DIY Air-Source Water HP

[launboy]

The Goal: Heat my hot tub more efficiently than the resistance heat(1350w) does to bring down running cost.

The Plan(for now): Get window air conditioner, remove condensing fan and bend coil downward so it hangs below unit. Put condensing coil in water in place where it will flow through. Eventual plan is to swap condensing coil for some sort of refrigerant-water HX.

The Unit: 8000 BTU Kenmore found on Craigslist for $20.




A test of it's heat-pumping abilities, I put it in the window backwards. Ignore the ghetto sealing job. (all measurements taken with digital IR thermometer with fan on high and after 15 minutes of operation)
Outside temp was 25*F, air exiting unit was -11*F.
Inside temp was 67*F, air leaving unit was 96*F.
Compressor temp settled around 135*F

The coil never got any frost on it in over an hour of running which was a surprise to me, must be really dry out there right now.




*Note* you can see the canister thing referenced below in this picture. The black thing right next to the compressor.

Then when I began disassembling the condenser side for water testing tomorrow I discovered this...

The coil had a nice blanket on to keep it warm. The lack of airflow probably explains the warmer then expected air inside. The clean parts you see are areas I had already pulled it of, it's almost 1/8" thick in some places.

Question, should I be worried about liquid flood-back to the compressor with such a low ambient outdoor temp? There was no frost indoors but the entire line from the evap back to the compressor indoors had thick frost on it right up to the connection on the compressor. The pipe temp at that connection was about 38*F but the temp of the canister looking thing(what is this btw?) just before the connection was about 14*F. The hot gas entering the condensing coil was around 122*F and leaving it about 86*F if that matters.

Will update with progress tomorrow. Comments, concerns and suggestions welcome!

Adam

[AC_Hacker]

You are going about this thing the right way...

Get as much of the crud out of the HX fins as possible, of course.

I would have to say, however, that I have no confidence in being able to re-position the HX into any other useful position without rupturing the system. There was another poster on the 'Manifesto' thread who said it always worked OK for him. So, I just say, "Good luck."

In the event that you do spring a leak, here are some things to think about:

• The heat exchanger that you don't use can be very useful in other projects, that you have not thought of yet, so save it. You will want to prevent debris an bugs from getting inside, so tape up the raw ends.
• The cap tube (the little copper coil that is inside your AC, that is about the diameter of spaghetti) is already properly sized to your compressor, and your refrigerant... it will also work with R-290, maybe not so good with R-134a. So take care that you don't break or kink it. It is made from a different kind of copper than the rest of the copper tubing, and is not very malleable. If you re-use your cap tube, the length is VERY important, so salvage the cap tube not at the small diameter, but at the larger diameter tube that it is brazed to.
• When the glorious day comes when you do re-assembly of the parts, do not try to use copper water pipe solder, it will not only fail early, it will make it very difficult to use silver/phosphorous-bronze rod that you should use. You will need to use a MAPP gas or equivalent torch to braze, propane works very poorly, acetylene works great if you have it.

No matter how your project goes, you will gain from your experience... best of luck.

-AC_Hacker

[Ryland]

Is there any reason that the whole unit would not work stood on end? does it rely on gravity and having the the top be the top to work right?

[launboy]

Thanks AC, I was able to bend the condensing coil on another window unit down below it carefully without rupturing the pipe, it was when I went to bend it back that the pipe kinked at the U and ruptured. As long as I'm very careful and bend all of the pipe a little instead of making one spot take the whole bend I think I'll be okay.

I did take all of those steps you mentioned into mind when I disassembled the old AC that i broke. Cut the cap tube out with some bigger tubing on both ends, and saved both coils. Unfortunately after sitting, that units compressor was frozen up, so there's no hope of using it for another project, but everything else was still good so I saved it.

Ryland, no I can't just turn the unit on it's side. I wish it was that simple, but unfortunately the compressor must remain upright to run properly and keep it's oil in the crankcase.

Adam

[AC_Hacker]

Originally Posted by Ryland

Is there any reason that the whole unit would not work stood on end? does it rely on gravity and having the the top be the top to work right?

The compressor needs to be upright for its lubrication to work correctly.

There is also a preference for the HXs, though not as drastic.

Usually, the condenser gets fed from the top and the refrigerant returns to the compressor, from the bottom.

The evaporator usually gets fed from the bottom and the refrigerant returns from the top, although I have seen the evaporator specified the other way, a few times.

Having the HXs oriented unusually probably will not kill your system. The worst you might find is sub-optimal operation.

The compressor is a different matter. If your compressor is not specifically designed for horizontal or inverted operation, and you try to run it in an unusual orientation, don't be surprised if it self-destructs.

-AC_Hacker

[AC_Hacker]

Originally Posted by launboy

Thanks AC, I was able to bend the condensing coil on another window unit down below it carefully without rupturing the pipe...

Most excellent.

Don't forget the photos, inquiring eyes want to know!

-AC_Hacker

[MN Renovator]

"The Unit: 8000 BTU Kenmore found on Craigslist for $20."
"The coil had a nice blanket on to keep it warm. The lack of airflow probably explains the warmer then expected air inside. The clean parts you see are areas I had already pulled it of, it's almost 1/8" thick in some places."

It looks like we've discovered why this unit was for sale on Craigslist, that dustcake made the thing next to useless for cooling the space of its previous owner.

[launboy]

Originally Posted by MN Renovator

It looks like we've discovered why this unit was for sale on Craigslist, that dustcake made the thing next to useless for cooling the space of its previous owner.

That is very possible, although the guy said it worked fine when he used it in his apartment, but the house he moved into had central A/C so he didn't need it anymore.

AC, the coil I was referring to that I successfully bent, I ended up causing to rupture after I bent it back to "air conditioner position" and tried to bend it down again. I was able to bend this units coil down and out into a good position though, with no kinks or breaks.


I have now water tested the unit and can say that I do believe it will heat my hottub.

This is my test setup: The A/C is sitting on its shell and the coil is just sitting submerged in a 15 gallon tub. I filled it with 75* water because I wanted to avoid issues with low condensing temps( I know many refrigeration machines have low ambient fan controls to keep condensing temps up). The spatula is what I used to circulate the water.


I ran the test for one hour twenty minutes and went from 73* to 105* where I ended the test. The temp rise averaged about 4* every 10 minutes even as the water got hotter. These temps aren't the most accurate because I was using an IR thermometer unfortunately.

By my calculations:
15 * 8.34 = 125 lbs water
105 - 73 = 32* rise in water temp
125 * 32 = 4000 BTU(1)
4000 / 80 = 50 BTU/Min
50 * 60 = 3000 BTU/Hr(1)
*1 watt = 3.41 BTU per hour*
*800w * 3.41 = 2728 BTU @ COP of 1 (2)*
3000 / 2728 = COP of 1.1 (1)

(1)Actually higher because I used an uninsulated plastic tub sitting on wet concrete which was losing BTU's the whole time.
(2)I wasn't measuring, so I'm using the rated wattage, 800w.

I had to stop twice for 10 minutes to defrost(so actual testing time was 1:40, with 1:20 of it being running time. Defrost consisted of turning off the comp and leaving the fan running.

So I had a COP of 1.1 which could be higher, or maybe the same depending on the heat loss of the tub and the actual wattage being drawn, but with an outside temp of 34* I don't think that's too bad.

Things to improve efficiency:
Have good water circulation through coil. I only had what my stirring with a spatula created, and I wasn't consistently stirring.
Harvest heat, Question can I steal heat from the compressor or is that a bad idea? During the test the compressor was around 110 on the bottom half and 135 on the top. The other thing I'm going to do is to steal heat from the jet pump on the hottub. It's 1HP and draws around 1300w. This should improve efficiency even in colder temps.

One last thing, after reading your post about fluid routing, I looked closer at my unit. The condensor is like you state most are, top in from comp, bottom out to evap. My evap is interesting though, its a dual layer coil with two runs. One entering at the top, and the other entering halfway down. The top one flows down the front layer halfway and back up where it exits. The bottom run does the same on the lower half of the evap. This pic below shows it better.


Adam

[MN Renovator]

When I was messing with small A/C fan motors that weren't connected to a load they still drew power, enough for it to be significant. If you can cut the fan from the circuit, it would reduce power draw more. I'm not sure what the expected amp draw would be from the unit at the temperature differences you have between the unit and the water. I would have expected more output than 3k BTU/hr but I have nothing to suggest.

[AC_Hacker]

Originally Posted by launboy

By my calculations:
15 * 8.34 = 125 lbs water
105 - 73 = 32* rise in water temp
125 * 32 = 4000 BTU(1)
4000 / 80 = 50 BTU/Min
50 * 60 = 3000 BTU/Hr(1)
*1 watt = 3.41 BTU per hour*
*800w * 3.41 = 2728 BTU @ COP of 1 (2)*
3000 / 2728 = COP of 1.1 (1)

Well, it is a start...

Your COP should be much higher, on the order of 2.5 to 3.x or so.

If you're gonna be doing this stuff, you need to have a way to actually measure what your A/C is actually drawing. In fact, you'll find that the power consumption varies a lot, and it can give you an indication of how well you have your setup "tuned". I have a Kill-a-Watt that I like to use.

You also need a reliable thermometer of some kind. A good digital thermometer is fine, the chief advantage is that you're less likely to mis-read the numbers, but having a digital thermometer is not necessarily an assurance of accuracy, you really need to do some tests and find out.

I have had some pretty good luck using digital indoor/outdoor thermometers that I got at the thrift store for a few bucks. The outdoor part you can use for your probe. The indoor gives you an ambient temp reading, always good to know when you're gathering data. And if you have several digital indoor/outdoor thermometers you can compare all the ambient readings and then you too can know that digital is no guarantee of accuracy.

I have gone through quite a few thermometers and have learned over time which ones I can trust and which ones I can not.

My standard is a lab grade mercury column thermometer. I take very good care of it so I don't break it.

Also, the liquid agitation thing is really a big deal. That's why liquid heat exchangers are the way that they are... you have an electric pump (usually electric) and it pumps the liquid through the HX at a constant rate, and you have a HX that is designed to take maximum advantage of the work that the pump does. The whole system has evolved over many, many years.

I do understand that if you don't have the means to get set up with the tools to get in there and wrestle with the HVAC monster on it's on terms, that you are doing the best that you can. But if you want to strangle every bit of performance out of the beast, you need to get the tools of the trade, and only after that and you have had some hands on experience, you need to introduce yourself to a bit of the theory that lies behind it all.

But it is amazing what you can do with just a few tools and a little gumption.

If you haven't already done so, you should look at what Acuario has done with his work. He keeps saying things like, "I had no idea what I was doing, but it works unbelievably well!"

And he is heating his swimming pool and now also now his house!

I'm not so sure that he can still say that he doesn't know what he is doing, because he keeps doing it so well...

But it is really great what you have already done... you actually tried something and it actually worked.

The first step of an interesting journey...

-AC_Hacker