Homemade Heat Pump Water Heater (R290)
 

I've Started on my new project, home made heat pump water heater. This project started as a metal recycling trip. I was loading up scrap steel into my dump truck when the neighbors noticed what I was doing. They came over and asked if I could take various things in as well. I ended up keeping two non working dehumidifiers and 3 hot water heaters. The HWH I'm using first is a GE 30 gallon electric unit. I lucked out in the fact that this unit uses just 1 4000 watt heating element on the bottom port. Apparently they made models with two elements with the 2nd being in a port toward the top of the tank. But instead they just installed a steel plug.
So first thing I've done after flushing and checking for leaks, was take out the plug in the top port and drill two holes thru it and tap them for 1/8"npt. Then I installed two 1/8"npt to 1/4"compression fittings in those holes. Next I bought 30' of 1/4" copper tubing and made a fairly sharp bend at the half way point and snaked 25' worth of it into the tank. Which was no easy feat, but persistence paid off. I then ran the ends of the tubing thru the plug and fittings and pretwisted the tube in the counterclockwise direction. Then threaded the plug back on, and snugged up the compressin fittings to make a watertight seal.
Last night I tapped into one of the dehumidifiers and found out it has nearly no r22 left in it. So I acquired some r22 and gauges, pumped in some refrigerant and the unit came back to life.
But never being happy with something normal, I grabbed a old propane torch, took the top off and brazed on a 1/4" flare fitting.
This morning I removed the r22 from the dehumidifier and hooked up my vacuum pump to the system for a half hour. I then took my new propane "torch" and refilled the unit with propane strait out of the Coleman bottle. The unit came back to life. On the r22 it was using 435-450 watts on R290 it's using 365-410 watts. This is where I'm at now. Next step Is to drain the system and plumb the coil in the HWH tank into the lines on the dehumidifier replacing to hot side HX. I'll try to figure out how to post pics. I don't have a computer hooked to the Internet, only my iPhone, and I'm not sure how to get photos uploaded.

Hv23t

[IMG]http://i649.photobucket.com/albums/u...01F7264C1C.jpg[/IMG]Here is a pic of the HWH plug with the HX tubing going thru it.

Here is my torch conversion. http://i649.photobucket.com/albums/u...048BA9B6E6.jpg

Dehumidifier running on propane!![IMG]http://i649.photobucket.com/albums/u...01A33380E9.jpg[/IMG]

And getting nice and cold

[IMG]http://i649.photobucket.com/albums/u...01AECE6A34.jpg[/IMG]

I'm interested! I have been watching on the forum for someone to attempt one of these, and up to now the only one who seems to have completed a project is ac hacker. He converted his unit to water on both sides, to run off his ground source loop. Xringer threw in the towel when he found a premade unit. Both have posted awesome efficiency numbers and general satisfaction with their results. So I am eager to see a hacked unit brought to a new life.

Now, a few questions.
What model or capacity is your dehumidifier? It seems that for DHW production, 6000 btu is all that is necessary.
What's your plan on frost control? These units are designed so that the evaporator will freeze up under room temperature PDQ. They then defrost, sucking more water out of the air while the ice melts. Good for drying the air, not so good for max heat output.

Good luck and godspeed!

Thanks for the encouragement,

You have some good questions that I'm not sure of the answers at this point. As far as frost control goes, I'm just going to initially just wait and see what happens. The dehumidifier has the cooling HX and the heating HX. I'm going to loop the cold to the hot essentially doubling the area on the cold side hoping it won't get quiet as cold. I disconnected the fan last night to see if the hot side would reach a usable temp to heat water. During normal operation with the fan on it was only reaching about 90 deg. And the cold side was below freezing. After disconnecting the fan I saw temps in the 140's on the hot side and climbing and the cold side was in the 34-38 range. So between the doubling of the HX space and the temp range increasing I'm hoping that freezing wont be as big an issue. I currently have a GE Heat Pump water heater that works great. Ive never noticed it going into a defost cycle. It may do it, I'll pay more attention in the future. I probably should pull the cover off and see what it looks like under there. I didn't have a chance to play with anything today, I had a rifle match to lose. I'll start replumbing tubing tommorrow.

Hv23t

[IMG]http://i649.photobucket.com/albums/u...5E830EF6FA.jpg[/IMG]

R290 Smell
 

Just an observation. I just drained the propane out of the dehumidifier so I could make the mods to the plumbing. The filter/dryer in the system took nearly all the sulfur smell out the propane. I didn't smell it at all, until I brought the hose up near my nose, and even then it was just a faint smell. Maybe if it had been in there longer it would be completely gone.
Hv23t

Got a late start today, so I didn't get as much done as I was hoping. I did finish all the plumbing and I vacuum'd out the system but didnt have time to recharge it, or insulate the lines. But here is where I'm at.
[IMG]http://i649.photobucket.com/albums/u...408C92AC98.jpg[/IMG]

[IMG]http://i649.photobucket.com/albums/u...40CB95422C.jpg[/IMG]

[IMG]http://i649.photobucket.com/albums/u...418FF40DE2.jpg[/IMG]

Hv23t Wrote:
Just an observation. I just drained the propane out of the dehumidifier so I could make the mods to the plumbing. The filter/dryer in the system took nearly all the sulfur smell out the propane.

When I was working on my GSHP I found the same. The propane became odorless. Your project looks great. I'm on the edge of my seat for the performance figures. We'll have to see if the eco-renovated propane enhanced water heater can give Xringers A7 any competition for operating costs.

Keep up the good work

Randen

Sweet! That unit looks like a million bucks to me. My wife would tell me to put the covers back on....
Your dehumidifier is just like mine. I'm anxious to hear how it's going to perform. My guess is that it depends on where this beast will live. Hopefully indoors where it won't get chilly.

Here you go... More wife friendly..
http://i649.photobucket.com/albums/u...2AD09D19F2.jpg

First test
 

I fired her up for the first time today..
I haven't really thought of how I'm going to conduct tests yet.
So this is probably meaningless, but for a first run I'm happy.
I'm not sure I trust me thermometer but here is what It was saying.
Time. Temp (water) f
0:00. 67
0:20. 81
0:40. 86
1:00. 90
1:20. 94
1:40. 98
1:51. 101 tripped thermostat
On water heater set
At 100 deg + or -

Hv23t

I left the water heater turned on overnight. It got down to 34 deg last night, which I know is way to cold for this thing to work. I just went out and the dehumidifier was off and the water was still at 115 deg. I went and checked the kill a watt and it read out .70kwh. So it ran for an hour and 20 minutes to keep it warm. I'm very happy it turns on and off using the thermostat that was on the WH.

Hv23t

Great mod
Congratulations ! on its success.

Wanna make me one? :)

Awesome project you have going on here! And your photos will be useful for everyone.

As far as figuring efficiency... if you could let the unit cool down for a while, like maybe a day or two, and measure the water temperature. It may be a bit tricky since the water temperature will be stratified somewhat, hotter at the top, cooler at the bottom. If you had a tiny circulation pump laying about, this would help. But leaving all this aside, just measure the water temperature before you start. Then run your unit with the Kill-a-Watt attached, until it shuts off, then measure the water temp again.

one BTU is the amount of energy required to raise the temperature of one pound of water one degree Fahrenheit

Look up on the Internet how much a gallon of water weighs, and multiply that by the capacity of your water heater... this will give you the number of pounds of water you are heating.

Then, look at the temperature rise from beginning to the end of the run... Let's say it was 67 F in the beginning, and 101 F at the end.

So, in such a case, your temperature rise would be 101 F - 67 F or 34 degrees F.

If your tank capacity is 50 gallons, and if the Internet tell you that the weight of a gallon of water is something like 8.3454 pounds per gallon, then the weight of your water is:

weight = gallons x 8.3454

weight = 50 x 8.3454 = 417.27 pounds.

Then the BTUs would be:

BTU = (weight of water) x (temperature rise)

BTU = 417.27 x 34 = 14187.18 BTUs

To calculate efficiency, you will have to have all the units the same, and since your Kill-a-Watt measures watts (where is a Kill-A-BTU?) and you will want to convert your calculated BTUs to watts, so:

watts = BTU / 3.412 = 14187.18 / 3.412 = 4158 watts

Next, you will need to know how many watts were consumed to heat the water.

if your Kill-a-Watt reads something like 2.34 KW, you multiply that by 1000 to get to watts:

watts = Kw x 1000 = 2.34 x 1000 = 2340 watts.

So, to find your efficiency, you would divide your power out by the power in:

Efficiency = power out / power in

Efficiency = 4158 / 2340 = 1.77

In engineering terms, the above number, 1.77 is called the coefficient of performance or COP for short.

So, you could say:

COP = 1.77

If you multiply that by 100, you get your efficiency as a percent:

% efficiency = COP x 100 = 1.77 x 100 = 177% efficient

Not a bad deal really, you are using one Kw of electricity, but you are getting 1.7 KW worth of hot water, just by preventing junk from going to waste!

But between you and me, I think your actual efficiency will be much greater than that!

Hope this helps...

Great work!!!

(* BTW, that is a nice looking car there in your garage, too! *)

Best,

-AC

NICE!! Very professional looking hack! :)

I got confused about how long the HX loop is.?. I don't know how long my A7 HX is,
but I think it's 18 to 20 feet long (36' to 40' total length), it's a smaller diameter too.

I think that it would be easier to seal the HX, if you didn't have to pre-twist the 3/4" 'plug' CCW.
The method used on the A7 is pretty good. See pic.
I'm pretty sure that I could make a plug in my shop, that would fit into standard 3/4" plumbing fittings. Maybe use a flare type nut fitting.?.

~~~
We don't use a lot of hot water (2 retirees) and the A7 uses less than 1 kWh per day.
Which I think is pretty dang good when compared to other heating methods..


Keep up the good work!

Looks great, nice project!

Thanks -AC
That helps alot. I'll move the unit into the garage so its in a more stable environment. I might be able to do a circulation pump, seems like I have a sur-flo around here somewhere. I'll see what I can get done this weekend.

Hv23t

Xringer ,

What is the temp in the area surrounding your a7?
The HX is 25 feet of 1/4 copper tubing coiled in the tank. I like the a7's fittings, but I'm not sure I could have made the bend at the end of the loop without kinking it going thru a 3/4" pipe. The 1" plug was a pretty sharp bend but I managed.
1kw a day is amazing, my GE Geospring water heater is rated at just under 1900kwh a year so about 5kwh a day. I wish I could hook the kill a watt up to it. (240 volts)

Hv23t

Hi Hv23t,
It's about 57F down at the A7 right now. I don't think it's going to get much cooler during the winter.
(It's 35F outdoors right now).

That general area around the A7 can drop down to 55F while it's running,
but the area warms back up a few degrees pretty fast.

I guess it's the slightly warm floor of the basement warming up the air.
So, my A7 ASHP is really a ground-source Geothermal unit.. :)

Maybe I should drill some holes in the floor and implant some long heat-pipes?
Of course they would be dripping wet during the summer months.. :o

Haha, I just shot my IR pistola down into the sump pump holes,
and the temperature two feet under the basement floor is 57F.

The year-round average temp here is 54.3°F and that does have
a large effect on the upper few feet of soil..
So, maybe we'll see the floor drop down to around 54F,
but I'm pretty sure the A7 won't notice it much..

Let the testing begin. I ran water thru the tank for a few minutes until the out put was stable at 56 deg. I moved the WH inside the garage where the air temp is stable and wind won't cool anything off quicker. The air temp in the garage is 60.3 deg f. http://i649.photobucket.com/albums/u...7813DDC04B.jpg
I will update with results when its done.
Hv23t

Well I finished the 2nd test of the system on Friday. I havnt had time to do the calcs on efficientcy till this morning. Here are the results:

Time (hr:min) H2o (f) Watts used

0:00. 56. 0
0:20. 65. 210
0:40. 73. 360
1:00. 78. 510
1:20. 86. 670
1:40. 88. 810
2:00. 89. 910
2:20. 95. 1090
2:40. 100. 1270
3:00. 105. 1450
3:20. 109. 1630
3:40. 115. 1810
4:00. 119. 1990
4:20. 122. 2170
4:40. 125. 2360
5:00. 129. 2550

I used the formula that AChacker gave me to figure out COP.
The COP Changes over time with temp so I just did it for total at each hour.

1 HR. COP=3.16
2 HR. COP=2.65
3 HR. COP=2.69
4 HR. COP=2.32
5 HR. COP=2.10

For my first attempt at something like this I'm very happy with the results. And I know there is room for improvement. #1 is going to do something about a defrost cycle. After about an hour the HX was just a block of ice. What air was able to work its way around it was coming out at 41-45 deg. The dehumidifier only went into its own defrost cycle once at the 1:40 mark. So I think I'm going to install a recycling relay that will turn the compressor off every hr for 4 to 5 minutes. #2 is going to be insulate the feed and return line to the tank HX seperately. Now one line is about 30-40 deg hotter and some of the heat is escaping to the return line befor reaching the tank. After reaching 130 the system shut down at the preset thermostat setting. I was playing with my ir temp gun and found that while the temp was 130 toward the bottom of the tank where the thermostat is located. The top of the tank was at 148 deg. So ac- was probably right about installing a circulation pump to get more accurate numbers. My plan this next week is to install this system inline ahead of the main water heater at my parents house. Essentially being a preheater for the existing one and making sure they don't run out of water if this one breaks down.

Hv23t

If you want a simple way to turn it off and back on every hour, maybe one of these will work? (at least for 20 hours)..
Timex Electronic 7 Day Timer for Use Inside or Outside The House | eBay

You can set up to 20 off-on cycles for each day.

I'm using one of these to keep my A7 AirTap off during the late night hours.
So far, it's working just fine.
You might want to use one setting to keep the ASHP off between midnight and 4AMish..
Since it's not likely to be needed during those hours..

I like these timers so much (and the price), I just ordered # 4.. :D

The reason your COP declines is because the temperature difference between your HX in the water tank and the water gets smaller as the water gets hotter... quite normal.

You are totally successful! For making a heat pump water heater out of junk that you dragged back from the jaws of hell, COP=2.1 is nothing to be ashamed of... and when you average them up, you certainly have genuine bragging rights!

It's possible that you could improve on the performance by using a larger evaporator HX... That just might mean another trip to the dump... maybe a re-design of the basic layout.

But if you built that car that's in your garage, you are no stranger to changing things around.

On the timer thing, you might watch and time the frost build up and melt down. I would suspect that night time would be the worst frost conditions. Then plan your on & off cycles accordingly.

BTW, if you're gonna run your unit with R-290, be sure that the area where it is located has the possibility of ventilation... I'm just thinking about the safety thing. The volume of propane in your unit is pretty small, but best to be safe.

Congratulations on a great job.

Best,

-AC

Tell me what the basement temperature is about Mid February...

-AC

Recycle Timers
 

I was looking last night for a recycling timer, preferably an adjustable one so I could fine tune my defrost cycle. I was just getting ready to order a cool DIY kit for $18, when I found a couple omron timers on eBay nobody had bid on yet. Well I just finished out bidding the others on both. I got one omron H3CA-8
http://i649.photobucket.com/albums/u...04F13F87DC.jpg
And the other is a omron H5CR-B, which is a real sweet heart with all the bells and whistles.
http://i649.photobucket.com/albums/u...04FA2F2B17.jpg
And all for less than $30 with shipping.
I love eBay.
Hv23t

But if you built that car that's in your garage, you are no stranger to changing things

Yeah I'm used to making things work. Unlike the car I'm hoping this thing will end up saving some money.....

Hv23 t

Today I went to my parents house and uninstalled their 50 gallon propane water heater. I've installed two 30 gallon electric units. One is hooked up in the normal fashion. The second which is my homemade heat pump water heater is hooked up in series as a preheat tank. If everything works out fine the heat pump will heat all the water and the other one won't have to do any work. But if it's to cold or they use to much water it will turn on and take over. I havnt got all the electrical worked out with the defrost cycle. In working on that at my house where I have all my tools. Then I will go over and install it after I think I have everything worked out. I'll get the kill a watt on it after its all buttoned up and see how effiecent it turns out to be.

Hv23t
http://i649.photobucket.com/albums/u...5B741E8823.jpg

What are you planning to use for a defrost system? Also, do you have a circulator pump or some system to distribute the heat from the HP tank to the other tank? Otherwise the resistance still has to maintain that tanks heat, although I guess the heat loss through the tank isn't much.

Adam

Great Job!
 

Congratulations on the good job.

It was very kind of the dehumidifier folks to make such an attractive shroud for your heat pump!

Really, that's about as straight-forward a solution to improving the efficiency of a water heater as I can imagine. Well done, really well done.

As I recall, you saved the water heater you used from scrap hell. So you might not know its age. At any rate, someone mentioned way before on the 'Manifesto' thread, that having a heat pump instead of 220 volt heating elements, would be much less stressful on the inside of the water heater core, because the temperature cycling is less extreme, and also because you don't have high amperage elements cycling inside the tank, that can accelerate corrosion. I have no way of proving those assertions, but I pass them on along because they sound so intuitively correct.

Keep us posted on the measured efficiency.

Also, it should be a pretty easy matter to keep track of energy costs over a few typical months and compare that with the cost to run the propane heater, from similar months.

Best,

-AC

Yes your right the resistance elements will have to maintain the temp in the primary tank. I cant imagine there being much loss there. I'm going to the hardware store tomorrow to pick up a couple tank blankets to insulate them even further. My defrost system is just going to be a recycling timer controlling the compressor. It will take some trial and error but my first setting will be 1 hr on and 5 minutes off. I've done it manually a few times and seems to work well enough. The refridgerant in the lines going to the tank are in the 100-130deg range. When the compressor turns off and that heat distributes itself thru out the system the ice melts pretty fast. I also just ordered a digital temperature switch that I'm going to set to bypass the heat pump altogether when the air temp gets below 45 deg f. And then it will use the 4000 watt 240 volt element, that is running on 120 volts so is only 1000 watts. And then of course the primary tank has its 4500 watts worth of heating elements to keep the warm water flowing.

Hv23t

Yeah the water heater was set to go to the scrap heap. The only reason it was being discarded was do to a remodel of the room it was in. It is 15 years old but we don't seem to have problems with rusted out tanks around here. My previous homes heater was 26 years old with no probs. the propane heater I just took out of my parents house was 29 years old. When I hooked the hose up to the drain port to empty it the water came out purfectlt clean, no signs of rust. And that was the first time it had ever been done.

Hv23t

This is basically what I found with my Hottub HP last winter, even in below freezing weather, with the condensor in 104*F water the ice melted on it's own pretty quickly, but the heat didn't last long enough to dry the coil so the melt water just refroze on it. If you keep you only run your system above 45 you shouldn't have that problem though. Honestly you could probably run it lower than that. It would be interesting to see where the balance point would be on a small HP like this. Most commercial units it's around 35*F.

Hv23t,

I think that having a temperature controlled cut-out is a good idea.

However, there is a curve that results when you plot COP vs. air temp, and COP drops as air temp drops. When the COP approaches 1, it is definitely advantageous to switch to resistance heat.



Here is a plot I found on the web just to show you the shape of the curve. The temperatures here are in C not F, and the place on the graph where COP = 1 will most likely be very different from yours, but the idea is very similar to yours.

I guess my point is that 45 degrees seems to me to be an arbitrarily high point to set your cut-out, and that even though the COP is declining, it's still a lot better than resistance heat.

But exactly what your curve looks like will depend on more factors than I can possibly guess. You can get a rough idea by monitoring the system as ambient temp changes.

By the way, there is a very timely conversation going on right now on the 'Manifesto' thread that concerns R-290 efficiency. You should check it out, as it pertains exactly what you are doing.

Best,

-AC

Thanks AC I'll check out the manifesto thread. I picked 45 deg as a starting point for no particular reason other than on my geospring Heater at 40 deg it switches to resistive heating automatically. So I was just using a starting point somewhere around its. I'd imagine they switch over then for recovery time reasons. But I'll play around with it to see what's best. Since its just a preheat tank as long as its over 1 cop I'm still ahead.

Hv23t

My fridgie friends from the UK all seem to be concerned about sulfur content in BBQ LPG. They seem to think it will harm the copper components long term. I've no idea but has anyone looked into it?

I'm not sure of the long term effects either. But when I first charged the dehumidifier and ran it for a couple days, I released the propane so I could turn it into the heat pump. The filter dryer seemed to filter out the sulfur as the propane had no more smell. So I'm guessing it wont be an issue,
Hv23t

Yeah, this can shorten the life of a system in general, the compressor in particular. I don't have any data as to how much. A properly set-up heat pump that is installed out of the weather can last decades. To get this kind of life, you'd want to go the distance with the best refrigerant you can obtain or upgrade, and scrupulous attention to water removal, etc.

But for hacking and learning, BBQ-gas works. But for your Opus, you'll want to use best practice.

The problem is sulfur + water which can make sulfuric acid, which is the real problem.

A good vacuum pump is used to create a vacuum deep enough that water will boil off, and hold the system vacuum long enough so that the water can be removed. This means a vacuum in the sub 200 micron (or equivalent) range. Good equipment can do this. A micron gauge will tell you if your equipment is good. I thought my equipment was good until I got a micron gauge, then I found out that my vacuum pump was not creating a sufficiently low vacuum. I now have an inexpensive, but new vacuum pump that does the job.

Also, some of our most industrious fiends have used HVAC filter/dryers to remove sulfur content. The noxious rotten egg smell (Hydrogen Sulfide) that is present in BBQ-gas is an add-in (Capton) for safety, because propane is odorless. So filtered BBQ-gas will be nearly odorless.

You should be able to locate 'refrigerant grade' R-290 that is properly dry and sulfur-free, but it's not cheap.

If you search the forum, you should be able to find postings of how others have done it.

Best,

-AC