Another hot water tank heat pump
 

Anyone that's been following my post on the solar space heating for shop had been introduced to the items that presented themselves to me. A slightly used hot water tank and a air-conditioner that was pushed out of the window landing on its condenser.
From here its not a great leap to know what's going to happen next. We're going to hack this into another heat-pump hot water heater.

The weather here in South-Western Ontario has been very rainy and I'm still waiting for some materials so work on the solar panels have been paused and in-door projects prevail. Saturday and Sunday produced a condenser with a threaded adaptor to replace the lower element. I'm hoping to be able to guide the tubing in and turn the threaded adaptor in. A prior twist counter clockwise of four turns and tighten into position should straighten the thing back.

Again keeping with the theme of the eco-renovators everything is scrap pieces that are cluttering up the place. Rain will be ending in two days and we can get back to the shop improvements.

In the photos a brass adaptor is turned up using the old elec. element as a model the threads are a 1" straight pipe thread sealed with a rubber ring.
The threads are test fitted in the tank before milling the hex details.
A copper cutting from a 1/2 copper pipe stub out is used to return the condensors end.
The copper tubing condensor is soldered to the adaptor.

Randen


http://ecorenovator.org/forum/attach...1&d=1382804276

http://ecorenovator.org/forum/attach...1&d=1382804276

http://ecorenovator.org/forum/attach...1&d=1382804276

http://ecorenovator.org/forum/attach...1&d=1382804276

http://ecorenovator.org/forum/attach...1&d=1382804276

http://ecorenovator.org/forum/attach...1&d=1382804276

http://ecorenovator.org/forum/attach...1&d=1382804276

Nice work. The turn-around at the end is well crafted... and then there's the threaded brass adapter.

Randen, you're relentless!

-AC

How many BTU is the poor victim rated for? Can you still use the old condenser, or did it let out its charge? I'm asking because I have a similar project thought up. I've heard of others running the entire A/C unit as an outdoor unit to improve performance, but haven't got much follow-up after the fact. From what I've seen, the two heat exchangers in the window unit together have roughly the same heat transfer area as a mini-split outdoor unit of similar capacity.

Once again, you are the man! You could sell those water heater tubes with bung on the internet for the fame and fortune, dude. Looks perfectly industrial to me. They could be repurposed as wort chillers or moonshine condensers for even better marketability. I wish I had a garage with a machine shop.

Jeff5may

The unit is 5kBTU I think? The condenser took a good hit and one of the elbows on the end is crushed but it held a charge. The condenser is bent perfect for the scrap bin. As far as running it outside not an option here in Canada. COP would take a huge hit in the winter.

Thanks for the cheer but spending forty hours a week standing in front of the machines and then time spent there on the weekend sucks. The up side after 30 some years making goodies for the industries things I make for myself can be made pretty with ones eyes shut. Selling these things on-line I think one would be hard pressed to make shop-rate.
I do get the expression when someone gets to watch the machining in action " man what I wouldn't do with these machines in my garage or basement" Well yes it is cool but you can never have enough. After you have a lathe you require a drill press then a mill then drills, endmills, turning tools and the list is ever expanding.

I have been bitten by a bug. Efficiency. After spending so much money on energy and finding these technologies so effective on the bottom line. I`m hooked. It took a oil bill that I couldn`t afford to make the tipping point. I knew about these things Heat-pumps solar, LED, CFLs. But I was so caught up on traditional thinking. I`ll do what every one else does. Work harder and pay that fuel bill. Replace that light bulb with another. The brakes had to be applied hard.

I have to thank people like Gary at builditsolar and the AC Hacker for setting the course. I`m impressed on what solar and heat-pump technologies can do for us.

Randen

Cut out the damaged part of the condenser and use it as a second evaporator. (Put it right after the expansion valve.)

Right, that's what I was saying. You don't necessarily have to put the unit outdoors, just plumb both heat exchangers in the window unit to function as one big evaporator. The increased heat exchanger surface area and airflow will drop the dT on the evaporator side, raising your energy efficiency. Since there are two fans tied to one motor, you could remove one blower wheel to maybe speed up the other, but using both wheels and exchangers (especially the larger, deeper condenser) would be preferred with respect to overall performance.

If the condenser is all bent up, it's a much wiser choice to just get rid of it. Trying to milk something that may or may not hold up usually comes back to haunt you sooner rather than later... just a few days after the thing is all buttoned up. I hate it when that happens.

Did you even say if your new hack would be air source or water source?

-AC

AC

It is indeed air source. First attempt at assembly failed. I put a twist in the tubes so when I went to turn the bung (adaptor) in the tank the tubes would straighten out again. Not so lucky. Tubes kinked!! dam!! Working on a better simpler solution. A photo will explain best.


Randen

Do you think there is enough tube in the condenser to be able to transfer the heat?

I remember when I made my first heat exchanger for the pool it proved to be way too small so heat transfer was minimal.

The second attempt used much more copper (10 metres) in 3 concentric coils and works wonderfully, so much so I've now built 3 of them, 2 heating the pool and the 3rd in the central heating.

The difficulty you obviously have is the limited size of the hole you have to insert the tube and that you have to screw the whole ting in place.

The tanks I've been using are much easier as they have a flat plate and flange washer with bolts to hold the plate in place. I guess you could machine up something like that with the correct tools and save yourself the problem of having to screw the unit into place.

Acquarioe,

I believe 10 meters is a magic number. The airtap hpwh unit uses this length of tubing for its hwt snake condenser. Many, many hobbyists have used this length for their water side exchangers. As long as the tubing is large enough for the refrigerant to flow, I believe one could heat or cool a skyscraper with a 10 meter exchanger!

Acuario & Jeff5may

Thanks for the heads up on the length of the HX tube. As we know longer is better. Already installing that lenth of heat exchanger tubing proved problematic. Using the baseline that the A7 install by Xringer I had modeled the HX lenth. The tubing I used is 5/16 od so a little more surface area by a 1/3 and about the same length. The proof will be in the end trial. Xringers A7 was 7Kbtu mine is a little smaller at 5Kbtu.


Xringer wrote: "I unrolled the HX loop and I think it was just under 15 long. So, less than 30 feet round-trip. The OD is just under 0.22" (5.6mm)."

Randen

Well I must admit it was pure chance/luck. The place where I buy it from sells rolls of un insulated 1/4" tube in lengths of 5 metres.

We need to keep in mind the operational parameters here.

As I recall, Acuario, you are using compressors that are in the 12,000 BTU range or larger, am I correct? Also, you are working to achieve an output temp of maybe, 80 degrees F, am I correct? So your compressors are larger and your lift is lower.

Randen's compressor is less than half that size, but at the same time the maximum temperature is going to be maybe 110F-ish. So his compressor is smaller, but his lift is higher.

Jeff5may, if you are right regarding the 'magic number', it's probably because of the lucky combination of smaller compressor and higher lift.

-AC

SHHHH...don't spoil the magic. It doesn't work if you explain it!

Hi guys,
Randen, I made the same fittings you made except for the end return. I suspect that you took a wall stub for roughing in sink and cut the end off for the spun end. It looks nice but I am a bit concerned about its pressure rating for refrigerant. I guess we will see. Also i would make sure the anode is really good or you will loose your refrigerant within a short time. Another option would be to coat the copper in an epoxy paint. Effectiveness may be SLIGHTLY lessened but it may prolong the life quite a bit.

I was making the HX for small solar systems and the problem I had was spinning the big coil into the tank so I made a union type fitting instead which was easy to make a water seal on.

About machining, landlord is selling the building so I got the boot. 600V, 3700lb lathe and mill has to go and I am looking for really nice smaller ones that can run on 240V and fit in the (5000ft down to 600ft) new, perhaps temp, space.

another hot water tank heat pump
 

The assembly of the copper HX coil into the HWT didn't work out so well. I had pre-twisted the coil counterCW and piled the copper tube into the tank like the A7 HX installation, but trying to turn the adaptor in caused the tube to kink. I'm going to try to cork-screw a different orientation of coil in and try to turn the adaptor in like the original electric element.

The length of the HX has been subject to some controversy. The length of copper is 22ft. of 5/16 and if what Xringer had indicated was in his A7 for a heat exchanger this coil may have more surface area with a smaller compressor. I say lets giver a try.:thumbup:

Randen

http://ecorenovator.org/forum/attach...p-sam_2417-jpg

http://ecorenovator.org/forum/attach...p-sam_2418-jpg

When I saw your fitting, I wondered about twisting it into the tank..

http://i46.photobucket.com/albums/f1...W/P1000706.jpg

The A7 comes with a two piece fitting, the only part that turns is the retaining/collar nut.
http://i46.photobucket.com/albums/f1...W/P1000707.jpg


And it will leak a little, if you don't crank it down tight.. (Easy does it)..
http://i46.photobucket.com/albums/f1...W/P1000710.jpg

PS: That little nut with the small tube for the heat sensor blub also leaked.. Had to be tightened a little.

What refrigerant is the compressor designed for and what are you going to use?

If it's a R22 compressor, R290 (or more accurately, R433b, a R290 based mix designed to replace R22) will be nearly a drop in replacement, but expect about 125F maximum hot water temperature. It would work very well for showering, but dishwashing effectiveness might be less than you like.

If it's a R410a compressor, you'll get about 2/3 the capacity after giving it the Davuluri Treatment, but 140F should be easy to attain. Note that you'll need to watch your discharge temperature if you do that. Counterintuitively, that's more likely to be a problem with lower evaporating temperatures.

In either case, you'll want a TXV since a cap tube would not be very well suited for such a wide range of conditions. You'll also want some sort of remote control to allow the temperature to be set as low as 105F in order to get more efficiency under more normal conditions, but allow it to be boosted for dishwashing. You might also want to look into replacing the rather inefficient PSC fan with a Delta with Cindy Wu technology. You'll be surprised just how inefficient smaller PSC fans are.

Randen,

Thank you for posting the details & photos of this attempt, even though it was not a success.

This provides a lot of information for all of us.


BTW, in the thread posted by Hv23t, The tubing was 1/4"... The diameter you're using is 5/16". It's possible that the slightly smaller tubing that Hv23t used was just enough more compliant to make the turn.

His fitting was almost identical to the one you made...


And his procedure was very nearly the same as yours:

Pretty similar...

Best,

-AC

I think the main difference is that the soldered joint turns the tube as it's being tightened, while the compression fittings do not. Use plenty of sealant (or Teflon tape) on the compression fittings since they tend to leak. That's especially problematic on the discharge line due to thermal cycling.

If you read the thread of the former attempt, you would see that in both cases (randen & Hv23t), tubes are being twisted inside the tank. The compression fittings that Hv23t used are for external connections, and have nothing to do with the tube on the inside of the tank

In the case of Hv23t, the the tubes did not collapse, in the case of randen, they did... what's the difference?

Using a threaded fitting to affix the copper tubes inside the tank is a very straight forward approach. Obviously, it can be made to work.

-AC

The tubes do twist around each other, but if they're soldered, they would twist themselves. While with compression, they would not.

He was using the compression fittings as seals. They would not grip the tubing until tightened.

This is an excellent point...



However, (and I am not referring to Xringer's airtap, I am comparing the attempts by randen and Hv23t) the twisting failure is taking place inside the tank.

Xringer's Airtap uses a custom fitting that would be difficult for most DIY people to make, unless they had a machine shop. It does turn out that both randen & Hv23t do have machining equipment at their disposal, but have chosen to use a more straightforward approach (that would be accessible to most DIY readers).

The Hv23t attempt worked, the attempt by randen didn't... what made the difference?

-AC

You guys reminded me of that leak I had on my A7..
http://ecorenovator.org/forum/geothe...heater-33.html

I just picked up another 1-peice fitting (3/4" iron pipe to PEX adapter. ID=0.533").
Here's the two piece 'Swivel' that I replaced.. (Also comes in 1")
SharkBite 3/4 in. PEX Brass Female Swivel Adapter-UC530A at The Home Depot

http://www.homedepot.com/catalog/pro...bf4d28_300.jpg

I wonder if something like this could be used? Make a brass slug with
two holes for the copper tubes and braze the whole thing in place.?.
Add a brass nipple to screw into the tank..

If the tubes were too large, maybe the 1" adapter could be used?
And it that was still too small, maybe the ID could be bored out a bit..

The soldered fitting applies a torque to the tubing, while the compression fittings do not. I think that's the problem. The tubes twisting around each other can be compensated for by pretwisting them the other way.

OK, I can't help myself. Here's the difference:

The Airtap unit Xringer installed is a pre-engineered, somewhat drop-in unit. The Water heater fitting is made so you can assemble the whole shabang without much chance of anything bending very much. The Heat exchanger coil slides straight in, and can be positioned while the plumbing is somewhat dismantled. It has a swivel fitting, so assembling the plumbing presents no further twisting once the tubing is snaked into the vessel.

When HV23T made his heater, he snaked the exchanger coil through the rather large hole, leaving a few feet of tubing sticking out. He threaded these pigtails through the fitting, then tightened the fitting while the pigtails were still free to not spin the whole assembly inside the tank. If anything, the ends may have spiraled around each other a few revolutions. Once the through-wall fitting was tight, he then tightened the compression fittings for a waterproof seal.

The heat exchanger randy made is all one piece. Besides the wrangling of fitting the tubing through the bung hole, he had to twist the entire assembly multiple revolutions to tighten the through-wall fitting into the bung hole. Since copper is somewhat rigid, more so the larger the tubing diameter, when the exchanger contacted another surface inside the tank, it tried to crease rather than bend.

Rather than force it in and hope for the best, randy wisely decided to stop and re-bend his exchanger so it will fit. I would have done the same thing. Trial and error, man. That's what it's all about.

For what it's worth, I don't think it will make a whole lot of difference what shape the exchanger is, as long as it fits in the tank. Lanky snake or twisty slinky, as long as you put the thermostat down low in the tank it will sense cold inlet water arriving. Once the unit fires up, convection will mix the water pretty quickly.

OK... I think I see what has happened.

There are two kinds of fittings that rely on compression...

This one uses a small band that is compressed at both ends...


This fitting is used very often in plumbing, for instance to provide a small tube to supply water for refrigerator's ice-making function.

I formerly thought that a compression fitting was a compression fitting, and I used this kind of fitting in some of my early refrigeration experiments. It turns out that this kind of fitting does not stand up to the vibration stresses and extreme temperature cycling that is the everyday life of refrigeration equipment. I know this because one of these fittings failed me and bathed the clothing I was wearing with spraying liquid propane. To say the least, it was unpleasant... to be more honest, I could have died.

Since that experience, I never even allowed the idea of the former type of compression fitting to enter my consideration as a valid type of alternative when working with refrigerant.

It may be possible to make this type work if the compression ring is brazed on (not soldered) to the tubing... maybe.


The other type of compression fitting is a flare fitting...


This type will stand up to the rigors of refrigeration, both to vibration and to extreme temperature cycling. It is commonly used in refrigeration work.

From the looks of things, I now think that Hv23t) has used the former type of compression fitting.

Not a good idea.

-AC

He wasn't using the compression fitting to seal in refrigerant. He used it to seal the water so it doesn't leak from around the tubes.

AC Hacker
The Flare and compression fitting both have their places. A lot of times I think the place for the compression fitting is to remain on the shelf at the hardware store. Their not very reliable. Any flex or vibration they start to leak. However a good flair fitting is golden.

I believe the success of Hv23t was because the tubing was thinner and therefore less rigid and not softening the tubing by soldering or heating the tubing removing its hardness.

I had thought of making up a fitting like Xringers A7 but with so many projects on the go I was wishing for a short-cut. I hope with the new cork-screw type HX I may have a winner. I will keep you posted. These could be relatively easy to make. We'll see how it works out.

Randen

I'd like to say that I went from first type to second type on my Air compressor drain (7 hp 190 psi tat averages 3 hours/ day run time). I was cracking a copper tube every 6~9 months, and now the second style has been troubble free for just over a year.
Suddenly wishing that I had plumbed my pre-lube (pre-start engine oil) pump with 2nd type on my TDI.

Look again.

-AC

Can't really tell from the photos. The tubing isn't discolored at the fittings, though. Either the tubing is solid through the tank fittings (i'm assuming) and compression fittings hold in water, or there is a break in the piping at each fitting (I'm hoping not) and the compression fittings hold in refrigerant and water. Only HV23t knows for sure.

Whether or not it was done that way this time, this is far too common an amateur mistake! High pressure lines like refrigerant and hydraulic lines do not act like a garden hose! You can't just use a barb fitting or a zip tie on them, they must have high pressure fittings and be crimped, flared or brazed/welded together. Pipe fittings won't work either, they must not have tapered threads. Teflon tape or pipe dope cannot hold back the pressure forever, either.

Quick-disconnect type fittings may or may not develop leaks right away, but if the connections are vibrated or moved, it can be assumed with certainty that they will eventually leak (this happens commonly in cars: brake and a/c lines and fittings are replaced during routine maintenance as cheap insurance against repeated failure).

As with any homemade design, you are better off over-building your contraption the first time through to eliminate Murphy's Law from having its way with your creation. If you err in your judgement, it will be revealed to you sooner or later. Trial and error is how many machines are perfected after the architects are done dreaming. Real-world results always beat blueprints and simulations.

another hot water tank heat pump
 

Well so far so good. I wound a tighter more neat cork-screw type HX. The copper cork-screw was wound into the threaded bung hole. It proved to be a little on the difficult side but the end product was good. It is a neatly placed HX in the bottom of the tank where it will perform the best. It required an extra manipulation through the top threaded holes to keep the coil from walking up the inside of the tank. With a flashlight in one hole steel rod in another and an eye looking down the third I managed to keep it where it needed to be. Tomorrow we'll braze up the Hacked air-conditioner fill the tank with water and see how it will function.:cool:

Randen

http://ecorenovator.org/forum/attach...1&d=1382804027

http://ecorenovator.org/forum/attach...1&d=1382804027

http://ecorenovator.org/forum/attach...1&d=1382804027

http://ecorenovator.org/forum/attach...1&d=1382804027

http://ecorenovator.org/forum/attach...1&d=1382804027

Interesting new approach!

-AC

AHA! So that's how you get a golf ball through a garden hose!:thumbup:

Seriously though, you are a wizard. It had to have taken lots of patience and skill to fit up. Much better than the junk they throw together on the discovery channel. Much more like 'Norm Abrams' style. It should live longer than the water heater.

The unit is up and running. Its using 5.4A at 110V so about 590W. Temperature start was 12 Deg C @11:30 am looking for 45 Deg C The tank is 175L . The Copper Cork-screw HX is working extremely well Temp in 58 Deg C out is 20 Deg C. So far no frost problem. Propane is the refrigerant of choice.

Randen

That amperage is a little high for a 5000 BTU unit at the condensing temperature you indicate. What are your high and low pressures and suction line temperature?

BTW, since the load on the fan motor is now about half, you might find that switching it to low would use less power but not slow down the fan by much. In that case, leaving it set to low would likely improve overall efficiency. Also, if you cut away part of the foam insulation so that it discharges air from both sides, efficiency is also very likely to improve. Also insulate the compressor and discharge line to avoid unwanted heat loss, just make sure the discharge line temperature stays below 90C or so.

Well 4.5 hrs in we hit 41 Deg C. As I only brazed on a low side schaider for the fill and very quickly put a charge of propane in, Yes its drawing a little more current than it should. The compressor hadn't run for a number of months so it started drawing about 6.7A on the first half hr. on start-up. Tomorrow we'll start again this time looking at the charge and current draw vs. temp

All looks really good!!

Randen

It's probably overcharged. Without a high side reading, you won't know how much subcooling you're getting. Also, trying to pull a vacuum through the expansion valve won't work.

NiHaoMike

Mike the unit is indeed over-charged but there are no leaks in either the water side or refrigerant side. The sizing of the 22' x 5/16 of copper cork-screw HX seems more than adequate. Its such a small unit and it took just a crack to charge and it hadn't been run for such a long time. The cold oil was causing some problems. Now that everything is flowing we can adjust the charge more accurately. The unit is such a small inexpensive unit a TXV was not in the formula, it sports a cap. tube. I would think after 1/2 hr. the vacuum should have worked nicely with the cap. tube.

The tank temp rose to 53 Deg C. Nice. Tomorrow after it cools we will adjust the charge and run it again. With the spare time while watching the tanks performance I will start the copper work for the solar collector in the "solar heat for the shop"


Randen