Horizontal vs. Vertical Orientation

[toadeater]

VR Zone made an interesting discovery in their review of the new Scythe Zipang heatsink. By turning the heatsink so the heatpipes were horizontal, they achieved a 9c temperature drop. If you have one of these Big Typhoon type coolers, you might want to experiment with the direction the heatpipes are facing if you're not getting good temps.

http://www.vr-zone.com/article...e_Me%21%21/5536-5.html

 

[DerwenArtos12]

Originally posted by: toadeater
VR Zone made an interesting discovery in their review of the new Scythe Zipang heatsink. By turning the heatsink so the heatpipes were horizontal, they achieved a 9c temperature drop. If you have one of these Big Typhoon type coolers, you might want to experiment with the direction the heatpipes are facing if you're not getting good temps.

http://www.vr-zone.com/article...e_Me%21%21/5536-5.html

I'd be willing to bet that they temperature difference is from one of two things, neither of which is gravity.

1) with the heatpipes in the vertical orientation they're practically touching one of the mosfet heatsinks which is going to bleed heat off to the cooler heatpipes and reduce their ability to pull heat away from the CPU.

2)Depending on the thermal compound and application method it's a fact that re-seating a HSF can improve performance if improperly applied or, with pastes like AS5 there is a curing time after initial installation before it gives it's best performance.

I've been using a downdraft heatsink with heatpipes for more years than I'd like to recall and it's been installed every which way you could think of with no notable difference to temperatures. This theory relies on heat being affected by gravity which, anyone with any knowledge of thermodynamics knows, is pretty much completely BS. Some heatpipes(afaik only some older ones) were filled with a chemical that was liquid at room temperature and had a boiling point of aroud 40C so that some very small degree of phase change could be used to transfer heat away from the cpu and into the fins but, with either orientation that effect would go right out the window. With the verticle orientation all the liquid would end up in the fin end of the heatpipe and with the horrizontal the liquid would self level and there would be both liquid and gas through the entire length of the heatipipe. So, I call BS on this article until I get some substantiated, repeatable, and 3rd party confirmed results.

EDIT: speelling and grammarr

 

[Nathelion]

Word. My sense of physics tells me that the only way they could get this result is if they were doing something wrong with their first mounting.

 

[imported_wonderwrench]

Well orientation can affect heat pipes because the way they work. I agree 9C is hard to believe though.
http://www.enertron-inc.com/en...s/heat-pipe-design.php

 

[DerwenArtos12]

Originally posted by: wonderwrench
Well orientation can affect heat pipes because the way they work. I agree 9C is hard to believe though.
http://www.enertron-inc.com/en...s/heat-pipe-design.php

That graph is only a 3mm heatpipe for one and it's a straight pipe, 180 degree curved pipes, let alone 360 degree curved pipes, completely changes the dynamics and more or less completely negates the assistance of gravity. Plus that was a 10W load on a 3mm heapipe and was only measuring out to 200mm. On top of that most of the high end coolers are using powder metal heatpipes which are more or less competely un-affected by gravity. IF that specific cooler used groove heatpipes and they were longer than 150mm then it could have made up to a 5C difference, if they were 3mm heatpipes with a full 10W load. Thermal resistance with even the lowest dualcores pumping out 80watts at load would be less effected, especially when considering that cooler has 6-6mm heatpipes and by their design I'd say they're probably powder-metal heatpipes.

 

[Ika]

I oriented my Big Typhoon so that the fins were parallel to the ground, so the rear fan could do some work pulling out heat from the heatsink.