Radiant heat not returning hot or able to reach set delta-t

CarlosMarrero

Having an issue with a new installation. I cascade 2 navien nfb200h boilers, I have an indirect water heater and a total of 14 Taco Viridian delta-t circulator pumps. Everything is working fine for 9 radiator zones and the domestic hot water but I have 4 zones that are for radiant floor heat which are not working properly. I'm using a taco 3 way mixing valve for the radiant zones. I have tried purging all the zones, to remove any air but still having issues with the return water coming back hot. I have the mixing valve set at about 120°, return water coming back around 80°. When I close off some of the loops in the zones, it appears to work better and get hot. Could there be an issue with the flow rate, any recommendations or help would be much appreciated. Thank you

DCContrarian

Your system is stunningly beautiful.

But was it engineered? Water coming back at 80F is not necessarily by itself a problem. But before the first foot of tubing was laid someone should have done calculations about flow rate, heat transfer, feet of head, etc. And the question would be what numbers were predicted, and what are you seeing now.

Just some ballpark numbers: if you're using 1/2" pex, loops are usually kept under 300 feet and flow kept under one gallon per minute. A 300' loop at one GPM requires a head of 11.3 feet at 120F. Most circulators can provide that. At 8" spacing 300' covers 200 square feet. With a 20F temperature drop at 1 GPM you're delivering 10,000 BTU/hr. Over 200 square feet that's 50 BTU/hr/SF, that's pretty hot, that's at the upper end of the range for heated floors that are comfortable to walk on. If your water is dropping 40F and your floors aren't piping hot the flow rate has to be really quite low.

Have you measured the surface temperature of the floors? From that we can estimate how much heat is being put out per square foot, which multiplied by the square footage will tell us the flow rate.

How long are the loops? What kind of pipe? What kind of circulator? From that we should be able to tell what the predicted flow should be and whether actual is meeting prediction.

hot_rod

A common radiant design would be .5- .75 gpm on a 1/2" loop.

The actual floor output is based the floor temperature vs the air temperature. How warm you run the slab or floor.

So 82° floor is about as high as you want to go for a comfortable residential application.

In a 70° air space that looks like this

82- 70 = 12 x 2 btu/ sq. ft = 24 btu/ square foot from that floor.

Are you running the circs in delta T mode? What delta T. Typically 10- 15° for radiant loops in a residential floor.

If the circ is not in delta T mode, running a fixed speed, the delta will be high when the floor starts up, cold. The delta decreases as the floor warms. On a design condition the delta should run close to the designed number, ideally 10- 15°.

I think that is around a 2.7 Cv mix valve, what is the total gpm you are trying to pull through it?

6 gpm would look like this, 11' head

GroundUp

Turn the water temp down and let it run. Problem solved.

CarlosMarrero

Hi, I'm not sure if the system was engineered when originally installed. I connected to the existing radiant zones that were installed by another contractor about 25 years ago. There are 4 zones that feed certain areas throughout the building. 1st pump is supplying 1200 sq ft. of radiant in the basement. There are about 10 loops of 1/2" pex going into the concrete slab. The 1” copper supply and return trunk lines are in the wall. They run along the entire side wall of the basement. There are 1 x 1/2" tee’s which branch off for the loops, each branch line has a ball valve. There are access panels in the wall for the valves. I closed off all the branch ball valves, I opened one loop at a time and purged them on boiler return purge valve for that zone. After running the heat for several hours, the basement floor did warm a bit but in certain areas (patchy). I still could not get the return water over 80 degrees. I even tried raising the water temp. to about 140 to see if that would help but still the water would not return any higher. However, it did heat up a bit quicker when I tried again, when I closed off some of the loops, had about 5 loops open.

The 2nd pump picks up the recording studio, which is about 300 sq ft. same issue, return hot coming back hot. ½” pex going into the concrete slab. The 3rd zone picks the master bathroom which is 150 sq ft. and the 4th zone picks up the penthouse which is 420 sq ft. There is 1” supply and return. Copper manifold in a closet which has 6 loops of ½” pex going into the floor.

I’m not sure how many feet are the loops; I’m assuming they should not be over 300 ft. I am using Taco Viridian Delta-t circulator pumps. I am running the pumps on Delta-t mode. I did have the delta t at 20 but I changed it to 15 to see if that would help.

CarlosMarrero

I will be returning next week to the job and respond to your punch list. Thank you all, I appreciate the help.

DCContrarian

Another thing to verify is whether the radiant zones have worked since they were installed 25 years ago.

hot_rod

You might try running the circ on a fixed speed. Delta T function, constrained delta operation, is not always the best control method.

The ∆T at which a hydronic circuit operates is always determined, at any time, by the circuits ability to release heat.

Basically the distribution system is in charge of the operating condition of the boiler.

You need the system to run until it reaches thermal equalibrium. You know this when neither the supply or return temperature is changing. That is when you measure the delta T. On a high mass system this could take hours.

This time of year it will be hard to have a load on the heating system to dial it in?

t

GroundUp

High mass needs time to respond and stabilize. Return temps seldom need to be more than 5* above thermostat setpoint, in which case 80* may already be higher than needed. Why do you want the return temp any higher? If you cut the SWT to say 95-100* and let it stabilize with a 15-20* delta, all is well and you've solved the problem.

HomerJSmith

I have a hard time understanding your problem. However, I will point out some issues that I see in your photo. I don't like using a Thermostatic Mixing Valve such as a Taco 5003-C3-G. I would much rather prefer a Taco I-Series set point mixing valve with a sensor. It has a much higher CV. A thermostatic mixing valve requires a temperature differential between the hot and cold ports to function properly. Another error I believe is the Taco Hi-Vent air valve connected to the intake to the pumps. It may be possible that when all the pumps are operating that you may be sucking air into the sys. I would tightly close the cap on the top of the valve essentially making it inoperative. It doesn't even need to be there. If you have air in the sys, you are not going to have any flow at all.

Are these four pumps, pumping into 4 manifolds with multiple loops or a single loop for each pump?

Mad Dog_2

Just Curious....how many SF is this building? Gorgeous pipe work...I'm salivating like Pavlov's Dog! Mad Dog