New pelt cooling design?

[Silver]

As the landscape changes again, we find that Mr. Leufken may have added something new to the game. Don't know how to paste a thread so you will have to search for it. The potential for pelting to very low temps may be within grasp. Now, for home built units along the same lines?

Direct cooling pelts.

[Colin]

What was old (http://www.overclockers.com/tips111/) is new again.

[Silver]

Did I bother to say "Thank You". Gone about as far as I can with water cooling and am starting to look at pelts or other forms of cooling. Thanks again. BTW Leufkin took out a patent on the the idea, three in all I hear.

[Colin]

It's impossible to get a patent when there were prior applications of a design. Marketing smoke!

[riprock]

...plus, unless they've changed the patent process in the last 10 years (which is possible), it used to take at least 6 months from application to award. Is Leufkin in the US?

His website is somewhat misleading...
"Our new design are is so unique I have obtained a total of 3 (three) patents pending for their design, application and uses in semiconductor and electronic cooling applications."

He has obtained patents pending??? No, a patent pending simply means you've submitted an application. I did a search at the USPTO website for applications since March 2001 and none were under "Leufken".

I'd be pretty surprised if any of his applications were eventually approved.

As far as prior applications that Colin stated, if you can prove that your design is a significant and is sufficiently original, then you might get approved. Again, I don't think he'll be successful.

Oh, and someone oughta send that article in to the USPTO with the reference to Leufken's applications.

[Diggrr]

I think direct cooling has one advantage, no interface surfaces for the temp transfer to slow down.
Now on to the one disadvantage I do see. The volume of Leufkin's block makes the water slow down (i.e. a drop in velocity). I am currently running Tim Whittaker's sewer cap design with a finned copper plate in place of the peltier unit and can get no better than 100 degrees. The rest of the system is in great shape...I just flushed it while assembling my new system. In my oppinion, you want the water to flow fast through the block, and slow through the radiator so the system is most efficient. This just can't happen when the block opens up inside like it does.
Don't get me wrong, I can't thank Tim enough for the quick, cheap solution he shared with the world, but it's not the optimum.
(Besides, I love people's expression when they find where it came from).

The rest of my system? Danner Supreme 250gph and a copper heater core from Ford (6x9x2) bigger than 'big momma'. And the whole thing is within a foot of height in my yy-cube.
I even put a biga** blower in the case to check airflow...no difference in temps...water stays at 79-81.

If only Leufkin's design were smaller volume...by cutting down on the height of that 'cavern' inside...hmmmm....maybe I'll just get my own pattent.:D

[ButcherUK]

There were some big threads on this over at [H], but they got deleted by the mods because most people told tom it wouldn't work... Anyway it probably wont work - peltier ceramic is likely to crack due to heating cooling in a wet evironment and if you get a leak pelts go BANG. Also it doesn't cool the edges of the pelt, could cause issues as ceramic doesn't exactlyspread heat well ;)

[Intraveinous]

Wows!!! So Mr Leufken basically looked at an overclocking and watercooling site that hasn't been updated for 5 years and goes, "HEY!!! I could probably file for a patent on that!!" RE:The link to an OC.Com tip, yep... that looks pretty similar... :D As far as the one without a pelt... I was ("OOOOHHH! You're putting water in your computer?!?!?!?!") watercooling Cyrix 6X86 chips like this when they were brand new... Similar way as that arctical, cut the top off of a sewer cap, siliconed it directly to the chip with 1/4inch OUTER diameter fish tank tubing providing water from a little 'Zen Rock Garden' fountain... Didn't even use barbs, drilled the holes in the plastic, fed the tubes in and siliconed around them... More fun than a barrel full of monkeys... My question is, if you have water flowing over the surface of your chip, what do you put on all the conductive contact points and bridges all over the top of AMD chips??? IT was one thing with Intel P133s that were ceramic topped, or the 6X86 and K6s that had a metal cap, but new chips don't have that kind of stuff all the time...
Oh well, my rant on marketing, patents, and old technology being renamed to seem like it's the newest greatest best thing...
Fav quote from the site: "Direct water cooling is not new- the Cray 9 used a group of 'sub computer' minicomputers submersed in a 3M product, FLUORINERT." If you replace the bit from the word Cray on with 'Hobbyists have been doing this since the days when a 200MHz processor was top of the line, I just decided to try and make some money off it" you'd be more correct... :D

Hooray for ranting...
John

[Diggrr]

They would both work, albeit with some prep work on the consumer's part. Preparing the chip with epoxy, sandwiching the tec with a cold plate against that o-ring.

http://www.spodesabode.com Great site!

He's done both, and one did go pop. The o-ring should fix this bacause of the lack of hard epoxy.

I'm just saying he could have chosen a more efficient block to consider waterflow.

One last poke...an aluminum block with my copper radiator...not gonna happen.:eh?:

[Silver]

Yes, I must concur that "Pat Pend" is not the same as "Pat". In this I am wrong. As of today Leufken is taking orders on this design. It does seem as though the design would be rather easy to replicate in copper should it be successful. I did read of another trying this and the pelt cracking. Seems the different contraction/expansion rates did'nt agree with each other. Should the gasket compensate for this would it not also allow for future leakage?

[RoadWarrior]

Originally posted by Colin
What was old (http://www.overclockers.com/tips111/) is new again.

Yeah, there ain't a lot new under the sun really, back when I started overclocking in the bad old 486 days, circa 94/95 I was at Uni with some comp sci majors, we were using heatsink compound, making blowholes, ducting with toilet roll middles jammed on our hsfs and talking about peltier devices and watercooling as possibles......

Has been quite amusing really to see all these things "discovered" since. :)

Not that the guys who did it don't deserve credit for the practical application and bringing these tricks and hacks to the attention of the wide world, it's just that they weren't the first.

regards,

Road Warrior