Questions about Scythe's heatpipe caps

Cooling Processors quietly

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
Fire-Flare
Posts: 422
Joined: Sun Jun 13, 2010 12:44 pm
Location: Seattle

Questions about Scythe's heatpipe caps

Post by Fire-Flare » Mon Sep 06, 2010 11:44 am

First, I'd like to know how tall they are.

Second, do the coolers perform better if the caps are removed? In my mind it seems they'd act as insulators.

danimal
Posts: 734
Joined: Mon Jun 08, 2009 2:41 pm
Location: the ether

Post by danimal » Mon Sep 06, 2010 4:23 pm

scythe makes all kinds of coolers, are you sure that they all have the same size of caps on the pipes?

no, you should never remove the caps.

bonestonne
Posts: 1839
Joined: Mon Feb 05, 2007 2:10 pm
Location: Northern New Jersey
Contact:

Post by bonestonne » Mon Sep 06, 2010 6:04 pm

As far as I'm aware, the caps are actually glued on for looks alone, the heatpipes do not rely on those caps for any sort of functionality. If that's changed over the past couple years, then by all means, leave them in place.

As far as performance is concerned, I doubt the caps have any effect, as they are glued to the end of the heatpipe above the last fin....heat wouldn't be able to travel up that far, and it would be dispersed pretty quickly if it did.

kater
Posts: 891
Joined: Thu Sep 07, 2006 11:20 pm
Location: Poland

Post by kater » Mon Sep 06, 2010 10:07 pm

I'v seen ppl safely remove caps from first versions of Ninja (e.g. SPCR team) as that helped fit the cooler in narrow cases.
Personally, I've never seen a heatpipe used in PC coolers that would not be closed by soldering. Soldering seems to be the only reliable method of keeping the magic stuff (aceton, alcohol, some other sh...) inside for a couple of years. A screwed on cap or a glued cap would be IMHO way too loose to keep the cooking stuff (presumably under some pressure too) inside.
So if you know you can remove caps w/o damaging the pipe itself, I'd say go for it. Just remember that you're working with thin walled cooper, a soft yet brittle material.

frenchie
Friend of SPCR
Posts: 1346
Joined: Tue Jul 08, 2008 4:53 am
Location: CT

Post by frenchie » Tue Sep 07, 2010 3:02 am

kater wrote:Soldering seems to be the only reliable method of keeping the magic stuff (aceton, alcohol, some other sh...) inside for a couple of years. A screwed on cap or a glued cap would be IMHO way too loose to keep the cooking stuff (presumably under some pressure too) inside.
+1

I've had all kind of coolers (CPU, chipset, GPU) from Noctua, Thermalright, Scythe, Zalman and Arctic Cooling. The heatpipes were always soldered closed. Like kater said, a screw cap would be too risky and harder to manufacture than just a dab of solder.

quest_for_silence
Posts: 5275
Joined: Wed Jun 13, 2007 10:12 am
Location: ITALY

Post by quest_for_silence » Tue Sep 07, 2010 3:03 am

kater wrote:ISoldering seems to be the only reliable method of keeping the magic stuff (aceton, alcohol, some other sh...) inside for a couple of years.

Why just "for a couple of years"?

Regards,
Luca

kater
Posts: 891
Joined: Thu Sep 07, 2006 11:20 pm
Location: Poland

Post by kater » Tue Sep 07, 2010 3:13 am

Well, ok, forever ;) As long as the metal rots away or the magic vapor loses its properties. True. OTOH, how long are you going to need even the best performing cooler?

I've had my original Ninja for some time now and will definitely keep it as long as I can, but ultimately, when CPUs and other heat generating PC equipment will be able to do away w/o coolers, like in the days of first PCs, we will no longer need them. This day is coming, and it's going to be a sad time for geeks like us :(
No, seriously, given the increasing power efficiency approach and importance, given that sand based chips only have a few years before becoming obsolete - it's just a matter of time. 10 years ahead and we'll be looking at our smoke belching & fire breathing machines as some hellish contraptions brewed by retarded cavemen, with just brute force in mind. I guess we'll be using those bio-green-goo-chips in no time. Try applying a GeminII to that :lol:

foreverdisturbed
Posts: 4
Joined: Thu Sep 09, 2010 9:21 pm
Location: USA

Questions about Scythe's heatpipe caps

Post by foreverdisturbed » Thu Sep 09, 2010 9:34 pm

Hello, I was a former employee of 18 years for Thermacore inc. and I can tell you that end caps on heat pipes have no function except to protect the end of the heat pipe. Solder is sometimes used because when a heat pipe is closed by means of a pinch off tool or also known as a cold weld it will create a razor sharp edge which is dangerous and because if it was bumped on a hard surface it could cause the pipe to open. Every company does it differently such solder, welding, epoxie, braze or just a glued on endcap. I hope this info is usefull and if anyone has any questions about any unanswered Electronics cooling questions just feel free to ask. I have built almost all types of heat pipes, vapor chambers, pumped liquid cooling, thermosiphons, loop heat pipes, air to air and liquid to air exchangers for both high and low cooling applications. I can tell you the things that no electronics cooling company will tell you.

kater
Posts: 891
Joined: Thu Sep 07, 2006 11:20 pm
Location: Poland

Post by kater » Thu Sep 09, 2010 9:41 pm

Welcome to SPCR!*

This kind of experience and know-how is always welcome among geeks :) So do drop by and share it with us.

* always wanted to say it...

foreverdisturbed
Posts: 4
Joined: Thu Sep 09, 2010 9:21 pm
Location: USA

Questions about Scythe's heatpipe caps

Post by foreverdisturbed » Thu Sep 09, 2010 9:47 pm

Thank you, This is a much better welcome response than I got on Tom's Hardware Forum. Two different people told me not to respond to an old thread which was not that old and then they closed the thread.

mark314
Friend of SPCR
Posts: 82
Joined: Tue Aug 14, 2007 8:02 am

Post by mark314 » Mon Sep 13, 2010 8:48 am

You used to work for Thermacore, interesting.

I worked with them briefly in the past - they drew up a quote for a heatpipe prototype but it was far too expensive. $3600 for 6 small-radius heatpipes...

So I will ask you some questions then, if anyone thinks I am steering this thread off topic I will move it to PM or start a new thread.

1. For a small-radius (say, 8mm) heatpipe, what is the theoretical maximum horizontal length that it will still be functional? What effect does the condenser and evaporator length have on that? What effect does sintered vs. wicked pipe have? Does efficiency have a sudden drop off at long lengths, or does efficiency fall linearly over increasing lengths?

2. How would you attach heatpipes to metal blocks at the evaporator and condenser? I imagine that soldering is the best, but is precise machining and high pressure sufficient? Should TIM be used as well?

If you're interested in answering, it would be awesome to hear what you have to say. Thanks...

Mark

foreverdisturbed
Posts: 4
Joined: Thu Sep 09, 2010 9:21 pm
Location: USA

Questions about Scythe's heatpipe caps/thermacore

Post by foreverdisturbed » Mon Sep 13, 2010 2:59 pm

First I will tell you I was a Tech not an engineer. What I can tell you that $3,600 is no suprise for a prototype. Thermacore is not known for being low cost, Most job's were high end.

Length on a larger diameter heat pipe/thermosiphon usally meant that is was a gravity aided system. This means that length & radius was not an issue. This type of system was always high power usally well over 700w to as much as 4000w. This would be expensive even in high quantity's so i would have to say that your quoate was not excessive for a prototype.

On small diameter heat pipe about 1/4" Dia. low power length usally was not over 12" without affecting performance. Radius would not affect performance. This is in a heat pipe using copper powder wick and could perform in any orientation.

Larger heat pipe/thermosiphon systems still perform better when using a powder wick but, reall only needs to be in the evaporator end because it is gravity aided. The reason for using powder the entire length is that it was easyer just to fill the pipe. Also you would still need some sort of wick at the condenser end not for performance but, to act as a muffler. Without the wick the pipe would ping (make noise).

Soldering the pipe at the evaporator end is the norm and the best way for performance. On the condenser end the best way for performance was to stack your fins on the tube then to expand the tube using an air expander which gives the best thermal performance and makes the fins nearly impossible to remove. Most companys used pressed on fins which is not as efficient that is why they also use solder or glue them on after sliding them on.

Yes, Precise machining and pressure would be sufficient if you could apply enough pressure. We did not need to use TIM during testing because we could apply pressure using an air cylinder (Bimba), But in a computer you could not do this so then you would need TIM.

Hope this answers your questions.

mark314
Friend of SPCR
Posts: 82
Joined: Tue Aug 14, 2007 8:02 am

Post by mark314 » Mon Sep 13, 2010 7:02 pm

That does answer some questions. Thanks! The heat pipe companies like to make their own assessments as to the viability of my heat pipe designs and ignore the engineering and research I put into it first.

I am working in the 30-60W range, not 700W :) However, my longest pipe is about 22" long. But I have 4 pipes total - which is overkill for the application. I guess the question is, 3 would suffice but I have 4. Will the fourth, extra long one, work at only partial efficiency or not work at all? In other words, if a heat pipe is long enough to degrade in performance, does it degrade linearly or does it just not work at all?

I've taken over this thread. We can continue on PM if you want.

M

foreverdisturbed
Posts: 4
Joined: Thu Sep 09, 2010 9:21 pm
Location: USA

Questions about Scythe's heatpipe caps/thermacore

Post by foreverdisturbed » Mon Sep 13, 2010 8:34 pm

It will work with four but, probably not nessesary at such a low power. Why so long? The length will be the biggest problem if the pipe orientation is horizontal. If the pipe is verticle length should not be a problem and 3 pipes would be plenty for the amount of power it has to keep cool. In a horizontal possition it's the fluid return were the problem of performance happens and the larger the diameter and length the pipe is the more it has trouble getting the fluid back to the evaporator without the aid of gravity.

We did build some systems that used two 3/8" dia. 20" long with a very special powder blend which worked well horizontal even 5 degrees against gravity cooling 100w. These ran 25c above ambient. The two pipes were bent in a u shape with a copper block soldered on the evaporator and pressed/glued on fins and this worked well enough for the costomer. This powder was a pain to work with though.

I don't mind talking here on this thread unless someone else has a problem with it.

Post Reply