duct, without risking the asthestics
plus hot air is lighter and more easily exits through the ducts

how bout this, get those old newspapers, tapes and scissors ... craft some of your 'origami' ducting shapes, tape it on the rad with fan and fire them up

[Sadasius]

You know Serpentarius that is exactly something I was thinking about in my next build and is a very good idea. Works on exchange systems in buildings and such, why not a case.

[Serpentarius]

You know Serpentarius that is exactly something I was thinking about in my next build and is a very good idea. Works on exchange systems in buildings and such, why not a case.

you can (partially) insulate the vents (assumed thin metal sheets) by using Plastidip spray for the external surface to contain and minimise the hot air losing its heat through convection on the metal sheets

it's a shame i junk out my ghetto duct years ago (build via alu drink cans)
i bet you guys never used a plastidip on the internal casing build ... plastic finishing with soundproofing .. just remember to spray a thin primer for adhesive, they tear easily with the smooth alu casing surface

[BrokenArrow]

In my research for the TJ-07 case, I came across a thread (by AndyM I think) that tested three different ways of orienting the fans.

The clear winner was having the air just pass on through the rads. Even though the second rad was getting the warm air from the first rad, it was better than the other orientations you mention by quite a bit.

[shazza]

Thanks all, for the ideas. Not quite sure where the air duct goes?

@ BrokenArrow - thx - I remembered seeing some comparisons but couldn't find them with my searching - I'll try again. (Not sure exactly what you mean by "air just passing through the rads).

[alacheesu]

@ BrokenArrow - thx - I remembered seeing some comparisons but couldn't find them with my searching - I'll try again. (Not sure exactly what you mean by "air just passing through the rads).

This post, maybe?

[ohms]

+1 for Option 2 SNiiPE

[shazza]

Thanks, alacheesu - I assume that's the post BrokenArrow was referring to, and that's the type of info I'm looking for.

It was this comment by Marci in crediki's thread that got me thinking about it:

Re: the case design - Radiator of choice as INLET... Enough fans to create higher airflow at exhaust than at inlet. Get negative pressure inside the case due to higher exhaust than inlet... will improve airflow thru any rad (effectively push/pull with a big-ass shroud). ...

But, since I have extra radiators lying around, was trying to figure out if it made any sense to add another one on the exhaust side, since the fans will be there anyway. I don't need the extreme cooling, more interested in the option to run the fans at lower speeds.

[Garrett]

I don't mean to hijack the thread but I've been thinking about this as well shazza, in fact I like the Corsair 800D case, put 2 Quad rads in the bottom and modify the side panels so that each panel has one of these installed: http://www.scan.co.uk/Images/Products/1006758-a.jpg

I know the PSU will fit in the top of the case... get my idea?

That way you'd create a "TJ-07-like" bottom compartment that does all the watercooling for you

Install the motherboard, right next to it you have a cilinder shaped reservoir and the pump under that, and you route the tubing to the rads
fan/rad/fan/rad/fan something like that, imagine having 2 quad rads there

[BrokenArrow]

Thanks, alacheesu - I assume that's the post BrokenArrow was referring to, and that's the type of info I'm looking for.

Yes, this was the thread I was referring too. Thanks alacheesu.

[Utnorris]

Why not do the separate chamber and add a vent to the front of the case? Pull air in from the sides and let it exit out the back and front.

[MoMo_RT]

On my TJ07 I just have both radiators pulling into the center then I made sure there's enough open space above them for the exhaust to escape through the upper sections of the case. This allows the fans to keep their pressure up and it moves air through the case so you don't end up with stagnant hot air building up.