The Homemade Heat Pump Manifesto

[maciej_pl]

Originally Posted by AC_Hacker

Hello maciej_pl, I see that this is your first post to EcoRenovator. Thanks for joining us here on the 'Manifesto' thread.
This is very good. I hope you also have photos that you will be able to share with us, too. I think you may have to make a few more posts before the blog robot will let you share photographs.

I will post photos

Originally Posted by AC_Hacker

Do you know of any tests that indicate this?

There are some researches for boreholes stating that boreholes with larger diameter pipes can provide more heat.

Originally Posted by AC_Hacker

Actually, the chart was from a study of flat plate solar collectors that had water as the thermal transfer fluid. The collectors in the study were about 1.5 meter by about 3 meters.

I don't know the design, but the fluid spends some 30 seconds in this comparatively small heat exchager, so it has to acquire heat fast. Therefore turbulent flow is advisable in this case. The sun-lit copper pipe heats to high temperature, so the temperature difference betwen the medium and the pipe is large. A small copper exchanger is enough to heat domestic water tank, but due to small exchange area the flow has to be turbulent. That's how I see it...

Originally Posted by AC_Hacker

First of all, I completely agree with you that heat moves very slowly through the soil.

I think the thermal resistance of the fluid medium does not have to be reduced by making the flow turbulent. The thermal resistance of the fluid is not an issue with ground source, because it is much lower than the thermal resistance of the ground. It is impossible to make the ground turbulent, obviously In my opinion the only solution to the comparatively high thermal resistance of the ground is to elongate the time period of heat exchange. In other words: because the soil gives the heat in small doses over time, we have to leave the fluid in the ground for longer. I think the longer the fluid is in the ground before coming back to the heat pump, the better.

I think the way to improve heat exchange is by adding some elements on the outside of the pipe, like 10"-20" bolts sticking out in all directions, to make the heat flow faster in the ground.

I like to think about my ground source as an underground pond. The bigger the pond (larger diameter pipe), the more heat I can get from the ground.

The other way of thinking about the ground source is to represent each pipe as pipe in pipe exchanger. The inner pipe is 40mm PE pipe with glycol, the outer pipe is roughly 1m diameter pipe with stationary soil. Obviously you can't turbulently move the soil, so in my opinion it is no point in turbulently moving the glycol.

--
Regarding my project: my house has 210 square meters, floor heating, well insulated, heating energy use approx. 25 kWh/square meter per year.
I put 600meters of PE 40mm (instead of 25mm or 32mm) to compensate for high thermal resistance of the sandy soil I have, i.e. to assure long heat-exchange time in the ground exchanger. I wonder if it will work

[maciej_pl]

Turbulence aside, check this article:
"Life Cycle Cost Analysis for Ground-Coupled Heat Pump
Systems Including Several Types of Heat Exchangers", by Ngoc Bao Vu, Seung Rae Lee¤, Skhan Park, Seok Yoon, Gyu Hyun Go, Han Byul Kang

( I can't post URL links yet )


The life cycle cost decreases with increasing pipe diameter.

In my case, the bigger pipe with more glycol requires weaker circulation pump (pump 25mm/6 meters instead of 25/8 meters head) due to halved pumping head required. That translates to half the pumping electric energy required: 100W to pump fluid through bigger pipe instead of 180W to induce turbulence in smaller pipe. And even with turbulent flow you can't get more heat because the slowest part in this 'device' is the soil with high thermal resistance

In the case of 40mm pipe instead of 32mm pipe, the extra glycol and bigger pipe will pay for itself after... less than 2 years due to lower initial cost of the pump and then lower energy bills.

[AC_Hacker]

Until the robot lets you post hyperlinks...

The link to the "Life Cycle Cost Analysis for Ground-Coupled Heat Pump
Systems Including Several Types of Heat Exchangers" document is HERE.

This is good information, keep it coming...

Best,

-AC

[AC_Hacker]

This morning, prompted by the conversation with maciej_pl, I found some interesting new resources:

One is a free paper from IGSHPA that describes best practices when installing a ground loop for a GSHP system. The title is "Closed-Loop/Geothermal Heat Pump Systems - Design and Installation Standards 2010 Edition" and it can be found HERE. Quite a bit if the paper is references to other documents, etc. but there are some very good advice sections that professional installers are required to follow. Since we are DIY, we should also be aware of the industry standards.

The other is a computer program for sizing boreholes, that is for sale by IGSHPA called, glhepro 4. The price for the full featured programs is pretty high, at around $500 to $700 depending on features. But a trial version is available for $50. As of this posting, I don't know what the limits of the trial version are. Informations is HERE. The manual for the program is HERE.

I do think that some of the online calculators over at Build It Solar are of great value, and they are free. In particular, the "Fuels Comparison Calculator", "Home Heat Loss Calculator", "Solar Analysis Tools", and also the "How much solar collector area do I need to heat my home?". These tool are powerful, easily accessible, and free.

It would be very useful if we could come up with a set of simplified, effective calculators, on the order of the BiS calculators, for those hardy EcoRenovators brave enough to build their own loop fields. A simplified Loop Field Calculator could be one of these. I took a look at the source code for some of the BiS calculators, and they are open source.

Best,

-AC

[maciej_pl]

Originally Posted by AC_Hacker

Until the robot lets you post hyperlinks...


This is good information, keep it coming...

Best,

-AC

I mean, we are not sizing the ground loop according to the thermal resistance and properties of the medium in the pipe, but according to the thermal resistance of the soil. This is the limiting factor, not the thermal resitance of the medium, which is half the thermal resitance of the ground anyway, no need to reduce it using turbulences.
The cost of the turbulence (and thus achieved lower thermal resistance of the fluid) is high pressure drop per meter. So it is better not to induce turbulences in loooooooong pipes of the ground source heat exchanger The soil will not reward turbulences, because the medium can get twice as much heat from the ground with no problem, as it has half the thermal resistance anyway.


The limiting factor of the size of the ground loop is the thermal resistance of the ground. One way to overcome it is to increase the time the fluid spends in the ground by thickening the pipe.

The other way is to use longer pipe. But longer pipe needs to be thicker anyway to keep the pumping head low.


A very large diameter pipe would be enough to exchange a lot of heat. A perfect example is a pond. It consists lots of water, and you only take small amount to feed the heat pump. The inflow temperature never drops. That is a perfect ground source.

On the other hand is a short pipe with small diameter and little medium. No matter how turbulent the flow in it, the soil will not provide enough heat, due to high thermal resistance of the soil.

Imagine a ground source loop of 600m pipe of 1 meter in diameter.
It consists 500m3 glycol. So you can use it for heat pump for days and days, and not deplete the source. Heat pump send glycol -5'C, and pulls +5'C from it. Turbulence is not necessary in the case of ground source heat exchangers.






--- that's how I see it I will have a chance to see how this hyphothesis works in reality in 2-3 months.

[maciej_pl]

I think this is my 5th post, so I will be able to post images now

After checking in enlightened sources, the turbulent flow is required to stop depossit from settling on the inside of the pipe...

[AC_Hacker]

Originally Posted by maciej_pl

...After checking in enlightened sources, the turbulent flow is required to stop depossit from settling on the inside of the pipe...

I think this is only a small problem.

If you were building an open loop system, with new water coming in from some source like a municipal water supply, or even worse a well, this would be a big issue. But with your closed loop design, this shouldn't be bad enough to cause problems. If you were able to use distilled water for your system, it would be no problem at all.

Also, you might investigate the possible desirability of anti-bacterial additives to the loop water.

-AC

[AC_Hacker]

I just came across a very interesting site that has about a dozen free calculating tools, spreadsheet, and programs that are all GSHP related.

Of particular interest is one called, E-PipeAlator08. Here's the description:

E-PipeAlator08 (395kB download) is a program for designing water distribution systems and computing losses. Program handles water and water-glycol mixtures, (temperature corrected) in steel, cast iron, polyethylene, PVC and copper (a new feature) piping. Equivalent lengths for common fittings automatically entered. Heat exchanger, valve, and flow control valve losses also considered. New GSHP piping example calculations and figures have been added.

The other 'tools' can be very useful, too...

-AC

[maciej_pl]

Originally Posted by AC_Hacker

If you were able to use distilled water for your system, it would be no problem at all.

Also, you might investigate the possible desirability of anti-bacterial additives to the loop water.

-AC

The water will be filtered by a reversed osmosis filter, with 25% ethylene alcohol.

[AC_Hacker]

Originally Posted by maciej_pl

"Life Cycle Cost Analysis for Ground-Coupled Heat Pump
Systems Including Several Types of Heat Exchangers", by Ngoc Bao Vu, Seung Rae Lee¤, Skhan Park, Seok Yoon, Gyu Hyun Go, Han Byul Kang

The life cycle cost decreases with increasing pipe diameter.

I was reading over the paper, and here is a screen shot of the beginning of the conclusion:

Your comment?

-AC