• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

More rad surface area or higher CFM?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Ross

Senior Intel-lectual
Joined
May 20, 2002
Well, I finally swapped out my EXOS for an internal setup using 2 Swiftech MCR-80 rads. I figured it would run a little warmer and most of the time it does...about 1-2°C after some playing around.

When it's cool/cold ambient, the setup does cool extremely well (plenty of cold air available to both rads...one on top and in front), but the gap seems to grow the warmer ambient it is ("normal" room temp...72-74°F). EXOS CPU idled at ambient with high Vc/OC whether ambient temps were 45-80°F. The new setup idles ~1-2°C over ambient at the same Vc/OC at <65°F ambient and that creeps up to about 3-4°C over ambient at around 70°F+, which leads to my question...

Would better cooling be obtained with more rad surface area OR more CFM on the current rads? My plan always was to use a BIM II, but jumped of the 2 of Swifty rads instead. After seeing the nice 3°C drop in cooling with both of them in parallel as opposed to series, I am wondering if I had a BIM II in front *and* one of the MCR-80 rads in parallel would be that much nicer...it would add 33% more surface area without taking up any more usuable space in my case from the current setup.

Option 2 would be just to add some higher CFM fans to the MCRs already in there. One came with one fan from Swifty and the other was a bare 2nd that I purchased, so I am using the ATC-111 Coolermaster stock case fans on 3 sides of the 2 rads. I cannot find specs on the fans, but everything I did find leads me to believe they are only 25-28CFM. The Swifty fan is 32CFM for sure and it does move more air than the stock case fans. I was contemplating changing all rad fans to either Evercool aluminum fans or Panaflo M1As or H1As.

Can anyone confirm that the Evercool 80mm aluminums are in fact 37CFM/27dB? I've seen 32 and 37 CFM specs on them, so I was just wondering...

So what do you think would offer better cooling? $50 for a BIM II or $30-40 in better fans? Yes, of course both would be better, but just one or the other please! :) Thanks...
 
You would be hard press to get more cfm though those rads with 80mm fans. Without alot of noise anyways. I'd pick the most powerfull 80mm fans I could find and run them of a fan controller. If you have to use those rads for space reasons, then I would definitly do the BIM2 and use the other two you got now in parallel with the BIM2. You cold cool a GPU and NB with one of those single BIM with decent results.
 
Thanks for the reply. Yeah, that's why I was thinking maybe the H1As @ 39CFM/32dB and they have a pretty high static pressure compared to most 80mms. One more kicker is that the ATC-111 has those nice grills over the front and top fans. They look very nice, but block a lot of air flow.

I suppose I could run a BIM II in front, MCR-80 on top and MCR-80 on the back exhaust. That would double the surface area, but that is also double the number of fans and a whole lotta things that MCP-600 has to pump water through....not to mention a plumbing nightmare :)

To do it with one or two rads between the CPU and GPU/NB, they would have be series in there somewhere. That would definitely make use of every single fan hole for rads currently in the case, just sounds like a whole bunch of work for the pump. Guess if I get the BIM II I can try it at some point just because I'll have all the pieces to do that...

So in short, you think I am better off with more rad area than better fans?
 
I think your better off with more rad area myself. Better fans will always help but not as much as the rad surface area. Better fans can always come later if you need to or if you want to.
 
Well the answer is a bit of column A, and a bit of column B.

The thermal capacity of air is around 0.05C/W per 35CFM (that's real CFM - not pretend fan CFM). This presents a lower limit on what temperature your water can ever get to given a 100% efficient radiator (which is impossible).

Sticking some 39CFM 80mm fans on your BIM's would likely yield about 16CFM or so (pretty close to that) through each radiator.

Given radiator efficiency and a typical unoverclocked CPU under load, I predict you'd be likely to see water temperatures about 6C or so above ambient with the two BIM, and approaching 10C if overclocking heavily.

Compare that to just a single 12cm radiator with two Panaflo L1A's (one on each side), which would give water temperatures of 4-7C above ambient, depending on CPU overclock, and would be quieter.
 
Last edited:
Ross said:

Can anyone confirm that the Evercool 80mm aluminums are in fact 37CFM/27dB? I've seen 32 and 37 CFM specs on them, so I was just wondering...

the evercool aluminium 80mms come in 3 versions -
these are evercools figures copied from the back of the box for my "M" versions:
L 2000rpm, 0.09A,24.81CFM, 1.42mmH20, <23dB
M 2500rpm, 0.13A, 32.4cfm, 2.48mmH2O, <25dB
H 3000rpm, 0.15A, 37.27cfm, 3.25mmH2O, <27dB

think pressure is the key to driving air through rads - if you must have 80mm ones then 38mm deep ones would be best...
 
Re: Re: More rad surface area or higher CFM?

pauldenton said:

think pressure is the key to driving air through rads - if you must have 80mm ones then 38mm deep ones would be best...

Has anyone been able to find such a beast? I'd be interested in a couple of those 7volted.
 
Re: Re: Re: More rad surface area or higher CFM?

SysCrusher said:


Has anyone been able to find such a beast? I'd be interested in a couple of those 7volted.

Delta 80x38mm FFB0812HHE
Specifications: 80x80x38 mm, 45 CFM @ 3400 RPM, 39.0 dBA, 2.4w .2 amp, 12v DC
Delta 80x38mm FFB0812SHE
Specifications: 80x80x38mm, 68.51 CFM @ 4900 RPM, 48.5 dBA, 6.96W, .58 amp, 12v DC
Delta 80x38mm FFB0812EHE
Specifications: 80x80x38mm, 80.16 CFM @ 5700 RPM, 52.5 dBA, 10.80W, .90 amp, 12v DC
Vantec 80x38 Tornado
Specifications: 80x80x38mm, 84.1 CFM @ 5700 RPM, 55.2 dBA, 9.10w .76 Amp, 12v DC

may well be other cheaper sources - i know little about US retailers that don't sell w/cooling kit to the UK...
 
Cathar--
Thanks for the info and the temps you mention over ambient are about what I am seeing (~3-4°C), overclocked big time, but only with warmer room temps.

So 2x "39 CFM" fans will only push about 16 CFM through the rad? So those weeny 25-28 CFM jobs that are in there now must really stink ;) Toss the car speaker grill type case grills over them and the air would just dribble through...which is about what it does.

I'd like the performance of a big rad/core and some 120mms, but the only way to fit it in my case would be to totally cut out the drive cage, which I obviously do not want to do...

pauldenton--
Thanks for the info...that explains the 32/37 CFM rating of the "same fans". The H's look like they have some "decent" static pressure (>3mm) like the H1A's (3.78mm), but are more quiet. I was thinking the same thing about getting air through the rads, which is why I mentioned the static pressure a few posts up...I think. Maybe it was in another post or PM...but I agree with that thinking.

How many of those Evercools do you have and how quiet are they really?

SysCrusher--
Yeah, I've got an 80x38 Tornado here for you...I'll sell it real cheap ;) I'll even include a small bottle of Exedrin with it LOL! It's somewhat tolerable at 7V, but has an annoying (to me) pitch even there...
 
Ross said:
Cathar--
Thanks for the info and the temps you mention over ambient are about what I am seeing (~3-4°C), overclocked big time, but only with warmer room temps.

Well I was talking water temperatures. The CPU will always be considerably warmer than the water, it's just that P4's don't report anything remotely approaching accurate temperatures. This is not to say that I doubt your honesty at all, please don't get me wrong on that, but there's no way in heck that a 3.7GHz P4 dumping ~110W of heat is sitting at what would effectively be the predicted water temperature, regardless of what the P4's temperature might read as.
 
Ross said:
So 2x "39 CFM" fans will only push about 16 CFM through the rad? So those weeny 25-28 CFM jobs that are in there now must really stink ;) Toss the car speaker grill type case grills over them and the air would just dribble through...which is about what it does.

I was saying that a single 39CFM rated fan would push probably 16CFM through a single BIM. i.e. 16CFM through each.

Three BIM's with a 39CFM fan on each would be about equal in cooling performance to the single 12cm rad with a low-noise fan on each side, so that may be an option for you to consider if you have room to fit another BIM in there somewhere.
 
Thanks for the clarification. I am/can only go by what SpeedFan/MBM tell me, so temps are all relative...and yes, Sandra estimates it right about 109W I think...good call.

Three BIM's with a 39CFM fan on each would be about equal in cooling performance to the single 12cm rad with a low-noise fan on each side, so that may be an option for you to consider if you have room to fit another BIM in there somewhere.

That is exactly what I am thinking about doing. I have 2x 80mm slots in the front where I was thinking about adding in a BIM II to replace the one BIM (MCR-80) that is there now and keeping the 2nd BIM on top (blow hole) in parallel with the BIM II. So, in essence, surface area of 3x BIMs without a bunch of plumbing issues to connect 3 separate 80mm rads.

So I'll definitely get the BIM II and maybe down the road contemplate adding in some better fans for all of them. Just FYI, currently there are fans on both sides of both rads, they just happen to be really weak OEM Coolermaster case fans except for the one 32CFM job that came with one of the rads...

Thanks for the input and I think my plan is now set :)
 
Ross said:

SysCrusher--
Yeah, I've got an 80x38 Tornado here for you...I'll sell it real cheap ;) I'll even include a small bottle of Exedrin with it LOL! It's somewhat tolerable at 7V, but has an annoying (to me) pitch even there...

LOL I just might take you up on that. One idea is maybe gut an old dead 80mm fan and use it as a shroud. If you there's room of course. An 80x25mm will do. That would help with the dead spot under the fan hub and decrease noise alot.

Thanks for the links pauldenton! I forgot about the tornado.
 
Ross said:

pauldenton--
Thanks for the info...that explains the 32/37 CFM rating of the "same fans". The H's look like they have some "decent" static pressure (>3mm) like the H1A's (3.78mm), but are more quiet. I was thinking the same thing about getting air through the rads, which is why I mentioned the static pressure a few posts up...I think. Maybe it was in another post or PM...but I agree with that thinking.

How many of those Evercools do you have and how quiet are they really?


i have 3 ("M") in my coolermaster 201 (replacing the stock fans - 2 intake behind the grill(with the foam filter removed) and 1 outtake) they are very quiet (sufficiently so for me not to bother connecting them up to my rheobus as i see no need to run them under 12v) - that's in a pretty quiet setup (120mm fsp psu, stock NF7-S northbridge fan, modded SK-7 + smartfanII on rheobus at just over 7V, sapphire "ultimate" 9700 hence no GPU fan)
 
SysCrusher said:

One idea is maybe gut an old dead 80mm fan and use it as a shroud. If you there's room of course. An 80x25mm will do. That would help with the dead spot under the fan hub and decrease noise alot.

Actually doing this will make things worse. A fan casing is about the worst shape you can go with for a shroud. Trust me on this one.
 
i have 3 ("M") in my coolermaster 201
Ah, so you know exactly what I am talking about. The 201 and my 111 are pretty identical inside. Thanks for the feedback on them. I would probably rather have a tad more CFM, but if I end up with 6 of them, they really just need to be quiet, so maybe the Ms are a good in between.

Carthar/Sys--
the fan mounts on the MCR-80s are not directly against the core...the surround that they bolt to actually stands about 3/8" or so above the core on each side. I am sure not as good as a real shroud, but better than right against it. In any event, the depth that shrouds would add front and back would make a little bulky for the case...
 
Cathar said:


Actually doing this will make things worse. A fan casing is about the worst shape you can go with for a shroud. Trust me on this one.

I never tried it myself. Just heard others using that but I never knew what the outcome was. Just the last thing to try I guess if nothing else would work. Can't be worse than no shroud I would think.
 
Cathar said:
...
The thermal capacity of air is around 0.05C/W per 35CFM (that's real CFM - not pretend fan CFM). This presents a lower limit on what temperature your water can ever get to given a 100% efficient radiator (which is impossible).
.......
A graphical representation may interest

CatRad.jpg
 
Back