Georgia Think OUTSIDE the box - closed loop in aqueous well

I have a geocool CFX060 that runs great. No kidding, it's been a tremendous boon for our home. But I want to do better. I would love to hear your thoughts.

I'm a private homeowner, not a pro - and this was a DIY install.

A brief history:
On initial install, the unit did not work. I had run a closed loop system sequentially thru two vertical aqueous wells. My thought process was that water would make a great thermal exchange media, and each well had over 200 feet of water in them at depths of 300-500 feet below surface in granite.

I thought my flow center wasn't powerful enough for that setup, but I knew my well pumps were - so I cut the loop at the bottom of one well and made it an open loop system. (The Vodka I used as antifreeze was not a problem in our aqueous well as it was so dilute.)

Long story short, the GeoCool was defective - had a clean break in an interior weld and had no coolant. The folks at Ingram were very helpful and paid half the repair.

The system works great off my well pumps and dropped our electric bill.

Now I'm considering setting it up as a closed loop system again. Although it works great now, it does drop our water pressure in the house while running. I don't think a larger pressurized water tank would be much help because of the total amount of water required for the AC unit when active.

So I have basically two questions for you smart folks:

1) If I went with a larger water reservoir, how large would it need to be?
2) What kind of flow center can handle pumping water thru 3500 feet of 1" pipe (two 500 ft wells and the horizontal runs) with a flow rate sufficient to allow this unit to function properly?

(*I currently have a QT flow center with 2 grundfos pumps)

 

You will waste a lot of pumping power, as you might waste a lot now. You just don't know it. What model number are your pumps? 26-99 or 26-116? The best would be 1.25" pipe for each well running parallel, not sequentially. Then bring the water in with a 2 " line. That way you likely only need a single 26-99 pump and save a lot of pumping energy.

 

26-99 pumps, but not using them, only using the well pumps - even with the wasted energy on extra pumping power we save $2-300 a month on electricity
The inlet on the geocool is only 1", would a 2" throttled down to 1" provide any benefit?
I could run parallel, but it would increase the run 300' or more.

I don't know enough about fluid mechanics yet to be able to figure the potential flow rates with these pumps over this loop system, but this is how I initially ran the system before changing to well pump water supply.

 

Would this layout provide any greater flow rate? I thought the 1" inlet of my geocool would be a rate(flow) limiting step.

 

See attachment as a suggestion. That way you can use the 2 pump QT to purge out each individual circuit.

All piping should be at least 1.25". What is the distance between the 2 wells, and the distance between the wells and the house? How much piping inside the house?

 

Just some overlap in posts here. Same concept. You want to be able to shut of the flow to each well from inside the house, that way you can increase to flow to the other circuit for purging purpose. There is a huge difference between a few feet of 1" pipe, or many thousand feet of 1" pipe. 100' of 1" pipe has 100 x the pressure drop (flow resistance) as 1' of 1" pipe.

 

Already run as 1" line, that isn't going to be changed.
Minimal pipe inside the house.
Distance between the 2 wells ~500', one of which is 50' from house, the other is 200' from house (They don't add up because I had to run the lines around the house.)
I had planned on using an autobalancing or manual balancing control to balance flow between the two systems if I go with a parallel set-up. (Mainly because they have built in flow meter ports, otherwise I could use ball valves to balance flow)

I'm still trying to figure out if I will have enough flow in series as parallel requires I run an additional line.

 

Yeh, no.

3500' of 1"
15gpm

=500+ feet of head loss

 

I'm not sure I understand fluid dynamics fully, but wouldn't any head due to the depth of the wells be negated by the mass of the water in the descending line? How do you derive 3500' of 1" at 15 gpm = 500 ft of head loss?

I know it must be related to friction in the 1" pipe and that I also have to factor in the head loss at the HVAC unit to get accurate calculations.

I was able to look up a formula that helps:
h=K*v^2/2g
h= head loss
K= manufacturer's published K factor
v= velocity
g= gravity

And this formula helps find K:
K=(fL/D)
f= friction factor (.00006 for PE pipe)
L=length
D=diameter
so my K is 3500*.00006/(1/12)=2.52
and h=2.52*15*15/2*315.63=0.89 head in feet for the PE pipe

More significant would be the head loss due to the 90 degree fittings and ball valves, I would think?

 

Those are closed loop numbers - no vertical head factored in.

Here's some rule of thumbs for 1" HDPE:
- no more than 8gpm
- no more than 700' circuits

 

Well, I guess I wasn't clear. I currently have an open loop system, but can readily convert back to closed loop - THAT is what I'm looking for flow rate calculations on - converting to a closed loop system. Rules of thumb being what they are, they would also not recommend a closed loop running in an aqueous well - but it works. (*Just don't use toxic antifreeze or red brass.)

I also understand the 700' circuit idea, but if the number of transitions to angles and fittings doesn't change - as I have previously demonstrated, the coefficient of friction for PE pipe is negligible. The difference in head reduction between 700' and 1700' of 1" PE is negligible according to the aforementioned calculations. Yes, you risk crimping or interruption of those lines, but if you bury them 3' deep as I did it's not much of a risk.

Bottom line, the system runs well off a standard well pump feeding 1" PE. I just want to make it run better (lower the electrical cost, reduce load on the wells, and decrease the risk of sediment deposition in the HVAC unit.)

 

Your calculations are wrong. I just didn't bother to point it out. Flowrate does not equal velocity for example.

My last example. You have about 12-14' of headloss per 100' feet of pipe at that flowrate. Pull this off numerous charts online - you don't even need math. Fittings aren't even included either.

I'll leave you to figure out the rest.

 

I appreciate your help, but flow rate is directly analogous to velocity.

https://www.boundless.com/physics/t.../overview-97/flow-rate-and-velocity-354-5036/

Perhaps there's an error in my calculations - your 12-14' of head loss per 100' appears to be a tabular result - perhaps you could share a reference?

 

http://www.engineeringtoolbox.com/pe-pipe-pressure-loss-d_619.html

And yes, V and Q are related by A, but not in the manner you've used them in the calculations.

 

Very useful reference - thank you.

Couple of questions: Friction would apply the entire length of the pipe whether ascending or descending into the well, and it makes sense that a well pump for a fresh water well could overcome that head, it has to get water into the house after all. It's also able to get the pressure in my house lines to 80 psi and hold it there. Why can't a flow control center with grundfos pumps generate similar pressure? Or can they? You see, the well pump runs the geo HVAC fine. So I'm trying to find out what kind of pump system output I would need to generate to make it work in a closed loop system. (One that doesn't have to overcome the ACTUAL head as well as the frictional and fitting head as the well pump does.)

Said another way: wouldn't it make sense that a closed loop pump could be downsized compared to a well pump that has to overcome 3-400ft of head before it even begins pumping water thru the pipes and fittings?

That's where I was looking to gain some energy savings. Even though I double the length of the pipe, that head loss can't compare to a 1:1 head loss due to pumping up an aqueous well pipe.

 

Chances are you are currently pumping a lot less than 15gpm on your hybrid setup. I picked 15gpm as that is what your equipment has all its ratings based on. You can supply it with less (especially with open loop setup) and still operate but not at the best efficiency.

I just threw the numbers out there to show it is fruitless to try and get 15gpm out of 3500' of 1".

As to the vertical head in open loop. You're returning it back to the same well it looks like, so you likely have a siphoning b0nus helping you out.

Honestly, you can make me crunch numbers all day, but you can't do much with your current layout without changing the piping.

So currently it works "good enough". I can't put in a system like that, but the homeowner certainly can.

 

I probably gain quite a bonus in heat exchange due to the large reservoir in my wells (6" bored filled with a minimum of 200" of ground temp water). If I go parallel I effectively halve that friction head on each run, now I see the benefit. I sure wish I'd run 2" now, but it is what it is.

 

Listen to Chris, he is right. If you don't want to change the piping, stick with what you have.

 

So there's no pump that can service a closed loop as effectively as my well pump? I find that hard to believe - and it's 1" from my well.

 

Your well pump is not effective (efficient)! It uses a lot of pumping power and head pressure to overcome the static pressure. You right now spend a large amount of system energy consumption on pumping power!