I installed a system designed by a company online and installed it myself in a new house I built. It has never got warm enough. It runs all day to stay at 68 degrees. I think the problem is flow as the return water is not even warm. The entire system, 6 loops some over 300' is 1750' of 1/2" pex over two levels and it is driven by a Taco 008 pump. The living room of the house has 24' ceilings and is mostly glass so heat loss is substantial. It is a closed system with heat exchanger and the first floor is slab on grade second floor is engineered wood. It is located in So CA. I was considering replacing the pump with a Taco 0011 which has much more head and capacity to move the water. Can anyone help with this as proper pump sizing?

running all day doesn't mean it's not working. not maintaining the temp you want means it's not working the way you want. do you stay at 68 degrees? are you trying to get higher?

depending on the heat exchanger you have, this pump should be adequate for a six loop system... should push something approximating 0.75 GPM/loop or more which is plenty, barring some obstruction like air, really obstructive heat exchanger, closed flow meters, etc.

if you are trying to get higher temps or you are not maintaining it's more likely that water temperature or your tubing installation method is currently inadequate to meet the load you're asking it to meet. do you know what your water temperature is, and whether it changes with the outdoor temp? do you know how the pipes are installed in the floor (in a slab, in a joist, in aluminum plates in a joist, etc)?

What is the supply temp and return temperature on each loop (while running)?

I need to get some thermometers and check the temp change on supply and return loop. It won't maintain 68 degrees at all. The ground floor is installed in concrete slab and the second floor is under engineered wood. The tubing under the wood is installed in routed subfloor with insulation under it. Those seem to heat up fine. The concrete floor covered in stone doesn't seem to get hot enough, especially in the room with all of the glass and 24' ceilings.

1/2" pex circuits should not normally exceed 300'. If the return water is not even warm, the supply temp is too low, the flow rate is too slow, or both. I bet this was designed by a "pro" not in CA, so you likely won't get any help from them or have any legal remedy either. Perhaps you can tell us who did this design and how they responded when you requested help.

One really can't properly size the pump (or determine the proper supply temp and proper zone balance valve settings) without having all the required design parameters...which would first include knowing or determining the heat loss of each zone. If you have the design parameters and are up to the task, consider using the free DIY software on our website:

Borst Hydronic Radiant Floor Heating Design Software

Thank you for all of your help. The company is out of business so help from them is not an option. I ordered some thermometers so I can gauge the temp on both sides of the system. The supply side seems pretty hot though and the return side just barely warm so I think it is a flow issue. I will see what the flow is in the different zones. Maybe a kink when they poured the slab? WHen I flushed it with a garden hose and looked at each zone the water was coming out pretty good but that is probably a lot more pressure than my Taco 008. I will try your design software. It was calculated and sent to the city with the plans so there might be a copy. The thing is, it is LA where we are talking about it being 50 outside and having trouble staying 68 inside a brand new house...not right. I thought if it was a flow issue, going with a stronger pump might solve it.

Yes, more flow rate and/or higher supply temp will certainly get you warmer floor surfaces...but maybe not the floors you want warmer. If your manifold had temp and flow rate gauges, you should be able to sort out a workable solution assuming the design wasn't flawed to the extent this isn't possible.

in a slab it should be entirely possible to effect this situation, however in a room like you describe you may run into a limitation of the floor to emit heat... you can only get the floor surface so hot before it's uncomfortable. we use 90 on hard surfaces as a max, which would give you about 44 BTU/sq ft/hr of output in a 68 degree room. it's not LIKELY but it's possible you're over this in a cathedral great room. though if that were the problem you'd maintain some portion of the time and fall short as it gets colder out. does that match your experience? or are you just consistently failing to meet room temp?

Is the slab insulated underneath? How long has the system been running? is there a "reset" control on the radiant or is it just provided a single, fixed water temp?

you may very well have a flow issue, but again, its unlikely to be pump sizing in this case unless that's a very restrictive heat exchanger. it's probably valving, air, or debris IF it's a flow issue.

I did some tests. The supply water is 120 degrees and after running an hour the return temp is 85. It is flowing at 1.5 gpm. Yes the room does have cathedral ceilings and the slab is insulated and so are the sides. It is Los Angeles so the ground is never that cold. It gets warm when the sun is in all day and when it cools at night, it can't keep up with the falling temps. It never reaches 68 degrees once it drops and is usually 63-64 in morning after running all night. I guess the pump is pumping enough water. Is 1.5 GPM sufficient ? How much should the supply and return temps differ? Thanks.

Your problem is response time, mostly affected by supply water temperature, floor coverings and controls. Another reason to be careful of online Radiant Floor Heating suppliers. Present company excepted of course.

First, a proper ACCA Manual 'J' heat load analysis. I had the same problem 20 some years ago when installed my first staple-up radiant floors. Now we use a heat load program and hardly ever go without heavy aluminum plates. You are suffering from the same issue, too much resistance to heat flux. In your case the extra load of the thermal mass, in mine the resistance the the finished floor to the heat below.

In either case a raised supply water temperature and better anticipation will likely help if not cure your ills.

how thick is the slab?

Yes. Thank you. I don't have any floor coverings there is stone over the slab but the slab is very thick. The pex is only a few inches below the surface as I tied it all down to the rebar before the pour but it is a huge amount of thermal mass. I am using a heat exchanger and the boiler is set to 130 and after the exchanger the water on the supply side is 120. I will try to get the temp up. I hope it helps.

I would try to see if retrofitting a floor sensor is feasible. this is a large part of what they are for.

I would need to look at the plans but it is probably 2 feet thick but the pex is near the surface.

uh, wow. that's probably not going to work, even with a floor sensor. you could maintain a minimum temp for comfort (room temp plus a degree or two) but you're never going to really effect the temperature of that much mass fast enough to respond adequately to a cathedral great room's shift in demands, even in LA.

You have a lot of mass in a low output radiator. That's bad for response time. If you want to separate response time issues from output capacity issues, turn the thermostat to maximum and leave it there for 24 hours. Then see how it does overnight (leaving the thermostat at maximum). If it still can't keep up, then you need more output.

I think you are right. The room on the first floor that has 10' ceilings gets nice and warm very quickly. It is just the great room that is the issue. Will increasing the flow and supply temp help the response time? Is it worth getting a variable speed pump and pushing more water through 10 degrees hotter? I can balance the other zones so that most of this goes to that great room.

well 100 sq ft of living room with 2 foot slabs is 200 cu ft of concrete. you're over 5200 BTUs/degree of slab temp per 100 sq. ft. of living room.

how many loops are in just the living room, and how many square feet is it? I may stand corrected...

It is 493 sq feet. It has 2 loops of 1/2" pex 12" on center

then I'd say you're at an hour or two per degree change on the slab. could be made to work, perhaps, but I'd consider floor sensing or a constant circulation strategy pretty essential. and even that won't move fast if you have rapid temperature changes... though I don't think you have too many of those in LA at least. and if the load is as high as it probably is in that space... let's just say it's not a given. a real load calc is necessary as badger said.