Push or pull config on PA120.3 Radiator?

[el rolio]

well uh, someone said in another thread that thermochill reccomends push config for their PA series of rads. Is this for slower fans? does it work better that way for the slower fans? is it a real difference?

here is my setup:


fans pull cool air thru rad in this config. thoughts on this push/pull concept?

[creidiki]

the diff is about 1-2% in favour of push. try both and see what wopks bets for you. fan speed shoudlnt make a difference because of the minimal restriction of PAs.

[el rolio]

any difference noisewise? and its kinda difficult to try the diff things out hmmmz, cuz of the way the tubes are attached n all. hmmz

[creidiki]

push is quieter generally - rad before fan causes cavitation more than puish causes noise because of backpressure, with PAs.

but if youre worried about noise id ditch those 38mm boys there.

[el rolio]

but if youre worried about noise id ditch those 38mm boys there.

j-mags black yate loons will arrive any day now.

so the plan is to take the time to switch the fan setup as best i can and install the neoprene pad that came with the rad but to do any and all improvements at one time. shroud for the win? shrug actually i no space for that.

[eXa]

dont need shroud!

[creidiki]

yeah, shroud is actually pretty useless on PAs, the inbuilt one is more than enough.

[-aDaM^]

From my experience PULL works best with a shroud, PUSH works best with no-shroud.

shroud just takes away the dead spot...

[Radical_53]

I'm just about to test that with my rig... the push+shroud version does give me quite an impressive performance, but I'll test pull later on to see if it's even better than that.

[el rolio]

hey let me KNOW! cuz if im doing push then basically my fans would have to go ontop of my rad which will be mounted to the case directly.

hey on that thought: if the rad is gettin attached to the case directly, what would be a good dampening cushion? do i need one? or can the metal of the rad touch the alu of my case-top?

[Radical_53]

Just get another gasket like the thing Thermochill offers. Or simply neoprene... doesn't matter normally.

I'll give you some scores as soon as I'm through with it. Using YLs there btw.

[JoeBar]

Keep us updated radical...

[Radical_53]

I sure will. What would you think would be the most interestng fan rpm for looking at the push/pull difference? low, med or high rpms?

[eXa]

anything else than yates on the 120.3 is silly!

[Radical_53]

Well, l/m/h with Yates

I'm testing them "low" at 600rpm, "med" at 900rpm and "high" at 1200rpm (which then is 9.5V on my scale). So none of them is really "high speed".

[el rolio]

sounds good to me radical, cuz i dunno if i would ever run em at that full speed (12v)

[Radical_53]

Ok... I'll have some nicer scores later on, but for now these should do.

To start, for my setup push/pull only made a "real" difference at really low speeds. I'm using the shroud all the time, maybe that helps there too.

I pute the PA 120.3 in to keep an Opteron 170 (280x10, 112% voltage), 2 x 7800GT 512MB (500/1400MHz) and NF4 chipset (1.6V) cool.

push:
4V, 600rpm: 7.8K
6.5V, 900rpm: 5.9K
9V, 1200rpm: 4.7K

pull:
4V, 600rpm: 6.9K
6.5V, 900rpm: 5.5K
9V, 1200rpm: 4.7K

So as you see, the more air goes through, the smaller the differences are. By the way, I was really impressed of the overall performance... I've never seen anything alike.

[el rolio]

thanks! keep it coming! definatley me doin the push config where my fans go on top actually makes it easier to swap out me fans, since al lthe tubes are all connected and stuff. the only major issue im about to have is WHAT IN HELL DO I DO with all them molex connectors ARGH

[Radical_53]

Yup, give me some time to do that I'll try to test it the "marci way", with the air temps and all that stuff too.

[nop]

Thanks for your contribution Radical_53.

Could you please,if you'd not mind, do test the difference between "with shroud" and "without shroud" ? Thanks.

[Radical_53]

I'll see what I can do. The one side of the rad is without a gasket right now, so it should be possible

[nop]

I'll see what I can do. The one side of the rad is without a gasket right now, so it should be possible

sweet

[eXa]

k as in kelvin? whats that compared to celsius?

[Radical_53]

1:1

If you measure temperatures and want to give the delta, the difference between air and water, you don't say °C but K. That's all. If you took absolute temperatures, like "24°C air to 28°C water", then Kelvin would be +273... as the Kelvin-scale starts at absolute zero, whereas Celsius starts at the freezing point of water at 1013mbar pressure.

[eXa]

ok. good to know.

did know that kelvin starts at absolute zero, but wasnt shure if it was 1:1 or what! but now i do!