PEX Tubing Pump Requirements for Parrot Rescue

[sprint48]

It's been a few years since I've been on Eng-Tips, glad to see it's still active!

I need some help to determine pump requirements for a PEX tubing loop buried in a concrete floor.

In floor heating for a parrot rescue (featheredfriendsforever.org) located in Harlem Georgia USA
The question is what do they need for a pump?
System Specs:
PEX tubing - 1/2" burial grade (orange)in ~ 4" concrete floor
Length - 500 foot closed loop with 3 gallon bladder accumulator
Bend Radii - large, sweeping, on the order of 24" radius.
Fluid - water - will add propylene glycol
Flow rate - 2-3 gpm to work with heater temp rise
Pump will pick up cool return water, send it thru a tankless heater (

http://www.titanheater.com/proddetail.php?prod=SCR2-N120)

then thru loop.
System will be manually bled to eliminate all air
Pump will see maybe 80F max temp.
The water heater is flow activated and the pump will be on a line voltage thermostat.
110v and 220v single phase available.
Ideally the smallest energy efficient pump is what they need.
Time it takes to get the flow in the loop up to 'ramming speed' isn't critical and it will run continously over night.
K factors - there no valves in the system, the pump and heater will be plumbed straight in line and connected to the tubing with (2) 90 degree elbows, 1/2 npt by 1/2" PEX compression and one 1/2" npt Tee for the accumulator connection.
The original system had a 40 gallon hot water tank heater and a large boiler recirc pump that was donated.
My direct email is birdrescuebob at aol .com If you have any other suggestions, I'm very interested!!
Solar power would be nice but cost prohibitive.
No, this isn't school or thesis.... I'm LONG past those days! Just trying to help some Veterans care for over 1400 parrots.
THANKS FOR THE HELP!
Bob

 

[bimr]

At that temperature, you have 83.68 ft of head in the 1/2" PEX

http://www.uponorpro.com/~/media/.../CDAM_Manual_0912_AppG.pdf

Look at a Bell & Gossett pump for 3 gpm @ 40 psi discharge pressure.

 

[sprint48]

Thanks bimr!

They had a B&G in there but they seemed to fail rather fast. I'll see about getting another one.

I couldn't get your link to work

 

[bimr]

Other pump manufacturers includeaco and Grundfos.

Link to design manual:

http://www.uponorpro.com/Technical-Support/Manuals.aspx

 

[moltenmetal]

That is way too much tubing in a single loop- you'll be wasting a lot of power pumping water against that much head. It would have been far better to have loops of around 150-200 ft length- then the normal wet rotor circulators will work well and last forever. But of course that ship has sailed...your options are either a lower flowrate or the high head, low flow pump- or pumps in series.

 

[bimr]

Other pump manufacturers include Taco and Grundfos.

 

[georgeverghese]

No details on the heat exchanger for transferring heat into the enclosure(s) to be kept warm, its line components and associated pressure drop? Or am I missing something here - PEX is just a variant on plain PE tubing...

Titan inline heater catalogue says nothing about pressure drop.

PPEG-water mix will be warm on the service supply lines, and cooler on the return lines; no details on split in line lengths between the 2 and the return temp - PPEG/ water mix viscosity will be higher on the service return lines.

Maybe no of bends/ other fittings in this circuit is somewhat under estimated?

 

[moltenmetal]

At least two wet rotor circulators in series will be required for this dP at this low a flow. Good point about the dP across the heating device, but I presume that is probably a tank-type water heater so the dP across it will be minimal.

 

[bimr]

Agree with moltenmetal. Perhaps it is possible to go back and split the 500 feet into multiple circuits (zones) with a pump on each circuit.

Most circulators are not designed for such a high discharge head. Probably the reason that the pumps are failing.

George, I believe this is radiant floor heating and there are no heat exchangers.

 

[georgeverghese]

Moltenmetal,
This is a tank -less heater ,so there must be some heating coils burning up pressure in this heater ?

Bimr,
Had to think about this radiant heat transfer : Thermal conductivity for PEX must be very low, so tube external temp must be low, even when inside temp may be 70-80degC. And radiant heat would require hot external surface temp, so the least I could say (without running into detailed calcs) is that using PEX tubing is a poor radiant heat transfer medium. It may have a high surface emissivity value, but that doesnt help much. Would guess there would be much more radiant heat transferred across with metal tubing with high k.

 

[MikeHalloran]

George, the PEX doesn't radiate to anything; it conducts to the concrete in which it is buried. The concrete, being a floor, heats the air above mostly by convection. Thermal conductivity is a relatively minor factor once you've reached steady state or nearly so. Because of the mass involved, it takes a while to change room temperature, but that's not a bad thing in a heater. If you've lived in a house with hot air heat, you understand.

Sprint48, too bad your system is already cast in concrete. The few radiant floor systems I have seen comprise shorter loops, less than 100' each, stapled in a fairly orderly serpentine, with a copper manifold and ball valves at each end of each tube. I think there may be more than one parallel PEX tube per zone, so that a single leaking loop may be isolated and forgotten.

In your case, with one long loop, your first leak means turning off the heat and ripping up the floor for repairs. As you have found, the usual hydronic circulator pumps are not well suited for the head. You may have to work your way through a Grainger catalog to find a pump with a higher head capacity at the desired flow. B&G does make other pumps.

Two or more hydronic circulator pumps in series will work for a while, but I don't think their impellers/seals are balanced right for the high system pressure, so one of them will continue to fail. Better, or less awful, to use two higher head pumps in parallel, with isolation valves for servicing them.






Mike Halloran
Pembroke Pines, FL, USA

 

[bimr]

George:

http://www.uponor-usa.com/Residential-Radiant-Floor-Heating/Homeowner/FAQs.aspx

sprint48:

I think many of us are in agreement. Find a convenient spot, break up the concrete to expose the pex, and install at least one new circuit.

 

[moltenmetal]

Sorry, missed that it was a tankless heater. Definitely some pressure drop through the tankless's heat exchanger. All the more reason to make the loops shorter and run more loops in parallel.

 

[sprint48]

Thanks folks!!
The original design was 2 separate loops but the guy who put the tubing in (former Marine and CEO of the rescue) got in a hurry and made one big one instead. The original design used the old 40 gallon conventional tank water heater, which finally croaked, and a B&G circ pump. I'm not sure there's any reasonable way to split the system without tearing up a lot of floor. Of course that would likely mean hitting the tubing and creating a leak or other 'buried' problems. The original B&G pump worked well but it kept destroying the coupler between motor and pump shafts.

Putting 2 pumps is series is workable as is a higher head unit. Speed thru the system (3gpm)is variable - the downside to faster is potential erosion of the PEX in the curves. Now that I have the head info it gives me a place to start.

If any of you are ever in the Augusta, GA area (you know, Masters Golf) the Rescue is in Harlem, about 20 miles west of town, and 7 miles south of exit 183 Interstate 20. They should be back open by the end of June 2015 and it is something to see with your family.

The other piping nightmare has been the Bio-pond that supplies drinking water to the small ponds in each of the outdoor aviaries. I think they finally got it re-plumbed the way I told them, which includes a sand filter and UV sterilization. They put 1" pvc "jets" in the side wall to get the 13,000 gallons to circulate. Told them it won't work and they are getting a hands on education in piping and hydraulics! I'll be back there this weekend to see if all the new piping is the way it's supposed to be.

Thanks again for all the help and if you have any other input, I'd really like to have it.

 

[moltenmetal]

You do not need to worry about erosion of the PEX in the bends- it definitely will not happen in a closed loop heating system. But you do need to worry about the pumps- if you run them too far to the left hand side of the curve, they will not last. You need the flow in order to get the required amount of heat into the floor without having the boiler exit temperature set too high to get there.

 

[sprint48]

Moltenmetal - you're right on the erosion - I spent 10 years in slurry pumps so it's a reflex.
We're going to mount the B&G pump on a dedicated mount so it can be properly aligned. Even though it's mounted to the motor, I think there was misalignment that failed the couplings.

They are having the 2015 Open House on Sat Nov 21 2015 if any of you are in the Atlanta > Augusta GA area, please come visit!!

Thanks again for all the help and feedback!!

Bob

 

[Compositepro]

Besides high pressure drop, the reason that one loop is a bad idea is that when fluid transfers heat, the temperature must drop. With only 2-3 gpm of flow, you will have a large temperature difference between supply and return temperature to be able to carry much wattage. This means a large temperature difference in the floor in different areas.

 

[sprint48]

Happy New Year!
Here's an update on this project.
Splitting the loop can't happen. It would require a lot more $$ than the budget has.
B&G pump - the short term coupling failure was due to the way it was being operated. The pump was on the thermostat instead of the heater. Pump will now run continously instead of cycling.
Titan said their tankless will not operate in a closed loop system.
Plan A - 40 to 60 tank gallon conventional electric water heater
Plan B - Hot tub/Spa inline heater. Currently contacting companies about those. Might be flow rate issues.
Here's a video from inside the area in question, one of the volunteers with one of the cockatoos. You might want to turn your volume DOWN before you watch it!

https://m.youtube.com/watch?v=uOmiWyWReik

Thanks again for everyone's help!
Bob