5 loop radiant floor, not enough pump? Cold return

SteveGPWC

The System is a Navien ncb250 combo unit with 3 zones with 3 007 pumps.  2 zones are hydronic coils for the air handlers, and the other zone is a radiant floor with 5 loops on a manifold with loops ranging from 280-290ft each. 

The manifold shows I'm flowing about 0.4 gpm on each loop.  Supply is 115 degrees and return is 65 degrees even after running for 45+ minutes.  This is in a mudjob overlay under porcelain tile.

Is my pump just undersized for the radiant manifold zone?

Pictures in subsequent post.

Thank you for any input!
Steve

SteveGPWC

Supply manifold
https://ibb.co/WWshjK4

Mixing valve, pump, flow check for radiant zone
https://ibb.co/BTBccf7

Return manifold to boiler and T for mixing valve 

https://ibb.co/JQs2VDy

psb75

Something close to .5 gpm per loop is the norm. Your .4 gpm per loop is not far off. But...your temps of 115°F out and 65°F back, indicates a delta T of 50°F. That is a "deep" delta. This indicates a slab struggling to "come up to heat". In a radiant floor with tubing in concrete or "mud" the delta should be closer to 10°F. Keep in mind: It takes a long time for a floor to "come up to temp."--and when it does, it shouldn't lose much more than 1-2°F to maintain a set point. It is also best if your sensor is IN the floor mass as opposed to an air temp sensor. I'm assuming your floor area is in the range of 1400-!500 sf.

SteveGPWC

I entirely agree that my Delta drop is significantly high and wrong and that's why I'm here.  The slab  space is 1150 SQ ft.  I am located Long Island, New York, and this is the first time the radiant over slab zone is being hooked up and utilized since renovations.  We have just come out of a deep freezing week or two obviously, and I wasn't sure if my slab was going to take a day or two to reach temp and then return temps dramatically improve or if I was pissing in the wind with a low flow rate and needed a bigger pump or if I have something else wrong in my setup.

Pump is a single speed taco 007-f5 with dedicated mixing valve on that circuit right before the pump with a taco universal 3/4 flow check after the pump.

Tube layout on some of the space:
https://ibb.co/9w2jx3F
https://ibb.co/yycvm49

psb75

The install looks good and your description seems ok --esp. at 1150 sqf. Now, I have to ask: since it is an over-pour of an existing slab...is the old slab INSULATED? How thick is that slab? Hopefully it IS insulated underneath and at the sides. Even if it is, there is a LOT of thermal mass there to bring up to temperature. The only remedy is long circulation time and...patience. Oh and regarding the T-stat: "Set and FORGET." Do not do "set-backs."

hot_rod

In a thin slab,the temperature should come up within a day, even at design.

Sounds like a flow restriction, air, plugged strainer or mix valve, circulator plugged, does it have a check valve in it? Pump flowing the correct direction?
The components are all correct for the application

GroundUp

Simply put, the slab is taking all the BTU you're throwing at it. A 65* return temp says the slab temp is below 65 and therefore the air temp is likely below 60 so it needs some time to heat up. Like 24 hours, not 45 minutes. If the ultimate goal is a 60* slab, you can turn the supply temp down to combat this large delta. If the ultimate goal is a 75-80* slab, you'll just have to wait for it to heat up. Unless of course the insulation detail is poor, in which case it will likely never heat up. To answer the question, no an 007 is not too small nor is .4GPM per loop out of line for a typical radiant floor.

hot_rod

29.4 x (115x10x.2) = 6762 btu/ °

29.4 btu./ hour to raise a cubic' of concrete 1°

1150 sq ft at 20 btu/ft load needs 23,000 btu/hr

Assuming a 2" slab.
4 gpm in a 300' loop @ 12" OC should be putting out 15 or so btu/ sq ft

If it is in an insulated space, less then design condition? 10- 12 hours should be plenty to raise it 60- 80F floor temperature.

If you know the actual room load, insert those numbers, I assumed 20 btu/ sq ft.

You will need closer to .65 gpm to get 22 btu/ sq ft. Assuming bare tiles, no floor coverings, 300' 1/2" loops?
So on design day you may fall a bit short, if my guesstimates (20 btu/ sq. ft.) are in the ball park?

neilc

GroundUp said:

Simply put, the slab is taking all the BTU you're throwing at it. A 65* return temp says the slab temp is below 65 and therefore the air temp is likely below 60 so it needs some time to heat up. Like 24 hours, not 45 minutes. If the ultimate goal is a 60* slab, you can turn the supply temp down to combat this large delta. If the ultimate goal is a 75-80* slab, you'll just have to wait for it to heat up. Unless of course the insulation detail is poor, in which case it will likely never heat up. To answer the question, no an 007 is not too small nor is .4GPM per loop out of line for a typical radiant floor.

Groundup saw it,
OP has only had the floor on for 45 minutes at the time of his writing,
is this correct ?
give the floor a day or 2,

Dave H_2

Living here on Long Island and the deep temp drop we went through, yes, 45 minutes is nothing when it comes to a cold concrete absorbing energy, more than enough circulator, you just need time. Try not to overthink it. Radiant is low and slow, not fast and furious

Dave H

SteveGPWC

Thanks gentlemen... The original slab from the early 1980s is not insulated, definitely not ideal I know.  The mortar bed over the tubing varies from 1-2" plus thinset and tile.

1/2" pex was spaced very consistently at 8" on center, except under kitchen cabinets I got a little sparse and I made sure my loops are all within less than 10' in overall length of each other.  The pex was marked, so I know for sure each of the 5  loops is about 280-290' long.  

I just got home from work so I fired up the zone and I'll let it run all night and see how it is in the morning. The same space is redundantly heated by a hydronic air coil (which I'll shutdown once the radiant is stable).   In those short 45 minutes, I would say 30% of the floor was noticeably warmer to the feet but not showing on the return temps guage... And yes, my slab temp is frigid, the ground was frozen for well over a week.

Full disclosure, this was a DIY job.  I did zero calculations, just made sure spacing was good and hardware was routed correctly, the circulator was my unknown as it was one I had on hand.

PS... Advice or input on an appropriate supply temp to my setup?  I can certainly add a floor sensor if you think it's very advantageous.

GroundUp

Without insulation, it's going to suck a lot of BTUs forever but it should get to a comfortable temperature at some point. Just let it run until it's warm- don't turn it off. 115* is good for now, but you may want to raise or lower it when the slab temp stabilizes a bit.

neilc

radiant floor is not something you turn on and off, and expect heating,
you want it to run, continuously, and let outdoor reset, and floor temp sensors, control water temperature to keep comfort while running continuous,
how, and are you controlling floor water temperature?

SteveGPWC

There's nothing automatically or variably adjusting my floor water... I have a Resideo mixing valve I'm manually adjusting for the 1st floor slab, while the boiler is set much higher to supply the hydronic air coil for the second floor... 

Open to input if my setup can reasonably be improved, but taking the slab up and insulating isn't an option haha.  I'm a newb at these high tech adjustable condensing boilers.  I'm sending 170 degree water to the air handler coils.

hot_rod

The wide delta is your friend for ramping up a cold slab, give it time.

With a 60 delta and 2 gpm you are jamming 60,000 btu/ hr into the slab. Assuming the boiler is keeping up🤭

500XflowXdeltaT

As the temperature come up the delta will close and the rate of heat input will slow

When it settles to a 15 delta
500x2x15= 15,000 btu/ hr is what you will be pushing into the slab

SteveGPWC

Just an update... It's been running for 12 hours at 115 degree supply with the stat jumped out so it wouldn't shut off and help get the slab to temp.  The 1150sq ft space is 74 degrees air temp  and the floor is roughly  80% evenly warmed to the feet.  The return water is 80 degrees.

I think my move here is to lower the supply water temp since the air temp is above ideal but the delta is so big still?  I'm happy it's working though!  Was nice waking up to a vast improvement over the 12 hours.  The kids are loving it since they were used to playing on frigid porcelain.

hot_rod

115 is way too hot, 90- 100 should be plenty for a bare concrete, tile surface. You should have a thermostat that reads both air temperature and slab. Slab temperature alone will cause overshoots on mild, maybe even cold days.
What room temperature are you looking for?
Your average loop, maybe slab temperature will be 97 with the numbers you posted. 80- 82 floor surface is about the comfort limit.
A multi speed circulator is nice if you want to tweak flow rates, close up the delta, 15 is a common delta now that it is warmed up in the room. With a 74 room temperature I’d expect 5- 10 delta🤔
Good that it is working, crank down the supply temperature.

SteveGPWC

We are typically comfortable around the 69 to 70° air temp mark.  I only have an ecobee thermostat hooked up for the air conditioning and hydronic coil heating... My initial thought was trying to figure out two stage heating with the radiant and the hydronic coil on the ecobee, but I'm thinking about abandoning that concept and just getting a dedicated radiant thermostat nobody will mess with and hook up a floor sensor. 

Suggestions for a favorite radiant floor thermostat?  I have redundant wiring in every which capacity going everywhere to do anything pretty much.  There's an overwhelming amount of thermostats on the market with floor sensors.

hot_rod

This is one of my favorites, Uponor is another. You need to use both the floor sensor and the air sensing. If you set just a floor temperature, anything over 70f could, or will overheat the space

With this stat you can run a floor temperature, but also a not to exceed room temperature. So run the floor sensor setting at 80f, and still have room temperature control at 70. They work together, floor sensor and air sensing. Plus they are radiant specific thermostats

GroundUp

To elaborate on Hot Rod's comment, you'd probably be looking for a 519. The stat itself is the same (518), but the 519 comes with the floor sensor and the 518 does not.