Fuggers Freevap

[chilly1]

Want to build a simple evap? Dimentions as stated use a coarse thread tap for the inside surface..

THis is evap is licesensed to the members to use at will. No commercial use permitted. THX.


http://www.blairwing.com/images/threadvap.jpg

[wdrzal]

great idea to increase surface area,probable build that in 30 min. :clap:

[gclg2000]

kewl

[eshbach]

is that a top view or a front view?

could you possibly provide the front view if that is the top?

[wdrzal]

thats a top and side view,simply drill 1 hole complely thru then 4 holes to meet it,then tap them.I then would use 2 brass set screws,I don't think you will find copper ones unless you make them, to block off the one hole that was drilled complety thru, then braze over them.

[saratoga]

Should probably put a hole directly over the core to minimize the thermal resistance incurred by the copper.

[wdrzal]

Should probably put a hole directly over the core to minimize the thermal resistance incurred by the copper.

man you gota explain that better

[wdrzal]

for those of you that want to follow fuggers design exactly, drill 13/64 holes and get yourself a 1/4 by 20 nc(national coarse tap to end up with the .252 dimmension.

[Epsilon]

man you gota explain that better
He's saying that the heatload is applied to the center of the block. He suggests putting a hole right over it to decrease thermal resistance.

[wdrzal]

that drawing is not to scale if you drill 4 holes and evenly space them you will have a wall thickness seperating them of only about 3/32" 4x.252=1.08 " 1.50"-1.08"=.42 divide that by 5 and that your wall thickness, not much room left for more holes or to shift them.

[wdrzal]

.I then would use 2 brass set screws,I don't think you will find copper ones unless you make them,

you can rob 1/4" copper set screws out of electrical lugs. :woot:

[FUGGER]

The premise is so that it is easy to make in many different variants.

Cheap blocks can be done in minutes.

copper block stock, drill holes, 1/4" tap, braze, done. They are thin, easy to mount, and work xtremely well.

The threads increase surface area and the idea is that the movement of coolant along the threaded path will increase capacity over traditional channels.

[wdrzal]

While you could just braze up the holes,I figure running set screws in the ends(since they are all ready threaded) will keep guys just learning to braze from filling the channel with rod.

[Epsilon]

Screws are the most easy way to shut the holes. Iron / SS screws will work too. Use a high content silver rod with flux and it will be fine.

[wdrzal]

you will only be able to get a few threads so you don't block channel,more if you file a u grove in the end and position it correctly. also the single hole can be drilled only from one side and if you don't go thru you will only need to block 1 hole.have to check to see how long 1/4" taps are. brazing shut will be quickest way.

but like fugger said all kids of different variants can be thought of.

[Ad Rock]

Hopefully someone here will make one and post some pictures of is completed and mounted. This looks like it would make a good GPU block, but I would like to see it mounted on a CPU as well.

[smurfmods]

This could be really usefull to everyone if there where some pictures of a completed block and some mounting ideas. Anyone have any copper stock to do a semi howto?

From the picture it looks pretty simple to make. How would this compare to the performance of a chilly1 block?

[Malachia]

who will be the first to realize one???

i already have in mind some varinats :D

[Comp-Freak]

Where do you need to mount the suctionline and cap.tube? Is this right:

http://www.xtremesystems.org/forums/attachment.php?attachmentid=30700&stc=1

But how do you make that smaller channels by the cap.tube?

[Malachia]

Ok, now this is the Xtremesystems free evap, and we can post here all the mods??? i already have a few ideas...

[FUGGER]

Please post variants.

I will try to get a few images up soon.

[n00b 0f l337]

How bout this. Still the one U loop, but longer time in the block.

[Tonic]

Screws are the most easy way to shut the holes. Iron / SS screws will work too. Use a high content silver rod with flux and it will be fine.

My two cents, copper and SS have different heat conductivitys, at freezing
and heating to ambient temperature cycles it should make risk of leak :/

[Malachia]

here we go fugger

if someone can paint it better will be not a bad think...

[Sneil]

i think noob's is a little simpler as you dont have as many u tubs that'll be difficult to insulate. just make sure you dont drill all the way through at the top with his.
well this will make things nice and easy! thanks fugger :woot:

[n00b 0f l337]

When I need to drill to a length, I wrap the tape on the bit at exactly the end of the length I want. As soon as I hit tape, I stop. Also, can you use a hand drill for this? Or drill press only?

[gclg2000]

would be much easier with drill press and clamp it down obviouisly if you can.

[runmc]

I like noob's idea.:)

[Chris_F]

Nice evap design. Much easier than the maze design which requires much more drilling and I suspect that they would perform about the same too.

[n00b 0f l337]

Yeah! And no real brazing of the top on. I geuss I may do that with my system to test it out. But that means suction lines will be coming off up and down and same wiht capillary. I Wonder if I drilled down into those spots from above, and just clogged those extra 2 wholes that I could reroute the lines... I'll be giving it a shot.

[smurfmods]

I'm having some trouble trying to visualize what this will look like. Anyone have a 3d drawing or made one yet?? How would you mount it. :confused:

I made one up for the Copper cap type. Maybe a picture something like this?

[cadaveca]

Take noob's idea and 45 it on the block, cap's in the inside. Lots of hole plugging, unless CNC'd.

[Sneil]

here's a crappy 3d image i made of noob's mod

[n00b 0f l337]

Yup, perfect sneil. Not crappy at all.

[smurfmods]

Good Golly, What an Excellent Job. Exactly what I needed. Now I get it. That's a nice block. Super easy!


*edited for swear words and PCness* :dammit:

[n00b 0f l337]

Hehe, my first good contribution to the forum I geuss?

[Sneil]

hey, now now fugger gets most of the credit ;) :p:
but indeed still a good contribution on your part :toast: :up: more than i can say for myself :p:

[n00b 0f l337]

No, you built the 3d model!

EDIT: Here's another block without a U.

[Sneil]

sweet :thumbsup:

[n00b 0f l337]

I like this idea. How do you clog the holes though? Here's another more complex double layer design with suction and the capillary on one side.

Hope ya like it.

Anyone wanna make me one for free? :stick:
:p:

[BugsBMD]

Just a little different then noobs. This this make it easyer to insulate with the cap and suction on the top sence on some mother boards there is not much space on the sides of the cpu.

you fill the holes with set screws and then braze them in.

[n00b 0f l337]

Yeah good point. I geuss with mine, if you had the bottom plate larger, tthen you could have the suction and evap stick straight up.

[meke]

@n00b 0f l337,

your idea is fine, actually I did a similar design but I can only show the outer picture. I design this for sli inner gpu (the first vga close to cpu)

it's a 2.5x5x5 cm copper block and drilled like your design. (I did this 1 month ago)

http://img.photobucket.com/albums/v217/meke_oocay/223K/evap.jpg

[Ad Rock]

Edit: Its late(early) and I didnt read meke's post clearly enough ( I asked for more pics).

[FUGGER]

Meke, thats it.

You can also get these pieces of copper stock from dangerden.com

http://www.dangerdenstore.com/product.php?productid=128&cat=49&page=1

Bummer there is no picture available

[n00b 0f l337]

Would you recommend the 3/4" thick? I'm thinking of ordering one 2" x 4" x 3/4" and one 2" x 4" x 1/2" so I can make some test blocks.

[Ssilencer]

Sorry to bother, but the Fuggers freevap is a Ketchak Nº4

http://www.devilmaster.org/modules/coppermine/albums/Ketchak/4/normal_alone.jpg

[ellsworth]

Sorry to bother, but the Fuggers freevap is a Ketchak Nº4

http://www.devilmaster.org/modules/coppermine/albums/Ketchak/4/normal_alone.jpg

this design is one of the original waterblock designs people made back in the day of old...

it works and it's easy to make with common tools, i think this is why fugger was sharing it with everyone.

[wdrzal]

story of a hvac guys life,a dollar short and a day late.

[chilly1]

well yes and no
yes simmilar and no barbs

[smurfmods]

Didn't see it answered yet... After you drill the holes, what do you use to block them off? Copper screws aren't very common and you don't want to use a different metal because it will rust between them. Solder might be possible but you risk clogging the system if it runs down the channel.

Maybe a small flat piece of copper soldered over the hole? That would work right?

[Marvin]

I think brass screws and good brazing rods...

[meke]

Yes that's what I did.

thin copper brazing over the full side.

[Ssilencer]

well yes and no
yes simmilar and no barbs

hehehe, have no barbs cause is an evap
And yeap, it works


this design is one of the original waterblock designs people made back in the day of old...

yes, in the day of old, Ketchak did the Nº4 :rolleyes:

[smurfmods]

can you get brass screws easily at the local hardware store?

What about insulation? It's flat on top so you can just put a flat peice of foam on that right? But what about the other sides? I think you can cut the holes out of the foam for the cap and suction then slide it on the side. Can you use RTV? Have to check the specs on RTV and hope it stays "glued" below -45.

Maybe a custom box made of wood or sheet metal. Wood may be out because of the thickness of the foam+wood may be too large. Trying to come up with some designs. Hope to post something in a day or two.

What's your ideas on the best way to insulate this? :idea:

[chilly1]

Ues a piece of round foam insulation and streach it around it.

[smurfmods]

There are a ton of different types of RTV. A lot of them go down below -50... Anyone ever use it to glue foam?

Thanks Chilly1, didn't think of that. Do you think we need a rigid enclosure around the foam?

[n00b 0f l337]

Heh, where can you get those giant OD foam tubes?

*Will edit my post in a minute with a new freevap design I thought up*

****EDIT: Here's the pic. Basically carve in from each side with a good sized bit, drill in thru top for suction and cap lines, then plug the sides. Finallize by taking quite a thin bit, and drilling in from the side connecting the two original wholes, this gives tonnnnsss of surface area, while cooling all over the block. Then i geuss add braze into the small wholes from the thin cuts... A good bit of work, but I'm hoping its rewarding.

[chilly1]

Great Idea! more holes only smaller..
I am going to make 2 of the best design that comes up on this page. I will do one for me for testing and another rfor teh originator of tee idea.

[n00b 0f l337]

SCORE! Yay, considering I've got most of the ideas and I need an evap. When will you make it?


EDIT: Here's another idea I was thinking of, just a back and forth in a circle. Dunno how great as in that time evaporation may be more on one side than another.

Also for my design with tons of small drill holes, you could also do 2 or 3 layers of those holes, not just one layer. That would be pretty sweet.

[ellsworth]

this is a design i built a long time ago but never got to test. it's about as cheap and easy as you can get and from my spectulation is a very good design.

it consists of 3-4 different sized pipe segments with slits cut out of the side (i used a dremel...very quick and easy). the segments are placed inside one another with the slits opposing it's corresponding inner pipe segment. this allows the refrigerant to be trapped for a long time due to the "maze" design. this design should be able to be placed at any orientation (flat on a table, such as a test bench, or on its side mounted to a cpu in an upright case.

this can be made with a dremel and a torch....THATS IT! no drill press and tapping tools needed. i think it would be even easier to build. :D

EDIT: UPDATED DRAWING

[n00b 0f l337]

Thats a classic!

[Sneil]

i like it, very easy and simple. a little tall for side mounting.. maybe they could be stacked 2 or 3 levels @ 1/4" or 1/2" thick per level
edit: this is still ellsworth's design

[wdrzal]

SCORE! Yay, considering I've got most of the ideas and I need an evap. When will you make it?


EDIT: Here's another idea I was thinking of, just a back and forth in a circle. Dunno how great as in that time evaporation may be more on one side than another.

Also for my design with tons of small drill holes, you could also do 2 or 3 layers of those holes, not just one layer. That would be pretty sweet.


small holes? refrigerant expands many multiples in volume when it "boils off".

[n00b 0f l337]

Eh, good point, still worth a try to see if it works well. It could start to evap in first section, then suction will suck it cold thru the wholes for surface area cooling the block plus the other area... Worth a shot.

[Chris_F]

Here's my design. All that is needed is copper caps, a small piece of tube, and to cut a round base plate. You just drill a hole into the sides of the caps to allow the refrigerant to flow from one chamber to the other.

http://www.xtremesystems.org/forums/attachment.php?attachmentid=31046&stc=1http://www.xtremesystems.org/forums/attachment.php?attachmentid=31047&stc=1http://www.xtremesystems.org/forums/attachment.php?attachmentid=31048&stc=1
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31049&stc=1http://www.xtremesystems.org/forums/attachment.php?attachmentid=31050&stc=1

[n00b 0f l337]

Damn, but thats not really a new design, thats quite used a lot.

[Sneil]

by whom? i like it, easy & cheap

[n00b 0f l337]

Check out the evap gallery.

[ellsworth]

with chris f's design, you need to make sure the holes you drill are the diameter of the suction outlet or bigger (or multiple holes) so that you can pull vacuum equally. you still want to trap the refrigerant, yet avoid causing any pressurized spots.

i think a big topic people overlook in evap design is how much surface area of refrigerant is being exposed to the full vacuum levels and where this happens. in my design, if you placed just a cap on the top and left all of the channels exposed, the entire surface area or pool of refrigerant is exposed to vacuum and allowed to evaporate at a more ideal temperature throughout the entire base (the heatsource) since all of the refrigerant pool is exposed to the same low vacuum levels.

for example, in chris F's design, the refrigerant in the middle cap would be under higher pressure than the outside cap simply because there is restriction on the exit and refrigerant is being forced in the inlet. this isn't ideal since the middle of the base is where a majority of the heatload lies. the outside layers are going to transfer heat more efficiently than the inside.

evaporators aren't all about fancy maze patterns. an ideal evaporator allows the refrigerant to transfer heat where needed the most as efficiently as possible. to do this, we need to trap the refrigerant at the spot we want (the center of the base) and expose it to the best possible vacuum levels to ensure the best and most efficient transfer of energy.

[Chris_F]

Yeah, not completely original, but after much thought I came to the conclusion that this would be an easy design for me to use.

I was planing on having a large hole for the refrigerant to exit, would that make a difference?

[n00b 0f l337]

THats what the original thoguht behind mine was with the thin tubes, to allow the main center to get most of it.

EDIT: Here's another design taking that into consideration. I also thought we were desiging easy to build evaps, like ones without a ton of brazing. :( But if we are off that then heres a new design.

[Sneil]

here's a revamp of previous. holes will be slightly larger. the two cap lines that split off would be at equal lengths.

[n00b 0f l337]

Yeah, was thinking of that a second ago. I dont think double cap line will be agood idea though. I can just think of problems with splitting it. Otherwise that may or may not work. Heh, just hope chilly1 or fugger likes the designs, as I really gotta get an evap block. :stick:

[Sneil]

the only problem you'd have is in blocking the tubes from brazing,which can be avoided easily enough. its not that diffucult to do at all and is actually quite efficient, eh chilly1? :stick: :D

I think we all could use a free evap :p:

[n00b 0f l337]

Yeah, stick a dowel or somethign in the main whole, then braze, then remove.

[Chris_F]

As a side not one of the main reasons I chose the design I did is because it's much easier to insulate than a square design.

[n00b 0f l337]

How is that exactly? Just that you can cut out that big foam tubes and shove it inside? With a square you can just get a ton of neoprene, cut out squares, and build it up around it...

Eh, yeah, probably easier, but its been done a lot.

[Sneil]

As a side not one of the main reasons I chose the design I did is because it's much easier to insulate than a square design.

not really, its just like saying its easier to drive across the street then it is to walk. small difference for a small job. just use some flat neoprene for a block. :slap:
;)
so is there a deadline for the evap "contest" chilly1?

[n00b 0f l337]

Heh we all just need to win an evap :)
I'm gonna work on these all tommorow...
So lots of surface area, easy to make, perferably no brazing of a top segment on, just like side wholes, little captures.... This is somewhat of a challenge but I got an idea or two.

[ellsworth]

brazing the cap tube to/in the evaporator deminishes or maybe even eliminates its reusability. what if you want to make a new system that has no cap tube or needs an entire new cap tube? another consideration.

[Sneil]

Heh we all just need to win an evap :)
I'm gonna work on these all tommorow...
So lots of surface area, easy to make, perferably no brazing of a top segment on, just like side wholes, little captures.... This is somewhat of a challenge but I got an idea or two.

no use even trying, Sneilvap takes all :D
j/k good luck :toast:

[Sneil]

brazing the cap tube to/in the evaporator deminishes or maybe even eliminates its reusability. what if you want to make a new system that has no cap tube or needs an entire new cap tube? another consideration.
you just braze up the cap tube, or braze a piece of 1/4" to the block and braze the cap tube to that.(pic) if you need to take the cap tube out just clip off some of the 1/4" and the cap slides right out of the evep.
if your changing the whole cap line just slide a new one back in to the 1/4" pipe, pinch around the new cap tube and braze. side cutters work great for crimping around cap tube

[Revv23]

splitting the cap tube wouldnt work.

i mean, maybe you could halve the diameter of the the tube at the split, but good luck brazing that....

i think noob's first design or fuggers are the two best.

Remember that we arent designing waterbloacks but evaps, evaps you want the refridgerant to stay in the block for as long as possible but also you you dont want it pooling up anywhere, this is why corners ( such as in a maze) and small channels are bad. if you use small channels such as in noob's second design your taking away all the room for refridgerant to evaporate.

Of course this is all theory, i could be completly wrong, but its a pretty educated guees, just look at why chilly1's spirals work so well, no corners, big fat channels.

[Sneil]

this is how you do it... a close up view of capillary connection. pressure won't be lost between the cap lines. again its not that difficult to do this. you have to be just as careful when you braze the cap line to the filter/drier. if you can do that you can do this

[Chris_F]

Why not just have two cap tubes come off the filter/drier?

[Sneil]

Why not just have two cap tubes come off the filter/drier?
you could but there's more room for unequal ammounts of refrigerant discharging between the 2 with longer lengths. closer to the evap is probably better. just do what this guy did, not hard :cool:
edit:pic from fatty's system

[smurfmods]

Some of these designs would require a drill press to do right. The many small hole design is one of the best so far. Any way to adapt this to make it with channels? The refridgerant may not cool the whole block the same. It will follow the path of least resistance right?

Might be tough to find small bits long enough and I don't have a drill press. sniff...

[Sneil]

for mine a drill press would probably be nessarythats

The refridgerant may not cool the whole block the same. It will follow the path of least resistance right?
the reason for two outer cap lines and suction at the bottom middle. hope chilly1 likes the design :kissbutt:

[matttheniceguy]

i think a big topic people overlook in evap design is how much surface area of refrigerant is being exposed to the full vacuum levels and where this happens.

This was the main thing I was thinking about when I was designing my evap. It is designed to sit upright (like the mobo is in a case), so there is a cup directly over the core of the cpu to hold refrigerant. The cap tube will be positioned to spray directly on the base of the evap touching the core. There are a couple other things I did which are a bit different than most maze designes. Where the suction line connects there is a 1/4" hole in the block, so you can pass a piece of pipe through the top plate and part way down into the block. This holds everything together for brazing, and reduces the amount of refrigerant that will sneak between the top plate and block, above the maze pattern. Also, the cap tube can be fed down into the block through the suction, and then fed through a small hole in the single copper wall seperating the suction line from the cup. This makes it easy to braze, and makes it nearly impossable to melt or braze shut the cap tube.

[Sneil]

cleaned up a little..
holes can be drilled larger with smaller walls in between if need be. threaded as well

[ellsworth]

cleaned up a little..
holes can be drilled larger with smaller walls in between if need be. threaded as well


with that design, the majority of the refrigerant is going to pass on the outside channels since it will be following the path of least resistance. the refrigerant will be taking the shortest path to the suction line. this is also a very straight through design. the liquid can be sucked up directly by the suction line instead of being forced to boil off over the heat source. this design would make a great waterblock however (if the inlet were opposite of where the suction line is).

just remember though, waterblock design is not evaporator design. other charactaristics have to be noted, such as vacuum distribution and we actually want to trap the refrigerant in the block and force it to boil off rather than having a straight through, high flow waterblock.

with evaporators, the cooling happens when refrigerant boils. how much cooling is a factor of how much refrigerant is boiling at what pressure. we want all of our refrigerant to boil in the evaporator instead of the suction line or even in the compressor.

[wdrzal]

maximum "wetted surface area" ,exposed to "lowest pressure" possible =good evaporator
mass flow(velocity) must also be taken into account,refrigerant moving too slowly can cause oil pooling. A good design also keeps the refrigerant in the evaporator,and not boiling off in the suction line.


edit:ellsworth we had same ideas, i didn't read your post above before i started to type mine.

[ellsworth]

my final submission: :toast:


http://www.xtremesystems.org/forums/attachment.php?attachmentid=31086&stc=1

[wdrzal]

what if mounted sideways? not so good. this is position dependant,good design would not be position dependant.

[STEvil]

Red spots are where I assume would be the best locations for refrigerant inlet/outlet.. you could put the suction on the side and run the cap tube across the top down into the center to fit low profile applications like video cards.

[n00b 0f l337]

Thats pretty kick ass stevil, I'll be posting 3 more designs later today.

[wdrzal]

keep in mind tooling and manufacture requirements for a given design.While you may make a kickass evap on paper,if building it is either too complex or costly it won't be practical.

[Sneil]

ellsworth is simple enough if you make it smaller and definatly not open at the top. His eather stacked or single would work, but definatly not so tall as would cause major pooling if mounted on its side and certainly an open top wouldn't work. if your talking about pressure building up at the center you just need to make the first "door" larger. As you can see this definatly covers the largest surface area and is very effecient when compaired to your standard maze due to the small/thin walls of copper tubing. Also refrigerant is first being applied directly to center on top of load. I don't know if I can count this as mine or not. If we alter someone elses drastically enough can we claim as our own? seems to be whats been happening. If so is this altered enough? :confused:

[smurfmods]

What's the benifit to stacking designs? Seems like too much of the refridgerant wouldn't be where it should be.... directly cooling the base. A maze would be more effective.

I like your completely original design Sneil. :D What do you use to make it with? Round copper tubing, caps, or milled out of a block?

For fun I'm working on a stacked design right now. I think you guys will get a real kick out of it.

:peace:

[n00b 0f l337]

Well here's another, though I drew up around 15 more on post-it notes and will probably take a pic of the designs tonight. I think this one would allow enough refrigerant expansion while limiting it from not blowing too much thru at any second. Notice the widening of the expansion area towards the center.

[Comp-Freak]

No offence, but wasn't the original idea to design a evap that is easy to build? (for starting phch builders?)
This one doesn't seem easy to me, or am I wrong?

[Ad Rock]

No offence, but wasn't the original idea to design a evap that is easy to build? (for starting phch builders?)
This one doesn't seem easy to me, or am I wrong?

Ya I think you guys are losing the focus of this whole thread, which is to come up with a easy to make yet effective evap design.

Noob how would you make that last evap you posted? All those Tri-angles look complicated and I dont know how you would even go about making it.

[n00b 0f l337]

Drill press, then braze on the top part. Yeah, this has sorta lost off topic, but I said already that if we were getting off topic here was my contribution. :(

I'll redesign some new easy ones, I liked my tons of tiny tube design, but it seems to have been taken away.

[smurfmods]

Here is my pyramid block!

It is actually very easy to build, mount and insulate. Just use the basic maze design and stack it.

[n00b 0f l337]

LOL, thats still quite a bit of brazing. Also how would you insulate evenly?

[FUGGER]

Ad Rock,

I knew that was supergreg. (http://www.vanderzande.com/nerdcult/media/Super%20Greg%20(mirror%20site%20of%20the%20origina l%20supergreg_com).htm)

ROFL

[n00b 0f l337]

LoL, freakish...
What happens if I post my next design and someone copys with just a little improvement...? :)

EDIT: Well its a revision off my very first small line block, with more cooling to the core. And you can just spin it and stuff. <------- :slap:

********
SNEIL LOOK HERE
********

[smurfmods]

LoL, freakish...
What happens if I post my next design and someone copys with just a little improvement...? :)


It was posted with a wink and a smile really. ;)

It's certainly possible to do, but why would you want to? I knew it was a bit off topic and in the wrong direction. I just wanted to show the other extreme... We really should stay away from multi-layered setups and work towards simple blocks.

I posed the question "why are multilayered any better than a simple block?" Anyone? I see complicated designs in a lot of forums. Is there a need for it?

[n00b 0f l337]

No I dont think so, complicated is for the complex.

[wdrzal]

LoL, freakish...
What happens if I post my next design and someone copys with just a little improvement...? :)

EDIT: Well its a revision off my very first small line block, with more cooling to the core.

a molecular train wreck.hmmm ,but would the turbulance stall the refrigerant over the core?


edit: why didn't thumbnail attach?

[n00b 0f l337]

I wasnt really thinking that. I was thinking more of one side goes more towards bottom of the block, and the other is above the other but not touching. But now that you think bout it, I wonder if that would be a good idea to just do two layers of X crossings.

[Sneil]

Ya I think you guys are losing the focus of this whole thread, which is to come up with a easy to make yet effective evap design.
you say you guys, i think you meant to say noob lol. that triangle one looks nex to impossible to build. the last i posted is very cost effective, effective, and easy to make. you dont need to stack it, one layer is fine.
don't know if i can call it mine tho :p: i think its drastically changed enough from ellsworth's to call it mine. his is a wreck waiting to happen once mounted in its side. ill let chilly decide

I like your completely original design Sneil. :D What do you use to make it with? Round copper tubing, caps, or milled out of a block?
round copper tubing

[n00b 0f l337]

when you say you guys i you just needed to say noob lol.

"you say you guys I you"

???

[Sneil]

???

[n00b 0f l337]

???

Well now you edited, before it said that other thing that made no sense...
W/E, back to evap designs...

[Sneil]

anyways, i think ellsworth and myself should both get credit for this one. ellsworth's design would be chucked if it wasnt for my revamp. 1 for me one for ellsworth? eh chilly??? :D :D
its cheap enough to make! :kissbutt: :kissbutt: :kissbutt:

[n00b 0f l337]

Three times the ass kissing for that one Sneil. I personally would just like to see if the "molecular trainreck" would work... I think its got some good potential for one of chillys cadcades if the tubes are enlarged a little bit more than in my design.

[Sneil]

Three times the ass kissing for that one Sneil.
he likes it


I personally would just like to see if the "molecular trainreck" would work... I think its got some good potential for one of chillys cadcades if the tubes are enlarged a little bit more than in my design.
you palce that sucker on its side and youve got problems

[n00b 0f l337]

How would I have problems on its side? Capillary on bottom, suction on top?

[Sneil]

sorry if it seems like im trying to destroy every evap you make, just making points for consideration. :toast:

now the reason this design should be destroyed.... hehej/k

[n00b 0f l337]

Eh, I think with the lines that small, the suction will be quite great thru the small wholes. No matter how ya flip it its gonna get everywhere. Remember, the ENTIRE system is full of freon, theres just no "empty" spots.

[Sneil]

the entire system is not full of freon, you'd be flooding your evaporator and nothing accomplish nothing. you need room for expanding gas. pressure could build in the top right courner of the block.
oh well, ill let someone else give their opinon, ive yapped enough

[n00b 0f l337]

Wel as I was trying to keep mine simple. Like no brazing a top on.
How bout this then?
:para:


EDIT:: Also in my old image, if you stopped the second layer short of the end and went upwards for suction, and then went upwards for cap on bottom layer, do you find any problems with it? No pressure or freon pooling? :p:
http://www.xtremesystems.org/forums/attachment.php?attachmentid=30787

[Sneil]

simple does not mean it'll work well. sorry not meaning to discourage you
someone else will give their opinion

[wdrzal]

refrigerant acts like when water is spilled on a red hot surface, it rolls along as it sizzles(boils off), as balls of refrigerant roll along they collide in the center, will this 1: cause it to join into a larger ball ?or 2: split in to numerious smaller ones?and instantly boil off. You may have stumbled onto something, can the velocity of the refrigerant be used to help vaporize it?

[n00b 0f l337]

Hehe, I hope it makes chilly1 think about it enough to make himself and me one. ;) I'm hopin for the numerous smaller pelets. But wdrzal do you agree with Sneil about pressure on top, and only cooling bottom half with my design? Cause if the refrigerant just keeps coming into the block as it does, then it would really halft to push thru all areas... Maybe lengthening that core are towards the suction a bit more would be a good idea? I geuss my stumblings okay... I went outside and played with my water cannons on this idea. Shooting a "ball" of water at another just makes a big splash... I love potato cannons.

[wdrzal]

I was just focusing on the collision aspect, try,four small cap tubes,one to each cornner and suction in the center, a x- crossing wreck , refrigerant meets in the center. just therorizing as I type. head on wreck LOL multi position. lol

[n00b 0f l337]

Well I dont think a 4 way cap is for any first evap... And that would just get sucked up, no collision...

[matttheniceguy]

This freon talk is driving me mad!! There are almost no systems around that still use old school "freon". There are a lot of different refrigerants in use in peoples phase systems, so just the term "refrigerant" would be a lot more accurate.

nOOb, in an evaporator there is a mix of liquid and gas. The refrigerant enters as a liquid and then boils into a gas. In most cases the boiling convection heat transfer coefficent is so high that fins will actually insulate the base. This is why you generally want to have a fairly large flat surface over the core for the refrigerant to boil on. This surface should be somewhat contained however, to hold a bit of a liquid pool over the core where it will continue to boil. In your designes with two large intake and suction holes, and then small holes connecting them, I believe a lot of the refrigerant would boil when it enters the first large intake hole. This would lead to trapped gas within the evap and inconsistant refrigerant flow over the core. A design with the cap tube directly over the core would be more ideal, as this is where you want the boiling to occur.

Here (http://jokes.cypressmedia.com/supergreg.htm) is supergreg in action if anyone is interested. That guy is soo sweet I want to crap my pants.

[wdrzal]

"Freon" is the trade name for "refrigerants" not any particular type, so Freon is still widely used, guess you will have to go insane!!!! :( my condolences. walt

[n00b 0f l337]

Okay, good explanation about the trapping of gas... So now I'm workin on a new design for this sorta stuff. Basically really only things that start at center and go outward work well... Damn.

[matttheniceguy]

"Freon" is the trade name for "refrigerants" not any particular type, so Freon is still widely used, guess you will have to go insane!!!! :( my condolences. walt

Hmm.. didn't know that... No big deal, I'm 95% there already. :wth:

[n00b 0f l337]

Following those guidlines I've devised a new evap. Problem is I can no longer design em in those guidelines without having a ton peice of sorts being brazed on... So I shall now show you....

Freevap 7. :rolleyes:
*at this rate I can be like a final fantasy series*

[wdrzal]

train wrecks to race tracks ,that is the pit exit road in the upper left.if you really want to win, I think you should get back to something simple.

[pythagoras]

"Freon" is the trade name for "refrigerants" not any particular type, so Freon is still widely used, guess you will have to go insane!!!! :( my condolences. walt

Sorry to be pedantic but only refrigerants manufactured by Dupont can be called Freon :D

Regards

John.

[n00b 0f l337]

Well I've left a few "simple" ones in the early parts of the thread. :) I'm building to more complex ones... Im just trying uber hard to win an evap. Cuase I dont have any of the tools, (drills or presses) to even cut my own.

[wdrzal]

while dupont does own the copyright,freon is the the "trade name" for all refrigerants., thats how trade names come about , like klixon is a company that makes thermal overloads,not the name of the part. but they are called "klixon's", many times company names are adopted for products.

[n00b 0f l337]

Like rollerblades for example... Or bubble gum...

[Sneil]

Hey here's a brilliant idea for your evap noob :slap: *to both our heads* it'll allow it to function the way you were hoping anyway. weather its still a good idea or not... ? you better thank me for this :kissbutt:

[Sneil]

and if you want some velocity like wdrzal was saying, have smaller diameter "shoots" for the two upper angled triplet channels(the ones heading towards the center downwards) to increase velocity for collision, and larger ones for the two botton triple shoots for suction and evaporation. will help the collision theory anyway

[Ad Rock]

Ad Rock,

I knew that was supergreg. (http://www.vanderzande.com/nerdcult/media/Super%20Greg%20(mirror%20site%20of%20the%20origina l%20supergreg_com).htm)

ROFL

Yep the one and only supergreg, Fugger.

Some great SuperGreg quotes:

" Hanging out with a fan: "I think her name was Mariha. I asked her about the 'h'-thing, but I really never got an acceptable answer on that one." "

" Wrapping up after a gig: "Roccy looks a little annoyed in this picture, but he isn't. Rashy is chatting with a lawyer or something." "

And finally:

" "Sometimes as a DJ U have these 'golden moments' when everything seems to flow. You never know when they are going to happen. It's funny, on this particular occasion I wasn't thinking about DJing at all, but about a sweet girl I once met at a bus station." "

Sorry for the OT but that site is just too funny.

[Sneil]

guess i had to be there? :yawn2:

[Ssilencer]

I think it would be great that someone with knowledge tell us how the refrigerant behaves in an ideal evap.

[matttheniceguy]

It's a horrably complicated subject involving conduction and convection heat transfer, and insanly complex fluid dynamics. Just think of all the work going into waterblock designes like the G4 or G5, and then imagine the water is boiling like mad. I'd say it's one of those things where people have learned mainly by experience, and trial and error, rather than true understanding. I know I sure don't have any.

There are FEA programs which are very good at solid body modeling (ANSYS, cosmos etc), and other programs that are very good at fluid modeling (fluent etc.), but I don't know of any which are good enough at both to tackle something like an evaporator. Anyone got mad skills with any of these programs and feel like a challange? I can do the solid body stuff in ANSYS, but I would need to know the fluid temperature and convection heat transfer coefficient at many different locations in the evap. Unfortunatly I don't have a clue how to find these, and I think they depend a lot on the solid body of the evaporator so there is no way to solve for them and then put those values on the evap.

[STEvil]

If the liquid metal that is being used in the new liquid metal heatsinks (claims to have extremely high thermal conductivity values) were able to maintain its properties even when frozen solid then it would be pretty easy to basically design a massive evaporator with that liquid (or more likely frozen now) metal sealed in a chamber of the copper or aluminum evap as the main heat transfer medium....

Or even to cut two seperate channels in an evap, one sealed to contain the "liquid metal" and the other would be for refrigerant flow.. it'd still be somewhat larger than current evaps to allow a lot of flow, but its gotta be a bit practical ;)

[pythagoras]

I contacted nanocool a couple of months ago and the metals only good down to -10C

Regards

John.

[Special_K]

hello - great thread =]

K R complete noob at any kinda cooling (even air), and has neither the materials, time, or willpower to even attempt at making one of these, but would just like to understand better, and perhaps help others understand, so go easy please =]

also, please understand, K R not criticising ANYBODY - rather, K R questioning EVERYTHING that doesn't make sense (yet!) to try and understand more

a few questions/rambles based on 'K theory'(murphy's law) rather than 'K experience'(oxymoron):

1. from limited understanding of the system (feel free to correct any errors), the refrigerant needs to totally 'boil off' into vapour inside the evaporator before returning to the compressor, otherwise damage to your compressor will occur. This leads to the question - does the original 'fugger freevap' have enough capacity for that expansion?

2. having asked if there is enough room for expansion for the process of evaporation, the next question would be - is there enough surface area inside the block for adequate heat transferal?

ok - so that's the original submitted design interrogation for the time being =p

now to some of the other submissions =]


3. somebody suggested it would be better to have the evaporating gas flow over the hotspot of the block, and it started K thunking "is it better to have the hotspot covered by gas or copper?" - would it not be better to have a solid chunk of copper directly above the hotspot with lots of fins (larger surface area) and lots of cooling vapour around it for the heat transferal process?

4. noobof1337's first submission showed a longer evaporation route, but it has a narrower channel at top-left than the rest of the route - why is that? noobof1337's U-Bendless design looks very good =]

5. the oil used to lubricate the compressor is going to mix with the liquid refrigerant and taken around the loop only to to be deposited and allowed to pool in the evaporators lowest point when the refrigerant evaporates. Now several people submitted 'multi-level' designs, which looked great on paper, but didn't take this unfortunate side-effect into account because their suction line was at the top of their block. Would this eventually lead to a blockage and perhaps catastrophic malfunction of the system if left un-checked or is the 'suction' great enough to overcome gravity in most cases?

6. could noobof1337's double layer design be fixed by switching the capillary to the top level and the suction to the bottom layer?

7. do you run these coolers 24/7/365 or are they run for an hour at a time for benching and then stripped and serviced?

8. ellsworth design looks very ellaborate but isn't he forgetting the prinicple of "path of least resistance"? instead of flowing through the designed maze, it would be immediately sucked straight upwards once leaving the capillary shirley?

9. would ellsworth's (and the other pyramid) design also fall foul of the pooling of compressor oil over time?

10. whatever design you use, shouldn't the evaporator bore (ideally) be ever increasing, or at least constant size, from leaving capillary to eventual out line? this is essentially how the evaporation process is induced (high pressure to sudden low pressure equals boiling and latent heat conduction) so introducing small diameter routes will reduce efficiency - correct?



that's it (for now) - thanks for reading and any answers given =]

[n00b 0f l337]

Sneil, I beleive I already said, thats how it would be mounted for on a mobo, and leave it as it for open benching or vga...

[Sneil]

@K
1. yes fuggers does. there is a very small amount of refrigerant in these sytems and the key is charging with the right amount of refrigeration for complete evaporation.

2.yes.

3.there is a solid chunk of copper over the heatspot. the copper is what heats up andd boils the refirgerant. refrigerant evaporates(changs from liquid to a gas) on the hot spot. fins not nessary and difficult to implement

4.?

5.there are oil separators in cascades to stop this from happening. a single stage doesn't get cold enough to become viscous enough to clog the capillary tube.

6. that is how my present evap works

7. 24/7 for single stage. don't have a cascade running but normally not 24/7. not just used for benching but gaming as well(cascades)

8. the gas would get sucked up, the liquid would not due to gravity ;) ellsworth's cannot be placed on its side whereas my revamp of his can and would follow the maze with the top capped off.

9. no

10. yes small diameter will reduce effeciency, gas needs somwhere to expand. it is nessesary to have a high perssure side so gas can condense to liquid. having a low pressure evap allows gas to evaporate at lower temps. ie water in a vacuum can boil at 40c or so instead of 100c

there u go

[Sneil]

Sneil, I beleive I already said, thats how it would be mounted for on a mobo, and leave it as it for open benching or vga...
well u shoulda corrected my corrections on your drawing ;)

still dont know how good that design would work. im sure fugger/chilly1 would have a better idea then me

[n00b 0f l337]

Right also, with flipping my design to be for an upright cpu (which mine normally isnt, gotta love cardboard) my design will still probably do better if the center circle is carved out a little more also...

[ellsworth]

another note on the oil pooling. people use oil separators in cascades and/or they add propane to the system to circulate the oil. propane is an excellent carrier of oil and will help keep oil pooling to a minimum...atleast it should. in my design gravity keeps the liquid pool on the bottom of the evap so that the liquid is forced to evaporate before it enters the return line.

[n00b 0f l337]

You can also add a few bends in suction line near the compressor to let the excess liquid evaporate.

[Sneil]

i think by bend he means a loop or something to that effect, not a kink in the line in case somebody thinks that's what he meant

[BugsBMD]

Here is another one. How would this work? I think that it could be made with out to much troble.

[n00b 0f l337]

What happens if I post my next design and someone copys with just a little improvement...?

Other than that, yours has no real evaporation over the cpu center. Your cooling the corners.


Also, what about the collisions of freon particles? Does that even happen or work? Just runnign on someone elses idea.

[Sneil]

i think chilly1 needs to make his descision :stick:

[n00b 0f l337]

Hehe, and he shoudl look back at all the evaps, not just the newest ones. :)

[chilly1]

Ok lets give this a closind date of June 1 and then Iwill compile all the canadates using the following criteria.
Suitability of design.
Simplicity.
Ease of manafacture.

These need to be able to be made with common tools available to most anyone.
I will personally make 2 of the best design IMO. I will also accept input on this from those here. I may end up making the best two and running head to head tests.

[n00b 0f l337]

What do you by suitability? As in it could do cpu or gpu?

EDIT: Time for a new design, simple, lots of expansion room, probably best with a shorter cap tube to let more gas in, that way some maybe makes it. I cant figure out though which way would be best for rotation. So I'll let chilly decide. :cool: Hope you love it... :wierd:

[Special_K]

wow - thx for all the replies

they cleared up alot of questions =]

/K puts on thunking cap again =p

[Special_K]

well, for what it's worth, this is how K would do it

http://www.xtremesystems.org/forums/attachment.php?attachmentid=31211

it's a simple maze design, with the capillary and suction line in the middle and out the top for easy mounting =D

mebe the top-left corner needs thicker walls to strengthen it or summat?

[Sneil]

here's a gooder, much potential in this one :D. suction at top. Flip it around if you want refrigerant to be trapped longer at the center.

EDIT:ok i went with the trend and colored it like the rest of ya. its orange cuz so is copper :cool:

[n00b 0f l337]

It will follow the path of least reistance though and not go to those 2nd,ary spots...

[Sneil]

it will it will. regardless of which path it follows it'll be effecient. if too much is flowing through the first it'll go down the second. if it doesnt go down the second then its already effecient having the one with your charge. can't loose. that's way its a gooder. if you want you can have one pathway on the sides but its smart to have the two

EDIT: the inner pathways are closer to the core heat load so its perfect having more refrigerant flow down that pathway then the outer pathway.

[Sneil]

another :woot:

[matttheniceguy]

3. somebody suggested it would be better to have the evaporating gas flow over the hotspot of the block, and it started K thunking "is it better to have the hotspot covered by gas or copper?" - would it not be better to have a solid chunk of copper directly above the hotspot with lots of fins (larger surface area) and lots of cooling vapour around it for the heat transferal process?

By this I think you are meaning, would it be better to have a flat thin piece of copper over the core, with refrigerant boiling on it, or a big chunk of copper over the core with lots of surface area so that chunk of copper can be cooled. Two examples of this would be a Chilly1 Evap, which is a big solid copper cylinder with fins around it's outside, or an evap made by brazing a copper cap to a copper plate (I think it's knows as a lardass evap around here?)

The main consideration between these two designes is heat load. The convection heat transfer coefficient of a boiling refrigerant is so high that fins (the big chunk over the core) will actually insulate the core. A thin copper base will provide less thermal resistance between the refrigerant and the core and will work better... sometimes. If you use a flat base over the core, the refrigerant has to remove the entire heat of the cpu from a small surface area directly over the core. If your heat load is to great, this won't be possable. As a general rule I would say, low heat load, use a flat base over the core, with a high heat load, use a large fin over the core (something like a chilly1 evap) because the large fin will provide more surface area and will handle load much better.

[STEvil]

req'd:

1 flat copper plate divided into 4 sections.
some copper pipe cut in half (to create "eyes")
drill bit, dremel, etc to drill holes or cut slots on back side of the "eyes"..

Putting in 4 cap tubes and 4 suctions might be a bit hard (well, not so bad with some ingenuity), but I think this would work pretty well in almost any orientation (except upside down).

http://www.xtremesystems.org/forums/attachment.php?attachmentid=31226&stc=1

[Sneil]

use little makeshift copper plates out of 3/4" copper tubing to cove holes on top and side. just cut a 1/2" length peice of 3/4" tubing with side cutters, spread open and cover holes on both sides and top of block.

2 cap lines can be used as well if necessary

EDIT: can flip around if that would give even better results

[smurfmods]

use little makeshift copper plates out of 3/4" copper tubing to cove holes on top and side. just cut a 1/2" length peice of 3/4" tubing with side cutters, spread open and cover holes on both sides and top of block.

2 cap lines can be used as well if necessary


That is the best design yet. I think even I could even build that. The drilled holes would have to line up pretty close for it to work right. All the benifits of a maze without a milling machine. :toast:

[matttheniceguy]

Well I thought I would break out the solid works/MS Paint and throw something together. It is a simple maze design that can be made by drilling 5 1/4" holes, tapping them, and then putting in 4 (or 5) plugs. The cap tube is put in the centre, and the suction is at the end of the maze, and can be sticking out the top or side of the block. This design is super simple to make, as it only has one captube and one suction line, and 5 relativly large and easy to drill holes (all right angles too). This design should work pretty well both laying flat and sitting upright.

[Sneil]

That is the best design yet. I think even I could even build that. The drilled holes would have to line up pretty close for it to work right. All the benifits of a maze without a milling machine. :toast:


thanks for that, glad to hear it :toast: here's another. dont know if its any better since refrigerant doesn't start at center, but here it is

holes dont all have to be on same side

can go 6 shoots with smaller drill bit, suction would be on bottm left


i like my last better however :cool:

[BugsBMD]

This one should be easy to build and should work well. And it will be easy to insolate. You can make the cap and suction easy by drilling a big hole through the top layer and then make a small hole the rest of the way to the bottom layer. Then braze the cap first and then the suction.

[n00b 0f l337]

Heh, I think you'd have quite a bit of brazing and drilling but still a bit simple. Only problem is main evaporation will probably be at mid corner area. Otherwise, theres a lot of awesome designs now... I don't think mine are as good looking, but I think mine are more simple... :mad:

[Sneil]

drilling is simple enough, it doesn't matter how much you have, just a couple more minutes out of your precious day :) imho its more then worth a few min
EDIT: you've got some nice ones noob

[n00b 0f l337]

Well I hope this ones the best. Tons of evap room, and a new design that I know someones going to steal...

So reader beware. I can find where you live. :p:

[Sneil]

well i don't know how easy it'll be to braze those caps in but this my work better maybe not.

NOT STEALING YOUR EVAP just so you know ;)

this is completly your design and this is only a suggestion for it.

keep in mind you'll need small caps to fit inside, they only look huge in the pic because the evap itself in the pic is huge. still lots of surface area and added resistance to keep refrigerant at center longer. made them round well, because caps are :wiggle:

note: be forewarned those angled holes will be much difficult to drill without a press.

pic of NOOB's evap

[n00b 0f l337]

The problem with that is, the caps arent like round, you'd have to use pipes, and that I geuss would be a little hard with that last peice so close to the back. But stil a good idea.. :p: :woot:

[Sneil]

what caps are you talking about then?

[n00b 0f l337]

Like if you put an end cap on a pipe. Its like a round end ish, basically..

[Sneil]

yea i know, what square caps are you using in the first?


The problem with that is, the caps arent like round,

:confused:

[n00b 0f l337]

Not square, the caps on its side.... I didnt really draw in the round part... I'll do that now.

[Sneil]

oh, got it...

[Sneil]

what else you working on noob, how's your system commin along? have you started getting some parts yet?

[n00b 0f l337]

*sorry*
Edited my chat with sneil.

[Sneil]

you can get a nice condensor/fan/shroud assmebly from drew for a good price, he does some nice work. chilly one too. just mentiond drew since i got mine from him and am more then pleased with it. what type of compressor?

[n00b 0f l337]

*sorry*

[Sneil]

nice, what gasses is the compressor made for? you can use any i know but just wondering what is says it's manufactured for

[n00b 0f l337]

*sorry*

[Sneil]

why? propane works great. its what im using atm
EDIT: oh well, back to evaps. not making it easy for chilly1 to have to sift through all this to find the evaps

[Sneil]

your safe as long as you have no leaks and use common sense when charging etc.

[n00b 0f l337]

Yeah, but im a 16 year old, I live with parents. This is what they think...
Propane = BBQ = Fire...?

/--\
[--]
[--] Thats a house

() Thats a bottle of propane

! Thats an upside down match...

................... Thats the crater of whats left of house.

Don't work too well in there opinion. I'll probably do it anyway, then leave it outside for a week running to prove it aint gonna blow up.

[pythagoras]

This isnt the sneil and noob chat show :slapass: . Evaps is the topic ;)

Regards

John.

[n00b 0f l337]

Sorry for our little chat. I edited mine to say im sorry, I hope sneil will do it as well. Here's another design, but I don't think its near as good as my last one. Maybe this ones okay for a gpu... Who knows, chilly knows...

[Sneil]

i blame it all on noob, he better have said sorry! :slap:

here's another. effeciency is the key and if all the holes make it complex chilly1 then so be it. but again its only holes that have to be drilled, a press may be necessary with the angles. but how hard is it for us to find adrill press we can use? anyway, probably my most complex but their only holes...

jk noob :toast:

[n00b 0f l337]

Eh' too complex...

[Sneil]

chilly will be the judge of that thanks

really i don't see its complexity, maybe all the twists and turns are just too much for you to handle.. :)

[n00b 0f l337]

12 drillings... Thats quite a bit with all the measurements and all.

[Sneil]

why? how long does it take you to drill a hole noob? how hard is it for you to drill a hole noob? i think your underminding chilly's knowledge and experience and wasteing everyone elses time with your expertise, or lack there of on the subject :slapass:

lets not see anymore blabbing unless there's an evap drawing with it. at least from u and i anyway alright?
ok :shakes: :p: :toast: :woot:

[n00b 0f l337]

What was the edit for? :)
Here's a new evap featuring my feature of shoving caps and tubes in holes and brazing em shut. I call it...

Freevap11...

[BugsBMD]

Here is another one.

n00b I think that your last one will not have enoff flow mabe make your holes in the copper pipe bigger.

[wdrzal]

some of these don,t make sense to me from a mechanical drawing point of view, also when I ? small holes ,it wasn't that they won't work only that a sufficient number must be present as not to create a restriction so you have a pressure increase inside one side of the evaporator.Think max wetted surface,exposed to lowest pressure,no restriction,but refrigerant has to maintain sufficiant velocity for oil return.These are my ideas of a good evap.finally easy too build was the original idea. remember what fugger said,drill, tap,braze,done.

[ZENNZZO]

I got the jest of this, as being an easy to make evaporator for the masses.
1/2" to 3/4" thick copper stock is easy to source, as is 1/4"...

My idea is to drill and tap holes, clear through the base block.
4 passes of equal distance and size, say 3/8".
Now Dremmel out the areas shown in the diagram.
Braze on 1/4" end plates.
Drill holes on top for CPU or use the ends for GPU.
Simple, sound and anybody worthy of building a Phase unit can acomplish very easy...
There you have it FREEVAP...Zennzzo style
***Not to Scale***
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31315
***Not to Scale***
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31316&stc=1
***Not to Scale***
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31317&stc=1

[meke]

I got the jest of this, as being an easy to make evaporator for the masses.
1/2" to 3/4" thick copper stock is easy to source, as is 1/4"...

My idea is to drill and tap holes, clear through the base block.
4 passes of equal distance and size, say 3/8".
Now Dremmel out the areas shown in the diagram.
Braze on 1/4" end plates.
Drill holes on top for CPU or use the ends for GPU.
Simple, sound and anybody worthy of building a Phase unit can acomplish very easy...
There you have it FREEVAP...Zennzzo style
***Not to Scale***
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31315
***Not to Scale***
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31316&stc=1
***Not to Scale***
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31317&stc=1


yeah, this was what I did.

It works, I posted a photo in the previous pages. :cool:

[meke]

Also, if you braze the suction and capline to the side, not to top helps this evap fit to SLI gpu's.

[n00b 0f l337]

My last one wasnt great, but it needs good sized holes. That way it gets some good evaporation done in the center before it gets suctioned.

[ZENNZZO]

Also, if you braze the suction and capline to the side, not to top helps this evap fit to SLI gpu's.
I mentioned that also...The biggest challenge I found with SLI was the 1.5" of room to fit the evap. and insulation.
I did it with shortened Chilly1 evaps and I used a copper cap with the capline and suction in the side...
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31327
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31325


My last one wasnt great, but it needs good sized holes. That way it gets some good evaporation done in the center before it gets suctioned.
Noob,
Have you ever seen an evaporator made for Computers other than in pictures?
I don't want to sound discouraging, but build your 1st. Phase rig using a simple design or one of chilly1s' evaporators and you'll understand what is needed to be effective not to mention a good working knowledge of how Phase actually works.
Good luck with that...~Z

[BugsBMD]

All the evaps that I have drawn so far are not to scale. Also all my designs include taping out all the holes that were drilled. And if any one wants to comment on my evaps when they are posting theirs it would be much appreciated as I don't know a lot about good refrigerant flow.

With this new one you would drill a hole and then tap it then drill another hole then tap it ……………....................... You can also control the flow through the block by changing the size of the green holes.

[Sneil]

noob your torch probably wasn't getting the blocks hot enough for the solder to melt properly, this can be difficult with just MAPP and near impossible with propane

zenzo's is pretty similar to my 5 hole block. no dremmal tool needed, just a drill and some copper tube for covering holes, sneil style :D

[n00b 0f l337]

I have built 2 evaps before, frankly it was quite hard with just a hand drill, and I couldnt manage to get it to braze together perfectly, so they went to trash.

[elminster]

This is essentially the same as matttheniceguy's from http://www.xtremesystems.org/forums/showthread.php?p=880360#post880360. So if it gets the vote, give matt the credit, not me please. (Although my idea wont make a difference if ur not allowed different sizes, so just ignore it now if u cant do it)

http://www.xtremesystems.org/forums/attachment.php?attachmentid=31334&stc=1

Essentially the same, with a plug put in on the first hole (A). Although I suggest it would be best to maximise (X) value, limited only by the thickness at top and bottem (also possible issues with plug size). By doing this, it will allow more heat transfer to occur closer the middle and not around the outside where its less of a priority. Also if this was placed flat, a tiny extra amount of pooling would occour in the center.
Although would be best to vary it so the more of it evaporates near center and small amounts around the rest, as it would conduct outward anyway.

(Y) would be (5/8" - (X)) /2.
Would take 2 drill pieces, 5holes.


Cappilary tube would be placed exactly in center (just do an X across the top of the block from corner to corner).
Depth of Drilling for Holes:
(A) 1.5" - 3/16" or 2/7"
(B) 1.5" - 3/16" or 2/7"
(C) 1.5" - 3/16" or 2/7"
(D) 17/16" - Y
(E) 1.5" - 10/16" or 7/8"

Also if you have the Suction tube comming directly out of the drill hole (A), you produce alot lower profile.

Other thoughts i had about design are HERE (http://www.xtremesystems.org/forums/showthread.php?t=63929) (other thread)

Edit: Anyone know the capacity (volume would be handy to compare) of chilly1's other mainly used evap?

[Sneil]

minimizing wall thinckness for evaporators is something to keep in mind as you'll allow for a larger surface area in your design. 1/16" inner walls and 1/8" outer is fine and maximizes effeincy with the larger surface area. just make sure you use a press or you'll end up with holes in those walls where you don't want em

[elminster]

minimizing wall thinckness for evaporators is something to keep in mind.
I was thinking about that... although if you minimise thickness to outside walls, edges and top are no good. Just more loss.
So to get the Absolute max you can get from just a drill press, you wanna drill the holes slightly offset toward the bottem. Need a "This way up" sticker (before you finish insulating and stuff, would be ok after) if both cap and suction exits at edges rather than top.
For example using the 1/4" holes, putting them dead in the center, vertically, you would transfer similar amount of heat from the top than the bottem (more from proccessor side as its warmer).
Although if you put it at a ratio of 1/3 distance from bottem to 2/3 away from top, then you instantly get less heat loss, more efficency and lower proccessor temps (but not evap temps).
Although i dont think anyone has exact values for minimum thickness that copper could take with any particular design, and grade of copper for set pressures.

Edit: Also offsetting it toward the bottem of the block would allow you to make a grove to place cap tube in, (from center,going toward the edge or wherever). Giving you a flat finish along the top, if your using a hard insulation it would help a bit.

[Sneil]

I was thinking about that... although if you minimise thickness to outside walls, edges and top are no good. Just more loss.
not sure what you mean. edges will be fine. walls of copper tubing aren't even that thick.

[elminster]

not sure what you mean. edges will be fine. walls of copper tubing aren't even that thick.
I meant on the inside of the block... for example,

http://www.xtremesystems.org/forums/attachment.php?attachmentid=31342&stc=1
As marked on top crossection:
Base- being Proccessor side
Top- being top part to be insulated
(A) Coolant closer to this part loses nearly all heat, with small amounts transfered back down to base, ALOT less on normal designs as its not made from 1 piece.
(B) Coolant closer to the side edges will lose alot around the edge, with a reasonable amount transfered to the base.
(C) The base (and closer to the middle) is the best parts to cool as a huge percentage gets transfered to the proccessor.
(D) Fins around the center transfer more percentage of the heat to the base than edge parts (B), as it can only transfer to base or top section.

1st Design would be crosssection of Matttheniceguy's

2nd Design would prove to better because less will be lost through the top, and more will be transfered through the bottem. Although this complicates matters on thickness and exact placement, taking more time. Although easest possibly way of gaining performance in a design.

3rd Design would be a crosssection of mine, which all drill points would be taken from center, speeding it up, yet allowing more to evaporate closer to proccessor and increase surface area, but losing more from the top too if upside down (on an GPU in normal case setup). Also the offset of hieghts between first and second hole would allow more to be evaporated in the center if its layed down flat (mobo on a table for example, not in a case)

Im quite sure thats correct, although i aint the best of people to put ideas across.

[ZENNZZO]

I would venture to say you want some thickness/mass in the evaporator for thermal transfer and stability...
...for example you can just use a finger tip to de-ice the suction line in a spot,
but you'll never do that to a Chilly1 evaporator.
I belive he has an 11mm base and a larger core that is solid copper

[elminster]

I would venture to say you want some thickness/mass in the evaporator for thermal transfer and stability...
...for example you can just use a finger tip to de-ice the suction line in a spot, but you'll never do that to a Chilly1 evaporator.

I doubt thats quite true.
Sure more copper will act like a fly wheel and keep it colder if you get big temprature peaks.
But It also loses more heat to the outside and less percentage of heat would travel the entire distance. One of the main reasons why even though it has the hieght away from the processor, it still works so well is because spiral stops liquid flowing out (unlike maze) and its capacity to allow near enough all liquid to evaporate.

The ideal goal would be to have as much of the liquid as possible, evaporting close to the center of the proccessor (in both hieght and position on the proccessor).

btw, someone correct me if im wrong. I dont wanna sound stupid :(

[ZENNZZO]

I doubt thats quite true.
Sure more copper will act like a fly wheel and keep it colder if you get big temprature peaks.
But It also loses more heat to the outside and less percentage of heat would travel the entire distance. One of the main reasons why even though it has the hieght away from the processor, it still works so well is because spiral stops liquid flowing out (unlike maze) and its capacity to allow near enough all liquid to evaporate.

The ideal goal would be to have as much of the liquid as possible, evaporting close to the center of the proccessor (in both hieght and position on the proccessor).

btw, someone correct me if im wrong. I dont wanna sound stupid :(you may have forgot to factor in the insulation around the block sheilding it from the ambient, only being exposed to the CPU/GPU... ;)

[wdrzal]

you insulate the top and sides so it doesn't matter much.remember your physics,heat moves to cold.the greater the difference the faster the transfer.

[ZENNZZO]

you insulate the top and sides so it doesn't matter much.remember your physics,heat moves to cold.the greater the difference the faster the transfer.
So then a thinner base would be more effective you think?

[BugsBMD]

Another one. Please let me know what is wrong with my designs so that I can improve on my new ones.

[Sneil]

back to inner wall thinkness
I guess its a matter optimizng evaportation by considering the factors of a good heat transfer to the liquid refrigerant. with smaller walls your minimizing heat transfer to upper block farthest from heat load, however the large surface area closer to the heat load would make up for this easily. these are factors of negligable difference that we dont need to even bother considering. The only reason you'd have the thin walls is to allow for larger surface area through eather more holes or bigger holes which considerably outweighs the benifit of thick walls. surface area prevails. also note the circular surface area optimzes heat transfer from copper to refrigerant anyways and it is only at a minute point between the circular holes where the wall is 1/16" thick

the function of the wall is more for trapping/guiding refrigerant over the surface area then anything

your design is fine. keep in mind we want the refrigerant trapped at the center as long as possible.
EDIT: nevermind i had your cap and suction backwards
looks good :)

my two personal favorites :)

[n00b 0f l337]

My problem was I couldnt keep it hot, my parents who not let me do it on a stove or anything.

[matttheniceguy]

About this base thickness discussion. With a thiner base the refrigerant directly over the core is more effective, due to a lower thermal resistance between it and the core. Unfortunatly, the rest of the refrigerant in the block will be less effective due to a higher thermal resistance between it and the core. With a thinker base, you increase the thermal resistance between the refrigerant directly over the core and the core, but decrease the thermal resistance for the rest of the refrigerant. The ideal base thickness would depend on the heat load. For a low heat load a thinner base would be better, but for a large heat load a thicker base will work better. This is why Chilly1 blocks are so good for crazy heat loads. They have a very thick base and a solid core.

[n00b 0f l337]

Ah, that makes sense.

[BugsBMD]

Another one. Again all holes tapped.

[elminster]

Unfortunatly, the rest of the refrigerant in the block will be less effective due to a higher thermal resistance between it and the core. With a thinker base, you increase the thermal resistance between the refrigerant directly over the core and the core, but decrease the thermal resistance for the rest of the refrigerant.

Im not too sure how evaporating it closer to the core could possibly increase the thermal resistance of anything. The only thermally resisting material within the block is the block itself, the actual liquid doesnt resist any heat what so ever.
How would having a thicker base "decrease the thermal resistance for the rest of the refrigerant". Ive tried to think of it in every aspect possible although its still beyond me.

Lemme put it in an analogy... If you hit your finger with a hammer and you get a bag of ice. Best way to transfer heat is to put ice directly onto your finger, but tats not possible cause it will go everywhere.
If you put it in a thin bag, it will cool down your hand faster but lose more heat elsewhere. If you put it in a thick freezer bag, its gonna stay colder for longer and take longer to cool down the bag, but isnt gonna help as much. So now what if you put a thin bag with ice in it and covered the rest of it with a towel to stop it losing heat. Effectivly you have better cooling with similar waste as the thick bag, but it wont take as long to cool and it will be more focused on your finger.

The thermal capacitance of copper will always be the same, the thermal conductivity will be similar enough not to notice any difference, the amount you have of it will just vary times it takes to cool, and heatloss through edges (Assuming fixed volume inside is being taken up by space for coolant). So the only way to increase performance is to get as the coolant to evaporate closer to the core, increase surface area for transfer, and insulate the rest better.
I noticed with chilly1's is has a higher capacity than most of the maze designs around, with a larger surface area as well, which is probely why youd notice the difference in heavy loads. Im betting most maze designs still have more fluid comming out into the suction pipe, where it evaporates and does nothing to help.
Personally i would have thought under heavy stress's the heat given out would evaporate coolant faster, hence less capacity is needed. Its gonna be impossible to judge the exact capacitance (volume) needed for a certian setup, so it would need to be slight excess, but enough so that all the coolant does evaporate inside and not in the suctionline.

oh btw, srry for disrupting thread a bit.

@BugsBMD - Your asking to drill 15 different holes, in a range different sizes, a points that would be time consuming to measure. Considering their gonna be threading the inside, that would take a very long time to make.

[saratoga]

Im not too sure how evaporating it closer to the core could possibly increase the thermal resistance of anything. The only thermally resisting material within the block is the block itself, the actual liquid doesnt resist any heat what so ever.
How would having a thicker base "decrease the thermal resistance for the rest of the refrigerant". Ive tried to think of it in every aspect possible although its still beyond me.

Lemme put it in an analogy... If you hit your finger with a hammer and you get a bag of ice. Best way to transfer heat is to put ice directly onto your finger, but tats not possible cause it will go everywhere.
If you put it in a thin bag, it will cool down your hand faster but lose more heat elsewhere. If you put it in a thick freezer bag, its gonna stay colder for longer and take longer to cool down the bag, but isnt gonna help as much. So now what if you put a thin bag with ice in it and covered the rest of it with a towel to stop it losing heat. Effectivly you have better cooling with similar waste as the thick bag, but it wont take as long to cool and it will be more focused on your finger.



You're overlooking the issue of surface area. Think heatsink, you need fins to transfer heat effectively to the air. Liquid refrigerant has astronomical heat capacity compared to air (orders of magnitude better), but still needs some surface area to work with. However the area it needs is very small, so thin bases are going to work better as you said.

With watercooling you get effective cooling maybe half an inch or so from the die (look at the design of the white water, cascade, G5, or other high end blocks - everything is over the die). With refrigerant, the specific heat capacity is many times greater then water still, so the benefit of surface area is greatly reduced, and the cost of each mm of copper between the core and the refrigerant becomes more significant.

The ideal evaporator probably looks a lot like a white water block with more room for gas to escape, but the refrigerant still sprayed directly over the core onto fins or through nozzels or something like.


The thermal capacitance of copper will always be the same, the thermal conductivity will be similar enough not to notice any difference, the amount you have of it will just vary times it takes to cool, and heatloss through edges (Assuming fixed volume inside is being taken up by space for coolant). So the only way to increase performance is to get as the coolant to evaporate closer to the core, increase surface area for transfer, and insulate the rest better.
I noticed with chilly1's is has a higher capacity than most of the maze designs around, with a larger surface area as well, which is probely why youd notice the difference in heavy loads. Im betting most maze designs still have more fluid comming out into the suction pipe, where it evaporates and does nothing to help.
Personally i would have thought under heavy stress's the heat given out would evaporate coolant faster, hence less capacity is needed. Its gonna be impossible to judge the exact capacitance (volume) needed for a certian setup, so it would need to be slight excess, but enough so that all the coolant does evaporate inside and not in the suctionline.


Agree 100%. I think modern evaporators are not well designed for the job. They have the required surface area, but spread it too far from the heat source to be as effective as they should be.



oh btw, srry for disrupting thread a bit.



Better to discuss it I think. Thats how the best waterblocks were designed. Hopefully threads like this will inspire something like the whitewater, except for evaporators.

[Bergo]

Well, I figured I might as well quickly make up a submission, I appologize if this evaporator design is similar to another posted, but I didn't read through the entire thread, but keeping with the "simplistic 30 minute (for someone familiar with machining) build" I drew this up, If I had some copper, I don't doubt that it would be possible to build in under an hour with not more than a drill press. Depending on a persons brazing abilities, they could either tap the ends and plug them, or braze in small copper caps. Although somewhat over-simplified, I beleive it would perform "good-enough" for someone with limited resources and skill constructing their first single stage direct die phase change cooling system, which this particular thread seemes to be aimed at. Although some other submissions did look a bit more promising, I beleive that this block could perform near, if not on par with other, more complicated, blocks. I beleive that it would be interesting to see the difference in performace on an over-simplified design such as this, and a more complex design such as any number of the designs that n00b has come up with.

[Evil_Spork]

Meke, thats it.

You can also get these pieces of copper stock from dangerden.com

http://www.dangerdenstore.com/product.php?productid=128&cat=49&page=1

Bummer there is no picture available
why not order a maze1 and use that.. it seems to be pretty much what were looking for minus the threaded channels. for that just use a dremel. ;)

[Bergo]

umm, I wouldn't recoment trying to "thread" (it's actually "tapping" I beleive) with a high speed rotary tool :P Whenever I"ve tapped anyhting (working with watercooling, I"ve only ever had to on schedule 80 PVC) I sjut do everything by hand using my tap and die set and a socket wrench. Also, the maze 1 is no longer available form Dangerden, would have to buy one privatly.

[matttheniceguy]

Im not too sure how evaporating it closer to the core could possibly increase the thermal resistance of anything

What I was refering to was the thermal resistance between refrigerant not directly over the core and the core. With a very thin base, there is low thermal resistance between the core and the refrigerant directly over it, but the thin base does not transfer heat outward well, creating more thermal resistance between the core and refrigerant in areas of the evaporator that are not directly over the core. A thicker base will increase the thermal resistance between the core and the refrigerant directly over it, but will transfer more heat towards the outer edges of the evap. If the surface area directly over the core is large enough to dissipate the heat of the processor than a thin base can be used, but with a higher heat load a thicker base must be used to gain enough usefull surface area to cool the processor.


You're overlooking the issue of surface area. Think heatsink, you need fins to transfer heat effectively to the air. Liquid refrigerant has astronomical heat capacity compared to air (orders of magnitude better), but still needs some surface area to work with. However the area it needs is very small, so thin bases are going to work better as you said.

Your absolutly right about the heat capacity, but more important than that is the convective heat transfer coefficient (which describes conduction between the evaporator and refrigerant, or waterblock and water, or heatsink and air). This coefficient ranges from around 25-250 W/m2 · K for air, 50–20,000 for water, and 2500–100,000 for phase change. source (http://www.engineering.com/content/ContentDisplay?contentId=41005014)

The interesting thing to note about this is that for a copper surface, fins will actually reduce the amount of heat transfer for convective heat transfer coefficients above around 5000-10000 W/m2 · K. In other words, in most phase change systems, fins sticking out of the base actually insulate it from the refrigerant. Were the block made of aluminum or some other less conductive mateial, the coefficient at which fins would become uneffective would be even lower. The reason a maze design or an evap made of caps brazed to a plate works is not because the fins provide more surface area, but because they restrict the movement of the refrigerant and keep it over the core where it will be effective.

I did a project on a subject similar to this in an FEA course a while ago. I have attached a PDF of a report I did, that shows the effect of the convective heat transfer coefficient on fin effectiveness. Most of the report has been removed so the file is small enough to upload, but the parts relating to this discussion have been left intact. If anyone is interested in what the rest of the report said, PM me and I'll e-mail the full thing to you.

[BugsBMD]

Here is one of my last. It is very simple all the holes are tapped. I made the holes a little to wide and did not have time to make them smaller.

[n00b 0f l337]

Seems thats the same as pretty much the last few.

[chilly1]

I noticed with chilly1's is has a higher capacity than most of the maze designs around, with a larger surface area as well, which is probely why youd notice the difference in heavy loads. Im betting most maze designs still have more fluid comming out into the suction pipe, where it evaporates and does nothing to help.
Personally i would have thought under heavy stress's the heat given out would evaporate coolant faster, hence less capacity is needed. Its gonna be impossible to judge the exact capacitance (volume) needed for a certian setup, so it would need to be slight excess, but enough so that all the coolant does evaporate inside and not in the suctionline.

WEll if you assemble your system so that teh capillary tube is wound around the suction tightly near the exit of the suction from the block that would make it a heatexchanger which would make the system more effecient. Then you charge by superheat and this would eliminate the liquid refrigerant from becomming a problem in teh suctionline.


Great ideas guys/

[Bergo]

yea.. that'd be the evaporator design I posted turned 90 degrees :nono:

[ellsworth]

What I was refering to was the thermal resistance between refrigerant not directly over the core and the core. With a very thin base, there is low thermal resistance between the core and the refrigerant directly over it, but the thin base does not transfer heat outward well, creating more thermal resistance between the core and refrigerant in areas of the evaporator that are not directly over the core. A thicker base will increase the thermal resistance between the core and the refrigerant directly over it, but will transfer more heat towards the outer edges of the evap. If the surface area directly over the core is large enough to dissipate the heat of the processor than a thin base can be used, but with a higher heat load a thicker base must be used to gain enough usefull surface area to cool the processor.



Your absolutly right about the heat capacity, but more important than that is the convective heat transfer coefficient (which describes conduction between the evaporator and refrigerant, or waterblock and water, or heatsink and air). This coefficient ranges from around 25-250 W/m2 · K for air, 50–20,000 for water, and 2500–100,000 for phase change. source (http://www.engineering.com/content/ContentDisplay?contentId=41005014)

The interesting thing to note about this is that for a copper surface, fins will actually reduce the amount of heat transfer for convective heat transfer coefficients above around 5000-10000 W/m2 · K. In other words, in most phase change systems, fins sticking out of the base actually insulate it from the refrigerant. Were the block made of aluminum or some other less conductive mateial, the coefficient at which fins would become uneffective would be even lower. The reason a maze design or an evap made of caps brazed to a plate works is not because the fins provide more surface area, but because they restrict the movement of the refrigerant and keep it over the core where it will be effective.

I did a project on a subject similar to this in an FEA course a while ago. I have attached a PDF of a report I did, that shows the effect of the convective heat transfer coefficient on fin effectiveness. Most of the report has been removed so the file is small enough to upload, but the parts relating to this discussion have been left intact. If anyone is interested in what the rest of the report said, PM me and I'll e-mail the full thing to you.


i have found this to be true. i have built several hollow evaporators with flat bottomed bases (very thin too) and have found them to work very well. when you have such small heat capacity in the copper like this, charging for a specific load becomes very important. i think if a system is charged correctly for the load the system is seeing, a hollow evap would work just as well, maybe better,than any maze design. of course the motherboard would be laying flat and the refrigerant will be pooled over the heatsource. the trapping of the refrigerant is done by gravity. a hollow evap is of course the easiest to make...2 caps, a small segment of pipe, and a flare fitting all brazed together. once i get my hollow evap mailed back to me ill use it in my new cascade i'm putting together and post up some numbers.

[ZENNZZO]

Could you post some pics of the construction of it?
;)

[BugsBMD]

Sorry about that I thought that mine was different enough because of the different direction and I made the holes much bigger. But I will leave it up to chilly1 to decide.

[n00b 0f l337]

Thinking on that hollow evap design, and wondering if it works really, I thought of this, probably the simplest thing to do period, as the only drilling would be optional. :)

Basically a copper block, and just braze on a 90 degree elbow fitting, put the capillary tube in the bottom. I also woudl recommend pushing a bit of the capillary into the fitting as well just for a little resistance, like coil it in a ball above the bottom area.

Its just an idea, you can also do other things, like other fittings, a tube at 45 degrees, ect.

[Bergo]

the problem with that is it doesn't give give the liquid refridgerant enough room to evaporate into a gas over the CPU, all this would do it braze the very end of the return line to a thin copper plate, most of the evaporation would occur in the return line, which wouldn't be of much benefit for cooling the CPU/GPU. you'd want to use a large diameter copper cap brazed onto a thin copper plate so that the evaporation hapens in the evaporator (sorry for being kind of over-obvious) a copper elbow wouldn't be large enough.

[n00b 0f l337]

Yeah, thats pretty much it. YOu can adapt that pretty much anyway. But then the refrigeration evaporation is in the bottom of cap, but not like near the cpu center. Its a bad idea, my other one with the big open center was better.
Forget it or something.

[wdrzal]

Guys I just looked threw everthing that was posted,the thing that stuck out the most is the lack of dimensions. Without dimensions lines on paper are just that,"Lines".The size and centering of the holes must be worked out to be sure they fit within the overall dimension .Everyone seems to just be drawing patterns of which way the refrigerant will flow without giving much thought to the dimensions of the design.

[n00b 0f l337]

Well, assuming its a 1.5" block that chilly started with, its pretty easy to size the actual picture to 1.5" in paint print it out and get dimensions from that. This is designing at the moment.

[wdrzal]

1 day left,do you think chilly is going to work out all the dimesions on everything that is posted,even if he had the time,thats the designers job.If he has to do the dimensions ,whos design is it ???

Whats the drill size, whats the spacing, whats the wall thickness, whats the bottom thickness, where are you placing the mass,ect,ect.......these questions can only be answered if there are dimensions on your drawings.

[wdrzal]

WEll if you assemble your system so that teh capillary tube is wound around the suction tightly near the exit of the suction from the block that would make it a heatexchanger which would make the system more effecient. Then you charge by superheat and this would eliminate the liquid refrigerant from becomming a problem in teh suctionline.


Great ideas guys/


a tip: since a round tube touching another round tube only has a small contact area, soft soldering the cap line to a copper tube greatly increases the contact area,giveng better heat X-change.

[ZENNZZO]

1 day left,do you think chilly is going to work out all the dimesions on everything that is posted,even if he had the time,thats the designers job.If he has to do the dimensions ,whos design is it ???

Whats the drill size, whats the spacing, whats the wall thickness, whats the bottom thickness, where are you placing the mass,ect,ect.......these questions can only be answered if there are dimensions on your drawings.
Here ya go...
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31602
brought from the last page...evidently some folks have more than one drawing...
***Not to Scale***
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31315
***Not to Scale***
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31316&stc=1
***Not to Scale***
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31317&stc=1

[wdrzal]

Here ya go...
http://www.xtremesystems.org/forums/attachment.php?attachmentid=31602


Thats a start, but a mechanical drawing is done in 3 veiws to be completly understood, top,front and side.

this drawing tells me there is 4 ,1/4" holes spaced .20 apart, total size is 2"wide X ? deep. height is .55 no other info like other hole ,how deep are the holes or how they interconect is provided. A block could not be made from that info.Not trying to critisize but teach you what is necassary so a machine shop could produce your designs.

Remember high scool drafting class. Isometric (3D) and 3 veiw drawings?

[Bergo]

Here ya go, Sorry it doesn't look a bit nicer, but I got bored and tooka break from physics homework to draw a quick sketch, with some values in place to aid in construction if my design should be so lucky as to be chosen. If the pic isn't clear enough I can write a more detailed description later tonight or tomorrow, and accompany it with a better diagram, but it seems fairly self explanitory/simple. Questions? Comments? Critisizm? all is welcome

Thanks,

Andrew (Bergo) Bergman

http://www.xtremesystems.org/forums/attachment.php?attachmentid=31610&stc=1