Little help for evaporator please...

[Unseen]

I have a friend that can build for my a evaporator in cnc. Which evaporator do u think is the best for a setup that can do -50c. The compressor that i have is 3/4hp. I am between the design of bakers and Dabits.

thanks

[bh2k]

dabit's one is the best.

[Unseen]

have u tried something like that?

[Twynn]

The design of Dabit's evap is quite o.k.. But on an overclocking-day here in the Netherlands it turned out to be quite impractical. The evap was too big to fit onto the A64 boards..

Maybe just make it, but a little smaller :)

[Unseen]

ok. we designed this... Please give me your advice.

[Unseen]

another one....

[Gary Lloyd]

The cross sectional area of all channels must be equal to or greater than the cross sectional area of the minimum suction line ID, or the flow will be restricted. The minimum suction line needed is 1/8 inch ID.

[chilly1]

Did you mean 3/8 inch gary....

Fugger and I are going to do a run of blocks in the near future...
The one we currently is an openchamber design, however I have a design the is also a maze, The trick with a maze is to get as much surface area with a large as possiable chamber for the surface to reject the heat into. Ideally you want the same amount of heat absorbed at all points in the block If you calculate the mass flow of the refrigerant needed to absorb at a pressure differance, subcolling and superheat, then you can derive the coefficient of expansion, This will give you the Cubic area you will need to have to provide that value of cooling Keep in mind the limiting factors involved. The delta caused by thermal conductivity of the metal you are using, the delat caused by the thermal barrier of the refrigerant to surface, the turbitidy of the refrigerant also plays a part in the effeciency of the system. So basically makr the interrior wall as thin as possiable to maximise the interrior cubic area put gouges on the channels insite the block to cause turbid flow as this will increase thermal contact with as many molecules as possiable, put as many fins as possiable inside the block, You will find that with your smallest cutting tool you will have to make multi[ple passes to get the depth you need and you will also have to exceed the limits of the cnc device (most only let you cut on a 1/2 ratio twice as deep as wide) You will also have to use carbide cutters to accomplish this and a high flow rate on the cooling fluid during the cut.

[Blergo]

chilly1 hit it right on the nail with that milling advise. dont ignore it, its very good advise. :) I spent 1 1/2 years working with CNC machines and normal machines (horizontal and vertical mills, slab mills, lathes, all sorts) when i was doing my NVQ4 in production engineering, was great fun. but copper is a right biatch to machine :)
owen

[Unseen]

hmmm...do u mean that is better to use dabits design than the above?

[chilly1]

Not nessassarily just keep in mind that you need space for the refrigerant to expand otherwie you will limit theh capacity of the block and also use care when milling.. I have broke quite a few drill bits trying to minimise the passes for a block. The only problem I see with your design is that the highest temperature (exiting) gas is on th eother side of the lowest temperature(liquid/gas) refrigerant. This will limit the effeciency of the system. The saving grace of your block is the extrusions that have simmilar temperatures on both sides and this may work better than having adjecent tubes in a simple back and forth pattern. I would run with it. What thickness are you going with? How will you get the top of the evap to seal the passes? What are the dimentions of the passages? PM me if this is information you will not speak of in a public forum.

[LardArse]

I dont think it is advisable to integrate the mounting holes directly into the block itself. Condensation proofing would be much tougher this way.

[embro]

:with:

And the larger it is, the more you need insulation, most of the setup using big blocks like this were using some kind of paste on the motherboard to limit condensation, most of the time this stuff can't be removed 100%, removing all chance of RMA and most of the time the result is not really esthetic.

Personally (using prescott) im making my block 1mm less than the actual size. Im also making a mold right now to surround the evap, it will be the holding plate for the evaporator at the same time. Easier to insulate and looks better(in my opinion).

good luck

[Unseen]

how about that?

[Unseen]

.

[Blergo]

WOW!! thats one hell of a design unseen :o how well does it perform? or habvent you tested it yet?
owen

[Unseen]

Another one.. I think that one is better.

Its only designs.... What do u think?

[Blergo]

they look like totally wicked designs, but what the inside looks like isnt important at the end of the day because after it is brased up it will never be seen again, get some made up and test them to see how well thet work :D
owen

[hatemi]

Are you planing on doing it with two layers? I mean there is two similar chambers on top of each other and separated with thin sheet of copper. The captube goes to the lower chamber and suction to the higher and the refriggerant goe through both...

[Unseen]

at the beggining i think one. What is your opinion?

[Unseen]

or that? plz help!!

[solo]

the one you posted before is better IMO

[Omega1]

They all look good but you can't tell by looks how they would perform,you would have to make and test all of them to see which has the best performance.

[TheDogFather]

If everyone sends me a evap, I'll be happy to test them.

TDF.

[Çhrist0ph]

what is an ideal size for an evap? 2"x 2"?
how tall should it be?

[Unseen]

ok take a look. My friend made this for me... what do u think?

[Unseen]

one more..

[Unseen]

and the last...

[Unseen]

.

[masterofpuppets]

Looks nice. Don't know much 'bout how good it will work though, since I'm not all that good at Phase-Changing.

[berkut]

should work quite ok

[BuLkHy]

for easy insulation....3 cm x 3 cm ..... height around 3cm as well.....cylinder evap shud work great...n insulation is a piece of cake by then....

[BuLkHy]

btw unseen...very nice design... u can make another layer ...2 layer evap..shud even work better since its not that tall.....

[boshi]

what about using designs very similar to good CPU waterblocks?

In fact, if that would work, what about trying to just use a waterblock like the maze4 ( brazed properly with the right size holes, etc. )?

That would take a lot of the work out of it, I think.

Anyone have any ideas on this?

[solo]

you cant treat freon like water, it just doesn't work

[Gary Lloyd]

Yes and no. Some of the water block designs would probably work well if used horizontally, gravity keeping the liquid on the base, and some of the designs could probably be adapted. The thing to keep in mind is that the liquid is evaporating and vapor takes up the vast majority of the space. We want to keep the base wet as much as possible especially in the center.

[Unseen]

The second evaporator will have two layers. Maybe i will use this in my vga.

[chilly1]

A single chamber with as tall a heat dissipator directly connected to the bottom and a chamber that is large enough so that you will have the best raito of volume to surface area so that you will absorb all the heat you can at the refrigerant flow rate nessassary and a gradually increasing maze design...

EDIT: putting a chamber on the top will not do very much for temperatures on the base of the block, the problem is conductivity if you look at the block as a whole the material removed from the bottom and the lack of integral contact with the top will lessen the amount of heat transferred to the refrigerant as it exits the evap.

[Unseen]

on sunday i will test it...

[RaVeN155]

HI!

What do you think about this?

http://zeus.polsl.gliwice.pl/~a10/fc%20bloczek.jpg

It propably be base of my friend evap but with 1,8mm pins and 1,2mm space between pins.

[Unseen]

Back again... look how my baby performs!

[HawainPanda]

lol, nice!

[fr33ze]

looks good unseen, let us know what temps your getting