heatpipe cooling upside down

Cooling Processors quietly

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
Posts: 4
Joined: Sat Jun 21, 2008 6:15 pm
Location: Denmark

heatpipe cooling upside down

Post by loldrup » Sat Jun 21, 2008 6:22 pm

I am builing a HTPC where the motherboard will be placed upside down, so the CPU cooling fan will suck air in from below.
Do any of you know if this will rule heatpipe based CPU coolers out of the game? They work by vaporisation of a liquid. When the liquid is near the CPU it vaporises and the vapor should drift upwards towards the cooling fins (because vapor is lighter than liquid), where it will cool down again. That's how I have understood the heat pipe principle.
But when the fan is upside down, the vapor will already be at the upper most part of the heatpipe - it can't escape to the fins. Is this true? I hope not..

/Jon, Denmark

Posts: 39
Joined: Fri Jun 13, 2008 12:03 pm
Location: Pullman, WA, USA

Post by Stravos » Sat Jun 21, 2008 6:28 pm

I think the pipes fill up with vapor quickly enough that it wouldn't really be a problem. The wicking material *should* keep the liquid distributed throughout the pipe as well.

Posts: 1266
Joined: Sun Oct 09, 2005 8:35 am
Location: Pleasanton, CA

Post by cmthomson » Sat Jun 21, 2008 6:31 pm

No, modern heat pipes don't rely on gravity. The working fluid evaporates from whatever part of the pipe is hot, and condenses on whatever part is cool. The wicking on the inside of the pipe spreads the liquid more or less evenly regardless of temperature or orientation.

So the short answer is: yes, modern heat pipes work just as well "upside down" as "right side up".
i7 4790K CPU@4.6 GHz, ASUS Z97-PRO, 16GB G.Skill 2400C10, Intel 335 240GB SSD + WDC EFRX 1TB, Internal i7 graphics, Antec P180 case, Seasonic X-400 fanless PS, Megahalems CPU HS, Nexus 3-pin & AC PWM fans < 600 RPM, AcoustiPack foam, homemade ducts.

Posts: 308
Joined: Sun Jan 25, 2004 7:50 am
Location: Moscow, Russia

Post by EndoSteel » Mon Jul 07, 2008 8:21 am

The operability of a heatpipe oriented upside down depends on it's length (or the length of it's vertical section, i.e. how high does the wick structure have to pump the condensed fluid). 10cm - you doubtly notice any difference, 30cm - your temps go ski-high.

They make special pipes for upside-down operation, with a very thick wick (this widens the pipe's outer diameter to 8mm while the vapour chanel stays similar to that of a 6mm standard pipe) and even then the heat transfer capacity is lower than that of a standard 6mm pipe oriented normally.

Posts: 406
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

Post by colm » Mon Jul 07, 2008 9:24 am

a popular heat pipe vid card cooler proves to me upside down doesn't matter. path of least resistance loses gravity. Heatsinks and upside down cpus may be an oddity, as cpus aren't normally upside down. Vid cards are...
then again, most cpus are sideways, vertical and heatsinking through anything attached.

Post Reply