What ya all been waiting for. The COOLEST Parhelia H²0 block

[spadnos]

There's a place about 40 miles from me that will deliver (for free or very cheap).

They have several Bridgeports and others (some of which used to be NC machines, so would be easier to retrofit with CNC) in the price range of $1000 - $3150 depending on features.

It's funny - I've driven through Brentwood a bunch of times, but never noticed it. That is, assuming that route 125 goes through it (just south of 101).

- Steve

[Dilitante1]

Yes it does, the store is on the East side of the highway, with a warehouse nextdoor full of industrial odds and ends and a ton of industrial shelving in all sizes.

[gt40]

something to keep in mind: if you are working in aluminum you may want to look for something with a high speed spindle (3000 rpm or higher), especially if doing fine detail with smaller cutters.
Also, There are plenty of table-top machines that will be more than adequite if you are only considering making smaller (less than 6" square) parts. These are much easier to retrofit and much easier on the pocket book.
Also keep in mind many of the larger bridgeport style machines require 220 or 550 volt 3 phase power - not something you will have available in a home.

[Wombat]

Re: Maybe soon (not SoonTM)

Originally posted by spadnos


Also, I've thought of things like using clear tops (Lexan or acrylic), and building in things like temperature sensors and LEDs.

The temp sensors might be nice, but with sensors already at the input/output of reservoirs, they won't give you much additional information. The Lexan/acrylic idea sounds like a bad engineering idea, unfortunately. First of all, you'd need to put a bit of work into getting a fit and seal, since you're looking at multi-part designs, and they'll expand/contract at different rates, too. Second, these polymers will be nearly useless in transmitting heat from the device to the water, so you're looking at much less efficient cooling.

[DukeP]

But they look Kewl!!!

Make an open window design, and place a small red (or fluorescent reactive) rotor in the center.
Presto - something to look at.
(Oh, and You can also tell if the water is flowing).



~~DukeP~~

[spadnos]

@Wombat: I would only use the acrylic on the top - the base plate would still be copper. Also, I would use a gasket between the parts.

Interestingly, a sensor near the CPU can be used in conjunction with sensors near the radiator to determine if the pump is working.

@DukeP: I just thought of that type of thing this morning - using some polarized thing like those "moving" pictures of waterfalls and the like.

- Steve

[Wombat]

spad, that doesn't matter. The metal conducts heat very well. So the base plate matters a little more than the top or sides, but not much.

The metal->water transfer is proportional to the surface area contact, among other things. By removing the top metal and putting acrylic there, you're essentially cutting the surface area down to 75% (assuming circular piping). No thanks.

And really, what are people going to see? The system should be airless, so they won't see any motion.

[spadnos]

Originally posted by Wombat
spad, that doesn't matter. The metal conducts heat very well. So the base plate matters a little more than the top or sides, but not much.

The metal->water transfer is proportional to the surface area contact, among other things. By removing the top metal and putting acrylic there, you're essentially cutting the surface area down to 75% (assuming circular piping). No thanks.

And really, what are people going to see? The system should be airless, so they won't see any motion.

Good points!

One of my plans is to make a unit that has fins inside. Think of a circular cone with grooves routed out from center to edge. (this would also be a splitter, like BladeRunner's Splitter-X) So, you have some fins that are thin and tall in the center, and short/thick at the edge. The intake would also be in the center, so the coolest fluid would strike the large surface area near the CPU core.

You're right - there would be no fluid motion to see. That's why the propeller / polarized disc thing would be involved. (seems silly, but people seem to like blinkenlights) There are also additives that glow in UV, so a UV LED in a clear top would be cool. Also, with a clear plastic top and an LED side-light, you can engrave logos or names in the top and have really custom systems with glowing graphics. (of course, you could also potentially make the waterblock with some custom shape, so the UV glow cooling liquid makes your logo - hmmm - maybe I should patent that )

That said, I do like the look of solid copper (especially polished to a high gloss). There are also aesthetic considerations for the fittings, and plenty of shapes to choose from to make things interesting.

- Steve

[Enak]

Why has no one made a solid Silver block.

That's one of the things on my list "to-do" once my CNC mill is running.

[spadnos]

Possibly because silver is about $8/oz, while copper is $1.60/pound.

The thermal characteristics of Silver are better, but the 80x cost is hard to justify for the 2% increase in thermal conductivity (and 4% heat capacity).

- Steve

[DukeP]

But ehm... What about the Keewlness factor??


~~DukeP~~

[Greebe]

coolness aside, stop and think a minute about the material you would be using... silver... which is a highly reactive metal (far more so than copper) you would have to seal it's surface and that would kill off any benefit gained.

[BladeRunner]

Silver is used in the Cascade SS, I have one here:-



The benefits over Copper are very small, 0.5°C. Actually measuring that type of performance gain accurately is a major obstacle, and only worth while once the block design is fully explored.

Silver would be a waste unless you had near enough squeezed everything performance wise from other sources. For instance you'd most likely get far greater than 0.5°C by placing the PC or the rad at floor level where the air is cooler.

Ok so it's nice to be able to say it's a real Silver block, but most casual glances will see it as aluminium...

Silver and copper are close enough on the galvanic scale so won't be reactive together, but aluminium should not be mixed with
silver or copper, even with corrosion inhibitors its not good practise. Copper is just more practical, and can be easily soldered together.

I'm also working on a parhelia cooling solution atm

[DukeP]

Ohh my goodness, thats a pretty piece of equipment...

[fetches the water and droolproof keyboard]

Im not sure I would even want to install that baby.
Just having it laying on my desk would be kewl enough.


~~DukeP~~

[Skee^]

Any Intrest in Buying this Piece of art? im in doubts of selling it.