Taco Rep Says I need their biggest pump...

OKAP

I'm trying to put together a simple radiant pex system to heat 5 large concrete fish tanks in my greenhouse, but my pump doesn't seem up to the task. I started on this back in March, then started a new job, now I'm trying to talk my father through finishing it.  Neither of us have any experience with radiant systems.  I know I'm way out in left field with this application, so I appreciate feedback from those more experienced.



I originally purchased a Taco 00R-MSF1-4IFC.  I don't remember what math I was looking at when I decided on that pump, but it doesn't seem to want to move the water in my system.  I remember thinking I had around 11' of head and needed around 10 gpm, but those details are fuzzy months later.  My father got tired of chasing air bubbles and called Taco, someone there told him he needs a Taco 2400-50.  That's a big step up, so I hope to get a second opinion.



I have five concrete tanks, each with around 280' of 1/2" radiant pex beneath a pond liner. Each loop comes and goes through a 1/2" radiant pex manifold (everhot ssm106 with the 6th branch short-looped). We are using a domestic water heater with 3/4" pex to the pump and manifold, and an RX15 expansion tank between the pump and water heater.  We are using around 40% RV antifreeze.  The thermostat is different here because I'm targeting the fish-tank water, so I use an aquatic temp controller to switch the Taco.  All 3/4" pex totals about 30', no excess fittings or restrictions.



Because I'm targeting the fish-tank water temp instead of ambient air temp, standard heat load calculators don't apply.  I'm taking a hobby approach to this, and realize this is not an optimized system.  But it should heat the fish tanks, however crudely, if someone here can point me in the right direction on pumps.



Any suggestions for an appropriate pump?  Any other ideas as to why this system isn't flowing?  Thanks for your help.

Alan (California Radiant) Forbes

Pumps

Looking at the pump curves for the 2400-60:



http://s3.pexsupply.com/manuals/1265990034174/25826_PROD_FILE.pdf



and the 00R:



http://s3.pexsupply.com/manuals/1249544554380/101-125.pdf



Pressure drop through the 3/4" PEX, manifold and water heater are minimal.  The biggest pressure drop is through the 1/2" tubing - at  300', it's about 3 ft. of head.  Even if you add another foot of head for everything else it has to flow through and double it for the antifreeze (3 + 1 x 2 = 8 ft. of head), it should still flow according to the pump curve, so I'm at a loss.



The 2400-50 seems way too big; even the 2400-10 seems large for your application.



Email or call Joe Mattiello at Taco:

joemat@taco-hvac.com

401-942-8000 X 484



He has always been very helpful.  Just make sure you know the total flow rate you need.

OKAP

00R?

Thanks Alan, I'll check in with Joe Monday.  Looking at that 00R pump curve, and seeing that you ballpark the head loss as even less than I did, do you see any reason the 00R would be unable to push the system?



And I get it that systems are usually spec'd on required flow rates, but there's really no way of knowing (prior to our first winter) how low the water temp will drop on its own, or the rate of transfer (btu's/hr) between the loops and the fish-tank water.  Without that info, is there a way to guess at required flow rate?  I assumed I would have to use observational data to improve function after implementation.



Anyone care to compare this system to what a comparably sized, weakly insulated, slab on grade system (5 x 280' x 1/2" plus glycol) would require for flow?



Thanks everyone for all the views.

PeterNH

How Deep?

Hello,

How deep is the water in the tanks?

Is it possible the pex got overheated and collapsed/ovaled some from the weight of the water in the tanks?



Peter

OKAP

Tank Depth

There is only about 2' of water in the tanks.  I had some concern about pex flattening so before I put the liner down, I left the pex in full sun on a 110+ day and tried flattening it with my heel, to no effect.  This stuff is very rigid.   Good point though, and I wish there was a way to inspect the loops now that they are in place.





To those of you who do radiant design and install:



If you wanted to install a simple closed system with domestic water heater and (5) 300' loops in a poorly insulated space with zero degree weather, what pump(s) would you expect to work well?



Thanks everyone for your time and input!

PeterNH

questions

What is the typical temperature of the water at the bottom of the tanks?

How do you know for sure there is no flow?

What is RV antifreeze?

Is it 40% or could it be a much higher % ?



Seems that even with the biggest pump going that won't cavitate, the most flow thru 300 ft of 1/2 pex with 40% glycol at 50* is going to be 1.5 gpm per loop

more realistically 1+ gpm  is about it,  and that should be doable with the pump you have now.



Peter

OKAP

replies

In-tank temps are currently 58F, 65F is the winter target temp.  RV antifreeze (propylene glycol) is colored and visible in the translucent white pex we're using.  The system hasn't been air-purged yet and the little bubbles visible around the fittings aren't moving.  Also the return manifold has flow indicators (gauges?) on each branch, which aren't registering.  The mix could be off, I didn't do it myself, and it was filled based on calculated capacity.  PG went in first, then water until filled, we were relying on the flow through the water heater tank to thoroughly mix the solution.  I can test it once it has run for a while. 



Thanks, Peter, for confirming that my pump should be adequate.  I plan to drive out tomorrow and see for myself if any other problems are obvious. 



If anyone has other suggestions, opinions, comments, or criticism, I appreciate any troubleshooting ideas you can offer.  I don't have internet access at the farm property, and it is 75 miles back to home when I run out of ideas.

CMadatMe

Btu's

What size water heater is this? To move 1gpm in system with 40% glycol 280' with 1/2 pex at 120 degree water temp my head loss is 12.3'. Just ran it thru Uponor head loss calculator.



The question really is. How many btu's do you need to move to each tank?

OKAP

Water Heater specs

The water heater is 40 gallons, 40k btu's.  I can't calculate heat loss without first-year observational data, so rather than worry about what's needed to target a specific temp, I'm wanting enough flow to take advantage of all 40k btu's.  Does that makes sense?



I have flushed and flushed and flushed the system, the Taco 00R now runs and the flow gauges on my manifold indicate about a quart per minute flow on each of 5 branches.  With a 1/2 hp transfer pump assisting, I can get about 1.5 gpm on each of the 5 (at 40 psi).  I've also diluted the glycol while flushing the system, so it's now closer to 20 or 25%.



I am using translucent pex so I could see even fairly small bubbles, and I'm no longer seeing any.  The system would seem to be running fine if it weren't for the low flow readings. 



Any opinions as to the accuracy of this type of gauge?  It's the red and clear plastic cone shaped piece on the lower (hot) manifold.  A plastic disc moves up and down on a spring inside.  Thanks!

CMadatMe

More Questions

What delta-t are you running? How many square foot in each slab? What centers did you run the tubing. Can run it like it was a snowmelt system. Just targeting a surface temp right? What is the inside temp of the area?

OKAP

Outside the box...

Delta-t based calculations are based on years of observational data on similar structures, so I can't apply them in this environment.  I don't even have a slab.  The tubing is spaced about 4" in an area roughly 2' x 44' per loop.  There is no surface temp to measure, I'm using a submerged temperature probe and aquatic thermostat.  The inside temp of a greenhouse varies wildly, but I'm trying to control fish-tank temp not ambient air temp.



Without comparative data, it's tough to imagine optimizing this system the way a domestic system would be optimized.  I will be able to make adjustments to flow rates and water temps, thereby tuning for an optimal delta-t, and could even add pump capacity or another water heater if needed, but the rate of heat loss will change with the seasons, so it will probably be a never-ending process.

CMadatMe

No

Delta-t means btus transfer. No delta-t no heat transfer. What's my entering water temp, whats my returning floor temp>

If I'm moving your 40,000 btus' and my delta is 20 then I need to move 4gpm

If I'm moving your 40 and my delta-t is 10 then I need to move 8gpm.

If I want a 30 degree delta-t then I need to move 2.6gpm to get you your 40K.

If my delta-t is 5 degrees then I need to move 16gpm.

Without forumlating what delta-t your running your questions cannot be anwsered correctly.

How would you be able to size a pump if you don't know what you want or need to move. In snowmelt we want a specific target point to melt just like you want a specific water temp. No difference. I still need mass to absorb the btu's I'm delivering. Your inside temp or worst case temp becomes my outside temp in snowmelt.

OKAP

Ok

Ok, I thought you wanted to know what my target for delta-t is.  As for what it actually is, I don't know, I haven't fired up the water heater because I'm still troubleshooting the flow problem.



Lower flow yields a greater temperature drop in the loop, I get that.  In most installs you want lower delta-t so the heat is distributed evenly, but here the heat could all be transferred in the first 50' of loop and the fish-tank water will still be evenly heated. I guess I'm saying I think even a HUGE delta-t would be fine so long as it allows for 40k btu's with the pump I have.



I can't tell you what the inside temp will be, because it hasn't been up through a winter yet.  Even in the winter, daytime temps will be over 90F sometimes, night time temps would approach zero if the thermal mass wasn't present.  I have 6000 gallons of water flowing through 30 yards of gravel, and 30 yards of concrete holding it all together.  That's one huge thermal mass relative to the space involved.  The first domestic or snowmelt installers probably didn't have delta-t tables, they just started with what works for others and adapted to the observed circumstances.



I'm not trying to get the calculated perfect flow rate and matched pump, though I would love to.  What I need to know is, does anyone see a reason the Taco 00R would only flow around 1.5 gpm total in this system, or is it possible the flow meters are junk? 



I can only think of more air hidden in the lines, a bad pump, or junk flow meters.  Would having only 40 psi mains pressure reduce my ability to purge the system?  Thanks for sticking with me on this, I appreciate the experienced input.

AntnyC

Basics 101

Go back to the basics. Check all valves are fully open on manifiold, what are you doing for automatic air purging? Do you have an auto feeder on that system with a PRV and backflow? What about thermal expansion?  Also a domestic water heater rated at  40K BTU/Hr is working it arse off to heat 6000 Gals of water!!I  You pumping issue may not be an issue at all. Even a simple system needs to sized correctly of your just wasting time and fuel$$

Interceptor

00R

I have a floor system using the same 3 speed 00R IFC pump and Everhot brass manifolds with 6 loops of 1/2" PEX, 260-300 feet each. My supply/return piping consists of about 10' of 1" copper and four 90s. On low speed I get .2 GPM per loop on the flow meters, .3-.4 GPM on high speed. I don't know how accurate the gauges are, but your system should have similar readings. With glycol and 3/4 PEX supply/returns I would expect your flows to be somewhat lower.

OKAP

Interceptor

Thanks for that info!  I have it working now at around 1.5 gpm through the main loop with the Taco set to HI, which sounds pretty close to what you are seeing.  The weather has been generous in the last few weeks, sunny days and nights just beginning to freeze, so I haven't had much demand on the system yet. 



When we get a cold spell I'll have a better idea of the temperature drop across the manifolds so I can begin to figure if this is working efficiently.  I'm thinking a drop of 10-30 degF would be appropriate, any advice/opinions on this?



Thanks again to everyone for talking through this with me.

Interceptor

Delta T

At your current flow rate of 1.5 GPM, a 30 degree delta T would indicate about 22500 BTU/hr being transferred. That's if my calculations are correct.

OKAP

Well then...

I hoped this would carry a little more heat than that...  Looks like I could have seen some benefit from better flow on the primary loop!  I suppose I could replace the 3/4 inch plumbing, use a bigger pump, or remove the glycol from the system if it isn't keeping up. 



For now though I'll just have to wait and see what my observed delta T turns out to be. 



I'm still surprised at the low flow of the 00R.  If the flow gauges are accurate and I'm only getting 1.5 gpm through the main loop, then this configuration has around 15 feet of head pressure.  I would have thought I was closer to 10 feet of head, which would be closer to 7 gpm, right in the middle of the curve.  



It is much colder here starting tonight, I should know more after the weekend.

PeterNH

Is

The 00R pumping away from the expansion tank?

Is there a schematic?

Does the flow change with all the loops closed, except for one?

No possibility the lines have a kink anywhere?





Peter

OKAP

I used this....

I built what is shown below (minus the air purge) from threaded brass fittings and installed it on the water heater outlet.  After the brass fill and drain assembly, I have about 10' of 3/4" pex to the pump, then another foot of pex to the hot distribution manifold.  From the cold manifold I have another 10' or so of 3/4" pex directly back to the water heater.  That's the whole thing.  Sorry no schematic.



When I have only my short loop open, which is just a 3' loop between manifolds, I see about .8 gpm.



From what Interceptor posted I am guessing either this manifold is overly restrictive or the flow gauges read low, as even his system with no glycol and bigger main loop is running fairly high in the pump curve.

CMadatMe

You Can Get 7gpm

Just run a 5 degree delta-t. One problem will only move 17,500 btu's. I think your too caught up in gpm instead of btus...If you really want to try and make this work. Get a Taco 009VDT pump and play with the delta-t settings. You will be guaranteed the delta-t as long as the pump fits the curve and the 009 looks like its in your wheel house..

OKAP

7 gpm

Thanks for the pump suggestion, Chris.  That one costs 5x what my 3-speed pump does, but I can see how that money could be recouped through long term efficiency.  I'm a little confused though, what is the reason you suggest I pursue 7gpm if it reduces btu transfer?



I don't think I'm hung up on flow, or at least not high flow.  I was surprised at the flow results I get from the manifold gauges, and came here to ask if anyone can guess why the flow is lower than expected.  I'm still of the impression I don't really know what my flow is.  I asked if anyone had experience with this type flow gauge and only Interceptor indicated he had used it.  No reassurances to accuracy though.



I'm not trying to make changes to the system at this point.  I've been very happy with how much heat the tanks pick up during daylight hours, and haven't had any significant load on the radiant system yet.  I think my max flow is less than expected, but until I can observe delta-t during a really cold spell, I don't know for sure I have a problem.  For now I'm just enjoying the moderate weather we've had.

CMadatMe

Not Suggesting

I'm not suggestion you run 7gpm. Just trying to get you to understand that flow doesn't always necessarily mean I'm delivering more btu's.