The Homemade Heat Pump Manifesto

[bigsmile]

I just recall something when I read about DX hp. I think the oil return problem isn't restricted to vertical arrangement of coil. Long run of horizontal coil may also have this problem. To furcilitate oil return, it is usually recommended that horizontal runs of copper tube be sloped a little. This may be difficult to arrange if refrigerant lines are directly used for floor heating.

[AC_Hacker]

Jay-Cee,

Welcome to the conversation.

I really do like your ideas regarding radiant floors, I'm going to have to think these over. I especially liked the modular idea with the 6" steel 'stud' and the cast-in PEX. I like it, I like it.

Originally Posted by Jay-Cee

I have noticed AC that you stress over and over in the post how "insulation, insulation, and insulation" should be one's first call to action so I will start there and also get the GSHP manual you talk about and begin educating myself further.

Yes, insulation. Don't know if you're familiar with the German 'Passivhaus' concept. In North America, they call it "Passive House", but it's different from passive solar heating, although Passive House does include passive solar heating. The reason I mention it is that they have made some serious headway regarding insulation and high-performance windows.

When the insulation is really excellent, heating becomes less demanding. And sometimes hardly even necessary.

Originally Posted by Jay-Cee

I have an incredible resource at my disposal in the form of a 15 acre lake positioned 65' from the back of my house and its 13' deep avg.

Wow, your job just got a lot easier.

Since you’ve made it through the thread, you know that doing a heat load analysis is in order. Probably even before you get serious about insulation. Here’s a free tool for you to use. It would be a very good idea to do an honest analysis on exactly the situation you have right now, I know it's tempting to fudge on the numbers, but do it just as things are right now, and save the results. Then do it again for how you want to do the insulation and compare. This will make a big difference, and it will go a long way toward shaping your thinking.

It's also best to consider that energy prices will get higher in the future.

Energy saved is cheaper than energy made.

I don't know if you found it, but there is a guy with a web site who had exactly the same setup that you have, and he did a successful install. His site is called something like "DIY Heat PUMP". I just searched for it and couldn't locate it. I wrote to him and told him I was doing a similar project, including building my heat pump from parts from an air-conditioner. I never heard from him again. I guess he thought I was kidding!

However, even though he had a pond near his house, and even though he had access to heavy equipment, and even though he bought a ready-made heat pump, and even though he utilized existing central air infrastructure, he said that putting in that heat pump system was the longest, most involved project he had ever attempted.

Oh well...

So, once again Jay-Cee, welcome aboard. Your enthusiasm is most appreciated.

As you proceed in your project, as you have discoveries, please share and if you have questions, please ask.

Best Regards,

-AC_Hacker

[AC_Hacker]

Joe,

Welcome, welcome, welcome.

Originally Posted by North_Pole_Guy

what I am interested in at the moment (its cold outside) is more pictures/drawings of how exactly you tied the tubing from the compressor to the heat exchangers. I saw the pictures of where you cut the tubing apart and what you reused.

OK, I tried to take a photo, but there is so much tubing and sensors going on I'm afraid it would confuse you badly.

So a drawing is in order:

I didn't draw the wires going to the compressor for clarity...

a.) This is where the refrigerant leaves the compressor, and so it's where the circuit starts. The compressor squeezes the vapor a lot and when a gas is squeezed (AKA: compressed) it gets hot. So if you touched the tube leaving the compressor, it will be hot to the touch. It might be very hot, so be prepared to pull your finger away quickly.

b.) The refrigerant then flows through an up and then down path into the heat exchanger. It's no accident that the tubing goes through this long path... the compressor vibrates when it is running, and having a long path spreads the repeated stress out over a longer distance.

c.) This is where the refrigerant leaves the heat exchanger. It is best to have the refrigerant enter at the top and exit at the bottom, so that gravity will act in your favor. When the refrigerant flows through the exchanger, it is cooled by the water that is flowing in the opposite direction (very important for efficiency). A lot of the heat that the refrigerant had gained when it was compressed, has been picked up by the water. The hot water can be used for something useful, like heating your home in Alaska. In the process of giving up its heat, the hot vapor begins to turn to liquid refrigerant(AKA: it condenses). That is why this heat exchanger is referred to as the condenser.

d.) I hope it is clear in the drawing that the tube from c-to-d is passing behind the compressor, and is not going into the compressor. The point "d" is where the refrigerant tube is brazed to the very small diameter tube (AKA: capillary tube, or cap tube). The tiny tube going from d-to-e is the cap tube, and its small diameter, along with its length acts to slow and regulate the flow of the refrigerant in the system. Sometime the cap tube is refered to as a 'metering device'. So pressure builds up between the compressor and the cap tube. All the tubing from a-to-e is referred to as the high pressure side (AKA: high side).

e.) This is the point where the cap tube is brazed to some larger copper tube. When the high-pressure refrigerant leaves the cap tube, it sprays out and instantly goes into a vapor state (AKA: evaporates). When it evaporates, it gets cold, very cold. Just imagine a hot day and spraying water on your face... same thing.

This is also the place where the refrigerant (very cold refrigerant) enters the heat exchanger. Again, it is a very good idea to have the refrigerant entering from the top and exiting from the bottom, so that gravity is our friend. While the cold refrigerant flows through the heat exchanger, it is taking heat from the water that is flowing in the opposite direction (very important for efficiency).

So the water that flows through this heat exchanger (AKA: the evaporator) will flow next to some water. The water will give up some of its heat to the refrigerant. And the refrigerant will carry this heat on through the cycle.

This is where we run our ground-loop water. So even though the water is pretty chilly, it's about 50 degrees here in my loop, the refrigerant is so much colder, that it take the heat from the water...

f.) This is where the refrigerant exits the heat exchanger. The refrigerant will carry the heat that it picked up, with it on it's way to the compressor. Again, there is a loop, in fact, there should be another loop that I left out for clarity. But you need to have them there for stress relief. As the refrigerant passes from f-to-g, it passes through a filter that serves to hold excess refrigerant, and also to filter out any unwanted bits. There is also a dessicant that will trap water that may have been in the circuit. Water and refrigerant are not a good mix.

g.) This is where the refrigerant re-enters the compressor. the refrigerant has been cooled by rapid evaporation, then warmed somewhat by water in the heat exchanger, but it is cool enough to keep the compressor running cooler... but it does pick up some heat from the compressor.

So there you have it.

Did that help?

Originally Posted by North_Pole_Guy

I was wondering what controls the cold side temp above freezing your exchanger up, or is that a function of how much water is run through it?

Great question. For sure, you need to keep an adequate flow rate going, or it will freeze. You also have to monitor the temperature so that if it approaches freezing the unit will automatically shut down before harm happens. But in your case, you'll want to run some kind of anti-freeze mix in your loop field. There are antifreeze liquids that will not cause environmental damage. You can probably google it as fast as I can.

Hope this has all helped.

Best regards,

-AC_Hacker

[North_Pole_Guy]

AC_Hacker
that is absolutely great of you, that drawing did it for me. that is what i was thinking but have not hacked into these much yet, I have a basic understanding about how they work but not alot of hands on yet. I think one of the missing links for me was the refrigeration regulation, on these small systems I now understand it is done by the capilary tube, this provides the pressure drop the refrigeration needs to get cold. In the natural gas industry they use a JT valve (joules Thompson) to liquefy natural gas, same thing here.
so on bigger systems do they use an adjustable valve to better regulate how cold the cold side is by varying the pressure drop?
Yes I would agree that counter flow in a heat exchanger is the most efficient at heat transfer. I will have to order up some flat plate heat exchangers off e-bay to get my experiment rolling soon, only a few more months before it warms up and other projects will take precedent over this.
I have been reading alot on construction in sub-arctic conditions for quite a while, the university here has published a cool manual
.cchrc.org/docs/best_practices/REMOTE_Manual.pdf
I think that super insulated is by far the way to go, as you stated energy is only going to go up.
I plan on taking my time and building a home correctly then heat with the best means to include radiant heat. Open floor plan that can be heated with a wood stove in a pinch but have the primary heat pump secondary wood/coal fired outdoor boiler.

I have a set of gauges and a home made vacuum pump from an old freezer now I need to get me a small set of torches and play around with getting the fittings to bottle up my refrigerant from my system. Up to this point I have only played around with R134A on a few vehicles, patching the leaks pulling a vacuum then recharging them, with pretty good luck.

Really looking forward to tearing into this project in the coming few weeks, thanks for the kick start, oh on that link there is a link inside it to some pilot projects here in the Fairbanks area where they installed GSHP systems this year, be interesting to see what they come up with for numbers.
Shoot if I could even heat a small building like my chickens house with this heater for cheaper than I can with oil I would be happy...
thanks again for the welcome, looking forward to reading more about your project and I will work on learning all I can and sharing it here..
Joe

[AC_Hacker]

Originally Posted by North_Pole_Guy

I now understand it is done by the capilary tube, this provides the pressure drop the refrigeration needs to get cold... so on bigger systems do they use an adjustable valve to better regulate how cold the cold side is by varying the pressure drop?

Yes, exactly so. I have an adjustable valve I'm going to try out on my next system, which I think will work ok. For the smaller compressors, it's a bit challenging because the valves have a stated range within which they work. It seems to be easy to get valves in the one ton and higher range, but not so easy when you want to go smaller.

Just in case you don't know this already, the cap tube diameter and size is precisely fitted to the compressor/heat-load/refrigerant combination. If you have a compressor meant for R-22 and you want to use R-134a, not only will your cap tube no longer be the right size, but your heat exchangers will be too small (R-134a carries less heat per compressor cycle than R-22), and worse, the R-22 lubricant is not compatible with R134a. R-290 has similar characteristics to R-22. This is one of the reasons I have used R-290 (AKAropane) which of course carries substantial personal risks to work with, and it is illegal in some parts of the world. There is some stuff available on ebay called Enviro-Safe 22a, which I am interested in trying, which seems to be a direct replacement for R-22. It may be propane, I don't know.

If your de-humidifier was charged with R134a to begin with, you're good to go.

Originally Posted by North_Pole_Guy

I will have to order up some flat plate heat exchangers off e-bay to get my experiment rolling soon, only a few more months before it warms up and other projects will take precedent over this.

I've been using a 500 watt compressor that previously had R-22. I used 20-plate heat exchangers that are about 8 in by 2.5 inches. It seems to be working out pretty well. Try to get heat exhangers that have pipe thread fittings on one side and "sweat fittings" on the other side.

Originally Posted by North_Pole_Guy

I have a set of gauges and a home made vacuum pump from an old freezer now I need to get me a small set of torches and play around with getting the fittings to bottle up my refrigerant from my system.

Don't forget that you need to flow some kind of inert gas through the tubing you will be brazing, otherwise the copper will oxidize and you'll have nasty flakes in yor system that could clog your tiny cap tubes.

And I like the idea of a homemade Freezer Compressor vacuum pump. I think that there are lots of people who would like to see how you built that... Please post photos!

By the way, the manual you mentioned is absolutely great. It really has some of the best building techniques I have see regarding insulation.

However, I haven't gone through the manual carefully, but I did not see any mention about methods to use to avoid Thermal Bridging, which the Passive House people are well aware of. Thermal bridging can reduce your effective R-value by 15%, check it out.

So good luck with your project... be sure to let us know how it turns out.

Best Regards,

-AC_Hacker

[RobertSmalls]

I read through parts of the REMOTE manual, and I like the fact that they're proposing walls a foot thick with insulation. However, I don't really see the advantage of this approach instead of putting the extra insulation in an offset studwall INSIDE the house. A staggered studwall allows for a conventional, durable, well-tested exterior construction method.

I'd like a house with thick walls, windows mounted flush with the exterior, and a deep window sill for potted plants or people to sit on.

[North_Pole_Guy]

Part of the idea of adding the insulation on the outside to to build a continuous pure insulation to provide 100% free thermal bridging. Another reason is to provide the cold area outside away from the interior to prevent condensation from forming and building mold and such inside the wall cavity. Here in the Interior you don't place any plumbing in exterior walls due to the freezing problems during extreme cold spells. This building technique allows you to put plumbing and other utilities into exterior walls.

This method would also be less labor intensive and less cost than the "Swedish" walls that you mention "staggered walls" or a wall within a wall.
This would allow you to frame with 2"x4"'s if you wanted to or else they say 2"x6" on 24" centers if you wanted to run any type of utilities in them to allow for more room.

Just plain 2"x6" walls here in the interior is not enough to be very energy efficient, there is a lot of thermal bridging at all of the studs which lowers the whole building R-value by quite a bit as AC_Hacker said.
So adding a few layers of sealed foam on the outside of the building stops all thermal bridging except around doors and windows.

[Jay-Cee]

AC Hacker,

Something else just hit me that may be more conventional and cut out some challenges AND cost while still being potentially just as efficient so I thought I would throw it out here for review and see what everyone else thinks.

When I think of the modular idea previously discussed with steel studs, I see the pro's as being:

1. Relatively cheap.
2. Readily available.
3. Ability to do one run at a time and can do solo if necessary.
4. Concrete slabs would be easy to vibrate/trowel to an even and consistent height.
5. Ability to manage how much concrete weight you want to add (at the cost of efficiency)
6. Sufficient density and mass
7. Slabs can be bonded to new subfloor with mortar enhancing thermal transfer.


What I don't like:

1. Overall floor height being raised at least 2", probably more like 2-1/2" and all the challenges that come with it. (stated previously, but I also wanted to add the trip factor if a person doing this has stairs going up or down anywhere in the house. We have all heard the saying "watch out for that last step...!"
3. Voids of either air, or wood, or foam insulation in between each slab takes away from the "even-ness" of the overall system.
4. Cost of concrete, mortar, additional substrate, furring strips, insulation (maybe) and steel studs could add up more significant than expected.
5. Long term potential possibility of termite/insect infestation without homeowner ever knowing because the floor is now above the bottom wall plate and has voids.

The height thing is a real dealbreaker in my mind, so I started thinking "how can a person place/pour a 3/4" - 1" concrete floor filled with tubing and it not crack or break up over time due to the shallow height??

So here was my evolution of the idea... What if you cut 5-1/2" wide strips of 1/2" or 5/8" concrete board (wonder board, hardi board etc...) and screwed them to your original subfloor leaving a 3/4" gap between each. Then run your 1/2" plastic (it would yield a run every 6" O.C.) in the gaps. Then tile DIRECTLY over the top (preferably a larger floor tile like 16x16 or even 24x24). As you are setting your tile you will fill each pipe run gap with mortar which will bond the tube, the cement board, and the tile all together at the same time. Most good tile mortar has latex flex agent added to it to allow for movement. The larger tile will yield a fabulous bond to the substrate below and there will be NO VOIDS. Thus, your total FINISHED Floor height will be: 1/2" cement board/piping + 1/8" mortar bed + 1/4" tile = 7/8" above subfloor and no challenges with doors/stairs. It seems this would yield the most uniform and "solid" type surface with the least amount of materials.

Note: I would put the strips across the floor joists and mortar them to the subfloor to ensure strength.

I also did some rough calculations about cost and weight for comparison.

Weight (using a 30x30 floor, 900 sq ft.)

Per Radiantec.com's DIY radiant floor installation guide (I cannot post a link)

They recommend on a floor joist system installation:
7/8" pex on 16" center (would seem appropriate for the modular slab idea)
or 1/2" pex on 8" center (my idea above mentioned 6" on centers)

so my rough calculations were this
Modular Slab design with metal studs:
* would need 25 runs at 16" OC approx 28' long (not sure how much room needed to bend 7/8" pex in a "U" shape). Plus the 2 ends at approx 30'
25 runs x 28 feet = 700' + (2 runs x 30') = 760' of "modular slab"
* avg weight of concrete/cubic foot = 143.38 pounds.
* .125' x .4375' x 1' = cubic volume/lineal ft. of 2x6 steel stud = .0546' cube
* .0546 cubic feet x 143.38 lbs = 7.83 lbs/ lineal foot of modular run.
* 7.83 lbs x 760' of slab = 5949 lbs.
* Then have to add concrete board on top of that. 1/2" Wonderboard claims on their website to be 55 each sheet (3'x5').
* 15 sq ft = 55 lbs or 55/15=3.66 lbs/sq.ft.
* 3.66 lbs * 900sq. ft. = 3299 lbs.
*Thus far total weight is 5949+3299= 9248 lbs.
Not included is finished floor material weight, water weight in pipe, additional mortar, wood furring strips, or metal studs.
Seems a little heavy, will be over 100 lbs./sq. ft. when all else is considered.

The new idea would only weigh the weight of the concrete board, 3299 lbs + the mortar bed and fill (not calculated) + the tile (not calculated). My thoughts are that it is a significant amount of mass, but certainly not so heavy one would have to question whether his joist system could hold it. I believe this type of weight would work on any home without question. If you think about it, its nothing more than installing a tile floor with the exception of "hiding" some pipes in the underlayment.

Cost:
additional 41.4 cubic feet of concrete-(.0546 cu ft/lineal ft. x 760 lineal foot)
Whatever you spend on metal studs.
Possible pricing difference between 7/8" pex of lesser length vs. 1/2" pex of greater length.

Final Note and Disclaimer: Please keep in mind that NOTHING has been mentioned about what method would or would not be more thermally efficient. I am completely uneducated in that area and hope that your responses will help me to understand! All logic here is approached from a "what seems easiest from a cost, construction, longevity point of view", and probably a biased one at that.... be advised.

[North_Pole_Guy]

AC_Hacker
I will try to insert a picture of my donor de-humidifier below, it has the capillary tube split into two parallel runs instead of one capillary tube coiled up.
So do I try to find some new capillary tube to replace the two with one so total length of the two?
or do I just try to fashion up some kind of fitting to braze the two in parallel?
I don't think it will by too difficult to try to braze them in parallel. I see at the local hardware store they sell a very small tubing that i could practice with, oh yah and I won't forget about the purge gas during brazing...
Joe

[North_Pole_Guy]

Jay-Cee,
have you heard a product called warm board?
looks like the way to go for new construction or you may even be able to add it to a home on an upgrade?
warmboard.com

Joe