Silver heatsinks.

B

BrainEater

[H]ard|Gawd
Joined
Jul 21, 2004
Messages
1,234
Heyaz !!!

I'm putting together some pure silver gas/liquid heatsinks.thought I'd share teh pics so far.

silver1.jpg


silver2.jpg


cut :

silver3.jpg


shaping :

silver4.jpg


Lapping plate :

silver5.jpg


starting;

silver6.jpg


Finished to 220 grit ;

silver7.jpg


Final slug weight 4.57 Oz , .999 silver.



I'm ready for the next step of machining.........more soon !!

:D
 
Thx. :D

I thought a 100$ slug for a heatsink was semi extreme as well.

That's ok tho.....the three solid silver heatsinks I'm building , are going into the most extreme cooling setup you've ever seen.

:p
 
i cant wait to see how these perform, are you planning on just running fans, or are you gonna go extreme and make them into water blocks for a watercooling setup?
 
:D

Here's the plan (so far)

1. Very carefully drill out 64 small holes , 3-4 mm deep in an 8x8 grid.

2. Purchase 11 or 16 feet of 14 gauge .9999 pure silver wire.Cut said wire into 64 , 2 or 3 inch pieces (I havnt decided on length yet )

3. Wire bits into holes , solder em all in.

---------------

The heatsink will be held in place by a custom made , adjustable tension bracket.

I'm also considering having the wire media blasted to provide a 'bumpy' (golf ball like) surface profile.

-----------

as far as intended use : I'll be trying the first heatsink on this e6600.

It's also being designed with sub-cryogenic submersion in mind. ;)
 
thx ! :D

------------

I should probably preface this by saying ;

I'm not expecting super low temperatures while running on air.There just isn't enough surface area....

I run a thermalright xp-90c right now.All copper , surface area in the square foot range.....I doubt many heatsinks beat it.Mine won't that I know.

However , I need this heatsink to work from ambient in air to -100c submersed....Heatsinks with heatpipes won't.

----

Lastly , it's certainly not cost effective.While I did actually just have the 10 Oz ingot sitting here as a paperweight (old project,unfinished) , the cost on this sucker will fairly substantial......it's probably going to be about 250$ worth of silver alone , not counting the effort.

Ah well....It will look nice.

:D
 
wow!! that is freaking craaazy !! I like it!!! :p
bling bling to the pc!! yeah!!! :)

i hope your next project involves some Gold !!!! :eek: :p
 
hehe

:D

No gold.It doesn't have properties I'm looking for.....

Diamond , well mebbie.....I need to get rich before I'll be able to try out my diamond waterblock....heheh

-------

Quicky update.....

My silver wire is on order......Might be a while tho dammit......

:D
 
have you thought about gold or nickel plating it when you're done? The reason is that silver oxidizes pretty quickly and that will start to effect it's temperature transfering properties. While a plated layer is so thin that it won't effect heat transfer and keep the silver from oxidizing. Just a suggestion from a jeweler.
Also how are you planning on soldering the wires? If you are using plumbing solder you'll probably melt the wire long before the solder flows. fine silver is very soft and hard to work with, which is why jewelry is made with .925 sterling silver. You may want to look into silver solder, get very easy flow. This is solder designed for fine jewlery work made from silver. It has a very low (for metals) melting point. This is going to be a very hard job to solder. From personal experience I once made a container out of copper that was in effect a heat sink, bewteen uneven heating and the trouble of getting the solder to flow, it was a tough project.
I don't know your experience with silver or soldering but a good tip is to make a mini oven out of brick, with enough of a space to get the torch in there and move it around. That way the brick radiates the heat back into the heatsink and helps to get the solder flowing.
It may not be what you initially intended but if the soldering doesn't work out then maybe try some thermal epoxy, like the AS5 veriety. I've used it to attatch heatsinks with no other way to mount them and have had decent results.
Good luck and don't get frustrated, i want to see how this turns out.
 
Wow ! awesome post !!

You've answered the questions I had intended to ask my jeweler......I've never soldered silver in my life , and well , I know the thermal properties of silver quite well.

It seemed to me it would be a challenging soldering job.

If I could get just a teeny bit of threading into the block and wires , I'd use thermal epoxy.....but it's gotta have mechanical attachment other than the epoxy.....I'm 99% sure that will fracture @ -80c.That is after all the purpose for it.

Thanks for the input !

-----

ps.....it'll mostly be under liquid so plating is not required ;)
 
So you're submersing the board in oil, and running chilled water through the loop like I'm prototyping too eh?

Goddamn, I thought I would be the first on this board to do that XD. I was originally thinking of putting the computer in a welded vacuum, then filling that vacuum with helium (For it's awesome thermal capacity) but preliminary testing showed a small gas leak through SATA connectors, even when encased in epoxy!

Since we're working to a common goal, maybe we could share advice with one-another? This will be my third submersion test... and you're well, famous for it :p
 
ok i see more of your reasoning now. So a couple more tips on soldering. do look into silver solder. It comes in different flow levels. basicly different alloys that melt at different temps. Use the very easy flow rating. The way solder works is that it flows towards heat. this is the major hurdle. Because the wire will heat up quicker than the base the solder wants to flow up the wire and not make a solid connection. You may need to do several test pieces.
How I would go about this is to actually drill through the base instead of just tapping holes into it. noe set your base on a hard flat surface and slide thw wires in, making sure they get all the way to the bottom. Now using super glue, glue all the wires into place. don't worry through the heating and a solvent later you can remove it, this is to hold the wires in place while soldering. After it's dried flip it over, so it is wires down. Now solder from this side. Now when you're heating to solder concentrate on the top. Don't let the flame linger or you risk melting the wires or base, but you'll need to get the base and wires hot enough for the solder to flow, both need to be hot. Because the wires will heat up first the solder will flow down into the base and because you drilled all the way through now the base isn't quite such a mass and can heat up quicker.
the silver will oxidize from the torch and heat. The easiest thing I can say instead of trying to buy and mix what jewelers call pickle, take it to a jewlery shop and ask if the jeweler can put it into their pickle, let them know that you only used silver, including silver solder. explain your process. They will probably turn you down but it's worth a shot. The reason they may turn you down is because they don't know if they can trust you and if any foreign metals get into the pickle it can screw it up to the point where it actually can copper plate any metal if ferrous metals are in it. If you need to mix your own pickle you can use swimming pool chemicals used for lowering PH. you can get it everywhere, just make sure it contains cupric chloride. Mix it up and then heat it up, it doesn't need to boil just get hot. Unfortunatly this will turn your silver white, but silver in it's natural form is white. you can try to buff it, but anything you use will reduce it's heat transfer abilities. The reason bars and coins etc... is shiney is because they are all put into a device called a tumbler where steel shot is vibrated or rotated over the piece which shines it up. Chqances are you won't have access to something like this, but if you know a jeweler who does their own manufacturing they probably have one.
after the pickling you'll need to file down the base flat and sand it down. It sounds like a lot of work and it is, but the results are worth it. I just don't think the current way you are thinking about constructing this will work out how you want. If you have any questions at all about metaworking with solver or other non-ferrous metals fell free to ask.
 
Arcygenical said:
So you're submersing the board in oil, and running chilled water through the loop like I'm prototyping too eh?
Click to expand...

Nope.. :D



--------------------

Thanks for the input Bigbadgreen...
I'll run this by my jeweler , see what he thinks.
 
AWESOME xD

might be do you want sell left piece of silver, silver waterblok will be rocks :D
 
Yea sorry there's no more updates ....I'm doing my research very carefully before I go and blitz a bunch of silver into the bin...hehe.

-----

I'm afraid I can't sell the other half.....that 10 Oz ingot was less than 80 bux several years ago.Silver's now above 17$ an ounce....And I need the other piece for 2 more heatsinks assuming the first one works.

You can go get your own 180$ dollar ingot at any major bank :p

-----

I'll know more over the next week.

:D
 
:eek: 180$
bt maybe ONLY 180$ (dolar value is funny low nowadays)

also any updates :D:p
 
Yeah I had an Aqua Computer 478 water block that was pure silver. I sold it on eBay for like $17.
 
No progress on the heatsink yet.

It's part of a much larger project , and it's at the bottom of the list for that project.

:D
 
Arcygenical said:
So you're submersing the board in oil, and running chilled water through the loop like I'm prototyping too eh?

Goddamn, I thought I would be the first on this board to do that XD. I was originally thinking of putting the computer in a welded vacuum, then filling that vacuum with helium (For it's awesome thermal capacity) but preliminary testing showed a small gas leak through SATA connectors, even when encased in epoxy!

Since we're working to a common goal, maybe we could share advice with one-another? This will be my third submersion test... and you're well, famous for it :p
Click to expand...


YOINK
 
Ehh... Now you've got me even more interested. What, pray tell, is in this solution of yours?
 
Nope.

-----

I'm far enough along , I can prolly let the cat outta the bag.

I am rebuilding the Thinktank to run pumped anhydrous Isopropanol @ -60 to -100 C.

It'll be awhile tho....Between the $1.16 per square inch I paid for the new glass , and sourcing the 'cryogenic adhesive' , it's proving to be a rather expensive project....

You'll have to wait for pics.

:D
 
Kewl, were you able to get your Ag pin fins attached? If not, I've been thinking about that.
 
No actually , it's just the silver slug still.I'm glad too actually , as I am considering a different design.....A heatsink / fluid block hybrid.....kind of like attaching 'liquid in' lines onto an air heatsink.No out lines.......

We'll see what happens.

:D
 
Ah ha, I knew it was IPA. Sounds like it's going to cost you a fortune. I wonder how the components will cope with temps like that.
 
:D

Ok.

Here's a teaser pic.There's hurdles and marathons before she runs, but.....

tt2_48.jpg


This is the last post on this thread by me......

You'll need to wait for more.


:D
 
You must log in or register to reply here.
Back
Top