DIY Air-Source Water HP

[launboy]

Originally Posted by MN Renovator

If you can cut the fan from the circuit, it would reduce power draw more. I would have expected more output than 3k BTU/hr but I have nothing to suggest.

I wish I could cut the fan out of the circuit but I need it to pull air through the evaporator coil. I'm actually considering adding a second blower from the dead 5000 btu unit. My hope is that this would pull more air through the coil and improve efficiency. I'm not sure if the second fan would pull enough extra air through to make a difference though.

I expected more output too, and I'm sure it was more because going off the 1* I lost during each 10 minute defrost I'll assume I had the 1* loss every ten min through the whole test which means I put in an additional 1200 btu's over the course of the test.

That addition ups my output to 3900 btu/hr, giving me a new COP of 1.43. Which still isn't great but it's better than the other number. Today's project is to get the cracked pipe on the hottub repaired and fill it so I can hopefully begin heating today.

Adam

[launboy]

Originally Posted by AC_Hacker

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

I fully agree, and may run another test in the tub with better measurements and a pump to see what kind of performance I get.

Originally Posted by AC_Hacker

I have a Kill-a-Watt that I like to use.

I just may invest in one of these.

Originally Posted by AC_Hacker

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 know what your saying there, I have three Indoor/Outdoor thermometers at school and they never read the same, close, but never the same. Two of them have humity sensors and those are even less accurate, one consistantly reads 10% lower then the other.

Originally Posted by AC_Hacker

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.

That's where I'm at, I want to do the experimenting and achieve the efficiency I know I can, I just don't have the means at this time. In the future when I'm more able, I'll definitely be improving the design with a larger evaporator and a real ref-water HX.

Originally Posted by AC_Hacker

If you haven't already done so, you should look at what Acuario has done with his work.

I've seen Aquario's work and it is an inspriation because he's doing what I want to! My system would incorporate solar into the mix too.

Originally Posted by AC_Hacker

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

Thank you! Hopefully it will only get better from here.

Adam

[launboy]

Today was a successful day! The HP is now installed and operational, and the best part of it is, it actually works.

Yesterday I built a bracket for the coil to hang from, prepped the hottub for filling, and put the unit in place. All that left for today was final checks and the fill. Aside from one loose hose clamp which was easily fixed, I have no leaks! This project has gone way too well for one of my creations, I hope nothing blows up on me.


View of the inside while filling. That hole where the hose is going in is where the filter sits and is the main intake for the pump to draw water from. When it runs most of the water is drawn through the coil on its way in.

The brackets holding the coil are actually the plastic frame parts from the things that slide to fill the extra space in a window, jsut with one side cut off so they're L's. They seem surprisingly sturdy, only time will tell how well it hold up.


View from the outside showing the evap and the pump, which I placed directly below the unit on purpose for ease of getting it's eat to the HP.


Frost was a definite issue today, as you can see, luckily it was 50* during the day(perfect day to do this) and only dropped down to about 42* when I took that pic at around 11:30pm after a half hour of running. Unfortunately the humidity was 66%, not good for ASHP's. On the upside, I've learned that as long as it's above freezing, simply turning off the compressor and leaving the fan on for about 10 minutes clears all the frost right up.

You'll also note that piece of insulation in the bottom of the pic with the space, that's what creates the "chimney" for the heat from the pump to be delivered to the HP.

Notes from today...

1. I need to figure out a better way to control the unit. To start it I have to dip the thermistor in warm water at first, then after a few min when the comp discharge line is warm I can keep it against that and it'll stay running. I don't want to have to deal with this anymore. Can I simply cut the wire and connect the ends together or will this kill my digital controls?

2.Defrost, this is clearly a problem that's not going away and since I dont have the dehumidifier control anymore I need to figure out a way to approach this. My simple idea is to run a timer to a small resistor which I'd wrap in against the thermistor. I'd set the timer to keep the resistor warm and depending on conditions, have it click off for 15 min allowing time for defrost. I don't know much about how to go about setting that up though so I'm not sure how well that'd work.

My other thought was to see if I can readjust the mechanical temp control from my other a/c to a lower temp and attach it to the line going from the evap back to compressor. This is the place that our self defrosting fridge and dehumidifier both had their sensors, and I noticed this line only gets frosted once the whole coil is frosted over. further testing is needed to see if this is possible. If so I'd set it up just like the other A/C where it just switched the compressor deirectly.

3. I'm not sure how much heat I was producing today due to not having enough time to get a good hour long test in and log some data, but my hottub is approx 350 gal and takes just under 3000 BTU to raise one degree. My goal is to achieve at least 2* an hour rise in moderate, 40* temps, giving me 6000 btu/hr output from n 8000btu/hr rated a/c. Does this seem reasonable or should I be able to get more? given 70* water and 30-40* ambient.

Adam

[launboy]

Forgot to add, so far I only have $30 invested in this project, $20 for the A/c, and $10 for some fittings I needed for plumbing. I think I'm doing pretty well so far!

Adam

[AC_Hacker]

You really don't want to mess around with the A/C when you are in the hot tub. That would be a really pathetic way to die... the headlines would be very embarrassing, especially if your friends were also in the tub with you. I certainly wouldn't want it on my conscience.

Originally Posted by launboy

1. I need to figure out a better way to control the unit. To start it I have to dip the thermistor in warm water at first, then after a few min when the comp discharge line is warm I can keep it against that and it'll stay running. I don't want to have to deal with this anymore. Can I simply cut the wire and connect the ends together or will this kill my digital controls?

I don't think it would be a good idea to short the wires. Maybe unplug the thermistor from the circuit board if possible, or cut one of the wires so you had an open would be ok... if that causes your A/C's brain to stop working, it would be easy to re-connect.

I don't know if you are into electronics, but if you could get the actual schematic of the controller board, you might be able to re-purpose it. Your controller board still thinks that it's job is to make things cool. Thermistors can have a positive or a negative temperature coefficient, I don't know which yours is. If you could understand the circuit, and get a thermistor with the opposite temperature curve, it could make your controller act like your A/C was a heater, which it actually is now. Another, simpler approach is to get an ohm meter, and see what the resistance is of your thermistor at say, 80F and find a potentiometer that would be close to that value at half-turn. then you could dial the 'pot' so your unit stays on.

Originally Posted by launboy

2.Defrost, this is clearly a problem that's not going away and since I dont have the dehumidifier control anymore I need to figure out a way to approach this. My simple idea is to run a timer to a small resistor which I'd wrap in against the thermistor. I'd set the timer to keep the resistor warm and depending on conditions, have it click off for 15 min allowing time for defrost. I don't know much about how to go about setting that up though so I'm not sure how well that'd work.

Yeah, your A/C (now heat pump) is getting all of it's heat from the little HX that keeps frosting up... you are dumping loads of heat out of the system because you have the other, bigger HX in direct contact with a very large volume of water... so you have loads of heat going out and little coming in. Obviously it will work, but if there was some way to supply loads of heat at the input side, there'd be less frosting, and more water heating (and bigger COP). It would be really great if you had an air HX that was MUCH bigger that what you now have.

If you dug a 40 foot trench about 6 to 8 feet deep and buried about 200 feet of plastic pipe arranged like a slinky and circulated water, that would really help, or dug a borehole about 100 feet deep and circulated water, or if there was a pond or flowing stream really close and you could throw the evaporator HX into that pond, that would help a lot, too.

Lacking these measures, having something that would periodically turn the A/C off for a period of time so the frost melts, and then back on might help.

-AC_Hacker

[launboy]

Originally Posted by AC_Hacker

You really don't want to mess around with the A/C when you are in the hot tub. That would be a really pathetic way to die... the headlines would be very embarrassing, especially if your friends were also in the tub with you. I certainly wouldn't want it on my conscience.

Oh trust me I wouldn't run this with people in there. That chance isn't one worth taking for me. I'll run the pump, but the a/c will be off and unplugged.

Originally Posted by AC_Hacker

Maybe unplug the thermistor from the circuit board if possible, or cut one of the wires so you had an open would be ok....

Thermistors can have a positive or a negative temperature coefficient, I don't know which yours is. If you could understand the circuit, and get a thermistor with the opposite temperature curve, it could make your controller act like your A/C was a heater, which it actually is now. Another, simpler approach is to get an ohm meter, and see what the resistance is of your thermistor at say, 80F and find a potentiometer that would be close to that value at half-turn.

I can unplug it, and tomorrow I think I'll try that. I also have a multimeter that does Ohm's, I didn't realize they came in opposite resistance styles. The simplicity of the pot is appealing though.

Originally Posted by AC_Hacker

Yeah, your A/C (now heat pump) is getting all of it's heat from the little HX that keeps frosting up... you are dumping loads of heat out of the system because you have the other, bigger HX in direct contact with a very large volume of water... so you have loads of heat going out and little coming in.

This makes perfect sense! I'm thinking an evap that was at least double this size would be nice. I consider this heatpump my beginning point to learn from so I have a little background knowledge when I custom build version 2.0. So far I've learned three things for sure, oversize the evaporator, double to triple its original size. The other thing is that I definitely need to build a tube in tube condenser, probably something like Aquarios design. The third is I need a reliable defrost control and a temp control actually designed for heating.

Originally Posted by AC_Hacker

If you dug a 40 foot trench about 6 to 8 feet deep and buried about 200 feet of plastic pipe arranged like a slinky and circulated water, that would really help, or dug a borehole about 100 feet deep and circulated water, or if there was a pond or flowing stream really close and you could throw the evaporator HX into that pond, that would help a lot, too.

Lacking these measures, having something that would periodically turn the A/C off for a period of time so the frost melts, and then back on might help.

Where I live I think vertical would definitely be the way to go. I'm about a 1.5 miles from Lake Michigan and as far as I know our soil is a few feet of clay but after that primarily sand. Being this close to the lake our water table can't be very deep. I can't find any info on our water table in Kenosha, WI.

For now, I'll just have to come up with a way to turn the unit on and off to allow for defrost.

Adam

[launboy]

One other thing, how do I handle defrosting in below freezing weather? Real heatpumps have reversing valves. I don't have that luxury.

Adam

[AC_Hacker]

Originally Posted by launboy

The other thing is that I definitely need to build a tube in tube condenser, probably something like Aquarios design.

If you can find copper tubing that isn't too expensive (or free) that's a good way to go.

Here are some HXs I got for cheap. If you get soft copper tube for inside & outside, straighten them out, insert the smaller inside the larger, then coil them up together... then get some copper Tees and braze the ends on... it's not so hard to do. I was lucky in that I found these HXs for cheaper than the copper to build them.

You might also haunt ebay for brazed plate HXs. Good chance that you can find a complete unit for less than you'd spend on copper to DIY. Look for brazable connectors (AKA: sweat connectors) on the refrigerant side, NPT or barb on the water side.

Originally Posted by launboy

Where I live I think vertical would definitely be the way to go. I'm about a 1.5 miles from Lake Michigan and as far as I know our soil is a few feet of clay but after that primarily sand. Being this close to the lake our water table can't be very deep. I can't find any info on our water table in Kenosha, WI.

If you have the available property to put in a trench with slinkies, that's much less work, much less that can go wrong.

But a properly sized bore hole will out perform a trench by virtue of it's depth... and more constant temperatures at greater depth.

Originally Posted by launboy

For now, I'll just have to come up with a way to turn the unit on and off to allow for defrost.

The reversing valve is really a great way to go, the defrost cycle happens very quickly... less than a minute.

I had a very small de-humidifier, and I noticed that it would just stop the compressor, and that apparently the refrigerant would just drain down into the evaporator coil, and that would cause the frost to melt pretty quickly... then the compressor would start again, etc. Worked pretty well.

When you get setup with tools, you'll probably start collecting various vapor-compression machines, and you'll have available parts. Like for instance, I have a 1 Ton portable ASHP with a reversing valve for the purpose of changing from heating to cooling. I also have a 65 pint/day de-humidifier with a reversing valve for the purpose of frost removal.

It's pretty amazing actually, that it is so easy and cheap to acquire this kind of hardware for experimenting. If you had asked me 5 years ago, what I thought was the chance that I would become an HVAC hobbyist? I would have thought you were out of your mind.

Now I am one.

-AC_Hacker

[launboy]

So after running for 5.5 hours which was probably closer to 4.5 - 5 hours when defrost time is subtracted in, I calculated an Average COP of 2.6 to 3.1 which I'm not exactly sure of because I'm not exactly sure if my heat loss over the time period was 4* or 6*. How does that COP sound with an avg 42* outdoor temp?

Originally Posted by AC_Hacker

If you can find copper tubing that isn't too expensive (or free) that's a good way to go.

You might also haunt ebay for brazed plate HXs.

Where'd you find those exchangers so cheap? I think I might have issues with corrosion or clogging and flow rate problems using a brazed plate exchanger on pool water. I need to flow at least 30 GPM through this. I'm thinking a tube in tube with a PVC pipe outer shell for ease of connectivity to pool plumbing, and an inner copper coil, much like Acuario did, seen here :http://www.tortosaforum.com/poolheater.html#_self though I’m not sure how to properly size that, any tips? My hacking unit will be the 8000 btu window A/C I’m using now.

Since I’ll have the circuit open, I’m improving my evaporator coil too. I’m just going to use what I’ll have, so my choices are 8000 btu evap coil, 8000 btu condensing coil, 5000 btu evap coil, and a 5000 btu condensing coil. I’d Like to only use two coils because the way the pipes are set up I could use two coils easily(the evap right now has 2 circuits). I’m thinking the 5000 and 8000 btu condensing coils because they’re both thinner, allowing air to flow easier and size wise, they’re just larger.

Originally Posted by AC_Hacker

If you have the available property to put in a trench with slinkies, that's much less work, much less that can go wrong.

But a properly sized bore hole will outperform a trench by virtue of its depth... and more constant temperatures at greater depth.

That really would be the way to go but right now, a little more than I want to undertake.

Adam

[launboy]

Maybe I'll be lucky enough to acquire a reversing valve, it seems like those complicate the system quite a bit though. How does that work with the cap tube?

Adam