Hydronic radiant floor heat

yes, I will have a electric heat pump (Geo Thermal) for forced air, the electric rates here are on a 3 tier level, can't remember at what usage the rates drop, but HVAC man said that because of the rate drop electric would be the way to go for the Geo system. Hence the electric boiler, it just made sence that if the Geo was going to get me into the lowest rate alone, then the boiler draw on electric would be at the lowest rate also. And the only gas bill would be for running a cooktop, and gas water heater that is tied to the geo unit. The Geo unit will produce around 90-100degree water then all the gas water heater has to do is raise that temp to 120degrees, not to far of a raise so it shouldn't cost very much to run. Plus I also have a electric water heater that is the "storage tank" for the geo unit, one element will be shut off in it so it will raise the temp from 90/100 to say 110 then into the gas to finish it off. And before you ask, YES this system was professionally designed by my HVAC guy, in hindsight I should have asked him about a design for the in-floor system.
Anyway I have installed a couple more temp gauges in the system that work this time (there as close to the flow as possible) I now know that I have a supply of fluid @ 95degrees and a return @65 degrees, so there's a 30 degree Delta-T, but I had to choke down the return line by 1/2 to get the temp up that far. I choked it down by shutting the return ball valve half way on the return manifold.
BTY the 1/2" line that is on the left in that pic is the bypass, i now know that by just doing that (linking the supply to the return) won't really work without a pump, however it does help get the temp up so some hot water must be getting through.Anyway in this round of testing I have it shut off so that I can really tell whats kind of temp I can return through the 6 loops.
Now I'll wait for morning and check the delta-t again hopefully it's higher along with the supply temp. I'll check in approx 10hrs this would have given it 12 full hrs at full bore, I know it will take longer, but I would think I should see some change in 12hrs.

The idea of a bypass is not a straight piece of pipe. It is a pressure regulated device to allow excess water to bypass the supply manifold as zones open and close if you have a fixed speed pump. You want the manifold pressure to remain stable for balance. The pump needs to be properly sized for the required flow. The thermostats need to understand high mass heating sources.

If the slab were insulated properly and tubing design were figured out properly, the geothermal would have been an excellent source for the floor.

Why not do radiant upstairs?

Has you HVAC actually done a heat loss analysis to determine what the heating load will actually be? Is there a de-superheater on the geothermal to give you DHW? Is the house going to be well insulated, good glass, tightly sealed? Money spent on this is once - fuel is forever. Has he calculated the operating costs of what he is doing vs either a single gas boiler or all heat pump? Was this the basis of the somewhat odd arrangement? I know it is too late for some of that; but that is what should have happened.

crater said:

I now know that I have a supply of fluid @ 95degrees and a return @65 degrees

Click to expand...

I think you will have to run the system long enough to warm up the ground under the slab.

If you get it to temperature it is probably going to be a real bear to control. When the thermostat says turn off the water, all that thermal energy you stored is going to tend to keep the slab providing heat.

You may find you want to install something else. Maybe get a bigger heat pump and use it for the areas with radiant.

Find out what percentage of time the boiler is being powered.

You can do that by putting a time clock on the heating elements, or by measuring the kWHr used over a measured period of time.

If the boiler is operating virtually 100% of the time then it makes no difference what the layout of the tubes is.

If the boiler is not operating full time then you can increase the heat transfer capacity of the tubes by increasing the temperature setting on the boiler.

You may find that you are losing a lot of heat into the ground under the slab. If that is the case your only solution may be to abandon the tubes and put in some different electric heaters. Heating the earth with electricity can be an expensive proposition.

The house has a R-21 fiberglass insul in walls and will be insulated to a R-40 for ceiling. The windows are a R-6 triple pane, argon filled, low-e combo. The doors are insulated steel and the glass packs are double pane argon filled with low-e. I have house wrap and will be insulating the box sill as a min. The basement walls will be furred in and insulated at a later time. So yes the house will be tight and well insulated. I am a 20yr experenced carpenter by profession, hence maybe I was over confident about the design of the radiant heat system. I make no claim to be a engineer or plumber/HVAC person, but have a pretty good concept on how it works. In hindsight my first mistake was Not getting a HVAC person involved with the radiant from the start, the second was thinking that the radiant design wasn't as important as it is, I thought that because this was just a secondary heat system and all I wanted it to do was supply a little heat to the basement and garage and that I didn't need to design it for a full blown heating application. Third I measured another buildings pex tubing for spacing not known what there heat source was and installed mine as theres was.
Back to nitty gritty.
Question? How does radiant floor system work at all if it doesn't have any insul. under it?
My neighbor has a approx 1500sf shop with radiant heat in the floor with NO insul what-so-ever under the concrete. He is running a 90K btu gas fired boiler that nets 68K btu's, his tubes my be closer but he is heating a larger space (12' ceiling) it takes him approx 36 hrs to heat his space up to 70d f.
his is also a slab on grade,it does have 4' footings however he also has 2-9'x12' and 1-16' x 12' overhead doors, maybe he's just luckier than me, I can't my basement over 50d f.
What would happen if I doubled the btu's of the boiler (new boiler) is there such a thing as oversized to the point of reverse affect, or just would cost a fortune to opperate.

Another ?
Could there be enough air in the loops to prevent a higher return temp? I can reach temp. when running 2 loops, When I turn on the other four is when temp takes a dive. I also have tried shutting off the 2 loops that I can get up to temp with and run just the other four, but can't get temp up either. My thought is that there may be a air lock in maybe two or three of the four loop zone, and I'm just not circulating, but wouldn't that mean that the supply and return temp would rise because there would only be water circulating though possibly 2 loops at a higher rate of flow (water is blocked in 2 loops so it has to go to the other 2 loops)?
explanation: I have #1 Zone has 2 loops/-> #2 Zone has 4 loops /-># 3 Zone(Garage) has 2 loops (i'm for now not running Garage loops)

I believe that the only place air could prevent circulation is at the pumps. But you always get circulation. Air should be purged because it makes noise and can lower transfer rates. I can't read what pump models those are, but I am guessing they are oversized for what you are doing. Is the area with the 4 loops likely to lose more heat? You saw it before the floor was poured. Sand in one place, rock in another, wet or dry - something like that.

How did you flush out the tubing? Do you have antifreeze in the water. Even when the house is heated, the garage could freeze. If you still just have water you could possibly do some simple tests to see if the loops seem right. You don't have controls on each loop at the manifold so you have to look at sets.

Do you have pressurized water on site. If so, connect the hose to the system. Turn off the boiler and pumps. Open a drain (return manifold would be good). Close that bypass pipe you are using. Get a bucket. Turn on the water and time how long it takes to fill the bucket for each circuit. Four parallel loops should pass twice as much water in the same time as the two loop. Check them out and you can at least see if there is anything real strange.

If you have a make-up water connection and it is turned on - has the system been using water? Maybe there is a leak.

Sand in center of floor, rock around all edges. I have Harvey's heat in for anti-freeze at a 50% solution (21gal of mixed water and anitfreeze) I am running a spirovent so I am assuming it would get rid of most of the air. The tubes were plugged at the ends at installation to keep out debris, so I did not flush tubes, just started the pumps, this is why I think there could be air in the tubes. I have also been told that I had more pump than I needed, they are Grundfos ups-15-58-FC. That is why a restricted the return line but slowing the flow did not increase the return temp, it just raised the supply temp.

Antifreeze in the lines reduces the efficiency of the heat transfer by around 20% if I remember correctly, so your boiler effectively becomes downsized by that amount. Not a lot of heat is available. You could have air still trapped in the loops. Depending on the pump, it may or may not have enough head to push through the entrapped air. Having the antifreeze makes it harder to try to flush the lines, since adding water will dilute things.

So if the effectiveness of the boiler has been reduced by 10k btu's (20% of 50k) then I'm really undersized as the heat loss should be 49,875 btu's + the under designed tubing layout and lack of insul.
Though antifreeze is nessary in this climate, so the boiler may have been under sized to begin with?
Assuming I am trying to heat a 1940sf space to 70d with no insul. under concrete and a tube spacing of 16 oc what would the correct sized boiler be. The design temp should be +5degrees f. The home will be well insulated with very good windows and doors.

The boiler is not really "downsized" by the antifreeze. Antifreeze lowers the amount of heat held in any given unit of volume of the transfer fluid. With correct flow rates, all of the heat available from the boiler can be used.

Go to the Slantfin website and download their free heating calculator. See if you can answer your questions with it. I have not used that one, but you should be able to find some basic information.

That assumes you can increase the flow without damage or excessive wear or turbulance (cavitation) which will mess up a pump. Given identical systems, one with antifreeze and one with pure water, the one with water will deliver more heat per time. All bets are off if you change the system parameters in other ways.

I went to the slant fin site d/l'ed there heat loss calculator. It say I have 87,149 heat loose(btu/hr). could this be anywhere near correct? I entered for a wall factor concrete that was partially exposed, mine is only exposed by about 2'. If I change the exposed wall factor to 8" of concrete below grade it changes the heat loss dramatically down to 55,325 (btu/hr). All else remian the same. In fact I have entered no insul for under concrete and 16" tube spacing.

jadnashua, this system has not been designed. No one involved knows about flow rates or any of the other good points you raise.

Crater, here is a link to a good discussion of pumping. http://www.grundfos.com/web/homeca.nsf/Webopslag/ALEE-5PAHSV

You pump is three speed. Have you changed speed rather than throttling the flow with a valve.

I use two of your pumps in my system. I have what is known as a primary/secondary configuration with a buffer tank. You don't really need any of that - just giving you an idea of pumps. My boiler is 155 KBTU (my heat load is only around 50 KBTU but I have other rather demanding short term loads). One pump moves all of the BTUs from the boiler to the short loop in the boiler room. The second pump moves all of the BTUs from the loop to a buffer tank. These are large diameter copper pipes so the "resistance" is lower than 1/2" tubing.

I have another larger (but not huge it crosses at X 32, Y 26) pump that drives all of the zone manifolds. Each loop is individually controlled by solenoid valves. The point here is pump sizing. I have one larger pump that services 6000 sq ft of heated space. It is probably oversized. My original intent was to use a variable speed pump instead of a pressure bypass, but excessive noise from the Grundfos pump caused me to abandon the effort.

Just a note, your pumps have check valves built in so there will be no backflow through them when multiple pumps are running.

Draw a graph with the X axis being flow in gpm, and the Y axis being pressure in psi. Make a grid equally spaced in both axises. The endpoints of the pump curve are just a bit under 20 on both axises. I would say 18 X and 19 Y. Draw a straight line between these two points. The actual curve is drawn in increments of 5. The curve is slightly bowed. If you want to try, the center of the curve passes about through the intersection of the 10 grid lines on both dimensions. This is for the highest of the 3 speeds. I have not found one for the other speed ranges. Try poking around the Grundfos site.

This curve will tell you what flow is possible at any pressure in the range. This is for water. Antifreeze will change things a bit, but I would just ignore this for the moment. Now you can examine the behavior of the loops using the pressure measured and the flow curve. The flow and the temperature delta will tell you roughly how many BTUs are being moved. That number should be adjusted for antifreeze as noted in a previous post.

Is the antifreeze you are using specified for radiant systems. It needs to be.

Concrete exposed to air in a cold climate is a really bad thing. You said you were going to insulate the concrete. That is what you want to calculate.