Dxgshpwh (ground source heat pump water heater)

[AC_Hacker]

Originally Posted by jeff5may

Guys, it was me who made the exaggerated claims.

jeff5may,

You are absolutely right, it is my mistake to confuse the moment with the refutation of stupid claims. My bad.

The fact is that MEMPHIS91 has taken the fully documented risk of trying a project that has incorporated so many advanced features, and with the encouragement and expertise of many people, he has it working.

MEMPHIS91, I think you are awesome.

Further, he has pushed the knowledge of DIY GSHP water heaters forward in at least three very significant ways:

1. MEMPHIS91 has come up with a great solution, using cheap commodity parts, to the problem of copper tubing exiting the water tank. Randen had a great solution too, but his ability to custom-fabricate a precision part may not be available to everyone.
2. MEMPHIS91 has proven that a DX evaporator can be done, and there does not appear to be any problem in this build due to DX. This is really BIG.
3. MEMPHIS91 has come up with a design that eliminated the problems of ice formation in the evaporator, and the design hassles associated with dealing with periodic icing.

Great, just great!

-AC

[stevehull]

Jeff, would a more extensive evaporator tube provide a better match for this particular sized scroll compressor? The point was made a couple times by Memphis that the water table is within feet of the surface. You mentioned at one point the advantage of thin walled parallel tubes being thermodynamically a better exchanger.

In this case, would a larger evaporator volume (actually surface area) maximize out the compressor size? Let's assume that it is in water and that it is not in moist sand.

I am very intrigued as I have a spare water well (4 inch diameter) that is quite deep (200 ft) and has water within 30 feet of the surface. What would your guestimate be of the BTU/hr collection for that situation? I could easily put in 400 ft of 1/4 inch copper tubing (down and up), but don't want to waste a lot of copper if I don't have to.

Memphis only had < 60 feet of copper in the water.

The other thing I could easily do is create a coil of copper (100 feet coil) that would occupy perhaps only 20 vertical feet of the water well, but below the water surface. Cold water, as a result of the cold evaporator, would sink away and the warmer water would be on the surface. This is a huge reservoir of BTUs.

Water temp is similar to Memphis at about 65 F.

I think we all agree that Memphis has done an outstanding job. I am intrigued to maximize COP for the least amount of copper, effort, etc.

Thanks in advance.

Steve

[jeff5may]

Originally Posted by stevehull

Jeff, would a more extensive evaporator tube provide a better match for this particular sized scroll compressor? The point was made a couple times by Memphis that the water table is within feet of the surface. You mentioned at one point the advantage of thin walled parallel tubes being thermodynamically a better exchanger.

In this case, would a larger evaporator volume (actually surface area) maximize out the compressor size? Let's assume that it is in water and that it is not in moist sand.

I am very intrigued as I have a spare water well (4 inch diameter) that is quite deep (200 ft) and has water within 30 feet of the surface. What would your guestimate be of the BTU/hr collection for that situation? I could easily put in 400 ft of 1/4 inch copper tubing (down and up), but don't want to waste a lot of copper if I don't have to.

Memphis only had < 60 feet of copper in the water.

The other thing I could easily do is create a coil of copper (100 feet coil) that would occupy perhaps only 20 vertical feet of the water well, but below the water surface. Cold water, as a result of the cold evaporator, would sink away and the warmer water would be on the surface. This is a huge reservoir of BTUs.

Water temp is similar to Memphis at about 65 F.

I think we all agree that Memphis has done an outstanding job. I am intrigued to maximize COP for the least amount of copper, effort, etc.

Thanks in advance.

Steve

The short answer is definitely maybe for a few minutes. The next shorter answer is it really depends on a whole lot of factors. Things like duty cycle, intended purpose, and level of complexity come to mind. A 4 inch static bore might not be enough to heat your whole house with. It may end up as a 4 inch, really long underground popsicle after the water gets too cold. I do know one thing, though: a pump and dump system would do the trick easily and reliably for not much pump power. A well full of warmth can be drawn from at will.

When this thread began, Jake clearly stated he had done lots of research, and had a rough idea of what he wanted to build. As it turns out, he had most of his rig figured out already, but needed a few blanks filled in before he was confident enough to start the build. The main unknown being the dx loop in the ground.

Jake plainly stated up front he knew the single loop might not perform. He indicated that digging more loops was no sweat. He said he had spare parts and compressors to throw in if a tiny one didn't perform well enough. After some construction details were decided upon, he manned up, went balls to the wall for a couple of laps, and there you have it. Less than two weeks later, the darn thing works like the plan said it should.

Sure, another borehole could be installed. It would definitely increase something other than the cost. But if he doesn't need it, why bother? This story isn't over yet, and the long-term viability and performance of his source loop has yet to be observed.

I don't believe our original poster is going anywhere fast. He has stirred up lots of activity and attention creating his misfit wonder.

[MEMPHIS91]

Ac, thanks, your work is pretty inspiring as well.

Jeff, Yes I did really fly through this project. Only reason was the wife is doing 4 summer school classes and is studying a lot. Gives me tons of time to tinker. And your very right, I do not plan to go anywhere. I plan to put some of the info I learned into more useful documents so someone doesn't have to read 150 post to find out the end result. And I have several more builds up and coming. My next one will probably be a small heat pump herb/food dehydrator. The heating element type one we have now pulls 500watts and its running 10+ hours a day. (We do aquaponics and have TONS of herbs and things growing all year) Plus its adding to the heat load of the house. I'll start a thread on the that soon.

The water heater is hooked up and has run some today. I'll post the weeks usage soon. Only 2 people in the house so it shouldn't be that much.

[AC_Hacker]

Originally Posted by MEMPHIS91

...My next one will probably be a small heat pump herb/food dehydrator. The heating element type one we have now pulls 500watts and its running 10+ hours a day. (We do aquaponics and have TONS of herbs and things growing all year) Plus its adding to the heat load of the house. I'll start a thread on the that soon...

Another great project idea, whether ASHP or GSHP. Your work indicates that in your area, GSHP/DX is not such a big deal.

Regarding the COP of such a project, as randen pointed out, the COP drops off as the mass of water that you are heating increases in temp (lower delta_T). If you do a heat pump dehydrator project, I don't anticipate a declining COP, since you may not have an increasing temperature to 'push against'.

* * *

I did some more graphs of your last report from your project, here's what they look like:

X = time, Y = tank temp

This one looks nice and linear, just like your prior run, only now we have more data.

X = time, Y = COP

This graph show that the COP declines over time, actually due to the increasing temp, seen in the chart above. Note the unusual dip in data that occured at about 120 minutes on the X axis. This affects all the rest of the charts.

X = tank temp, Y = COP

This graph directly show that the COP declines as tank temp increases.

time vs. hole temp

This graph shows the decline in the hole temperature over time, which is the result of extracting heat from the ground and 'pumping' it into the tank. Your heat pump will run periodically, so heat from surrounding soil will gradually flow into the 'heat vacuum' that your fiendish device has created. I would expect that in the long term, there will be a declining 'saw-tooth' temperature pattern that will eventually stabilize at some temperature point that is higher than the low point of this curve. A data logger could capture that.

X = hole temp, Y = COP

This chart shows how the DX hole temp affects the COP.



CONCLUSIONS:
No big surprises, that I see. Don't know why the odd reading from the 120 minute mark. I had readings like this also, and I'm still not sure what it is about.

The hole temperature decline is expected. It does trend toward a leveling off place, over time.

The water heater will run periodically, and between runs, the hole temp will regain heat from the surrounding earth.

I'm glad that MEMPHIS91 put temp monitoring into this build. My early experiments used exactluy the same kind of equipment You can learn a lot form this simple equipment. I thrashed around for quite a while , trying to find a reasonably priced temperature logger.

I finally settled on one that works and is cheap, called Multilogger. It assumes some tech skills are already in place, but it has worked for me.

There are other loggers out there which are more accessible to the general public.

You don't need a logger to build a project and make it run well, but they can be very useful to see trends over time.

* * *

Best,

-AC

[MEMPHIS91]

Thanks AC,
I would like the look into a data logger and did some looking into the one you posted.

I will monitor your assumption about the ground getting use to the borehole temperature swings, it already seems as though the ground heats up faster, then again its been 90-94F here lately.

7 days of use and the total power used with thermostat set at 120F is 16.1kwh

That comes to 2.3kwh a day with power at .11, that is 25 cents a day for hot water. We use hot water to wash the clothes with, for hand dishes, for showers (I normally have 2 showers due to my job), and we have a dish washer. So it may only be 2 people but we use a lot of hot water. $7.75 a month doesn't sound bad at all to me.

Thanks again for ya'lls help. I'll keep this up to date on how much power it uses long term.

[AC_Hacker]

Originally Posted by MEMPHIS91

...I will monitor your assumption about the ground getting use to the borehole temperature swings, it already seems as though the ground heats up faster, then again its been 90-94F here lately.

I hope you are keeping periodic records of ground temperatures... it will be very useful.

If you decide to go with the multilogger, I can elaborate on what I have learned.

Originally Posted by MEMPHIS91

7 days of use and the total power used with thermostat set at 120F is 16.1kwh

That comes to 2.3kwh a day with power at .11, that is 25 cents a day for hot water. We use hot water to wash the clothes with, for hand dishes, for showers (I normally have 2 showers due to my job), and we have a dish washer. So it may only be 2 people but we use a lot of hot water. $7.75 a month doesn't sound bad at all to me.

Thanks again for ya'lls help. I'll keep this up to date on how much power it uses long term.

If you really need 120F water, then all is well & good. If you can get by with water that doesn't need to be that hot, reducing the target temperature can yield surprising reductions in energy use. Also, an insulation wrap on the tank will help.

-AC

[MEMPHIS91]

Originally Posted by jeff5may

In your setup, a txv is going to be trying to be a fixed orifice, running full bore, until something bad makes it close. This something bad is going to be excessively high liquid line pressure or a flooding evaporator. You might as well just install a CPEV (constant pressure expansion valve aka flowrator) or spring regulator. If you do get a valve, make sure it is adjustable and rated for high temperature operation, close to your borehole temperatures.

As I suspected, you are quickly moving in the right direction. Trimming your cap tube dropped your discharge temperatures across the range and dropped your suction superheat as well. You just got closer to your optimal cap tube length. The heat flowed into the tank faster, taking 3 hours instead of 4 for a 30 degree rise. My guess is your power usage decreased the same amount.

Sorry for the multiple long posts, not trying to hijack the thread. Not trying to tell you what to do, just trying to give you the info you need to make good choices. You are just steamrolling through this project, and I hate seeing fast doers getting bogged down in details and theoretical maybe-mess.

Are you sure you haven't done this before?

I just was rereading over things, I guess I missed some information here.
Can you explain more about the CPEV?
And yes I am sure I've never done this before.

[AC_Hacker]

Originally Posted by jeff5may
In your setup, a txv is going to be trying to be a fixed orifice, running full bore, until something bad makes it close. This something bad is going to be excessively high liquid line pressure or a flooding evaporator. You might as well just install a CPEV (constant pressure expansion valve aka flowrator) or spring regulator. If you do get a valve, make sure it is adjustable and rated for high temperature operation, close to your borehole temperatures.

Originally Posted by MEMPHIS91

I just was rereading over things, I guess I missed some information here.
Can you explain more about the CPEV?
And yes I am sure I've never done this before.

jeff5may,

I have had serious problems finding sophisticated metering devices for equipment as small as MEMPHIS91 is working with, especially metering devices that are tailored for R290. I found inexpensive ebay options to be vanishingly rare for tiny machines.

I had to search for months to find a TXV that was designed for R290 in that size, let alone a CPEV.

Have you checked availability??

-AC

[jeff5may]

Ok, here comes the disclaimer again. I am not an HVAC guy, so I don't play by the rules as far as common practices and carbon copy responses. As experimental rigs, you can expect not to break even on your investment in equipment. The thing may squeal and let out magic smoke at the end of the day. Then again, it might work like it should. Success usually doesn't come easily the first time through. Nothing is certain, guaranteed, or implied to work as it should on paper. Your skill, tenacity and determination is required for any amount of success.

That being said, the pro grade tx valves can be made to work in a wide range. They have field replaceable parts: a body, a powerhead, and an orifice. Some of them can be changed from fixed superheat to constant pressure or adjustable superheat by changing parts in the body. Most of them can change capacity by changing orifices.

Sporlan/Parker and Alco/Emerson have equivalent products that cross to each other. Danfoss, not the same parts physically, but the valves will work in the same spot and do the same function.

All of them can be increased in capacity range by adding a distributor at the outlet. You can also add a piece of cap tube to choke them down in capacity, or to limit the full bore flow rate. When I built the "dirt cheap window heat pump", I found a txv in a 3 ton heat pump outdoor unit with a distributor. When I looked up the part number for the valve alone, it was rated at less than or equal to half that, right in the range of my compressor. I rigged it in without the distributor and it worked well.

There are lots of ways to make these valves do things they were not originally assembled to do. Nihao mike and Randen both have threads on here showing how they rigged up what to make it do it. My favorite mod is randy and his water hose spigot superheat adjuster. That tx valve just begs to be tweaked at least weekly, just for fun.