# Horizontal closed loop what if question

Discussion in 'Geothermal Loops' started by kk4ej, Apr 3, 2019.

1. ### kk4ejNew Member

Lets assume a system has a loop design of 4 x 3/4" x 700 ft loops. Other than the obvious answer of expense of pipe and digging, would adding a additional loop ( making 5 in total ) provide a better "reserve" to use versus using longer pipes, such as a 4 x 3/4" x 800ft. It would reduce the flow in each pipe and provide a much larger area for exchange...

In the software, I basically see about a 2 degree change in the max EWT....

Is there something else Im missing?

My background is electronics and using the logic of heat sinks, Ive never had a circuit give problems by having larger heat sinks, but lots of issues with sinks too small.

Thanks in advance. This is a learning experience for me, like a big curious kid.

2. ### waterpirateWell-Known MemberIndustry ProfessionalForum Leader

My experience doing vertical loops is that adding a extra loop to a field makes a small difference in performance as the software bore out. It becomes a matter of economics. Cost of more loop vs. performance gain. As you are your own installer adding a extra loop only adds material and time costs.
Eric

Adding another loop reduces the flow rate down each individual loop. It is a good idea to insure the flow rate is high enough to keep the Reynolds number >2500 if possible, but not so high as to create unnecessary flow resistance. So if adding another loop reduces your Reynolds number too low you may get less heat transfer per loop and may not gain much if any advantage over 4 loops. How many tons is the unit? Maximum expected total loop flow? Water alone or water plus ? in your loops (anti-freeze affects the Reynolds number and flow resistance).

4. ### docjenserWell-Known MemberIndustry ProfessionalForum Leader

It depends on the amount of water your are trying to pump through (how big is your heat pump), the pump you are using, the amount of heat you are trying to pull out of the ground, wether is variable speed or dual or single speed (are you pulsing the loop field), as well as the antifreeze you are using (different Reynolds numbers and viscosity).

It is balanced against your cost and what your current design performs at. If the 2 F difference are between 25F and 27F, going to 27F is worth a lot. If it is the difference between 402F and 40F, not so much. It also depends on if you are using slinkies or straight pipe.

We migrated from using 800ft lengths to 600ft and added 2 circuits for every size for better pressure drop results.

We are using 5, 6 and 7 circuits of 600ft pipe to serve 3, 4 and 5 ton heat pumps, designed for the full load, but we are also designing with a dedicated water-water for DHW on the same loop field.

It comes down to the question what you want to achieve.

5. ### kk4ejNew Member

The system I hope to get going shortly is a 3.5T unit, desuper, 10.5 GPM, 4 x 700ft x 3/4" loops, inside manifold and 26-99 3 speed 2 motor pump, non pressure, and glycol. Trench will be 24" wide, single loop per trench, 6ft, ( out and back )

The reason I have 700 ft and not 600 is the loop field is 75 ft away from the house and a limited space to get the pipes into the basement. I wanted a inside manifold.

It makes sense with the reynolds number, too slow you lose thermal, too fast you lose thermal. So finding the sweet spot seems to be the key for max benefit.

Thanks all, I have learned alot just reading the forums and exploring this. I wanted to do a GEO for the house when it was new in 03 but no one in the region knew anything more than it was super expensive. Now Im glad I waited, because the equipment and process has come a long way. Having my own backhoe and lots of space makes mine a much easier.

6. ### Eric KurtzMember

IMHO, that Reynold's number thing on your average home GSHP system is not a big issue. I would estimate that more than 50% (may be as high as 75%) of the energy transfer in my ground loop is being done exclusively thru conduction (ie. when the unit is idle). I have approx. 90 gal in my ground loop and on stage 1 it is pumping 7gpm. That means it has to run for 12 min just to circulate all the fluid thru the HP. With 3/4" pipe, I don't think turbulent flow makes much difference. The amount of pipe in the ground and the distance those feet are apart from each other is much more critical. If you would put 10 circuits in without much overlapping pipe, your conduction surface and gallons in the pipes just went up so much that it would more than make up for the loss of turbulent flow. Where turbulent flow may be more important would be in a larger system that is actively circulating fluid all the time. then yes, you want to maximize your heat transfer.
Just thinking out loud and trying to learn, so hit me if I'm wrong.

waterpirate likes this.
7. ### docjenserWell-Known MemberIndustry ProfessionalForum Leader

You are wrong, in part.
First, not sure if turbulences fulfill the definition of convection. Technically, all the turbulence does is mixing the water on the outside wall with the colder water on the inside pipe (in heating mode), thus moving colder water to the pipe wall (which could be convection), creating a higher delta T between the water at the wall and the pipe, increasing heat flux (conduction).
So the heat transfer is less to a non flowing loop.
The turbulence only has significance at higher load situations when you need the capacity the turbulence adds. And yes, you can compensate for the lack of turbulence with a larger loop.
The Reynolds number of 2500 is a bit arbitrary, you can see some mixing already way below that.

PS, you don't loose heat transfer with too much flow.

kk4ej:
If you use a single 26-99 pump flow center, with glycol, you should get around 9.5 gym, which is fine for a 3.5 ton heat pump, but it would save you a lot of pumping power. Reynolds is about 1200. What are the loads you are having? Heating and cooling.

PS: you don't loose heat transfer with too much flow. Just efficiency due to higher pumping power used.

8. ### Eric KurtzMember

Good points! I was not saying that it is not something to look at. Probably better to say that I would say that the arbitrary number of 2500 is more pumping power than needed for the little bit of gain you have over, say 1500 or even a 1000.
I like your point about the higher Delta T helping heat transfer.
A question though.
If I pump thru my loop at a higher flow rate, it will lower my EWT temp due to fluid spending less time in the ground heat sink. right? Or will turbulent flow make up for it? My gut feeling is that the the little bit of gain in 3/4" pipe is not going to come close to making up for the less time in ground contact. I should do a test sometime to see if I can quantify that, but maybe someone already has.

9. ### docjenserWell-Known MemberIndustry ProfessionalForum Leader

It depends if the loop is sized so well that it is saturated and picked up all the heat it can, even at a higher flow rate. We play a lot with variable speed pumps, which reduce pumping at lower load situations, and it can get funny during low load situations due to low pumping power, without turbulent flow.We sometimes see a reverse temp shift, where the EWT raises despite higher heat extraction, which we attribute to lower turbulence. So you need some kind of turbulence.

Which is countered by the fact that the variable speed run 2.5 times as much, with permanent draw on the loop, without any loop pulsing.

Al the stuff which no software in the world actually accounts for. Many things going on in the borehole which we don't know much about.

The biggest thing we continue to not care about is the 3 gpm/ton. For me it is just a rating number, since the rating formula does not account for the added pumping power you need to get to 3 got. So you pay 5 tomes as much for pumping, to get a 3% better rating (COP might go from 4.7 to 4.8). but in real life you want to throttle back the pumping, to save 150 watts, even if it cost you 30 watts on the compressor to a slightly colder leaving water temperature.

10. ### kk4ejNew Member

Doc, If I understand the concept here, turbulence would be desired to break up the skin effect of the fluid inside, thus making more heat transfer, which is our point anyway....???

Loads my AC guy gave are: Heat 30,700 Cool 28,100 and sized a 3.5 ton unit. I had a 3 ton ( air unit ) and it seemed to handle the load fine, but I am adding a small edition that will be fed from this new unit so he upsized to allow for the new area. According to the Loop Link program, I am 117% on heat and 31% oversize on cooling. 2 stage HP should idle along shouldnt it?

I have a QT 26-99 arriving today with my pipe, Hope to be digging in loops soon.

11. ### docjenserWell-Known MemberIndustry ProfessionalForum Leader

Skin effect is usually an electrical within a conductor.

heat exchange happens at the pipe and heats up the water next to the pipe wall. The water in the middle of the pipe remains cold. The warmer water at the wall reduces heat exchange, since it has not much temp difference to the outside wall anymore. Thus it needs to be mixed with the water in the middle of the pipe for more heat exchange.

Your numbers look all good from the distance. What brand heat pump?

12. ### kk4ejNew Member

Skin effect is in the electronics world, which is my background.

I'm still debating a geocool versus a Miami heat pump. Identical specs but difference in money. I'm wanting 2 stage, desuper and 10 year warranty is attractive.

A local vendor quoted a water furnace, install only duct interface and I did the wiring and plumbing for 16,000. I laughed at them.

Thanks doc for your feedback, I took delivery of my pipe and flow center today, just in time for some heavy rains and thunderstorms.....

13. ### docjenserWell-Known MemberIndustry ProfessionalForum Leader

Who makes Geocool? Not sure if I would be crazy about Miami Heatpump but also not sure how much the product has been influenced by Bosch.

Geocool is a brand sold by Ingrams.

If you like low tech, and no bells and whistles, go for Miami HP. It's loud. It works just fine. Parts are pretty generic, readily available. Good luck if you're out of warranty with the other brands.

16. ### kk4ejNew Member

Im open for suggestions on other brands to look at....

NC73, "its loud" as in compressor noise or vibration?

Headers are built, now waiting on the rain to end so I can get to digging.....

17. ### docjenserWell-Known MemberIndustry ProfessionalForum Leader

Ingrams might sell it, but who makes them?

Some brands you could give to me for free...and I would not put it in. The old Florida heat Pumps were like that.

Usually, I enjoy the quality long after I forgot about the price.

18. ### kk4ejNew Member

ClimateMaster Tranquility 22??? Alot of bad reviews, makes me wonder.