Steve hull , the water heater is set to only 120. I'm very sure it could go higher though

Which ever way you go, oversize the refrigerant to water heat exchanger! Much like the evap coil you revised in your dehydration unit, too much is almost enough. This is your heat transfer mechanism, and with more transfer comes higher efficiency. Just a few degrees less drop from inlet to outlet on either side saves you money on your power bill.

You also have the freezing point of water to stay on top of. During the winter, when you need heat the most, your pond water may be precariously close to freezing. A few extra square feet of transfer area could keep you from using a backup heat source. A few less could spell disaster due to a rupture.

Well if i'm going to be saving so much money not buying so much copper, I don't mind building it over sized. I've done some looking but can't seem to find where anyone has build a coax coil for 4 ton. Any bookmarks ya'll have stashed away that could save me some hours on google?

I just lost out on a dead water cooled unit with a 5 ton coax coil. The unit this was in got sent to the dumpster as the scrap value was too low. Was about two hours too late . . .

My suggestion would be to find a ClimateMaster or WaterFurnace dealer in your area and then bring over goodies to allow you to scrounge through their boneyard. This is how I got my one ton unit that I rebuilt.

I took the shop guys out to lunch today for their help and they will be looking for all kinds of goodies for me. Cost me about $57.

Worth EVERY penny . . . .

Steve

Totally doable BUT there are ZERO local installers for those systems. My closest bet would be Memphis, TN..... and I really don't like going there much less making friends. I do have tons of friends in the HVAC business, maybe they know people from areas that use them more. It would be worth a road trip I think.

But if I did make it this is what I was thinking. Pipe in pipe. Center pipe is for water with larger pipe around it filled with R290. Do several of these tied together.
http://ecorenovator.org/forum/attach...1&d=1437014715

There have been several DIY HXs similar to your drawing, done here at ER.

Acquario did a very interesting unit, that used a plastic shell and had multiple copper coils running inside, for the refrigerant. His first try had insufficient copper tubes, so he rebuilt it with more tubes, and got the results he wanted.

http://ecorenovator.org/forum/40959-post9.html


Also, randen did an interesting one wherein he used large flex pipe for the shell and copper inside for the refrigerant. You'll need to search for that one. The part that made that one work was machined brass end plugs that he could braze the refrigerant lines to, and ot also had fittings for water. With your amazing hack brain, you'll likely come up with an off-the-shelf solution.

Your biggest problem will be maintaining a hermetic seal of the refrigerant carrying portion.

The pressure involved in the water portion will be in the very low psi range, whereas the pressure of the refrigerant will be very high, and will be subject to severe temperature cycling.

So, if you're gonna do tube(s) in shell, think through your tube structure, and how it must be constructed, then think how you will need to wrap a metal or non metal structure around it, that will allow your water flow and maintain a seal at 10 psi or so.

Also, there is some flexible pipe material that is intended for connecting natural gas to a stove or something. This stuff is interesting because it has a high surface area, and would encourage turbulent liquid flow, which will increase efficiency. I have seen it in stainless and also in copper. The copper variety should be easy to braze. The stainless could be brazed also, with care, high-silver brazing rod and proper flux.

Rave on!

-AC

Here is a guy who is building a counterflow tub in tube, with some interesting additions.

Please note that this HX is for liquid-to-liquid not refrigerant-to-liquid, so the issue of hermetic-sealed refrigeration-grade quality doesn't come into play.

https://www.youtube.com/watch?v=DVf-lTFpR2c

-AC

Thanks AC!
So after a bunch of brain storming this is what I have come up with. I suck at drawing 3d so bare with with as I explain.

I have 100 feet of 1 inch rubber industrial hose. Its rated for high pressure and high heat. My idea is to run 3/8 (or smaller/bigger) copper tubing inside of it and then coil the entire thing into nice loops. The ends are threaded so I would use a metal female 90 and drill a 3/8 hole so that the inner tube can come straight out and then braze that. It really would be like a tee, with the water flow making a 90 degree turn and the copper continuing straight out. I would do that on both ends of the hose. Giving me 1 inch of water flow over 100 feet of 3/8 copper tubing.
When I thought about it, it really is simple. My first idea that I posted had the refrigerant on the outside therefore the crazy pressures where acting on the copper tubing as well as the water pipe. When you put the water on the outside you only have to make the outer tubing water tight. I think this could even be done with pex, only downside is the pex is a pain to loop into tight loops, but if size didn't matter than pex should work great.

The main question is now (and what I am currently researching) how much surface area of copper tubing do I need with how much flow of water at 55-65F temp water.

EDIT: I just watch the youtube video. This is very very close to what I was thinking. Awesome idea with the wire on the outside. I wonder what would be better, parallel or counter flow? More research to do.

Question about flow answered. COUNTERFLOW
Parallel and Counter Flow Designs Heat Exchangers | Engineers Edge | www.engineersedge.com

Counter flow always better.

I can't tell you right out of the gate how much surface area, but if you are looking at examples, you may come across water-to-water examples. You might be able to do an extremely general estimate from those examples, but refrig-to-water will require more surface area.

Come to think of it, randen's first unit was about 2.5 or 3 Ton. He had a diy HX like you are considering. Check his specs... they worked.

If you are going with open loop (pump & dump), brazed plate will not do, it will quickly foul. But if you are going closed loop, Brazed Plate HX are very much worth looking into... but tube in tube will be OK, too.

-AC