The Homemade Heat Pump Manifesto

[RB855]

Included driers are usually very small units with very low capacities. I intentionally removed the existing one (directional, no filter) and installed a much much higher capacity bi directional filter type. The factory drier only worked in a/c mode! Not in heat. This way it will work in both heat and defrost (or if I ever change my mind and turn it into a heatpump single room conditioner)

The immersion plate exchanger was just a offer of someone may have a use. It would be very easy to build a tube in shell type HX out of it. No idea what pressures they could run at, as this particular one was all aluminum.

Took a real camera with me to work today and grabbed a handful of new shots. I should draw up a diagram of how exactly its all hooked up for yall (edit, there's now a poor drawing). Had it running as a watercooled a/c today as the humidity was nasty. Filled the pan up with condensate :-P I would like to note its still a long ways from what I could comfortably call finished. I have a bracket built to hang the BPHE's, there's a lot of insulation to go, installing a proper pump, and whatnot. I have a line out for a new evaporator as I damaged a pass reducing one of the circuits to 10 pass's instead of 12. Will also be redesigning the distributor to be all equal length, as there is evidence of unequal flow through the 3 circuits.
Dont poke too much fun at my brazing. I started off using the wrong stuff. Very very high temp melt high strength. The bonds are probably stronger than the copper itself, but resulted in some sloppy connections.



The reason for the waste heat cooler isnt because of the BPHE not removing the heat so much as when your entering water temp is 80-125* (tank mixing causing your pickup water to be warm). It also is a last step to improving the liquid quality to the txv by making sure the liquid is cold as it can be short of using a reclaimer on the vapor return. Its almost useless when the water inlet temp is low as the BPHE are way overkill for the given job and extract almost all the heat. low flow rates still get only a couple degree delta.
For those of you building ground source, I wonder if you could increase your "loop capacity" by including a small ref to air to add/remove some of the heat before sending it to your loop? If your heating and your evaporating temperature is 30, and its 50 outside, there would be gains to be had as it wouldn't be removing as much heat from your loop, no?

[Vlad]

Here is some info about my water to air HP/AC unit:

I got this unit for free with dead compressor (it was 265V anyways...).

Originally unit was installed in office building with water loop that had boiler for heating and cooling tower for cooling.

It was: 1 ton, 265V single phase, R22 (evacuated), cap tube, original charge 13.5 oz.

Now it is: 1 ton, propane, 2 txvs (with check valves ), 240V single phase, charge 7 oz.

I have some R 22 but I wanted to see propane performance. I distilled some auto propane. First I purged r22 tank(oops I wanted to say propane tank) (in the morning open valve very slowly just crack it and let non condensibles go. Make sure you don't shake the tank before you open it.) After I put tank in chest freezer to freeze it (it will help to keep butane and some moisture inside). After I started vapor recovery. I did added filter/drier.

So after this I got 15 oz of distilled propane. The smell is almost gone because of filter/drier.

Now I have my unit running 16°F subcooling, 15°F superheat, High side 115 Psi, Low side 63-65 PSI.

House is getting COOOOOL

[AC_Hacker]

Originally Posted by Vlad

Here is some info about my water to air HP/AC unit:

I got this unit for free with dead compressor (it was 265V anyways...).

Originally unit was installed in office building with water loop that had boiler for heating and cooling tower for cooling.

It was: 1 ton, 265V single phase, R22 (evacuated), cap tube, original charge 13.5 oz.

Now it is: 1 ton, propane, 2 txvs (with check valves ), 240V single phase, charge 7 oz.

I have some R 22 but I wanted to see propane performance. I distilled some auto propane. First I purged r22 tank (in the morning open valve very slowly just crack it and let non condensibles go. Make sure you don't shake the tank before you open it.) After I put tank in chest freezer to freeze it (it will help to keep butane and some moisture inside). After I started vapor recovery. I did added filter/drier.

So after this I got 15 oz of distilled propane. The smell is almost gone because of filter/drier.

Now I have my unit running 16°F subcooling, 15°F superheat, High side 115 Psi, Low side 63-65 PSI.

Very interesting conversion. So it sounds like you replaced the old compressor with an equivalent R22 model that was rated for 240V, right.

If you have some time, it woulld be nice to see some pics of your new machine. I assume you went with the refrigerant-to-water HX that was already in the machine. I'd be interested to see some details on that HX.

If you can figure any way to measure the efficiency of your new unit, we'd all like to know that.

Originally Posted by Vlad

House is getting COOOOOL

To my way of thinking, on of the best things about your new heat pump is you're not only cooling the house, you're also storing heat for the winter.

... a real double win!

Good work.

-AC

[RB855]

Pulled some new numbers. Using an open loop (read, city water)...
1 gpm flow
Water entering 85.8*f
Water leaving 103.5*f
~8900btu hr
Liquid temp 87.3*f
Hot gas 121.8*f
Ambient 86.8*f 54%
Evap discharge air 74*f
Draw 2.4amps ~550w @ 220v (1725 btu equiv)
Est Gain 5.2:1

Using very inaccurate amp clamp, total system current is 2.4amps @ 220v
So, say 550w. Pump not included as I'm using city atm for steady state testing.
----+edit
Lowered the gpm to get it closer to intended operating conditions
0.62 gpm
Water entering 88.7*f
Water leaving 115.7*f
~8500 Btu/hr
Liquid temp 92.7*f
Hot gas temp 134.5*f
Ambient 94.5*f 45%
Evap air 74.9*f
Draw 2.9amp 640watt @ 220v (2180btu equiv)
Est Gain 3.9:1

There seems to be a pretty significant efficiency hit with the temperature increase, Even though the liquid temp only went up a couple degrees. I didnt have gauges on it at the time so don't know what the pressures were running. About a 400btu/hr loss, and almost 100watts more power. Granted the incoming water temp had risen, as well as the ambient. But, it looks like I can go with significantly less water flow than I originally imagined. Seems 1gpm is more than suitable, vs the 5+gpm I thought I needed. Suppose the oversize BPHE are paying for themselves in this respect since the delta is reduced to 2* at 1gpm.

[AC_Hacker]

Originally Posted by RB855

Pulled some new numbers. Using an open loop (read, city water)...
1 gpm flow
Water entering 85.8*f
Water leaving 103.5*f
~8900btu hr
Liquid temp 87.3*f
Hot gas 121.8*f
Ambient 86.8*f 54%
Evap discharge air 74*f
Draw 2.4amps ~550w @ 220v (1725 btu equiv)
Est Gain 5.2:1

Using very inaccurate amp clamp, total system current is 2.4amps @ 220v
So, say 550w. Pump not included as I'm using city atm for steady state testing.
----+edit
Lowered the gpm to get it closer to intended operating conditions
0.62 gpm
Water entering 88.7*f
Water leaving 115.7*f
~8500 Btu/hr
Liquid temp 92.7*f
Hot gas temp 134.5*f
Ambient 94.5*f 45%
Evap air 74.9*f
Draw 2.9amp 640watt @ 220v (2180btu equiv)
Est Gain 3.9:1

There seems to be a pretty significant efficiency hit with the temperature increase, Even though the liquid temp only went up a couple degrees. I didnt have gauges on it at the time so don't know what the pressures were running. About a 400btu/hr loss, and almost 100watts more power. Granted the incoming water temp had risen, as well as the ambient. But, it looks like I can go with significantly less water flow than I originally imagined. Seems 1gpm is more than suitable, vs the 5+gpm I thought I needed. Suppose the oversize BPHE are paying for themselves in this respect since the delta is reduced to 2* at 1gpm.

Good information, very useful.

Curious thing with pump velocity... when the volume goes up, you may notice a reduction in temperature, but because you're moving more water, you may be transferring more heat (more BTUs/sec) even though the outgoing temp is lower.

Also, when you connect your unit to a water tank, and the heat pump is circulating water in the tank, instead of heating constant temp city water, your delta-T will gradually go down, and your efficiency will also decline, similar to this graph:

It will be interesting how it all plays out.

But your initial tests look very promising!

-AC

[RB855]

I hope to keep the tank turbulence down so that the water can stratify to at least some degree. I fully expect semi heated water to return to the unit, which was the purpose of the waste heat circuit I added (which in other worlds could be used as a inlet preheat or such). My early tests were using a 12 gallon bucket and there was a sweet spot around 110-115* inlet water temp, with some 3-5gpm flow. Before and after that point the energy returns dropped. I don't think Ill run another temp/time test till I have the unit hooked up to a proper tank. Bank account is hurting a little right now, so food and bills come first. Maybe next month I'll be able to sort out my tank connections and get some plumbing in. Still lacking a proper pump, so will have to do with whats on hand till future notice.

[Vlad]

I was looking for my old electronic PT chart but couldn't find it.
Tried many online PT charts but they mostly for regular refrigerants.
I was looking for R290 (propane).

Finally I was thinking what if I try Android apps. And here it is KoolApp by Danfoss. Very easy very accurate ton of refrigerants and it is right here in your cell phone or tablet .

[Vlad]

I installed 1 ton unit in 3000 sq ft section of my house primarily for heating fresh air in spring-fall season. This unit is running now in AC mode on city water (temporary). It is big enough to keep my house cool but it is not very hot outside. I have couple of options to increase AC capacity for hot weather.

1. install bigger unit. But I don't need more heat.
2. add extra unit . Need more space and plumbing and etc.

I was thinking about using kind of 2 stage cooling and even 2 stage heating arrangement.

This is just an idea:

Add water/air coil. This can be used refrigeration coil from walk-in cooler or...

Run water from underground loop through water (ex refrigerant) circuit.

In AC mode hot outside/inside air will pass through this coil first and after precooled go to heat pump/ AC.

Now AC will receive much colder air and will cool it more down.


By doing this we only need little more water flow to supply extra water to extra coil.

[AC_Hacker]

Originally Posted by Vlad

I installed 1 ton unit in 3000 sq ft section of my house primarily for heating fresh air in spring-fall season. This unit is running now in AC mode on city water (temporary). It is big enough to keep my house cool but it is not very hot outside. I have couple of options to increase AC capacity for hot weather.

1. install bigger unit. But I don't need more heat.
2. add extra unit . Need more space and plumbing and etc.

I was thinking about using kind of 2 stage cooling and even 2 stage heating arrangement.

This is just an idea:

Add water/air coil. This can be used refrigeration coil from walk-in cooler or...

Run water from underground loop through water (ex refrigerant) circuit.

In AC mode hot outside/inside air will pass through this coil first and after precooled go to heat pump/ AC.

Now AC will receive much colder air and will cool it more down.


By doing this we only need little more water flow to supply extra water to extra coil.

OK...

So your idea is to use water only (no refrigerant machinery) to cool down incoming air that will go to your air conditioner, as a kind of booster?

If that's correct, Randen tried something like this, but he did not have the AC following the water cooler, and he had uncomfortably high humidity. Looks to me like your arrangement will handle this nicely by isolating the humid air from house air.

And you can direct the condensation water into your wife's garden!

By the way, did you get your loop all finished up, pumps & piping and all?

-AC

[Vlad]

Originally Posted by AC_Hacker

OK...

So your idea is to use water only (no refrigerant machinery) to cool down incoming air that will go to your air conditioner, as a kind of booster?

If that's correct, Randen tried something like this, but he did not have the AC following the water cooler, and he had uncomfortably high humidity. Looks to me like your arrangement will handle this nicely by isolating the humid air from house air.

And you can direct the condensation water into your wife's garden!

By the way, did you get your loop all finished up, pumps & piping and all?

-AC

OK last few days I had a chance to test my AC as it is + 90F outside. This is a real test. 1 ton AC for 3000sqft house is not too much. When outside was 78F-80F It was OK inside but 90F is a killer.

So I added a huge coil that was used in walk-in cooler. I ran water through this coil. It precool mix of fresh and return air and this cooled down air is going to AC after. There is not much of TD so this coil almost doesn't remove moisture(it does, but much less then AC).

What a difference. We can live again. I used a small vent room as a plenum. Later I will add air filter and some controls to control fresh air mix and temperature.

AC, my ground loop is not connected yet. No pumps no manifold... I just temporarily use city water. This will give me idea what to expect from ground loop.

Here are some pictures of my Heat pump.
I had to add this funny dome because new compressor is higher then old one (nothing can stop DIYers )



Here is an old compressor:


And here is a new one placed on old's place:


Here is air coil with cap tubes(There were 3 cap tubes in parallel now I have 2 TXVs):


Here I unbrazed them:


Here I opened holes for new distributor tubes (I used 1/4" tubes):


Here I brazed distributor tubes to coil:


Here more pics of the same distributor:




This is a final view:


This distributor I made from 3 pieces of 1/4" copper and one short pies of 5/8" copper brazed all together using Sil-Fos 15%

Here is one very useful tool for connecting gauges with minimal loss of refrigerant:



You have to use it especially if you work with critically charged systems (all cap tubes). This tool allows to connect gauges and press shredder valve after your gauges already connected. Also you can depress shredder valve on high side first and open both valves on your gauges. High pressure will equalize with low side. After you depress low side shredder valve and after you disconnect your gauges. This allows to minimize refrigerant loss every time you hook up your gauges.