Whatever happened to the "thermosyphon"?

Hey all:
Yes, it has been quite awhile since my last post, mostly due to work constrictions and other outside influence (divorce, moving, etc.)
I guess I have been gone too long to realize what has happened to overclocking as a whole (still using the old Abit nf7-s and Corsair XMS), but i was interested to see what actually became of Joe's "thermosyphon".
I saw a few articles and it re-sparked my interest in this concept. I had a semi-working model on this computer for awhile, but i disassembled it when i put the computer away, for fears of leaks and such.
I'll be making the jump to 64 bit soon, and i was kind of leary about attaching a huge heatsink to my CPU, so water is the best option. Although if a working thermosyphon is available, that would weigh considerably on my purchasing plans.
Any info or links would be appreciated...

I have wondered the same thing as I haven't seen anything at all on this since that article.

here is another (which seems to be dead-on-the-ground)
http://www.patentdebate.com/PATAPP/20070025078
patent http://www.wipo.org/pctdb/en/wo.jsp?wo=2007016365

Wow Bill, good to see you! How's things?

Hi Scott, watching your testing and amazed to see you trudging on; Congrats. Too few provide raw data + the conditions, to be analyzed by others.

After my abort with ArticChoke, I built 4 thermosyphons which all worked so-so ($4K in lessons); I am aware of other more developed designs which work very well indeed. Not cheap, not really DIY w/o say $10K and a scientific education.

OMG ! its.... its... its BillA !

Long time Bill !

Good to see you still visit us here once in a while.


Miss your work.

WarriorII said:

OMG ! its.... its... its BillA !

Miss your work.

Click to expand...

The WC community needs your work.

Bill come back to pro!

Hey Guys, Thanks and all that . . .
but life moves on, and I'm done with WCing
(although I would like to get back into thermosyphons, I think this is the future)

Hi Ben
Joe was presented with an either/or choice, he opted and you can assess the results
let it go, but then where to go ???
no answer, sorry (the engn forums have defined 'languages')
EDIT: christ, are you in NH from TX ?

BillA said:

Hey Guys, Thanks and all that . . .
but life moves on, and I'm done with WCing

Click to expand...

Have to ask...
Is there another endeavor where we can be uh, informed/inspired/amazed by your uh, musings?
...or are you completely off the internets?

WOOHOO, wlecome back BillA!

Well lookie here...I dissappear for a couple of days and who appears out of nowhere to grace us with his presence....lol

Good to know you're still around, BillA...

Keep the faith!!

Good to see Bill around.

I've always wondered this, though. With a Thermosyphon, they're designed for a specific heatload, correct? What happens when your heat load is below/above that design load?

With above, I'd imagine it's the same as a TEC--you get heat "runaway."

With below, I really don't know. Would the Thermosyphon still work? I guess that's my main question.

Any input welcome. Thanks.

TJ
variable results, but that is a limitation of the boiling plate design - can be circumvented
a ts will have a higher C/W at low loads, no problem as the heat load is low also

Thanks for the reply Bill. I still think it's a very interesting technology, but one that won't catch on with the crowd that frequents this forum unless this can change:

Not cheap, not really DIY w/o say $10K and a scientific education.

Click to expand...

People here are willing to pay for parts, willing to DIY, and willing to learn (ie. education) but............$10K is a boatload of money to lower a couple degrees when blocks of copper, a pump, and an old heatercore still do well enough. Well, good enough for me, anyways

Hi Bill, hope you're OK

Major problem in a thermosyphon is the boiling plate effectively, but condensing parts have to be good enough to avoid high pressure rise which leads to higher boiling T°.

The base plate design have to be specific to :
- reduce as much as possible the surface superheat in order to allow the fluid to boil very quickly/efficiently with minimal T° difference between fluid saturation T° and plate T°
- to boost the critical heat flux as much as possible to avoid film boiling which is bad (-> insulation by the vapor film onto the hot surface, too much power density) and keep an efficient bubble boiling.

Base plate must be as "dirty" as possible, with tiny pins/structures/holes to provide lot of rough surface, so lots of nucleation sites for bubble development. The same rules have to be applied to evaporators used in phase-change systems.

Some firms sell copper plates with a tiny structure above made of copper foam (Metafoam). They could achieve very good boiling due to the sponge structure.



There's a lot of studies on boiling because it's not so simple when you want to optimize any design, but you can make a good/classical system without spending 10 k$

Well,...
I've been thinking of working up another proto, possibly with metafoam as the base material, and something other than copper as a "block" material (to allow the gasses to escape the block before re-condensing). In my test blocks, I had problems with getting the gasses out of the block and to the exchanger, which was possibly caused by the expansion rate of the gasses overcoming the vacuum in the chamber. I also tried making a "radiator" type exchanger, where I soldered a 2" long piece of "baseboard heater" to the plate, fitted it with a schrader valve, added methanol and put it under a vacuum of 28.6mm Hg. It worked for awhile, until it went into a sort of "cold-cycling" loop, which i believe was caused by either having too much coolant, or loss of vacuum. I did have word documents on all the different setups and trials, but they're gone now.

BTW Bill, you might want to change the link in your profile...