Big Radiator ?

[JToolz]

Ok I have a radiator that can fit 2 120mm fans. I want to set it up as a push pull setup with 2 120mm fans on each side. Does anyone know of any place I can find shrouds that will mount on a radiator like this with places to mount up 2 120mm fans? Thanks

[JeffPH]

this is how to do it













same concept tho

[JToolz]

what did you use for the metal and that foam stuff?

[JeffPH]

its an anodized aluminum bar.

used a self-adhesive draught excluder for the seal.

[JToolz]

Originally posted by JeffPH
its an anodized aluminum bar.

used a self-adhesive draught excluder for the seal.

Where did you buy that aluminum bar from?

[Death ++]

im gonna be making a shroud for a rad thats that size, but im gonna use a Bawls box

[JeffPH]

Jtools i got mine from an aluminum shop that does windows hehehe for houses hehehehe

[JToolz]

hmm thanks I will try there but do you think I could score that at lowes or home depot?

[JeffPH]

Jtools : am from the philippines hehehehe. but i think u could try home depot.

[JToolz]

haha oops sorry didnt notice but thanks for help!

[Jabo]

I think you would be much better assembling your fans with 4-pull config and making an appropriate shroud for them (tapered). In this way you'd be able to maximise the use of your fans doubling CFM going through your rad (tried it myself and difference is staggering, for radiators air flow is the most important thing!)

[8-ball]

Jabo, could you explain the 4-pull config please.

Also, doubling up on the number of fans will not double the cfm through the rad. The same principles apply here as they do with using multiple pumps.

Fans have a p-q curve much like pumps with a maximum static pressure and a max free flow cfm.

Two fans in series will have the same free flow cfm but double the effective static pressure.

Equally, rads will have a p-q type curve much like those of water blocks.

8-ball

[Jimbo Mahoney]

I don't think Jabo means 2 fans in series.

I think he means make a BIG shroud for FOUR fans in a 2 x 2 grid, each pulling, rather than 4 fans in a 2 x 1 grid stacked.

Hard to explain!

[Bra!nFreeZe^]

Damn nice work there JeffPH looks really clean!

[Jabo]

Originally posted by Jimbo Mahoney
I don't think Jabo means 2 fans in series.

I think he means make a BIG shroud for FOUR fans in a 2 x 2 grid, each pulling, rather than 4 fans in a 2 x 1 grid stacked.

Hard to explain!

Bl00dy h33l I knew tlepaty existed


8-ball-> agreed m8 using water loop analogy here you could compare this to splitting flow into equal lines (equal Y-splitter). Each branch would have same pump in-line and pump outlets would be branched in by another Y-splitter. It';s best config for two MCP600's (they have enough head but not enough flow )
Similiar behaviour of gases and fluids can be observed in piped/ducted flow scenarios. It is expalined by mass conervation law.
Imagine a 22mm pipe and liquid flowing through it. It flows at certain speed, right? Now imagine a reducing fitting from 22mm to 10mm pipe. To comply with aforementioned law flow speed HAS TO increase.

Same with air and fans. If your rad has 'air entrance area of 120x240mm (2x120fans) and you use 2 fans to pull or push air through you'll get 135CFM (Ys-Tech 120mm bad boys) through rad . Add two more fans at the other side and you still get 135CFM but higher 'back pressure'.
Installing all 4 fans at one side in pull config (less turbulence and dead spotting compared to push config) you'll get 270CFM being pulled through!

Radiators performace depends almost solely on air flow (coolant flow is almost negligible according to newest BillA's 'private' testing), or to be more precise how many air units is able to make contact with heat exchanging area in given time unit. What it means is that he more heat dissipation area nad the more 'air flow over it' the more energy can bi dissipated.

I hope it explains a bit clearer 8-ball
I also tested it myself and temp difference (CPU) is staggering. Water temp diff is in a region of 5 to 7 deg C! (crappy measuring equipment and all I can quote are rough deltas.... )

[8-ball]

Originally posted by Jabo

Installing all 4 fans at one side in pull config (less turbulence and dead spotting compared to push config) you'll get 270CFM being pulled through!

No you won't.

135cfm is the max free flow cfm of the fan, much like the max free flow rating of a 1250 is 1200lpm, but you don't get that do you. It's EXACTLY the same with fans.

No matter what radiator you have, you will NEVER get the max rated cfm of the fan through it.

You are quite correct in that two identical fans in series will have effectively double the static pressure and the same free flow cfm, and two identical fans in parallel will have double free flow cfm but the sam max static pressure.

But you must also remember that a greater static pressure is required to flow more air through a rad.

Which configuration will yield the best performance is down to how restrictive the rad is. There is a point where either configuration will yield exactly the same flow rate. For a more restrictive rad, then fans either side in series will give better flow rates than your parallel configuration. Less restrictive, and the parallel configuration will offer better results. This is made harder by the fact that there is very little data on the restrictiveness of radiators, and it is harder to measure than for a water block. (apparatus for measuring cfm is more complex than that for water, and you can't see how long it takes to fill a bucket with air can you?)

See the graph I have attached. The blue curves are the p-q curves for 3 fan configurations.

One is for a single fan, one for 2 in series, and the other for 2 in parallel.

The black curve is for a hypothetical radiator of a given restriction whereby either configuration will yield the same results. The red curve is for a more restrictive radiator (most likely the fins are packed closer together, or the core is deeper) where the series configuration will yield better results. The green curve is for a less restrictive radiator where the parallel configuration as you have described it will work better.

In fact, 4 fans in a combination of parallel and serial, thus doubling both the max free flow cfm AND the static pressure, will still not double the airflow through the rad due to the fact that the resistance to flow of air through the rad increases with flow, hence the positive increase in gradient.

I hope this makes sense.

8-ball

[Reflex1]

wooo sum1 did their homework.

great work jeffPH tho ....i'm gonna do something similar to that once i get all my WC parts in.

[Jabo]

Quick response, I like that

I loved your parallel with bucket filling dude ! He He He !

1. Were did you get your graphs ffrom?

2. I must admit I was not approaching the subject in as much detail as obviously is needed here

3. Air flow and liquids flow has one fundamental difference - gases are compressible. Therefore they just 'pack themselves' more densly when confronted with a restriction (radiator) and still pass through in the same numbers but slower just to explain it quickly and it layman's terms.

4. Putting two fans in series will not double back pressure, in fact it will not increase it at all (unbelievable but true)
All you get is lower static pressure loss.
To increase staic pressure all you can do is get tronger fan/noisier fan

5. Hands on experience now.
I built 'Cooling Tower' enclosure for my radiator to properly duct fans and banish any pressure loss as well as to reduce noise.
Fans used:
a) 1x 135CFM 120mm Ys-Tech pulling

b) 4x 43CFM 80mm Ys-Tech pushing

a) - was pulling 0.9375 CFM/ cm^2
b) - was pushing 0.6718 CFM/cm^2

Then I changed to all push config and water temps dropped by 6-8 deg C.
Radiator used is extremely restrictive (will post a pic when back at home).

Now, according to 8-ball theory I shoull be much better off with push pull or series config since my rad is of extremely restrictive type (very deep, core, ~120mm, and <1mm between fins).

Anyways, that's my opinion on the subject

[Jimbo Mahoney]

Originally posted by Jabo
Then I changed to all push config and water temps dropped by 6-8 deg C.
Radiator used is extremely restrictive (will post a pic when back at home).

Is the theory of pulling air through a rad incorrect then?

Or is it a case of try-it-and-see?

[Jabo]

Originally posted by Jimbo Mahoney
Is the theory of pulling air through a rad incorrect then?

Or is it a case of try-it-and-see?

No, soryy for being misleading, it is all correct and even if sometimes you may not notice any substantial improvements in temps (your original push setup was very well implemented i.e. appropriate distance between rad and fan(s) preserved wirh air tight shrouding) you should be able to notice lower noise levels (in theory at least) and it'll be easier to clean dust

My example was based on comparison between push-pull config and one sided installation.

[JToolz]

I want 2 fans on each side.

[Jabo]

Originally posted by JToolz
I want 2 fans on each side.

If you want it go and get it

[JToolz]

Originally posted by Jabo
If you want it go and get it

Ahh sarcasm eh? lol. I was referring to building a shroud to mount up two 120mm fans to the radiator side by side. Since all the discussion back and forth about it, I was just wondering if someone has built one before.

[Jabo]

Weapon did, search for his thread about it, it titled sth like' dual 120mm heater core' or sth
A bit of sarcasm is always good
The key in building a shroud is to provide aprropriate distance between fans and rad's surface - the more powerful fans the more space they need to kill-off turbulence and get rid of dead zones

[Jabo]

My Saab heater core piccies as promised





and another one