May be I am missing something here, but Fourier's law of heat transfer and total capacity of the loop are not really the same thing.

If you extract the same amount of heat (your building load does not change) out of lesser ground, your ground temperature has to decrease further. While the resistance is less due to the high conductivity, your surface area is significantly less, can result in the same overall heat transfer.

However, what appears to have been disregarded is the fact that the ground is much colder since the same amount of heat had to be extracted from lesser thermal mass of ground, and therefore the refrigerant pressures have to be higher, resulting in lesser efficiency of the compressor/refrigerant circuit. In order to maintain the same delta t (or better as you claim) the refrigerant temp has to be much lower.
The better conductivity of copper will result in lesser heat transfer resistance but will not increase your delta T if the ground is much colder, which it has to be if you extract the same amount of heat out of half the ground.
A loop is not a steady state environment.

The principles are correct and accounted for. Values (like "much" and "half") are open to debate.

Lower ground temp (caused by shorter loops) means the liquid temp will be lower. Less delta T from the compressor to the ground (no heat exchanger, copper pipe vs plastic, grout, etc) means the liquid temp will be higher - do it right and these can balance out and produce the same amount of heat at the same efficiency.

If you are willing to accept lower efficiency, then of course you can shorten the loops and not offset it with better HE/liquid/tube/grout conductivity.

We are talking about significantly lesser ground temperatures, given that vertical DX loops are often less than half as long. So what is the delta t between the liquid inside the water loop and the ground? 2-3 F? Now how much do you gain with the copper? 1-2F? By how much will the liquid temp be higher now? But how much will the ground be colder if you extract the same amount of BTUs out of half the ground? Specifically if the borehole conductivity by far outpaces the ground conductivity, in other words the ground cannot transfer more heat to the copper pipe, what to you think will happen to the ground around the copper pipe? Temps will simply drop quickly. While the heat transfer equation is relatively simple, it assumes a relative constant delta T between the refrigerant and the ground. But it ignores that the in order to keep that delta T constant, the refrigerant temp has to get low and lower. This is when your efficiency tanks, and refrigerant pressures and the strain on the compressor increases.

Hi docjenser,
Fourier's law is what gives you the loop capacity. Of course to calculate the actual system capacity it gets slightly more complicated and you to account for the convection in the fluid as well as the compressor operation (standard vapor-compression cycle in the case of DX). And, as you pointed out, you also have to account for the fact that it is not a steady state environment. But as far as what happens below the ground, Fourier's law is all you need.

What you are saying would be correct for two identical geothermal systems with one that has shorter loops than the other one. Intuitively, the one with shorter loops will have less capacity and may have a lesser average efficiency since it will be run close to full capacity more often.

However we are not comparing two identical geothermal systems here. EarthLinked systems use copper loops and refrigerant (R410A). R410A simply allows for a higher temperature differential than an antifreeze solution does, they have different thermal properties. If you plug that in Fourier's law, it tells you that you can reduce the length of the loop to achieve the same capacity as a water-based system.

Everything I've said so far is not new to this thread and based on your post, I'm sure you understand it already. Now a common misinterpretation after all of this is to think that, delta T being higher for DX, it means that "the ground will be depleted faster" or that "since you are working with less ground, your capacity and efficiency will tank faster". I have heard that many times and it is really the equivalent of saying that the loop is undersized (despite the fact that we sized it with Fourier's law). It is not. The big misconception is to think that we are working with less ground to achieve the same capacity, that's just physically impossible. The matter of fact is that, for a same capacity, refrigerant-based and water-based use the same volume of ground. Refrigerant-based has a further reach away from the loops (delta T) and water-based has longer loops. If you look at the ground in terms of your heat source/sink, then that heat source/sink looks like a cylinder around the loop. For DX that cylinder is shorter and thicker, but with the same volume than the narrower and longer cylinder found in water-based systems. That's the correct way of interpreting the bigger delta T.

Hope this helps and thanks for a great discussion,
Gregor

This is fairly off subject, but it's not possible to use any type of plastic lines with any of the common refrigerants?

Gregor,

indeed a good discussion.
Now I disagree with you that Fourier's Law is what gives the loop capacity, since it only accounts for conductive heat transfer, but not for convection like water flow through the ground etc.
But lets stick with Fourier's Law and the heat transfer through a hollow cylinder, or a multilayered cylinder, and its application for DX and water loops.

As you pointed out in your geo outlook article (http://www.geooutlook.org/epub/GO2015No4/files/assets/common/downloads/page0005.pdf).

1) "Fourier’s Law shows that the heat transfer rate is directly proportional to the magnitude of the temperature difference across the layer, the heat transfer area and the thermal conductivity of the material it is traveling through, and is inversely proportional to the thickness of the layer."

2) "Copper ground loops have a very high thermal conductivity (k), meaning the overall thermal conductivity of the ground loops is slightly higher than that of plastic or other materials. "

3)" Since we are trying to achieve a fixed and pre-determined conductive heat transfer amount (Q) to meet the load demand of the home, having a larger temperature gradient and a better thermal conductivity allows for a minimized surface area (A)."

To 1) The better conductivity of copper is almost irrelevant since the thickness of the layer is very thin and the conductivity is excellent. Because of it there is almost no delta T between the inside of the copper pipe and the outside, there is almost no heat transfer resistance of the copper pipe. So the surface area of the heat exchange is mainly occurring in horizontal direction, now you have the cement (or the grout) and then you have the earth, which is really a bad conductor of heat. Making your cylinder shorter but thicker, you not only decrease your area of heat exchange, which you correctly argue could significantly decrease the area of heat exchange. The problem now becomes that the cylinder which is much shorter (lesser area) is not only much much thicker, but also has a much lesser amount of conductivity.
If you now plug the numbers into Fourier's Law for multilayer cylinders, you see that the denominator in the formula is mostly made up by the thickness of the ground layer and (in an indirect proportional way) by the conductivity of the ground layer. Large thickness and bad conductivity make the denominator very large, to the point that the nominator (delta T and surface area) becomes less important. The copper layer (and to a certain degree the cement layer) becomes almost irrelevant, because they have almost no resistance, but the much thicker cylinder of poorly conductive ground has a huge resistance.
In other words, you can make the delta T very large, but if the ground is unable to transport the heat to the (highly conductive) pipe due to a bad ground conductivity and the long distance it has to travel (thick cylinder) through that badly conductive material (ground), it will still result in very poor overall heat transfer.

Now, according to Fourier's law, the only way you can suck more heat out of the ground is to make the delta T even higher, meaning colder refrigerant, lesser efficiency and more stress on the compressor.

Which is why a long and thinner cylinder gives you not only more area, but also a much thinner layer of "insulation" (ground) your heat has to travel through.

Hi docjenser,
I apologize for my delayed response, it has been a busy summer. But I really have enjoyed this conversation.

Regarding 1), I understand your point but the conductivity of copper does matter as compared to that of HDPE (~800 times less than copper). It’s true that in a DX system alone, it’s so conductive that you can just ignore that layer in your sizing calculations, but that’s not true for a water-based system. So it does give us an advantage.

Not very important in the scope of this discussion, but I wanted to point out that research shows that axial heat transfer actually represents up to 15% of a geothermal system’s capacity!

“The problem now becomes that the cylinder which is much shorter (lesser area) is not only much much thicker, but also has a much lesser amount of conductivity”. It is actually the opposite! As mentioned, copper is 800 times more conductive than HDPE and our loops are NOT 800 times smaller. So our “Cylinder of heat transfer” has a better conductivity.

I think you’re misinterpreting Fourier’s law. First, you have to keep the comparison in mind: water-based vs refrigerant-based. The ground and grout conditions are the same, so all of that is fixed in Fourier’s law. Now if your delta T is bigger, you can reduce your surface area (shorter loop) to obtain the same capacity. That’s why DX has shorter loops.

Also, I don’t understand the passage about nominators and denominators in Fourier’s law. All of the factors in Fourier’s law have equal impact: if you double one of them (k, A or dT), the capacity doubles. So everything is equally relevant.

I think what causes you to misinterpret this, is the idea that the ground will not keep up. That will never be the case in a properly sized system where you ask from the system what it was designed for. Again, asking from a smaller surface of ground the same amount of energy doesn’t mean you’re taxing the earth. Water-based systems (and their “level of ground taxing”) have just become the arbitrary reference in everyone’s mind.

However, you are absolutely right: the ground is the limiting factor. Refrigerant-based systems simply harness its limited capacities more efficiently (in terms of compactness of the loop) than water-based.

Now, I think a lot of what you are saying comes from a good understanding/intuition on how the ground behaves/reacts. The only thing that seems to mislead you (and other installers I have talked to) is failure to appreciate the MAJOR difference between using refrigerant (R410A) versus water as a heat exchange medium. That major difference is that refrigerant – contrary to water/glycol – goes through an isothermal (constant temperature) phase change in the ground that allows for a higher temperature gradient and a higher capacity per foot of loop.

When I said “R410A simply allows for a higher temperature differential than an antifreeze solution does, they have different thermal properties”, I should have gone into more details and explain that refrigerant, contrary to water, undergoes a phase change in the ground. Phase change is the most efficient way of transferring heat, simply because your delta T remains unchanged, ensuring the same level of transfer all throughout. That is not the case at all with water-based systems: the more energy they absorb, the hotter the water gets and the less heat they are able to transfer.

Most geothermal installers I talked to have the conviction that running a geothermal system too long“taxes or exhaust the ground”. That’s actually not true. The ground is basically an infinite heat source (sink in cooling mode, but we’ve been focusing on heating in this conversation). What happens is that in a “traditional” geothermal system, the more heat you absorb, the less you are able to absorb heat (because your delta T decreases). So in fact, when you’re running a geothermal system too long, you’re not taxing the ground but the system’s transfer capabilities! Refrigerant-based systems like ours do a better job had maintaining that delta T and therefore that transfer capability.

And that’s inherent to our systems, that’s why we are able to maintain the same compression ratios as water-based systems (I say that because if you try to artificially increase your delta T, that will hurt your compression ratio and your efficiency).

Sorry for another long response!
Gregor

Hi Greg,

lets get right into it.

"Not very important in the scope of this discussion, but I wanted to point out that research shows that axial heat transfer actually represents up to 15% of a geothermal system’s capacity! " Love to have a reference for that statement. Where are the remaining 85% coming from? Heat comes from below and from the sides, and in a vertical loop field mainly from the side.

Regarding 1), I understand your point but the conductivity of copper does matter as compared to that of HDPE (~800 times less than copper). It’s true that in a DX system alone, it’s so conductive that you can just ignore that layer in your sizing calculations, but that’s not true for a water-based system. So it does give us an advantage. "
Sure, but the advantage is small. The wall thickness of 1.25" SDR 11 pipe is 0.11 inches. Plug it into Furier's law, account for the much larger surface area due to a larger diameter of the HDPE pipe, and multiply that with the longer boreholes (again adding to the surface area, and you realize that in the big picture it does not matter that much.

"The problem now becomes that the cylinder which is much shorter (lesser area) is not only much much thicker, but also has a much lesser amount of conductivity”. It is actually the opposite! As mentioned, copper is 800 times more conductive than HDPE and our loops are 800 times smaller. So our “Cylinder of heat transfer” has a better conductivity.

First I am not sure if you are really mean what you are saying, namely that your loops are 800 times smaller because a very thin layer in the equation has 800 times the conductivity.
http://www.geooutlook.org/epub/GO2015No4/files/assets/basic-html/page-5.html# The cylinder we are discussing includes the ground, even according to your own publication. As you mentioned above, think about DX as a thicker but shorter cylinder. Here lies the problem. You are trying to draw the same amount of BTUs out of a shorter borehole., which will result in the moisture or water next to the borehole to go through the phase change which will release a large amount of BTUs. You are making ice!
Now you created a layer with actually a better conductivity (water is an insulator compared to ice) but as a layer next to the borehole is much colder than water next to the borehole, plus it is now frozen and all the convection of flowing water is gone (good luck plugging that into Fourier's law) and the only way to get more BTUs to the bore hole is to draw them from further away through a thicker layer of ice. Meaning the ice next to the borehole has to be really cold, which is the only delta T which matters to the medium in the pipe wether it is R410A or water/antifreeze mix.


"I think you’re misinterpreting Fourier’s law. First, you have to keep the comparison in mind: water-based vs refrigerant-based. The ground and grout conditions are the same, so all of that is fixed in Fourier’s law. Now if your delta T is bigger, you can reduce your surface area (shorter loop) to obtain the same capacity. That’s why DX has shorter loops.

Also, I don’t understand the passage about nominators and denominators in Fourier’s law. All of the factors in Fourier’s law have equal impact: if you double one of them (k, A or dT), the capacity doubles. So everything is equally relevant. "
No I not think that I misinterpret Fourier's law, I simply disagree with you that conductive heat transfer is all what defines loop capacity in a geo system. Again, you will reduce the temperature of the ground next to the borehole much more compared to water based systems, and you have to draw heat from further away which you can only do with much higher delta T between refrigerant and ground. In your multi layer formula in the outlook magazine you can see that the denominator is made up by the thickness of the layer divided by its thermo conductivity. The earth being the thickest layer here with the bad conductivity makes the denominator very large, which will result in very low conductive heat transfer. You are not just changing length and delta T, but also thickness of the layer.

" I think what causes you to misinterpret this, is the idea that the ground will not keep up. That will never be the case in a properly sized system where you ask from the system what it was designed for. Again, asking from a smaller surface of ground the same amount of energy doesn’t mean you’re taxing the earth. Water-based systems (and their “level of ground taxing”) have just become the arbitrary reference in everyone’s mind."
No, I understand that soon or later the ground will move as much heat to the borehole as the system extracts, so this process is in an equilibrium. it is just a matter of creating a higher temperature differential to pull more heat from further away. While you have indeed the advantage of the phase change keeping your refrigerant at a similar temperature throughout the pipe in the borehole, the water based systems gain 5 degrees F, but that is factored into the design. The issue is that soon or later your ground has to get so cold in DX systems that the refrigerant has to get so much colder draw the heat from further away. This the refrigerant has a nice heat transfer capacity, but also a much lesser volume. At the end the direct use refrigerant and the higher conductivity gives you some advantage to get a higher delta T, but you are driving the ground down in temperature much quicker and further than with water based systems, which will impact your efficiency and life expectancy of your compressor.
You can draw the analogy with cast ion baseboard heat and in-floor radiant. Much better conductivity of the ion, much lesser surface areas, much higher temps required to get a higher delta T, for which heat pumps inherently pay a price. Low temperature floors run simply more efficient due to the lower temps needed and lesser delta T.
I get the isothermal phase change advantage of the refrigerant, but the temps around the borehole deplete so quickly that your delta T can only be maintained with colder refrigerant, which will cost you COP. You get 2.5 F average delta T out of the isothermal phase change, not much if your ground is 10F or more colder right next to the borehole.
No question the capacity per foot of loop is higher, but if you shorten the loop by so much you are loosing to much footage.

A "traditional system" is also driving the temps down next to the bore hole.
While the ground never exhausts (unless your are running the ground below 20F in our applications), the system will also remain a delta T between water and borehole, through reducing the water temps. The efficiency will decrease, but it will also do so in your system, just much quicker.

Is there really any question that avoiding a heat exchanger, refrigerant in the loop, copper tubing and grout have some thermal advantages? And that other things being equal, shorter loops have some thermal disadvantages?

The only things I consider interesting are "how much" and the proof that supports this.

I agree, the question is not wether copper has a better conductivity, or wether shorter loops perform less than longer loops, but wether overall system seasonal and peak performance and reliability makes it a good system for the customer, thus providing value.
DX systems are around for a while but have not penetrated the marketplace in large numbers. On and on the perception is created that Dx performs superior, mainly with the argument of linear heat transfer, which really is a small portion of the equation for overall system performance and reliability.
Gregor tells us that Fourier's law of linear conductivity is what gives the loop capacity, and for what happens under ground Fourier's law is all we need. He thinks that myself and other installers he has talked to are mislead. Hmmm....
Gregor, forgive me for asking: Do you have the annual performance data on the geo systems you have installed and designed. Because we monitor every one of our residential systems, and stream live data of about 20 of them live on our website, we know the performance.
You should be able to pull this off pretty quickly and actually add proof to support your claims.

Greggor! You're killing me here! I've now been successfully selling refrigerant based geothermal systems for over 25 years. I've enjoyed an essentially competition free marketplace and have built a large customer base with great references. While everyone here is arguing theory, formulas, and hyperbole, I'm selling and installing systems with documented performance and reliability. And now you are trying to convince a few talking heads that DX works and that they should be selling it. STOP! Please, let them continue to focus their efforts in being creative in their efforts to differentiate themselves from one another.

I understand why you say that Jonr. As far I know, such a comparison does not exist. It would be quite a complicated task since one would need to make the comparison between all the different systems and loops in varying ground conditions. I am not sure how much we would learn from that or if it would even be interesting. More limited case studies are interesting at the local level, but won’t help comparing two technologies.

The whole reason I intervened in this thread was to explain (with the help of Fourier’s law) that there are a thousand way of designing a geothermal system to achieve the same capacity. DX and water-based are just two different solutions to a same equation. Whether we are talking about the geothermal industry or any other, it’s not because one technology is more dominant that it is the only option or the best option. I have seen quite a few people on this forum saying “DX can’t/doesn’t work” instead of saying “I am only familiar with water-source geothermal”…

It’s not rocket science; EarthLinked has been doing it for 30 years and our oldest systems are still in operation. DX really is a standard heat/cool unit with its coil buried in the ground.

Which brings me to my next point and back to docjenser: I didn’t jump in this conversation to try and convince you. I was just shocked by all the misconceptions/misrepresentations on this forum regarding refrigerant-based geothermal and the amount of unfounded opinions. As an Application Engineers, I felt it would be good to share some of the basic science behind the technology, in order to give the right thinking tool to anybody coming to this forum and reading this thread. Fourier’s law is the perfect tool for a simple comparison. But of course, if you want to calculator loop and/or system capacity, it gets much more complicated and you do then have to account for convection and the vapor-compression cycle. But DX is obviously more efficient at that since we don’t need a second heat-exchanger like water-source does. So really, no need to go that into that much detail. The whole point I have been trying to make is that there are thousands of ways of extracting or rejecting heat from the ground; water-based and refrigerant-based geothermal are just two of them that have existed for decades.

But FYI, all of our PRIME Series (two-stage) units do come with an EarthLinked Diagnostics & Monitoring system that allows our contractor to remotely monitor their units. So yes: EarthLinked Contractors, such as MI_GSHP_Guy, have all the real-world data needed to support all of EarthLinked “claims” regarding capacity, efficiency, run time, compression ratio etc.

Gregor

P.S: sorry, I forgot a word in my previous post, now corrected to “our loops are NOT 800 times smaller”

Posted By EarthLinked Technologies on 26 Sep 2016 03:29 PM
But DX is obviously more efficient at that since we don’t need a second heat-exchanger like water-source does. So really, no need to go that into that much detail.

Gregor

Gregor, I would not mind being convinced by you. I am just critical of claims being made by highlighting a portion of the equation (such as the high conductivity of copper) to create the perception of higher system efficiency over the season. If you have real world data about annual system efficiency why don't you publish them here or elsewhere, and make it available so we all can learn. We do this with our systems.
I admit that I never put a DX system in, my experience with DX are about 12 service calls where people with failed systems has either compressor failures, or lost refrigerant in a leaked ground pipe, and were asking for quotes on a new water based system. Now, I do not know why they failed, whether bad installers or whether soil issues were responsible for the loop leaking.
I have seen as many badly installed and failed water based systems as I have seen failed DX systems, so water based systems have bad installers, equipment failure or other issues too.
BTW, what is an application engineer? What is your experience with water based systems?

such a comparison does not exist... won’t help comparing two technologies.

I guess we will just have to strongly disagree on this.

Was wondering from installers how the 2-stage DX equipment fared this past heating season?

How was auxiliary engaged? Using the compressor control board which relies on a timer? Using control logic of a heat pump thermostat?

If the system was monitored, how did the ground respond over the course of the heating season?

Thanks in advance.

Mini splits have pretty much destroyed the geothermal market.

Posted By anon on 14 Jun 2017 05:05 PM
Mini splits have pretty much destroyed the geothermal market.

Nah- not being included in the 30% federal tax credit extension for solar etc has probably had a bigger impact on the geo market than ductless mini-splits. The 30% discount from a fairly big number kept geo viable, if not always competitive against other options. Without the tax credit it's more expensive, often too expensive to make the sale. (It was already a tough enough sale even WITH the tax credit in many markets.) Now that fossil fuel prices have relaxed (for now) and the tax credit has evaporated, the upfront cost of geo is a higher hurdle to clear, with a much longer "payback" than a handful of years ago.

The zoning/sizing problems, heat distribution, and the aesthetics of mini-splits continue to make them a bit harder to sell in the US, but they're slowly starting to become "normal" enough in many local markets.

My WaterFurnace 5 series keeps my house comfortable throughout, I can't see how a mini split would be as comfortable.

My WaterFurnace 5 series keeps my house comfortable throughout, I can't see how a mini split would be as comfortable.

Posted By anon on 14 Jun 2017 05:05 PM
Mini splits have pretty much destroyed the geothermal market.

Not sure if that is the case, but I would agree that the discontinuation of tax credits and the falling prices of natural gas had a noticeable impact on the geo industry. But long term I do not see any alternative to geo in cold climate. Given the state of the technology, and assuming that electrically driven heat pumps are widely implemented, one thing is certain: Mini splits will destroy grid reliability due to their peak demand in cold climate.