New design cap in cap evaporator

[DetroitAC]

Well, I'm working on some what I would call "high performance" evaporators, but I realized lately that there is maybe a need for cheap, easy to build evaporators.

I've got a first prototype of a baseplate for a cap in cap style evaporator. The nice thing about the cap in cap concept is that you can buy the caps locally yourself, drill the passages, braze them up yourself however you'd like.

The problem is that you have to have access to a machine shop, and be able to get a chunk of copper, and then deal with the caps sliding around while you try to braze them.

I've made a plate that is machined to accept Mueller copper plumbing caps. There ae bosses that will mate to the ID of the caps, and the height of the bosses is such that they will all nest with a gap across the top roughly the same as the radial gap (except the two smallest, these were just too short, so the top gap is large). I've machined to leave the same gap (0.004 inches) as occurs between copper pipe and fittings, so brazing (capillary wicking of the filler) should be quite easy and strong.

I've also added three "teeth" that engage to create a very light interference fit. It's hard to see these teeth, so I've put a little pencil lead on them in some of the pictures.

The only strange part happened with the 2nd largest cap. This one was much shorter compared to it's length than all the others, so I had to mount it much higher than the others, and grip it by it's outside diameter (you can see one tall "wall" in the pictures, and one set of inside teeth). I'm not sure how well this is going to work, as the OD is certainly not controlled in production as tightly as the ID.

I may also need to make these steps taller, due to the lead-in chamfer of the caps. I have noticed on the prototype that the teeth only grip well for the first assembly (they are soft and small). I may need to make them bigger, add more, or maybe just forget them. The steps alone will help keep the caps in position unless the evap gets really banged around before brazing.

This one is 1.75 inches square base. I think I'll also make a smaller version that eliminates one cap, and will be 1.5 inches diameter round base, but same basic design.

You'll notice, I haven't drilled holes to establish the circuitry of the evap, no mention of where the suction is, where the inlet is. I think deciding those things is part of the fun in building your own evap. These caps also leave a lot of room for adding more heat tranfer structure, modifying the baseplate, etc.

http://img70.imageshack.us/img70/1614/evapv8010cropnj2.th.jpg (http://img70.imageshack.us/my.php?image=evapv8010cropnj2.jpg)

http://img486.imageshack.us/img486/1278/evapv8005cropmy8.th.jpg (http://img486.imageshack.us/my.php?image=evapv8005cropmy8.jpg)

http://img231.imageshack.us/img231/7963/evapv8004cropjo6.th.jpg (http://img231.imageshack.us/my.php?image=evapv8004cropjo6.jpg)

[THE JEW (RaVeN)]

Very nicely done. I'm always interested in cap in cap designs as I don't have access to a lathe or CNC :(

[Unknown_road]

well easy to build? Not everybody can machine the base plate that way ;)

[DetroitAC]

well easy to build? Not everybody can machine the base plate that way ;)

What I mean, is that the baseplate is very simple, and takes little machine time, so it's cheap. Then the rest of the evap is easy to build, in the end this means it's possible to build an evap for ($ cheap), and with only a hand drill and brazing torch, not much time invested in fabricating or tracking down materials.

[Rockhammer]

never mind

[andL64]

this is a very nice idea maybe i ll try making some of these ;)

thx

[expansionvalve]

Great looking design DetroitAC, simple, easy assembly, good on the pocket I should imagine, I very much look forward to seeing one in use, the first cap in cap design I have seen built this way I think, I very much look forward to seeing how she performs :)

[Brettbeck]

Really nice design DetroitAC.

Probably one of the nicest looking cap-in-cap evaporators i've ever seen :)!

[DetroitAC]

Thanks Brett, it is nice looking 'cause it's a bit of a contradiction. Precision CNC machining for an otherwise cheap and simple design!

[epion2985]

Its not very different from any other cap in cap evap. Is there any proof of actual performance inclease to justify the shiny base plate machining :)

[DetroitAC]

Its not very different from any other cap in cap evap. Is there any proof of actual performance inclease to justify the shiny base plate machining :)
No, there is no proof of any performance increases. That'd be like putting a Kia Sephia through the paces on an F1 circuit, to see what she'll do. It's not about performance, it's about basic functionality and cost.

[Blaster]

my 1st evap was a cap in cap :)

what i dont like in them is that there will be brazes inside :p

[Jack]

Why not use a round base, instead of the square piece you're using now?
Square evap will be harder to insulate :)

[n00b 0f l337]

This also makes it hundreds of times easier to braze.

[DetroitAC]

Why not use a round base, instead of the square piece you're using now?
Square evap will be harder to insulate :)

I agree Jack, I will make a 1.5 in diameter round version with one cap eliminated (so 5 caps). I've also found that I can reduce total machining time by machining down two of the caps, which allows me to make it all internal bosses to mate with the cap ID's. It'll be cheaper to make this way, so cheaper to sell.

Thanks n00b!

[pythagoras]

Hi Detroit,

Could you do me a favour and build me an evap to test on your calorometer when its done?

Simple tube with two end caps brazed on the ends. I have a horrible feeling that it wont be the worst performing:(

Regards

John.

[Brettbeck]

So are cap in cap evaporators still as effective if they are mounted sideways in a case?

[jinu117]

Yup... if it's going to be cap in cap might as well reduce number of caps to braze on. It's not cheap to get decent sized caps (a lot of them) nor easy to find on some sizes. Kind of defeats the purposes.

[wdrzal]

Copper Caps are relatively cheap,at least up to 1-1/2".

I can picture a design that only uses the outer cap with the inner chambers made with sleeves.

This would require to braze the bottom in a traditional method but put filler rings around the inaccessible inner sleeves ,so the evap can be flipped upside down and internal brazing can still be achieved to the top single outer cap. Heat control would be critical in this method.

[DetroitAC]

Hi Detroit,

Could you do me a favour and build me an evap to test on your calorometer when its done?

Simple tube with two end caps brazed on the ends. I have a horrible feeling that it wont be the worst performing:(

pythagoras, If you could send me a good description of what you want (by PM), I'll send you a quote to fabricate it and braze it. I don't think I can quote testing time until I have the calorimeter running and I know how much time setup and one condition requires.

I've redesigned, here is a 1.5 inch diameter round version. I made the three "teeth" on the perimeter of each step much wider, they are now about 0.1 inch wide, not really "teeth" anymore more like "lands". That worked much better, I had to tap each cap into place on the block with a plastic hammer (light taps), a piece of wood would work also. I could also pry the caps off and put them on a second time with OK fit. I doubt you can see the "lands" in the pictures, I didn't mark them.

http://img246.imageshack.us/img246/2876/evapv9005croppo5.th.jpg (http://img246.imageshack.us/my.php?image=evapv9005croppo5.jpg)
http://img246.imageshack.us/img246/7895/evapv9006cropyv0.th.jpg (http://img246.imageshack.us/my.php?image=evapv9006cropyv0.jpg)
http://img246.imageshack.us/img246/1498/evapv9007crophg2.th.jpg (http://img246.imageshack.us/my.php?image=evapv9007crophg2.jpg)

[THE JEW (RaVeN)]

Very nicely done. Have you decided on how to route the gas flow thru the caps?

[DetroitAC]

I can actually see several ways of doing it, and arguments why each has advantages.
Once my calorimeter is up and working, I may build them several ways and see which works the best. i also might not get around to testing them for some time. I'm working on some other designs that I hope are high performance, and I think it makes more sense to spend development efforts on those designs first.

[wdrzal]

Thanks Brett, it is nice looking 'cause it's a bit of a contradiction. Precision CNC machining for an otherwise cheap and simple design!


Copper $2.00...... copper caps $3.00 ...............CNC milling machine $250,000.00













Someone to make you parts.................."Priceless" :D

[DetroitAC]

I wish I had a machine that nice Walt :cool:

Here's the latest revision, thanks to n00b of l337 for some good suggestions.

This time, the caps get closer to the cpu as they get smaller. Smallest cap in the center is only 0.08" (2mm) above the cpu. Caps are held by steps that engage with their OD, so the caps will need a bit of cleaning up with a file to take off dings, and make a slight chamfer. Like before the "teeth" stick out 0.002 from the base circle to interfere with the cap and hold it inplace for brazing, while the base circle makes a 0.004 gap on the diameter for a proper brazing gap.

The last cap has an odd height, and is supported by it's ID so that no caps need to be cut down for this design. The caps do need to have holes drilled for the suction line, injection line, and circuiting the flow. A hand drill should be fine for this.

I've also added a step on the OD to hold a 1.5" snap ring.

BTW, I didn't push any of the caps down into the base, as you can only do that once and have an interference fit. the copper is soft, and the teeth don't hold as well the second time. Even so, you still get the steps to hold them in position if you do screw up the assembly.

With some of my other designs, I've been adding chopped up stay silv 15 pieces to the interior, and it works fine as long as the part gets hot enough to flow the filler. I *think* pieces evenly spread around each cap might work to braze the evap in one shot with purging. I wish I could get brazing rings of Dynaflow or 15, still working on that one...

This is very quick to machine, very low cost, I'll post in the for sale forum once I find it...

http://img504.imageshack.us/img504/7973/evapv12001qf4.jpg
http://img213.imageshack.us/img213/8081/evapv12002mc3.jpg
http://img165.imageshack.us/img165/5382/evapv12003tf4.jpg
http://img512.imageshack.us/img512/2668/evapv12004mj5.jpg
http://img512.imageshack.us/img512/4223/evapv12005qq7.jpg

[n00b 0f l337]

The final caps on that last outsidge ridge? Hmmmm thats going to be a bit akwards to braze, the braze may flow into the ridge for the clip... Should perform well though for a budget evap, how many caps go into this design?

[DetroitAC]

Yeah, i thought the same thing about filler flowing into the groove, so I made the groove twice as wide. It still might get partially filled, but there won't be that much to file away.

I put the steps to hold 5 caps, of course any of them could be eliminated for a simpler evap.

[wdrzal]

They make silver solder "rings" or sometimes called "wedding bands" the could be placed around the caps during assembly and theywould be sure to stay in place. By some SS wire instead of the flat sticks and make your own.

If you have the time ,making a high frequency induction heater would be a great way to braze your deign. I been thinking of making one. but dont really have a use for it.

[n00b 0f l337]

My cap in cap setup arrived today (quick shipper!)
Here's some more pics just because ;)

http://img518.imageshack.us/img518/2747/dscn0381dq5.jpg
http://img518.imageshack.us/img518/8215/dscn0379zl0.jpg

[DetroitAC]

I brazed one of these up today. This one is circuited with a 3/16" tube injecting to the smallest cap, suction out of the largest. I'm using flare fittings since this one will be for performance testing on my rig.

Each cap has a ~2mm hole drilled near the base, and these are 180 degress rotated from the previous, so the flow keeps high velocities near the base. The tops of the caps aren't sealed, so no doubt some flow goes over the top, but around the sides is the easiest path. The exception to this is the smallest cap, this one had a 1/4" hole drilled (radial gap to the injection tube) instead of the tight fit holes in the rest of the caps.

I used a plastic face hammer to tap the caps into place. For the two largest caps, i had some trouble getting them straight and started, so i used an arbor press. I'm gonna have to make that easier somehow... machine chamfers maybe.

The injection pipe is a 3/16 tube access fitting. I cut some notches with a dremel so that I could just push this all the way to the base (no worries about things shifting and getting brazed shut). I measured the tube length and the depth that it would go in, so that I'd know when it was fully "seated".

The suction flare fitting I milled away to leave only a little piece remaining. Same purpose, to keep that flow passage open if things shift during brazing. I cut some homemade braze rings out of stay silv 15 that I annealed to make softer and easier to bend. I think it was way too much filler, but it worked really well during brazing. I put in a picture of my purging rig, I used a Dwyer flowmeter that I had lying around, it has a built in needle valve that lets me dial the flow very low, and basically not worry about losing too much nitrogen wile I go off to do something else and let the part cool. You can see the little black ball by the 2 liter/min mark. I use 0.5 lpm to braze, i just turned it up so it shows up in the picture.

After it cooled, leak tested to 250 psig :D (overkill, but I needed to check the proof pressure of the part)

Next, I'll swap the circuitry and inject to the outer cap and, suction to the smallest.
http://img512.imageshack.us/img512/7122/evapv12014ic7.jpg
http://img426.imageshack.us/img426/7446/evapv12021jc2.jpg
http://img426.imageshack.us/img426/6793/evapv12022ws4.jpg
http://img244.imageshack.us/img244/2581/evapv12023kd9.jpg
http://img181.imageshack.us/img181/5740/evapv12011jy5.th.jpg (http://img181.imageshack.us/my.php?image=evapv12011jy5.jpg)
http://img426.imageshack.us/img426/5817/evapv12013ug1.th.jpg (http://img426.imageshack.us/my.php?image=evapv12013ug1.jpg)
http://img512.imageshack.us/img512/3402/evapv12016qp2.th.jpg (http://img512.imageshack.us/my.php?image=evapv12016qp2.jpg)
http://img124.imageshack.us/img124/2426/evapv12017kf7.th.jpg (http://img124.imageshack.us/my.php?image=evapv12017kf7.jpg)
http://img124.imageshack.us/img124/2171/evapv12018ra7.th.jpg (http://img124.imageshack.us/my.php?image=evapv12018ra7.jpg)
http://img109.imageshack.us/img109/1390/evapv12019vj8.th.jpg (http://img109.imageshack.us/my.php?image=evapv12019vj8.jpg)
http://img124.imageshack.us/img124/8995/evapv12020om0.th.jpg (http://img124.imageshack.us/my.php?image=evapv12020om0.jpg)

[n00b 0f l337]

Nicely done with the rings of braze. I'm going to do a 3/8" pipe through the dead center for suction (and just drop in a 1/4" pipe attached to a NPT for input of purging gas) and do one cap at a time. Between each cap I'll let it cool and get rid of all outside oxidation. It's probably easier that way for me since I dont have a nice brazing station and I only use a mapp torch.

[wdrzal]

I used a plastic face hammer to tap the caps into place. For the two largest caps, i had some trouble getting them straight and started, so i used an arbor press. I'm gonna have to make that easier somehow... machine chamfers maybe.


Put the caps in the freezer for several hours , then heat the base to serverial hundred degrees.


The cap should drop in, then stay there with a tight mechanical fit.;)

[DetroitAC]

Yeah, I reckon you could do that Walt, it'd get roughly 0.003 more clearance on the diameter (back of envelope calculation). That's not exactly convenient though.

I think I'll try to fix the parts first. Ideally the parts should do what they are intended to do, with no "workarounds".

[DetroitAC]

This is basically the same thing as the above sample (which was for testing), but in a form that is suitable to go in a DD system with a suction tube in the center of the evap and with an attached capillary tube.

This can be built either:
inject liquid to the center cap and suction from the outside cap,
or inject to the outside cap and suction from the center cap

It looks almost identical in both cases, except in the capillary either goes into the suction line, or into the outside cap.

http://img175.imageshack.us/img175/1870/evapv12051ir7.jpg

[n00b 0f l337]

Will you be selling the item in the brazed form, also, did you braze it with the steel clip on?

[DetroitAC]

Will you be selling the item in the brazed form, also, did you braze it with the steel clip on?

Yup, I made another post to the for sale section, and I got some cadmium plated snap rings, so they won't rust like the carbon steel. They've been brazing pretty well with the rings of stay silv 15, every one so far has been leak free. I braze without the snap ring, and since the filler is coming from the inside through the braze joint, it is thin and flowing when it gets in the snap ring groove, so it's very little change to the diameter.

[DetroitAC]

Circuiting for the Inject to the center cap configuration:

capillary is brazed to the smallest cap, shielding gas runs through the capillary, so i just need to clean up the outside of the cap
http://img127.imageshack.us/img127/6927/evapv12044cq8.jpg
A hole is drilled in the side of the center cap near the base after it is cleaned up, and all of the other caps are drilled with holes near their bases.
http://img174.imageshack.us/img174/6739/evapv12053uc3.jpg
Each cap is pressd into the base, and has braze rings around it, and a small bit of filler metal drilled and slipped over the capillary to braze it to the cap.
http://img109.imageshack.us/img109/1390/evapv12019vj8.jpg
http://img128.imageshack.us/img128/3428/evapv12046pe8.jpg
circuiting looks like this, each hole is out of phase with the previous by 180 degrees, so the refrigerant has to travel around the perimeter of each cap to get to the next hole. Since the holes are near the base, this hopefully makes the semicircle path close to the base lower resistance than going over the top, so hopefully the liquid stays near the base where the conduction path is to the base is short. The outside cap has the suction pipe at the center, so the flow probably goes somewhat around, somewhat over the top to exit.
http://img127.imageshack.us/img127/7863/v12evapcircuitcapcenteree0.jpg

Cicuiting for the inject to the outside cap, suction from the inside cap, is mostly just the reverse. Holes are smalles in the outer caps, biggest in the inner caps. The suction pipe is flared and staked into place in the smallest cap, so this evap is brazed all at once.
http://img282.imageshack.us/img282/118/evapv12032bd8.jpg
http://img127.imageshack.us/img127/4931/v12evapcircuitdl2.jpg

[Marvin]

Just receive mine from DetroitAC.
Very well machine finished, i loved it.
pics : (i know i will have a lapping time)
http://img120.imageshack.us/img120/8680/p1010014ji2.jpg
http://img120.imageshack.us/img120/7830/p1010008ec3.jpg
http://img120.imageshack.us/img120/5371/p1010009br8.jpg

thanks

[n00b 0f l337]

Its easy to braze isnt it?
I'm going to get em cut down a bit and used for some GPU's.

[Marvin]

Cleaned and lapped
http://img219.imageshack.us/img219/2020/p1010021ea3.jpg
http://img219.imageshack.us/img219/8205/p1010023xl6.jpg

[n00b 0f l337]

Did you braze with the steel ring on? I brazed without it on and now I cant get it on. Luckily I got o-rings coming.