hydronic floor heat complicated

MIradfloor

ok first off thanks for any and all help with my system!!

house was built in 2016 slab on grade 1/2 pex 12" spacing all but 1 loop is around 300' 6 loops total

hot water tank is also used as boiler k030-16m-np6 brazed heat exchanger for in floor primary loop ( I can take pics )

outdoor wood/coal boiler b3-32a-20 brazed heat exchanger for that boiler loop

3 zone valves 3 stats although I have now made the living rm stat centrally located the 1 main stat for all 3 zone valves ( didn't lay in floor tubing correctly for separate zones. I have pictures before the concrete pour.

circ pump for floor loops is a 007-f7 taco pump

if I remember correctly I only get about .25 gpm on each loop if I turn down the other 4 loops and only do the 2 loops that are the farthest away ( babies room/ office ). I can get .5 gpm max

problem is I can't get the farthest 2 loops to make good heat, living stat at 70deg and the babies room is at 65-67 max wife is not happy with that for the baby! lol. rooms are carpet with correct padding for radiant heat. rest of house is vinyl flooring.

I know you guys will need more info but basically just looking to get some opinions on the system design if its way off or close to correct.

thanks
Rich

MIradfloor

MIradfloor

MIradfloor

MIradfloor

hot_rod

Is the circulator a multi speed version, I would set it up a speed, sounds like you might not have enough pump for all zones and the plate HX?

Are the two plate HX in series so flow always goes thru both? Kind of looks that way in the pic. If so that adds even more pressure drop.

Gordy

If your flow gauge readings are accurate that’s 9.5’ of head at 1.5 gpm to all zones. A 008 with a steeper curve for higher head applications might get you the flow you are looking for.

The only sure way to size the circulator is to calculate your system head. To do this correctly you should have a design flow rate needed.

MIradfloor

no speed control but I do have a few of the grundfos UPS15-58FC I can try?

yes they are in series with each other but are on the primary side of the floor heat loops.

the big plate is the OWB and the small plate does the in floor

Gordy

A 15-58 might get it on speed two, or three. Didn’t know you had them laying around. Worth a try since you have them. Start on speed 1 but I think that might be close to what you are getting with the 007.

Gordy

The 15-58 flange to flange is 1/8” bigger. Hopefully you can squeak it in the piping.

MIradfloor

what do I need to do to get a design flow rate?

MIradfloor

I'll give it a try! worst that can happen is I take it out. thanks!

MIradfloor

got the 15-58 in and the flow is better but still not .5gpm ill take it for now. I would like to get the correct pump in there though if its possible to figure the out. thanks for the help!

Gordy

Did you try speed three?

Gordy

You need to figure out what is the flow rate you need. Then figure the head loss at the design flow rate of the longest loop plus piping, accessories, and equivalent length of fittings from the circulator, and back. And the heat exchangers head loss.

Once you know that you find a pump where that head loss, and gallons per minute needed intersect on the pump curve. Preferably in the middle it.

Gordy

So throwing an example out. If you need .75 gpm on each loop. That's 6gpm for the main piping to the manifold. If the head loss at that flow rate is 11' then you would look on a pumps curve chart find 11 in the vertical head column to the left, and 6gpm in the bottom horizontal column. If those two points intersect in the middle of the pump curve. That's the one you want.

MIradfloor

yes it on speed 3

MIradfloor

how do I figure out flow rate requirements?

Gordy

Did you try all speeds?

Rich_49

Might also wanna turn that circ above the Flat plate so it is horizontal

Gordy

As an example:

You would need to know the heat loss for the zone you are trying to heat first. Typically by design we use 10 degree delta for radiant floors.

So if the zone is 10000 btus with a 10 delta you would need to supply 2 gpm.
Gpm= btus/(500x delta t)

If that zone has two loops the 2gpm would be divided by the number of loops.

Now from there you would calculate the head loss for the system piping as stated above at that flow rate.

MIradfloor

speed 1 was like u said about the same as the existing pump speed 2 bumped it a little more and speed 3 pushed it almost to .5 and that is with 4 of the loops adjusted to .3gpm

MIradfloor

rich_49 I questioned that too but according to the manual as long as the pressure is above 20psi it can be in that position, my well is at around 40-60psi so should be good.

MIradfloor

MIradfloor

picture represents the 2 loops for the bedrooms/bath and part of the living room.

Gordy

You are correct if pressure is maintained at 20 psi. But it still likes to trap air.

If you are getting .5 to loops I'd wait to see how it works. It's a radiant panel in concrete? So it will take some time to adjust because of mass.

.5 gpm is pretty typical of a well designed panel. Some times less which is not working for you.

Rich_49

MIradfloor said:

It appears that this guage is installed on the same pipe as the circ . the guage reading is less than 20# . If I am wrong please forgive me

MIradfloor

yes concrete 3 1/2"-4" I believe on top of the foam board

-5 deg last night. still 66 deg in those 2 rooms and 70 in the rest of the house. starting to think there is an insulation problem.

Rich_49- that gauge is for the OWB loop which is an unpressurized system.

Gordy

The slab jnsulation looks solid, how about the envelope.

MIradfloor

2x6 walls blown cellulose. 1"foam board exterior. R40 cellulose attic. stud seams caulked, insulated windows.

MIradfloor

proper vent and truss hips all spray foam as well

Gordy

I would say you are good on the envelope, and floor for insulation.

MIradfloor

educate me please or did you mean to say "would"?

Gordy

Would sorry about that.