The Homemade Heat Pump Manifesto

[pick1e]

Originally Posted by AC_Hacker

Also, you have to consider the geometry of heat flow. In the case of a bore hole, you have a pipe in the center. When you first begin your extraction, you get more heat from the earth immediately surrounding the pipe, then it is heat-depleted and needs to get heat from earth further away, etc. And as time passes, there is progressively less and less heat extracted, and the curve gets flatter.

AC Hacker,

Actually, as I look at this graph again, it actually confirms what I mean. In this test you were adding heat with a coffee cup heater, right? The graph shows the temperature leveling off... Think about it, if the borehole wasn't able to keep up with sinking the heat, the T would be shooting straight up and your water would be boiling! Instead, as it reaches a higher delta T, it is leveling off, so the borehole is actually keeping up and reaching equilibrium.

But as you pointed out, whether that would remain true for weeks or months is certainly a different story. My point is simply that the rate of heat transfer through the soil seems like it should depend on the delta T, not just be static.

[AC_Hacker]

I think the problem here is that you might be reading the "Temp F" part of the chart and thinking that it means "delta T". It really is the temperature of the water in the test box. This test is usually done by measuring the temp of the water going into the loop and the water coming back from the loop, averaging them mathematically and plotting that.

I made the simplifying assumption that since the water coming back from the loop was splashing and swirling around in the test box, it was physically averaging itself, so there was no need for separate measurements and an averaging computation.

Originally Posted by pick1e

AC Hacker,

Actually, as I look at this graph again, it actually confirms what I mean. In this test you were adding heat with a coffee cup heater, right?

That is correct, about 200 watts as I recall... I also added in the power of the pump, which was about 50 watts.

Originally Posted by pick1e

The graph shows the temperature leveling off...

To be really correct, the slope of the curve is getting progressively smaller, in a logarithmic fashion.

Originally Posted by pick1e

Think about it, if the borehole wasn't able to keep up with sinking the heat, the T would be shooting straight up and your water would be boiling!

Language actually matters.

Do you really mean to say, "shooting straight up"? This would take an infinite amount of heat. I only had 250 watts.

Originally Posted by pick1e

Instead, as it reaches a higher delta T

I think you need to tell me what you actually mean by "delta T"...

As I consider the graph, in the first part of the test, the delta T (difference between the temp of the water in circulation and the temperature of the earth surrounding the pipe), the delta T is highest at the beginning of the test and more heat is transferred, as reflected in the lower temp in the box.

Then because heat travels so slowly through earth, the delta T becomes gradually smaller, and there is a temperature rise of the water.

Originally Posted by pick1e

it is leveling off, so the borehole is actually keeping up and reaching equilibrium.

Again, words actually matter.

The thing that amazed me about the test is that the borehole did not reach equilibrium. I expected it to do so in about six hours.

The test only ran 157.3 hours, enough for my purposes, which was to determine how much heat I could expect to pull out of the hole, on an ongoing basis.

But in 157.3 hours it never reached a balance.

Originally Posted by pick1e

My point is simply that the rate of heat transfer through the soil seems like it should depend on the delta T, not just be static.

When I said,

Heat moves through earth very slowly. I have heard various rates, but it's in the neighborhood of 6 months for heat to move through 16 feet of earth.

or,

The thermal transfer rate through soil is very slow, on the order of 2.6 feet per month.

...I thought that as someone with a technical background, you'd be able to recognize a simplifying generalization that pointed at a larger idea, when you saw one.

-AC_Hacker

[Vlad]

Originally Posted by pick1e

But, we are on a DIY thread, not a mass production thread. So there are far more variables that can be considered, not the least of which are soil conditions and resulting installation types and resulting labor. And you're equating copper embedded in concrete with copper in soil, which is totally different. Concrete is extremely alkaline which will corrode most things including your skin

I'm not concerned with the longevity of the system, I don't plan to live here more than a a few more years. So advising me to not TRY copper tubing for my project based on the premise of longevity isn't a reasonable argument.

I guess my point is saying "there are some points to consider before choosing a ground loop material" is far different from "copper won't work."

I am not talking about commercial production. DIY for me means better then commercial. For this reason I built my 4200 sqf house myself. I am tired of seeing low quality jobs and cheapest materials. I choose the best material and my house will last much more then I or my kids need. I might sell it after 5 years but I will sell it with good feelings.
In your particular case I would buy portable heat pump(like portable AC). Because you do not need much heat and when you decide to move you will be able to take it with you to your next garage.
GHSP have very long payback period.

[Vlad]

Originally Posted by pachai

As a follow-up to the idea of using two bore holes
near each other with an underground link.

(At this time, I am not evaluating whether
proximity reduces efficiency.)

(note, I worked for a few years in Rockefeller Center,
a complex of 19 commercial buildings that are
connected by an underground mall :-)


Now, being an IT guy and a DIY home renovator,
and with 15 years of experience "fishing" wires
with the wrong tools and the right tools, the next
part is easy.
Fish a pull string from one hole to the other
and pull it up. Tennis ball may help.
Or, put an elbow on the 2"pvc shopvac extension.
As in Perl, there is more than one way to do it.

Attach an end to a pipe and pull ( gently).
(Lubricate - with water?) Actually, push the
pipe down while guiding it with the pull string.

The Advantage to this is, I can have a vertical bore
with no weld. I hate to harp on this, but I TRIED to learn to weld to make battery racks, until my friend
described the metallurgic drawbacks of welding. Drilling and bolting makes fewer stress points.
And requires fewer skills.


Another thought...
We originally thought to have a U tube in the hole.
Now we have an "l" in the tube.
Maybe we could have an || ?

Two parallel loops in passing through
the same PAIR of bore holes.
The question is, should water in the pipe
flow the same direction in both,
or perhaps they should be in opposite directions - one will
be transferring the most heat in the top of the hole
and the other in the bottom? (or, in my design,
one is doing more in hole A and the other in hole B)

My reasoning with two loops passing through each
pair of boreholes is to not forfeit the linear footage.

The above question of flow direction highlights the
thought I had that each pipe should be labelled
on each end that will emerge into the basement..
but even labelled IN and OUT so that you know.

Also label your diagram to match, so that if a
leak is detected at the manifold, you can
refer to the diagram to see just what to dig up.

Hope this help spark more thought on this...
Seth

Sorry about this but I think you do not know the difference between wire and pipe. For me was challenge to push 2 3/4 HDPE pipes with very compact u-bent on the bottom into 5 inch straight hole. you are trying to turn pipe 90 degree down in the hole. LOL

[pick1e]

Originally Posted by AC_Hacker

I think the problem here is that you might be reading the "Temp F" part of the chart and thinking that it means "delta T".

Noper, I am extrapolating

The temperature difference (delta T or dT) is the difference between the ground temperature and the water temperature (or in a more general sense the temperature difference between hot and cold, causing heat transfer). It is what causes the transfer of heat across the boundary of the tubing.

I assume you started with tap water at roughly the same temp as the ground, so the original delta T would be 0. There is no difference. So if you just let the water sit there without heating or pumping, it would remain at 54 degrees because with delta T of 0, no heat transfer will take place.

Q=mCdT If dT=0, then Q=0.

But, if you look at the data, at 4 hours you have 79 degrees in your water. You don't have a reading for the ground temperature surrounding the tubing, but say for the sake of argument it was raised to 60 degrees (it should start to rise due to the warm water nearby). Now your delta T is 19 degrees. Then, at 7.5 hrs, your water is at 85. If your ground were at 65 degrees, your delta T would then be 20.

Since you are adding heat to the water, you are creating a difference in temperature, causing heat to transfer from the water, across the tubing walls, into the soil. The soil is soaking up the heat you are adding due to the temperature difference. And that transfer of heat is directly proportional to the temperature difference.

The point I was trying to make is that even though you are adding a constant (linear) amount of heat energy from the resistive coil of the coffee heater, your rise in temperature of the water is not linear. In the first half hour your water heats up 6 degrees, but at the end it requires 74 hours to heat the water 6 degrees.

That means the ground is catching up with the heater coil. The only way that can happen is by a growing difference in temperature, allowing more and more heat per unit time to transfer through the tubing. Eventually, because of the temperature difference, the rate of energy absorbed by the soil will equal that of the energy being added by the heater.

Here is what I mean graphically:

Originally Posted by AC_Hacker

To be really correct, the slope of the curve is getting progressively smaller, in a logarithmic fashion.

Indeed, yet you are adding heat linearly. This means the soil is catching up.

Originally Posted by AC_Hacker

Do you really mean to say, "shooting straight up"? This would take an infinite amount of heat. I only had 250 watts.

No, not literally. It will shoot up, straightly, but over a little. See the graph I sketched. If the amount of heat transferred to the soil per unit time was not increasing, then your water temperature would progress linearly until it boiled, in about 13 hours.

Originally Posted by AC_Hacker

the delta T becomes gradually smaller, and there is a temperature rise of the water.

Here is where I think you are misinterpreting the results in order to explain them according to what you are thinking. If the rate of heat transfer were not accelerating, plotting the water temperature against time would give you a straight line since you are adding heat at a constant rate.

For example, put that pump and heater in an insulated box with water and no outlet or inlet and plot the temperatures again. Of course you would expect to see a reasonably straight line with a positive slope. Now, imagine if you connected the ground loop and your soil was absorbing the heat energy at a constant rate. Then you would still have a straight line, the slope would be smaller.

Originally Posted by AC_Hacker

The thing that amazed me about the test is that the borehole did not reach equilibrium.

Maybe not completely, but pretty close. The temperature was rising, but not much. You didn't see the rapid rise in temperature at the end that you had at the beginning. In the final days you were getting what, 1 degree rise per day? At the beginning you had 6 degrees per hour! What you were approaching near the end of your experiment were the Thermodynamic magic words:

Steady State

Originally Posted by AC_Hacker

...I thought that as someone with a technical background, you'd be able to recognize a simplifying generalization that pointed at a larger idea, when you saw one.

I recognize oversimplification of a complex process.

[AC_Hacker]

Originally Posted by pick1e

Here is where I think you are misinterpreting the results in order to explain them according to what you are thinking. If the rate of heat transfer were not accelerating...

Good luck on this one.

* * *

I hope your work on your project goes well. Don't forget to share photos.

There may be a different site for this kind of work in the future. Do you wish to be notified when it comes on line?

-AC_Hacker

[AC_Hacker]

Originally Posted by pachai

Attach an end to a pipe and pull ( gently).
(Lubricate - with water?) Actually, push the
pipe down while guiding it with the pull string.

I think it would be a good idea to get your hands on some HDPE, if that is what you want to use, so that you can understand the nature if it's flexibility, and also the minimum bending radius before it begins to collapse.

Originally Posted by pachai

The Advantage to this is, I can have a vertical bore
with no weld. I hate to harp on this, but I TRIED to learn to weld to make battery racks, until my friend
described the metallurgic drawbacks of welding. Drilling and bolting makes fewer stress points.
And requires fewer skills.

Sorry to hear that you had bad experiences welding. I did too. Welding HDPE is much, much easier. I hope you saw the part of the blog where I show how I made a HDPE fusion welding device for less than $10.

If you knew how easy, fast, cheap, strong and long-lasting HDPE welding was, you just might change your approach to this project.

Originally Posted by pachai

Another thought...
We originally thought to have a U tube in the hole.
Now we have an "l" in the tube.
Maybe we could have an || ?

Two parallel loops in passing through
the same PAIR of bore holes.
The question is, should water in the pipe
flow the same direction in both,
or perhaps they should be in opposite directions - one will
be transferring the most heat in the top of the hole
and the other in the bottom? (or, in my design,
one is doing more in hole A and the other in hole B)

My reasoning with two loops passing through each
pair of boreholes is to not forfeit the linear footage.

The above question of flow direction highlights the
thought I had that each pipe should be labelled
on each end that will emerge into the basement..
but even labelled IN and OUT so that you know.

Also label your diagram to match, so that if a
leak is detected at the manifold, you can
refer to the diagram to see just what to dig up.

Hope this help spark more thought on this...
Seth

The Ground Source Heat Pump manual I refered to elsewhere goes into some detail regarding having more than one "U" in a hole. There is a gain, but it is smaller than you might wish, something on the order of 7% gain.

When you consider the difficulty of installation, it might be easier and cheaper to just dig another hole, and install more pipe.

If you are really space challenged, then gaining the extra % could be worth it.

There may be a different site for this kind of work in the future. Do you wish to be notified when it comes on-line?

-AC_Hacker

[pick1e]

Originally Posted by AC_Hacker

Good luck on this one.

I think it's a bummer that there is so much good and helpful information here, and yet some bad information that you have no interest in correcting. So I guess you are more interested in showing off than helping other DIYers.

How many people are going to follow what you have done, and waste their time graphing useless information and calculating what they think is thermal conductivity from an erroneous equation that was grabbed from some kid's failing school paper instead of trying to understand some thermodynamics? (You could ask the paper's author how they arrived at that equation but they didn't even put their name on the paper.)

Instead of all that work and getting a "k" value of 0.569 (which should be between 1 and 2 for soil, even the bad paper says that on page 5.) you could have saved a lot of effort by simply using the thermal conductivity of HDPE pipe, 0.50, since it is your bottleneck anyway.

[AC_Hacker]

Originally Posted by pick1e

I think it's a bummer that there is so much good and helpful information here...

Thank you for your splendid vigilance.

The great thing about DIY is that you now get to build your system your way and demonstrate by the performance of your system, the superiority of your approach.

And as to the testing methodology, you also get to develop a truly relevant method. Not only I, but the whole GSHP industry stands to benefit from what you will do.

I am very interested in the results of your efforts.

We are very fortunate to have someone of your discerning perception working with us.

Don't forget the photographs...

Best Regards,

-AC_Hacker

[pick1e]

AC,

My qualms with some of your methods aside...

You have a lot of good resources linked here and there throughout this huge thread, have you considered compiling them- for example at the bottom of the first post?