Expansion valves vs capillary tubing?

[AirConditioner]

I'm building a very simple experimental water heating/cooling system just to practice and test various ideas before I build something larger.

I have a 5000 BTU R22 air conditioner, less the R22. (damaged condenser) I'm going to remove the compressor and attach a 25' coil of 1/4" copper tubing for both the evaporator and condenser, each of them simply dipped into a 55 gallon drum full of water. One drum will get hot, the other will get cold. It will be charged with propane instead of R22.

My question is about the expansion valve. It currently has a long capillary tube, and if I am thinking correctly that will no longer be sized correctly if I modify the evaporator and condenser, right?

I could:
1. Save the existing capillary tubing and use it with the new system.
2. Modify the capillary tubing to make it work with the new system.
3. Replace capillary tubing with thermal expansion valve.

If I chose option 2, how do I know how to make it the right size? Trial and error? Should I just make it longer or shorter until I get the desired results?

If I chose option 3, what happens if I use an expansion valve rated for more BTUs than my compressor? Most of the ones on ebay are a ton or more, not 5000 BTU.

[TechShop]

There is a less complicated approach. Why not duplicate the design conditions of the original copper condenser, thus eliminating the need to change the metering device.

First, measure as closely as you can the width of the condenser. Particularly, the length of each pass of copper tubing.

Next count up the number of cross-tubes in the condenser and note how they are organized. Often, there is a T or Y on the inlet and outlet; splitting the flow into 2 or 3 parallel circuits. (Probably two circuits on a little 5K).

You can multiply the length of the cross tubes by the number of tubes to determine the total length of copper necessary. Then cut that copper into equal length parallel circuits as required. You may even be able to salvage the T or Y fittings from the original condenser.

As for your last question about your 3rd option... I believe the reason you don't see small metering valves is that most manufacturers do not use them on small systems. I believe it is simply due to the extra complexity, and added cost of manufacturing. The reliability goes down and the cost goes up. I believe these little window units are mostly filling a price point and the cap tube is the inexpensive, reliable choice.

[TechShop]

I measured an LG brand 5K unit I have here in the shop.
It has 2 parallel circuits of approximately 19 feet of copper that measures .215" OD.

Since that tubing is smaller, I would do the following math to match the internal volume using 1/4" copper:

area of .215" OD copper = .1075^2 * pi = .03631
area of .250" OD copper = .1250^2 * pi = .04909

ratio of volume will be = .03631 / .04909 = .7397 or 73.97%

So due to it's larger diameter, I only need 73.97% of the length if I'm using the larger size 1/4" copper tubing.

19' (per circuit) * .7397 = 14.05' (per circuit).

That means a 28 foot total length of copper cut into two circuits. In my mind I could justify whacking that down to by 1.5' per circuit to use a 25 foot coil. By my math, I would suggest you make two 12' circuits out of a 25 foot coil and stick with the existing cap tube.

Of course, you should probably check those measurements against your condenser and also use the ID of the copper, not the OD.

[superlen]

Since I stumbled onto your thread first (Edit errr second. - Tech posted in while I was typing), I'll post my thoughts on this. Warning, I'm not a refrigeration guru by any means, but I have done enough research to toss about ideas like I know what I'm doing.... I don't. Anyone feel free to correct me.

I too, like most, have pondered the effects of changing the size of your heat exchangers (HEXs - evap or cond) . If you spent any time on hvac-balk you'll know that you can never ever ever change the size of either HEX on any system ever. Why would you? what are you a dumbass? we all know they have to be matched...and installed by a licensed hvac contractor.. you shouldn't even be asking this question, only cheapskate diy guys wonder this.. you should only touch the thermostat after a licensed hvac tech has approved you. Of course by the fact that we are in this site, we know all that is BS.

Having made fun of that position, there obviously is a reason you can't just change one major piece of the puzzle and expect it will work perfect. It might, might not. In your case, if you changed both HEX from air to water & somehow managed to keep the amount of heat rejected/accepted approximately the same as what the original system was designed for, your stock cap tube should be sized appropriately. It's when you drastically change the heat transfer rate on the HEX is when you now have to consider what the ramifications are.

A cap tube or TXV both do the same thing. They are the metering device (MD) that sets the flow rate of refrigerant through the system. More flow, more heat moved, less flow less heat moved. In a cap tube system it is sized for a certain amount of BTU transfer of both the evap and the cond. It shouldn't matter what the physical evap or cond is, water, air, cooling tower, chunck of earth, solar panel, ect., as long as that HEX moves the BTU Rate that the MD is expecting things should be ok. Note: It's not the size of the HEX, its how many BTUS/HR it can move. A 4t 10seer AC from 1990 has an outside condenser size of X & a big powerful fan to move a lot of air across it. The same 4T 16 seer system from 2015 has a much larger condenser, so the fan can move less air. The same amount of BTUS/HR are moved with both condensors, yet the 16seer with it's smaller fan uses less total power, hence the efficiency boost. In theory you can have an air condenser that is MASSIVE and convection alone would move enough heat (no fan - very efficient... a DX system operates similar to this with earth conduction). In practice other variables with this fanless system cause issues, sheer size, mnf cost, oil transport, more friction of refrigerant over more lines, others I haven't thought of. I still am wanting to play with this exact concept on my home AC (use existing condenser coil to dump as much heat as possible by natural convection - no fan - then let a second water condenser take up the remaining btu rejection required).

So back to your question and system, I think the way to approach it is like this:
1. Start with evaporator and figure out if you're new evaporator is going to move more heat or less heat than the original. If more, you will need more flow rate (bigger diameter, shorter tube). If less, the opposite.

2. Make your condenser design so that it can move the same amount of heat (or more) than your evaporator.

3. Start experimenting.. I think the above two statements will keep you close to a working systems. Others, with more real world experience than I should chime in.

Your setup is going to be a little more difficult to predict as you have two tanks of water that start out a certain temp. There is no flowing water in this case so the rate of heat movement is not fixed. It will start at one rate and as the water heats/cools the rate of heat movement will decrease over time which of course would change what size MD you need.

Another note as for using too large a TXV. It will hunt around if it is over sized, modulating from fully open to fully closed. A TXV is more forgiving of load changes and mis-matched HEX than a cap tube is, but even still it needs to be somewhat close. I *think* on larger systems 3,4,5T you can go up or down 1T and still be in range. On a smaller system, this would of course shrink. The TXV valve datasheet will list details that should help.

Hope some of this was more helpful than confusing, and again anyone feel free to point out errors in my logic.

Len

[jeff5may]

Aircon,

Welcome to the madness!

My suggestion is this: set up your test rig with the barrels and copper tubing. Make your connections to the suction and discharge lines at the compressor with an access valve in each and a sight glass in the suction line. Chop out the cap tube, leaving a few inches of tubing at each end. Slide flare nuts on each end, and flare the ends of tubing on both sides of the cap tube. Use a flare union to connect the cap tube assembly to the evaporator tube. If you will be using a filter dryer, get one with male flare fittings and connect it to the cap tube. Put a flare nut on your condenser outlet tube and flare the end. Connect the flared end to the filter and pressurize the rig. Vacuum, purge, vacuum, charge.

The cap tube you have will work for a first shot to prove operation. Let us know what happens.

[AirConditioner]

Thanks for the info everyone!

Originally Posted by superlen

If you spent any time on hvac-balk you'll know that you can never ever ever change the size of either HEX on any system ever. Why would you? what are you a dumbass? we all know they have to be matched...and installed by a licensed hvac contractor.. you shouldn't even be asking this question, only cheapskate diy guys wonder this.. you should only touch the thermostat after a licensed hvac tech has approved you. Of course by the fact that we are in this site, we know all that is BS.

LOL! I totally know what you mean. HVAC guys and electricians both think they are god. I can't say I've ever hired either one of them to do something for me! (Nor would I dare ask them for advice)

Originally Posted by jeff5may

My suggestion is this: set up your test rig with the barrels and copper tubing. Make your connections to the suction and discharge lines at the compressor with an access valve in each and a sight glass in the suction line. Chop out the cap tube, leaving a few inches of tubing at each end. Slide flare nuts on each end, and flare the ends of tubing on both sides of the cap tube. Use a flare union to connect the cap tube assembly to the evaporator tube. If you will be using a filter dryer, get one with male flare fittings and connect it to the cap tube. Put a flare nut on your condenser outlet tube and flare the end. Connect the flared end to the filter and pressurize the rig. Vacuum, purge, vacuum, charge.

The cap tube you have will work for a first shot to prove operation. Let us know what happens.

That's what I've been starting to think, just put the cap tube on flare connections, then it's easy to change out as I experiment. I've got a bunch of copper tubing, flaring/swaging tool, and other odds and ends on their way from ebay.

I was going to attempt to copy the same copper tube size in my loop roughly as what the original HXs had. Although I was assuming it is 1/4", I should measure to make sure it's not some oddball size. The way it was plumbed from the factory, the hot side is one long 1/4"ish tube, the cold side is two shorter circuits in parallel.

My end goal is to have a machine that will make me a big tank of either hot or cold water during "off peak" hours when power is cheaper, at as high of efficiency as possible. Then I don't need to run heating/cooling equipment during peak electric rates, just circulate the water I already saved up. May also experiment with geothermal, I'm lucky enough to have an extra well I'm not using.

[AirConditioner]

Let's say I put in a cap tube that I think is just a little bit too large. Could I just carefully add a crimp in the cap tube after it is installed to fine tune it?

[mejunkhound]

make me a big tank

How big a tank?
5000 btu/hr for 8 hrs = 20,000 btu = about $1.00 worth of electricity for resistance heat, only 30 cents for HP input.

You mentioned 55 gal bbl. About 500# of water. Say you can cool the water down to 75F during the day getting the heat out, heat it up to 125 F during off peak

That is 25,000 BTU; however, you have also cooled your other bbl down 50F, etc..
do you see where this discussion is headed?

BTW, go TXV, about 3X too big is still OK, esp if you find a cheap one. Forget cap tube unless just for 'playing around'. Crimps in tube wil work also.

[jeff5may]

Browsing around on ebay today, I found this:
Sporlan EGV 1 2 C 5638275 1 2 Ton TXV 5" Cap Tube with Clamp R 22 150302 | eBay
From one of my favorite hacking sources, Surplus City Liquidators.
Further down the page, another gem, same dealer:
Alco BAEB1 2HCA 71070050 1 2 Ton R 22 TXV with External Equalizer 60890 | eBay

While you're at it, buy some more hardware. They combine shipping. If you don't especially like ebay, they are a brick and mortar store with a website and a 1-800 number.

Another suggestion I have during the trials/proving process is this: instead of flare unions at either end of your cap tube assy, use ball valves. These can be bought as a set of 3 for use with charging hoses. To swap out metering devices, simply valve off the device and break the seal on the downstream flare. Swap out devices and install the upstream flare. Crack open the valve slightly to purge the device, then install the downstream flare. Leak check and run the new device.

For rough guesstimating cap tubes, it is better to run multiple lengths in parallel if you are going to crimp them off. It is easy to crimp off a cap tube, not so easy to un-crimp to reverse the action. That being said, I usually do the ball valve thing and connect a chain of shorter tube assemblies in series with flare unions. Flare nuts and unions are not free, but they are reusable, and so are the cap tube assemblies once you make them.

[AirConditioner]

Originally Posted by mejunkhound

How big a tank?
5000 btu/hr for 8 hrs = 20,000 btu = about $1.00 worth of electricity for resistance heat, only 30 cents for HP input.

You mentioned 55 gal bbl. About 500# of water. Say you can cool the water down to 75F during the day getting the heat out, heat it up to 125 F during off peak

That is 25,000 BTU; however, you have also cooled your other bbl down 50F, etc..
do you see where this discussion is headed?

Eventually I think I would like to have a 275 gallon tote each for the hot and cold side, and a larger compressor, maybe multiple compressors. In the winter I need a lot of dehumidification in an unheated building, I want to use the cold water for that. I can use the hot water to heat a different heated space. Then, in the summer, I'll cool the heated space, and probably use a pump and dump type geothermal system for the hot side. Don't have humidity issues in the summer...

I want to use the 55 gallon drums to experiment with since it will be relatively easy to weigh the water and accurately calculate the BTUs on each side.

Somehow I hope to also pull DHW out of this too. Every Sunday and holiday power is "off peak" for a full 24 hours, so it would be nice if I could take advantage of that.

Originally Posted by jeff5may

Browsing around on ebay today, I found this:
Sporlan EGV 1 2 C 5638275 1 2 Ton TXV 5" Cap Tube with Clamp R 22 150302 | eBay
From one of my favorite hacking sources, Surplus City Liquidators.
Further down the page, another gem, same dealer:
Alco BAEB1 2HCA 71070050 1 2 Ton R 22 TXV with External Equalizer 60890 | eBay

Very nice! With ebay it's just a matter of actually finding that thing you know they must have...