My geothermal cooling loop seems to work

[Greg Hiller]

FWIW, I tested my 'new' cooling loop today. It works, but not as good as I was hoping. After running the loop for an hour or so, I was able to measure water coming out of the loop at 72 F, not as cool as I was hoping. Water running into the loop was at 74 F, and running at about 1 1/3 gallons per minute. My tank was running about 82 F. This works out to about 1300 BTU/hr, which is what you get with a small chiller. Still, the pump only draws 95 Watts. That can be compared to electrical power to the small chiller which would be 289 W, plus the electricity of a pump to run water to it. So, my loop is about 3X more efficient than a small chiller. Of course, early in the cooling season it would be a bit more efficient, and late it would be less. The cost for the parts of the loop was about $65 or so, plus the cost of the pump. The pump was on the expensive side, I think it was about $200 (a special high pressure Iwaki from Cole-Parmer). So....if the loop were to run for 5 months continuously, compared to a chiller, I'd save about 846 kWH, which where I live is about $93 a season.

If I'd been smart I'd have put more of the aluminum tubes in the ground, but I was getting kind of tired of digging!!

 

[Matt L.]

Greg,

Could you describe your cooling loop, or point me to where on this site you describe it? I am curious about your use of aluminum tubing, how deep you buried the tube, the diameter of the pipe, the length of pipe, and whether you circulate tank water through this pipe, or water to cool tank water?

Matt

 

[Scuba_Dave]

But wouldn't a chiller cool more then 2 degrees?
I'm not sure that you can call that a "success" ?

Since the water going into the llop is at 74, I assume it is a seperate loop from the tank
Right now the ground is fairly cool. But when summer hits my inground pool jumps from from low 60's to close to 80

How deep did you bury?

 

[Greg Hiller]

>Could you describe your cooling loop, or point me to where on this site you describe it? I am curious about your use of aluminum tubing, how deep you buried the tube, the diameter of the pipe, the length of pipe, and whether you circulate tank water through this pipe, or water to cool tank water?<

I used four of the aluminum heat exchangers sold here (http://www.aquaticeco.com/index.cfm/fuseaction/product.detail/iid/9248/cid/2224). I linked them together in parallel with some tubing and some tees. In the sump I have 4 - 25 ft lengths of 1/4" plastic tubing also in parallel. I have a high pressure pump that circulates through the tubing.

>But wouldn't a chiller cool more then 2 degrees?
I'm not sure that you can call that a "success" ?<

Depends upon how you look at it I suppose. I can calculate that the loop is cooling the tank, and I can calculate that it is 3X more efficient than a chiller. I can also calculate that it is about as effective as a small chiller costing about 3X as much to purchase. I think that's success.

>Since the water going into the llop is at 74, I assume it is a seperate loop from the tank<

Correct. It uses DI water in the loop.

>Right now the ground is fairly cool. But when summer hits my inground pool jumps from from low 60's to close to 80<

When you dig deep enough the ground temperature is the same year round. I believe in this area it's around 50, or maybe 53 F. I looked it up once, but it was not easy to find. The highest point of my loop is probably about 18 inches below grade, the lowest about 3 feet I'd say. The water enters at the top of the loop and leaves the bottom, so the last thing the water 'touches' is the coolest part of the loop.

Regarding the efficiency of the loop. If the ground temp is 50F, and the water temp going through the loop is 73 average, that means the driving force for heat transfer is 73-50=23F. If the loop in the sump were more efficient then I might get the temp of water entering the loop in the ground up to 78, maybe 79. If we assume 78, then the driving force would be 78-50=28F The increase in heat transfer would be about 20% then. I can probably increase the efficiency of the loop in the sump a bit by putting it in a higher flow area, but as you can see, the problem is with the loop in the ground. It works, just not as well as I would hope.

I had to dig by hand the hole for the loop. So the thought of digging some more again to enlarge the loop is not particularly appealing. There are a LOT of large rocks in my yard that are hard to move. If I had known back when the room was being built, what I know now, I would have had a bunch of these tubes in parallel buried in the ground.

 

[Moe_K]

Is there any way to do put tank water through the geo-loop? I suppose rusting is the reason you don't go this route?

 

[Scuba_Dave]

OK, I see what you are doing
So the loop is working - but maybe like you said you may need to improve heat transfer from the tank water to the holding tank

Good to know it works. I'm still going back & forth on my setup
The time to install a ground loop will be when the foundation is dug - not after

I do have a chiller I intend to install as a safety device
A backup to the ground cooler

 

[Matt L.]

Is there any way to do put tank water through the geo-loop? I suppose rusting is the reason you don't go this route?

Corrosion of metal components is one reason. Fouling of the water is another reason. From what I know, metal components provide much more efficient heat transfer than plastic ones. Also, in order for the cooling loop to work adequately, the water probably needs to move at a certain speed, and over the length of the loop, the tank water might foul.

When you dig deep enough the ground temperature is the same year round. I believe in this area it's around 50, or maybe 53 F. I looked it up once, but it was not easy to find. The highest point of my loop is probably about 18 inches below grade, the lowest about 3 feet I'd say. The water enters at the top of the loop and leaves the bottom, so the last thing the water 'touches' is the coolest part of the loop.

Do you think the loop would have cooled more (not that it didn't cool enough) if the line were deeper? I heard that in Boston, the constant ground temperature is 54F. Also, would you have been able to use metal pipe in the ground for better transfer?

Matt

 

[Greg Hiller]

>Do you think the loop would have cooled more (not that it didn't cool enough) if the line were deeper? I heard that in Boston, the constant ground temperature is 54F.<

Ahhh....54F, I knew it was something close to that. At this time of the year, even the upper sections of the ground are still pretty cool. The loop is on the north side of my house as well. I think it probably would have been better to get it a little deeper, but I don't think that's the over riding problem here. To get more heat transfer I probably just need more of the aluminum tubes over a larger surface area of ground.

> Also, would you have been able to use metal pipe in the ground for better transfer?<

Not sure I understand. The pipe in the ground is aluminum.

 

[MarkO]

Greg, any reason you need a 95Watt pump? I wonder if there's a more efficient pump? Maybe you can get better results with a lower flow rate? I also assume the pump is not submersed in your sump? That would kinda defeat the purpose at it'd end up adding 95Wa of heat to the tank.

As for ground temp. mid 50's is a good estimate, but keep in mind that in the northeast, frost can penetrate 3-4', so the ground temp is not all that uniform, especially at shallow depths.

 

[RichConley]

Greg, it sounds to me like the 74degrees is the problem, and you need a better heat exchanger.


Also, with passive heat exchange, that pump probably isnt helping much. (it creates a decent amount of heat itself)

 

[Liquid]

>But wouldn't a chiller cool more then 2 degrees?
I'm not sure that you can call that a "success" ?<

Depends upon how you look at it I suppose. I can calculate that the loop is cooling the tank, and I can calculate that it is 3X more efficient than a chiller. I can also calculate that it is about as effective as a small chiller costing about 3X as much to purchase. I think that's success.

Forgive me if I'm making a wrong assumption here, but don't you have to what the throughput of the chiller is in order to determine which is more efficient/effective overall? Yes, the chiller may take 3X when it is running, but if it runs less than 1/3 of the time then you are less efficient overall, right? Or, put another way, if the chiller chills the same amount of water down more than 3x as much, then it is a more efficient way of cooling. Or am I missing something?

 

[MarkO]

If the goal is to cool your tank more efficiently, then I think you need to focus on the power consumption of the pump. 95 watts seems very inefficient to me.
Many of you are talking about chillers, but another thing to consider is the heat that is given off by a chiller needs to be removed. Using an AC unit to cool the room the chiller is another factor.

 

[Liquid]

Well in talking about efficiency, power consumption is relative to the amount of cooling power the machine provides. For example, you could have a machine that draws 1000W when it is on, but it is so powerful that it only needs to run a 1 minute an hour. It would be 6 times as efficient as a cooling machine that draws 100W and is on all the time.

 

[Greg Hiller]

Greg, any reason you need a 95Watt pump? I wonder if there's a more efficient pump? <

The pump I'm using is designed for a high back pressure, low flowrate operation, which is what this is. The high back pressure is mostly generated in the 1/4" diameter lines in the sump.

>Maybe you can get better results with a lower flow rate? I also assume the pump is not submersed in your sump? That would kinda defeat the purpose at it'd end up adding 95Wa of heat to the tank.<

The pump is external. A lower flowrate would cool the water going through the external loop more completely, and heat the fluid flowing through the sump loop more completely. This would not necessarily result in an overall higher rate of heat removal from the sump. Remember, the cooling is a function of how MUCH water you cool in addition to how cool you make it.

>As for ground temp. mid 50's is a good estimate, but keep in mind that in the northeast, frost can penetrate 3-4', so the ground temp is not all that uniform, especially at shallow depths.<

The loop will be shutdown in the winter and blown out. I might even but an antifreeze type solution in the line before I blow it out.

>Greg, it sounds to me like the 74degrees is the problem, and you need a better heat exchanger.<

I thought I explained this. Maximum cooling that the loop will perform is dependant upon both the rate of cooling in the sump, and the rate of cooling in the ground. They will be equivalent, then must be by definition. Heat transfer is equal to the some heat transfer coefficient (which will increase slightly depending upon flowrate) and the difference in temperature across which the heat transfer is taking place. In this case the ground is 54F, the water going though it is 73F. 73-54= 19. In the best case the water going into the loop would be 79F, even then, the average temp in the loop would be much lower, but for argument, let's assume 79 F, 79-54=25F so in the best case you would go from a driving force of 19 to 25, an increase of 30%, but in reality probably much less.

>Also, with passive heat exchange, that pump probably isnt helping much. (it creates a decent amount of heat itself)<

Sure, but the you have to circulate it somehow, and whatever heat it adds is removed.

 

[Greg Hiller]

>Forgive me if I'm making a wrong assumption here, but don't you have to what the throughput of the chiller is in order to determine which is more efficient/effective overall? Yes, the chiller may take 3X when it is running, but if it runs less than 1/3 of the time then you are less efficient overall, right? Or, put another way, if the chiller chills the same amount of water down more than 3x as much, then it is a more efficient way of cooling. Or am I missing something?<

You're missing something. I can go to the Championlighting, or Marine Depot, or whatever site, and look up the heat removal capacity of any chiller I like. This removal capacity is based on what the chiller will remove when it is ON. This is how I performed the calculations. A chiller equivalent to my loop would consum slightly more than 3 times the electricity to remove the equivalent quantity of heat in BTU/hour, or any unit you like.

>If the goal is to cool your tank more efficiently, then I think you need to focus on the power consumption of the pump. 95 watts seems very inefficient to me. <

If you know of an pump that delivers 4.5 gpm (max) that can pump against a very high backpressure (cannot remember the PSI off hand) let me know.

>Many of you are talking about chillers, but another thing to consider is the heat that is given off by a chiller needs to be removed. Using an AC unit to cool the room the chiller is another factor.<

For me that is less of an issue since the sump is in the basement, but the basement will heat up, and the chiller will become less and less efficient.

 

[Matt L.]

but I don't think that's the over riding problem here. To get more heat transfer I probably just need more of the aluminum tubes over a larger surface area of ground.

I agree. Now I'm allergic to thermodynamics (seriously), but I'm wondering if those heat exchangers are wrong for the ground?

I'm not saying they aren't the coolest things I've seen on AquaticEcosystems in a year. I'm just wondering if they still aren't large enough (in terms of surface area) to transfer heat between internal water and external soil, despite their cool ruffles.

They look to me, at least, like they'd be better in the water (which of course they can't be for saltwater), and that even after all this rain, if they are in the soil above the groundwater table, they wouldn't be able to disipate heat in the soil.

I felt that for a heat exchange loop, you'd need a really long length of metal pipe, preferably below the water table.

When I thought of how I'd do my heat exchange system when I can own property, I'd want to sink a larger diameter pipe straight down into the ground. This, of course, would be costly and expensive, as it would be akin to drilling a well. Then, I'd insert a smaller diameter PVC pipe down the middle of the larger diameter metal pipe. Water would flow down one and up the other, with the PVC acting as sort of an insulator. To pass water through the system, very little head would need to be applied, as there is no change in static head and if the velocity is small, no measureable head loss.

I did misread where the metal coils were, thinking they were in water. Relying on plastic to exchange heat in the system water is also going to cost efficiency too.

Let us know how this goes, for I am very interested,

Matt

 

[Greg Hiller]

>I agree. Now I'm allergic to thermodynamics (seriously)<

I got an A+ in heat tranfer , but that was 22 years ago (man, seems impossible), and it doesn't seem to have helped me too much!

>I'm not saying they aren't the coolest things I've seen on AquaticEcosystems in a year. I'm just wondering if they still aren't large enough (in terms of surface area) to transfer heat between internal water and external soil, despite their cool ruffles.>

I agree in general, however, I'm not sure what would be better. I'm not sure that the problem is in resistance of heat transfer to the surrounding soil, or the rate at which that heat is further dispersed to the ground even farther away from the loop.

>They look to me, at least, like they'd be better in the water (which of course they can't be for saltwater), and that even after all this rain, if they are in the soil above the groundwater table, they wouldn't be able to disipate heat in the soil.<

I believe that Aquatic Ecosystems mentions they are designed for use in water. Actually, if you've even been to my house you will see that there is a small pond in my back yard. The pond is only about 12 inches deep, but it is close enough that when I was digging my hole, towards the very end I did 'strike' water.

>I felt that for a heat exchange loop, you'd need a really long length of metal pipe, preferably below the water table.<

That's correct, but you still need a large surface area, both to transfer heat out of the water in the loop, and a large area of soil surface to absorb and disipate the heat. When the pipe becomes long enough though I would think it would become quite expensive in terms of materials (I might be wrong on this, I don't know what pipe costs). These aluminum exchangers are so inexpensive that I could have used quite a few more without increasing the expense very much. The problem is the effort involved in burying them all.

>When I thought of how I'd do my heat exchange system when I can own property, I'd want to sink a larger diameter pipe straight down into the ground. This, of course, would be costly and expensive, as it would be akin to drilling a well. Then, I'd insert a smaller diameter PVC pipe down the middle of the larger diameter metal pipe. Water would flow down one and up the other, with the PVC acting as sort of an insulator. To pass water through the system, very little head would need to be applied, as there is no change in static head and if the velocity is small, no measureable head loss. <

Matt, interesting idea. I still think you might run into problems with not enough surface area of your metal pipe. For me, drilling deep would not be an option as there is a LOT of rock below my house (sounds expensive, too). There was a lot of blasting done to build most of the houses built up near me. For all of this work you have to take into consideration the cost of your project vs. the cost of a chiller, and the electricity long term to run it. There are lots of efficient heat exhchangers out there, but the ones built with things like titanium run into big $ very quickly. I tried to do this project on a shoe string budget as much as possible.

>I did misread where the metal coils were, thinking they were in water. Relying on plastic to exchange heat in the system water is also going to cost efficiency too.<

Not really very much, as I described in the earlier calculation. Even if they were perfect, there would be a maximum increase in heat transfer of less than 30%.

>Let us know how this goes, for I am very interested,<

What I really need is to get 10 guys to come over on a Saturday and dig a giant hole for me!

Question for you Matt....if instead of running into 4 of these aluminum exchangers in parallel (with all the connectors, etc.) I ran into 8 in parallel, how would the pressure drop through the loop as a whole change (grade in this class was not as good! ).

 

[NateHanson]

Matt your idea of running a cooling loop with concentric pipes wouldn't work very well, because the heat would transfer from the hot descending water to the cool ascending water. In fact, big salt water fish like tuna use this sort of counter-parallel blood flow to keep their body heat on the inside in cold waters. The hot arterial blood runs adjacent to cold venous blood in the extremeties, and that allows the arteries to pass their heat to the veins, rather than dumping all that heat into the cold water around the extremeties.

The short of it is that your concentric design would be a very efficient way to prevent tank heat from dissipating to the earth.

 

[Matt L.]

Question for you Matt....if instead of running into 4 of these aluminum exchangers in parallel (with all the connectors, etc.) I ran into 8 in parallel, how would the pressure drop through the loop as a whole change (grade in this class was not as good! ).

The pressure head would reduce by one quarter.

The pressure head h required to move the water at flow rate Q through n cooling loops is equal to Q/nth the flow passing through one of the loops (technically, the loop with the greatest resistance, but let's assume they are all equal). The pressure head h for the system is roughly a function of (Q/n)^2. So if you increase the number of loops n by two, you reduce the pressure head by a factor of four!

Matt your idea of running a cooling loop with concentric pipes wouldn't work very well, because the heat would transfer from the hot descending water to the cool ascending water.

The inner pipe would have to be well insulated. Otherwise, you would have to use a U-tube, which would be harder.

Matt

 

[MarkO]

Wouldn't 1/2 the U tube need to be well insulated as well?