Radiant floor return water too cold

Mkflyy

Just finished hooking up my radiant heat system yesterday 8 loops 300 ft long 18kw on demand water heater. It’s been 24 hours and the slab is only up to temp in about a 10 ft area close to where the pex goes into the floor. Upon returning the water is cool around 50* going out at 140*. Would this be a circulation issue?

Mkflyy

HomerJSmith

If you have 8 loops at 300' long, going out at 140 deg, depending on the temp differential between the environment and the floor, I would expect the water's coming back cold. I assume it is 1/2" pex tube. Increasing the flow if the heater can cope would help or you could increase the output temp. I guess all the loops don't operate all at once? A lot of things can explain it.

Mkflyy

Yes 1/2” all the loops are on and it’s getting warmer I guess I just didn’t know if it was common for it take this long to come up to temp across the whole pad. Building is 72x30 half of it has a finished insulated apartment the other half is not finished. Outside temp has been around freezing 32 high was 41 today

Derheatmeister

Maybe a circ. problem....Was a heatloos calculation and a system design done?
What is you head and what is you circ able to overcome? ...15-58 or a 26-99
Not a big fan of 300 foot loops..
BTW...For the US : 18KW x 3413BTU = 61434 BTU/HR ..

Mkflyy

No full disclosure I just did it all myself and have nearly no experience. Pump is 15/42 would that be undersized for what I’m asking it to do?

kcopp

15-42 plus 300 foot loops. Boiler not piped w a boiler loop and a system loop. Those manifolds also tend to have a good amount of pressure drop. As well as a tankless water heater having a high pressure drop.
Finally, those push fit fittings will restrict you quite a bit. Copper would have been a better choice...

hot_rod

What size slab are you trying to heat, 2100 sq. Ft.? How cold was the slab when you started?
Do you feel a temperature difference at the beginning and end of each loop. Did you get the loops purged well?
You are probably only pushing 4-5 gpm through those 300’ loops, it could take a few days to warm that much mass. 26 yards of concrete in a 4” slab that size X 4000lbs per yard!

Mkflyy

What would be my best fix a bigger pump pressure seems to be saying pretty steady 18-20psi

Mkflyy

@hot_rod yeah just over 2100 sq ft slab was cold sub 40* it’s heating it slowly but not to far from where the pex goes in the floor maybe a 20 sq ft area after 24 hrs. Sounds like maybe my pump is undersized so slow flow could be making it a slow process?

hot_rod

You have a massive concrete flywheel that you are trying to get moving, maybe 100,000 lbs worth! It will take some time to ramp up. Your limitation in how much energy you can move into the slab is all about tube size and flow rate. The pump you have is adequate for the size and length of the loops. As @kcopp mentioned that heater could be a bit of a flow resistence.
.50-.65 gpm per loop is a typical design flow X 8 loops = 5 gpm

Mkflyy

@hot_rod thanks for the info we have some warmer days coming hopefully once I get the entire thing up to temp it won’t take as much to keep it up to temp. 

Hot_water_fan

That first electric bill 

hot_rod

If you are a number crunching man, here’s the deal
Heat capacity of concrete is 29.4 btu/ cubic foot
So to raise a cubic foot of concrete 1 degreeF takes 29.4 btu

30x72x.3 (29.4)= 19,051 btu to raise your slab 1 degree F!
Possibly you are putting that full heater output into the slab, so maybe a couple 2-3 degrees per hour temperature rise into a 35f slab

What are you paying for KWh.😚

GroundUp

That pump would be adequate (not ideal, but adequate) if you had used the proper heating device and/or piped the one you have correctly. 90* delta with 18kw output means you're only flowing .17GPM per loop which is the source of your huge delta, which is a direct result of piping a tankless water heater in series. It will need to be repiped in a primary/secondary fashion to bring the flow up through the floor loops, with an additional high head circ to circulate the tankless loop only. Also, that blue pex does not have an oxygen barrier and should be replaced anyway.

hot_rod

Looks like 99 supply according to the temperature gauge. Thermostat shows 49, probably close to what the slab is, so maybe a 50 delta, not unusual for a cold slab starting delta.

That heater doesn’t look too restrictive. They don’t show a pressure drop graph
I calculated it as 8’ of 1” tube, 5 ells, not much pressure drop at the 4-5 gpm  you need to move through it, .50 gpm per loop.

is the blue Pex 3/4”?

Hows it look after a days worth of power?

kcopp

Did this ever catch up?

Mkflyy

After 52 hours the slab was up to temp only in a small area and actually lost a few degrees temp over night. I just think I don’t have enough flow. I tried to re purge the system this evening and fired it back up. Trying to only run the 4 loops on the apartment side to see how that does. It seems my flow meters are not working correctly. I’m going to run those 4 loops over night and see what happens. 

Mkflyy

@hot_rod yes the blue pex is 3/4

Mkflyy

@GroundUp forgive my ignorance I’m not opposed to what you’re suggesting to gain more flow but I also don’t understand. Do you have an example of how to make a primary and secondary loop?

hot_rod

If the flow meters don’t move at all when only 1 or two are open, you may not be purged adequately. Purge one loop at a time.
Flow going in the correct direction? White knobs turned up CCW. Do those flowmeters turn, they are valves also.
Any strainers in that heater restricting flow? Is it a multi speed pump? Are the loops labeled so you know both ends when you are at the manifold connections?

Item #2 in this view looks like some sort of flowmeters or flow regulator? Could be choking down flow through the heater

A true electric boiler would be the better appliance for that job.

Mkflyy

@hot_rod Circ pump only has one speed in the model I have. I can’t seem to find an electric boiler in stock any where. I don’t have LP or gas available I looked into gas it was going to be expensive to run a new line to the shop. 

hot_rod

Are you purging thru those valves at the bottom of the boiler? Around the boiler, the boiler piping, manifolds and loops. Plenty of pressure from a hose?

Checked all the above suggestions?

If so, next easiest would be to get aGrundfos 26-99 circ. It will swap out where the one is now without any changes in the piping.

Or buy another circ like you have and put them together in series.

Mkflyy

I have been purging from the manifolds them selves on the left side of the manifolds. I have purged all 8 lines individually twice. After running for 10 hours last night the water leaving is about 135* and returning at 60* I don’t know whether that is in line for a cold slab? I assume once up to temp there would be less heat loss. 

hot_rod

Just to be clear, when you purge. You have a hose connected into the system somewhere to pressurize it?
Purge from the return “blue” manifold like this?

Any movement in the flowmeters when you are purging? Opening just one at a time.

Since you are getting some temperature increase, I think you are down to a larger circulation. I would still look for a multiple speed version so you have some adjustability.

Mkflyy

Yes so a total of 3 hoses 1 from my glycol solution container to my Wayne 1/2 hp in line pump then a hose from the pump to supply manifold then a hose from return manifold back to solution. Even with the 1/2 horse power pump the flow meters don’t seem to be right I have a good flow of fluid from the hose back to the solution though 

HomerJSmith

First, are you sure that the manifold valves (input and output) are open? Also, is there a chance that you didn't put any under slab insulation or edge insulation in your install? That could be a mitigating cause. 300' is the max for 1/2" pex and then you add glycol to it and the pumping requirements can go up to 12% depending upon the glyco concentration. Basically, you would need to increase the supply water temp and increase the flow. 3/4" pex supply and return to and from the manifolds is undersized for an 8 port manifold. You have expansion joints or control cuts in the slab and a way to protect the pex at the joints? Just some thoughts.

GroundUp

Oops, I missed it the first time. Your flowmeters don't work because the manifolds are piped backwards- the lower one is supposed to be supply. This also adds a significant amount of resistance to flow. It needs to be repiped to fix that part anyway, so at the very least you need to switch those around and should get rid of the non-barrier pex while you're at it. It may possibly flow decent once that's taken care of, but I've run into quite a few of these in the field with the EcoSmart tankless WH and they do have a huge pressure drop so primary/secondary is a good idea unless you switch to a real boiler instead. Something like an Electro EB-MS-20 or EB-S-18 will modulate and only use what's needed, and is also available online if you can't find one locally; they are what I always replace the EcoSmart heaters with. I don't know how to share diagrams, but a quick Google search of primary secondary piping should give you an idea.

HomerJSmith

Good call GroundUp, It should be easy enough to just switch the red and blue valves at the manifold union.

GroundUp

HomerJSmith said:

Good call GroundUp, It should be easy enough to just switch the red and blue valves at the manifold union.

Switching the valves isn't going to push the flow in the other direction. The circ needs to pump into the lower manifold with the flowmeters in it.

hot_rod

Did you switch the red and blue manifold valve? Red is usually supply, but they are backwards as @GroundUp mentioned.

Blue from the pump needs to go onto the lower manifold, whit on top. Easy switch with Shark fittings

HomerJSmith

From the temp reading on the gauges, I assume that the red valve is being pumped into.
It really doesn't matter if the whole valves with the pex is shifted or just the pex is shifted with new connectors, but the color of knobs, the input and output would be wrong. You could always switch the knobs in that case.

Manifolds are assembled properly (red supply-blue return), however, this manifold was assembled wrong, it appears. The valves are attached to the manifold with unions, I don't see why one couldn't just switch the valves since the red valve is being pumped into.

hot_rod

Switch the piping, easy enough. Then switch the handles. Red on bottom, blue on top.
Flow meter will work now.
I don’t know that flow changes much. With the flowmeters wide open, and the white knobs open, it’s a fairly wide passage regardless of flow direction.
I think, as @GroundUp mentioned, must be some restriction devise in the heater.

A high head circ would be easier then a repipe to p\s, which would probably require a high head heater pump anyways.

GroundUp

@hot_rod have you ever run into an Everhot manifold piped backwards like this? I'm not certain why, but the restriction is quite evident. Possibly holding the venturi shut? I don't know the science behind it but I do know that I've had several sets of these with similar issues and once flipped around, the flow was significantly increased. But yes, the pressure drop through this particular tankless is also pretty serious.

GroundUp

HomerJSmith said:

From the temp reading on the gauges, I assume that the red valve is being pumped into.
It really doesn't matter if the whole valves with the pex is shifted or just the pex is shifted with new connectors, but the color of knobs, the input and output would be wrong. You could always switch the knobs in that case.

Manifolds are assembled properly (red supply-blue return), however, this manifold was assembled wrong, it appears. The valves are attached to the manifold with unions, I don't see why one couldn't just switch the valves since the red valve is being pumped into.

Yes, the red valve is being pumped into. But that is still the return manifold. The lower manifold with the flowmeters is the supply manifold, regardless of what color the valve handle is. Fluid flow must enter the lower manifold and exit the upper. These manifolds do not come assembled, and whoever put them together simply put the red handled valve on the return side- that makes no difference at all from a performance perspective.

HomerJSmith

For some reason, I don't feel I am being understood. Just switch the valves with the pex, easy to do because of the union. Yes, the red valve is the supply, un-connect it from the return manifold and re-connect it to the supply manifold with the flow gauges.

GroundUp

HomerJSmith said:

For some reason, I don't feel I am being understood. Just switch the valves with the pex, easy to do because of the union. Yes, the red valve is the supply, un-connect it from the return manifold and re-connect it to the supply manifold with the flow gauges.

The circulator will still be pushing into the return manifold unless it is repiped to the supply manifold where it belongs. The valve plays no role in the backwards piping. They are the exact same valve with different colored handles. He must cut apart the piping and run the outlet side of the circ to the supply manifold instead of the return manifold as it is now. The fluid is flowing the wrong direction through the loops and manifolds. If he wants to flip/flop the valves, great. Personally I'd avoid the potential leak point and swap the handles.

hot_rod

HomerJSmith said:

For some reason, I don't feel I am being understood. Just switch the valves with the pex, easy to do because of the union. Yes, the red valve is the supply, un-connect it from the return manifold and re-connect it to the supply manifold with the flow gauges.

I, for one am not following you, homer? How will moving the valves change the flow direction?

hot_rod

GroundUp said:

@hot_rod have you ever run into an Everhot manifold piped backwards like this? I'm not certain why, but the restriction is quite evident. Possibly holding the venturi shut? I don't know the science behind it but I do know that I've had several sets of these with similar issues and once flipped around, the flow was significantly increased. But yes, the pressure drop through this particular tankless is also pretty serious.

I’m just looking inside a Watts SS and a Caleffi. With them disassembled it just a hole through them. The valve and meter just screw the opening close. But I have not disassembled that brand to know for sure. The manifold correction is easy for the OP.
The heater may be not as easy. I’m Just suggesting the easiest attempts first.

HomerJSmith

I guess I'm missing something. This is the way I see it as it is. It won't change the flow direction in the pex piping supply to the manifold just change the flow direction in the manifold and infloor pex.
The float flow indicator position. Open-Closed

Having just installed a Caleffi twist flow manifold, I adjusted the flow by twisting (turning) the flow gauges to the flow that I wanted. I don't know if the Everhot gauges work the same, but if it does than those gauges show that the flow is shut off. Maybe they should be rotated to an open position.
I may have that all wrong. This is the balancing valve according to the catalog.

Mkflyy

@HomerJSmith a lot to catch up on here your last picture is correct that is currently how I have the flow going. I’m confused by the varying opinions is that not correct. The supply should be the manifold with the flow meters? In which case I would need to change the way I have it piped to have the circ pump pushing water to that bottom manifold first vs the way it currently goes to too manifold with white flow adjuster.