Foam Board Mod

[UsualSuspect]

ok i did my homework. I found with the article with mod to a zalman 7000ACu here. Yes same concept .. basically keep the heat from recirculating.
ther's been a debate whether it's better to suck hot air or blow cool air. My position, better to blow cool air. fans take in air from all directions at the input and shoots it out in a cone of turbulent air at the output. (you feel the air more at the output than the input). A hot surface will generate a skin of hot air that sticks to it. turbulent air cuts through this skin. Turbulence is our friend. Embrace turbulence.

 

[BonzaiDuck]

It always seems that in terms of temperatures, "blowing" trumps "sucking" by maybe 1 or 2 degrees Centigrade.

But the other strange thing about the ducted Zalman mod, it's hard to say whether the fan -- which is sucking -- is creating the effect of blowing or sucking, because the air comes through the heatsink fins in the same direction that the proprietary fan draws it in. In that guy's project, he still gets a significant improvement in cooling that more than cancels the supposed inefficiency of sucking air off the heatsink instead of blowing it on.

As to whether this post should be "PG-17", "PG-13", or "X-rated" -- it depends on whether the reader has his or her thoughts in the duct or the gutter. (Joke. ha ha.)

 

[KysrSose]

Damn! you beat me to the joke.

Apples to apples, how would his configuration perform if he simply flipped the fans to blow into the heat sink?

Can't say i agree that the air travels in the same direction in this new configuration vs the standard (Zalman). I'm thinking it'll get the air directly in front of the intake, this creates a void then air rushes in from all directions except that of the fan to fill it. ( i think that' how it works, i'm still trying to figure out this newtonian and bernouilli gaga) I'm not sure what's going on along it;s rims though. You don;t want it sucking ( air gets heated by the fins - sure the fins gets cool but you'r blowing into the core of the heat sink) You don;t want it to blow ( too close to the intake, loops are bad). you don;t want both ( loops are bad). The best choice i figure is to keep iair stagnant at the rim.

Disclaimer: I'm no thermodynamic physicist or aerodynamics engineer. My views are probably 100% misguided, oversimplified and false. But i had a lot of fun doing the research and even more fun applying what i've learned. Red Green is my Bob Vila.

 

[KysrSose]

UsualSuspect and I are in synch with this school of though ( we should, we're the same person ... said the voices in my head? ). So dont feel like you're being ganged-up on.

 

[KysrSose]

Here's some very good "sights" if you want to know more about ventilation and fans Helped me alot

Vent-Axia

Sunon Fans

Comair Rotron Fans

A few tips

enlightenment is irreversible.

 

[klah]

Originally posted by: KysrSose
Ok I take it back.. Dell dimension XPS .. not quite the same. Under that duct is a heatpipe heatsink. those twin 90mm look like high power fans will just brute force suck (pull system) the cool air into the 'duct' through the sides and front. effectively keeping ambient air = room temp around the heat sink. It does not look like it's forcing air to pass through the heat sink - passive cooling? (i don;t know, i can't tell what the bottom of the duct looks like). I'm sure it works, just in a different way.

This foamboard mod will leave no other way for cool air to escape without passing through the fins of the HSF - well maybe through the power supply fan but that's another issue that could be 'covered'.

Here it is with the duct removed: http://common.ziffdavisinterne...z=1&i=25273,00.gif

from: http://www.extremetech.com/art.../0,1558,1626729,00.asp

 

[BonzaiDuck]

I think that was the point with the designer's "cross-current flow" idea, because the fins are supposed to conduct heat from the heatsink base so that air will remove the heat, anyway. OTHER THAN THAT -- and the guy's rather respectable figures for cooling with very little fan noise, I can contribute to your perspective in adding that high-pressure at both the fins and the heatsink base is preferable to low pressure, because more molecules of air remove more heat per unit time. Isn't that the definition of pressure? Denser, more closely packed air molecules? And to get pressure, you have to "blow" instead of "suck". But again, it gets back to that "volume versus flow" idea, if you move more molecules per second over the surface, you still remove more heat, and that's what the Zalman modder's nozzle does.

 

[BonzaiDuck]

Anyway, that's an interesting arrangement for the Dell XPS, and it's probably pretty quiet. I'd just rather see an intake fan on that heatsink, and it would seem that you could modify the dell's heatsink shroud and do it -- and if it didn't work, cover it again with aluminum or some other kind of tape!!

It's CFM's versus noise. I think Usual Suspect is right -- some turbulence is a good thing . . .

 

[lobadobadingdong]

Originally posted by: BonzaiDuck
Anyway, that's an interesting arrangement for the Dell XPS, and it's probably pretty quiet. I'd just rather see an intake fan on that heatsink, and it would seem that you could modify the dell's heatsink shroud and do it -- and if it didn't work, cover it again with aluminum or some other kind of tape!!

It's CFM's versus noise. I think Usual Suspect is right -- some turbulence is a good thing . . .

the xps's are quiet, though they get stupidly hot. (I've seen them run as high 85C on speedfan under full load)

 

[BonzaiDuck]

Well, there you are. The OEM's take what they can from some good cooling ideas, then implement them with a minimalist approach so they can market a whisper-quiet machine.

 

[lobadobadingdong]

bringing back an old thread, but I just did this to mine (modified copy...hope you don't mind) 10C drop on my cpu and northbridge, and 8C drop on my video card.

 

[BonzaiDuck]

I finished my semi-permanent version using three pieces of foam-board -- a flat panel flush with the CPU fan-grill; an "exhaust-fan-box" that connects the exhausts to the main "chamber", and a bend side panel that mates the box with the flat panel. I used little 1/2" Velcro patches at certain strategic corners of the assembly to hold it together and add rigidity. I wish I could've got the flat-panel closer to the mobo, but this is an XP 120 setup with a 120x38mm fan. I got a few degrees C improvement in the CPU temp, but I was more impressed with the drop in chipset and AGP temperatures. I still have a few tweaks of improvement to make.

For the money and the trouble, it's not an idea or result that I'm going to discard. It works.

Akira34 is working on his adaptation for a V1100B Lian Li case. He reverses the airflow direction of the CPU fan, sucks air from underneath a flat-panel covering the motherboard, and ducts it to the exhaust fan that sits perpendicular to the CPU fan in the case rear. I can't wait to see what he posts as results.