Hydronic Floors & Hot Water Heat

[Redneck in training]

We have floor heating with geo heat pump in the house and love it. When we added another building (shop and guest room) I originally also wanted floor heating but at the end I decided to go with two mini-split units and invested the saved money in the solar panels. I am pretty happy with the decision. The shop is 36X63X12 and has no windows therefore 32000 BTU mini split is able to keep it cool at 70F even at 96F outside and reasonably warm at 65F in the winter. Since the PV also feeds our house (2500 sqft heated and AC, 1600 sqft garage is heated only) our average electric bill is about $120/month. We have two propane fireplaces as a heating backup in the house that we use sometime for ambience. We buy about 400 gal of propane about every four years.
The mini split units are relatively cheap and easily installed DIY. I paid $180 to certified HVAC company just to commission it because it was a condition for warranty. Another advantage of not having tubing in the floor is I can bolt anything to it without a worry I could damage the tubing. The disadvantage of the mini split in the shop is you need to check the filter often and clean it.

 

[Aquamoose]

I'm debating electric versus propane.

One comparison chart showed that if electricity is costing .15 cents per KW, propane would have to cost .93 cents per gallon to match. I pay .15 cents per KW and $2.00 per gallon.

Anyone have data to share?

My take is that electric is simpler. No moving parts besides the temperature setting knob and easier to install with no flue needed.

Raspy,

You & I are a somewhat apart on some points. My research consisted of some info from 3 different radiant floor heating companies that supply the parts & custom manifolds.

For instance, all have advised to pressurize the lines and compare the pressure readings at least 24 hours, 48 is better. Any drop indicates a leak and should be addressed BEFORE you call for concrete.

Another instance is the pressure tank in the ceiling setup. There are dedicated in-line devices that “burp” air out.

Also, the thermostat sensor in the concrete in a sealed end tube is right (fold 8” back & wire tie it), but it should always be in the area of the return side for a proper reading. There are also thermostats that are more suited for radiant floor applications as opposed to regular thermostats.

 

[Aquamoose]

I should also add about no insulation. I completely disagree with you. It is necessary to have proper insulation over the entire area. Why waste heat downward? Flat panel mesh was very efficient. When using fiber, it’s not necessary to use rebar except the perimeter but I used them only for the apron leading into the shop at 12” deep at the edges with good base gravel. Saves material costs and it’s faster to install.

 

[rScotty]

I should also add about no insulation. I completely disagree with you. It is necessary to have proper insulation over the entire area. Why waste heat downward? Flat panel mesh was very efficient. When using fiber, it痴 not necessary to use rebar except the perimeter but I used them only for the apron leading into the shop at 12 deep at the edges with good base gravel. Saves material costs and it痴 faster to install.

I think what Raspy said was that insulation was only needed at the perimeter of the slap IF the underlying soil is dry. I see his point. If the soil is dry sand/gravel, there won't be much heat transfer to the dirt anyway, and what there is can be thought of as heat stored, not heat wasted. The perimeter is different.

Books and ads tell me that fiber in the cement really adds to the strength of the cement, but of course stronger concrete still doesn't do the same thing in a finished product as rebar does - so each job is going to be different that way. I confess to being old fashioned and prefer rebar with standard mix concrete....mostly because I understand it better...

One thing I have never liked about PEX in a pour is that the cement guys always end up walking around in the wet pour and I worry about what they are doing to the PEX.
For this workshop project, that's what I really want to know....what brand PEX to use?? and how to protect it from the steel and the cement workers?? Once the PEX is in correctly, the rest of the system can be anything.

I use an overhead thermal expansion tanks with a dedicated air burper in the line. I was suspicious of that way of doing things, but after ten years it has worked as advertised.

On manifolds, for our house I used a custom hand-made manifold soldered up out of large diameter copper tubing. The house has a lot of zones, and I think that each loop's valves and controls are more important than the manifold & loop geometry. It makes me feel better to know I can isolate any one loop from another.
The one small hydronic pump ends up running such low fluid velocity through the system that manifold design doesn't have much effect on flow rate. Anyway, each loop has a thermostat along with valves to regulate flow so it is easy to fine tune the flow to match what is wanted in different rooms. Of course it is nice to have a low-resistance manifold and equal length loops, but if the system has individually tunable loops it is easy to overcome the inevitable differences in flow.

For the question about electric vs. propane..... There is really no reason at all why you can't have both. The real difference is not so much the price per KW as it is the speed at which they heat the water i.e. BTU/minute input. Electric is way slower unless you have some really large amperage feeding the system. And large amperage can be expensive to install. Some power companies want a surcharge for over 200 amp lines.

But what you can do is install both types of heater and plumb them so that you can switch from one to the other or run them together. Plumbing for either/or heaters is a cheap option. While doing this plumbing, might as well put in some extra connections in case you ever go solar or oil fuel, too.

For hydronic electric heat I like to use a large traditional domestic electric hot water heater. If you want a faster response - and can handle the amperage draw - wire the heating elements to go on at the same time & offset each element's thermostat appropriately. Or use it in normal mode. That's what I do - although I tried the other. I like the Rheem tank HWH. For the on-demand propane you have a large field to choose from. Consider using domestic hot water heaters. There are specialized hydronic heaters, but so far I don't see a reason to go that more expensive route.
rScotty

 

[teejk]

We used Rehau brand pex here (5/8" in the shop, 1/2" in the house). Benefit of the larger diameter is longer loops. Downsides are that fittings are not generally available in the hardware stores and any leftover tubing is not really good for anything whereas the 1/2" can be used for potable water at least in the Rehau brand (I can't speak for other brands). As for abuse during the pour, I think the stuff is much tougher than you think plus the loops are spaced 12" apart and our concrete guys were careful to avoid stepping on it. Another benefit of the Rehau brand is that the lengths are marked every foot...easy to know when you are at a loop's limit and time to return to the manifold and start a new loop.

 

[Raspy]

A couple of follow-up points:

I have never built a system that needed an air scoop/ burping device of any kind. A simple air vent and a flushing tee with shutoff valve is all that is ever needed. These are placed back at the boiler and not at the manifold. Remember that these systems are closed loop. Once the air is out, all that special air handling equipment is just sitting there doing nothing, forever. The only thing you need to do is get it burped enough to start running. Then it will clear itself completely over time by placing a single air vent in a good location, or an open expansion tank up high. Bubbles are the first problem, and they are relatively easy to deal with. Entrained air is the next problem and air vents do not do anything about that. Oxygen in the system, absorbed into the water will get used up in a slight bit of corrosion initially and then the corrosion stops and you have a perfect closed loop system that will run for many years without rust. Oxygen barrier tubing almost completely prevents oxygen from permeating into the system and a good additive is best to prevent minor corrosion issues.

In my house, I put the expansion tank up in the attic area so it would accumulate the last of the free air in the system and provide a place to fill the whole system. It is an open tank with no bladder. In fact, it is a sandblasting tank of about 6 gallons. I did not want, or have any reason to fill my system with a fill regulator.

As far as pressure testing, I may not have been clear. ALL systems must be placed under pressure before the pour and held there during the pour. Code says a100 PSI hydrostatic test. That isn't necessary for function though. A 60 PSI air test is fine. Sitting overnight is my general rule. 99% of all leaks are at the manifold, above grade, but be sure you find any that occur and fix them before the pour. Do not leave the tubing exposed outside for any longer than is necessary. I've had squirrels eat holes in it overnight.

As far as tube spacing goes, 1' is a good general number because it heats fairly well. And when tied to rebar is fairly easy to walk around in during the pour. A 9" spacing heats better, but gets more abuse during the pour. The rebar spacing should always the same as the tube spacing so the tube can be tied alongside the bar. Never run the tube along by itself in a concrete pour. Do not worry about the position of the the as far as being close to the surface or deep down. Consistency is better and it all heats well anyway.

I have drilled many holes in my radiant slab to bolt things down, to glue in vertical rebars to build a glass block wall, to add more plate hold down bolts and to install thresholds. You just have to study photos and draw a heat map on the floor to match the photos. It's scary, but completely doable.

Back to thermostats. A conventional setback thermostat is as good as you can get for comfort, timing and efficiency, with a few minor exceptions. A NEST thermostat is generally a waste of money and causes a loss of comfort, in most applications. For slab sensing a setback thermostat with remote sensor ability is the way to go.

Being in the business of installing these systems means sometimes work must be done in a more conventional and consistent way. That's fine. But on your own system you can afford to be more creative and fine tune the design more to your own needs and the house design.

With dry gravel under the slab you get excellent support for the slab. There is no heat loss straight down under the middle of the house. There is conduction, to a certain degree that increases the mass being heated. That is not a disadvantage or an energy loss. Remember, I made my slab 8" thick and it works very well with all that mass and no insulation under the slab. I do recommend insulation around the perimeter, say, two feet in from the edge under the slab and slab edge insulation And, of course in the walls. I also alway hold the heat back from the edge of the slab by about 1' from the inside of the walls. This reduces slab edge losses too.

When positioning a slab sensor tube in the slab, the exact location is not important as long as it's between some heating tubes and not off by itself, but it is only detecting an average temp at it's particular location. There is no "proper" reading that represents true comfort as slabs are different temperatures all over their surface. What we want is a reading that we can measure for a relative and consistent number. That's all. So the exact temperature is not important. Consistency is. I have a thermostat lying on the floor in one location that I use to monitor a representative spot. I know where it should read to be pretty comfortable in the house. Comfort is a fleeting and different situation for everybody and every situation in every room. Radiant rooms can and are very comfortable with quite cool air temps, for instance. So looking at a wall thermometer often has limited meaning. Reading the temperature of a radiant return line in the slab, with a floor sensor, is pretty much meaningless. A representative area that has had a chance to settle in over hours and is similar to the area you want to spend some time in, is much more accurate as a measure of whether you'll feel comfortable or not. And that spot is mainly valuable because you can set your floor thermostat to watch it for you.

I overwhelmingly use 1/2" PEX (5/8" OD) in new installations. But in my own home with an 8" slab I used 3/4" PEX (7/8" OD) throughout. I used number three rebar (3/8") on 12" centers in both directions and I added extra bars wherever needed. All tied with 6" or 8' rebar ties. Using bulk tie wire leaves a lot of sharp ends that can puncture. The 3/4" PEX is extremely rugged and allows a lot longer loop if needed. Longer loops are not always helpful, but sometimes. A stronger tube is less likely to get damaged during the pour and a thick slab gets a lot of stomping around during the pour.

So much of this endeavor is simply common sense, and trial and error. It takes into account mass, various temperatures, energy, the nature of water, lifestyle and time. Often, as in using thermostats, the thing you expect, is not what happens. An example of this is setting a thermostat to a certain temperature and expecting to have the house arrive there and stay there. Not gonna happen. Why not? That is where the fun begins. Another one is "average" floor temp. Slab vs room sensing and why. Insulation where and why. High vs low mass. Predicted and proper recovery rates (this is a big one). What the best energy source is. Weeding out inaccurate or poor advice. Understanding that there are too many variables to just expect a one-size-fits-all approach will be best for you.

 

[ovrszd]

Thank you very much John. I appreciate that you took the time to explain things in an easily understandable way.

 

[teejk]

A couple of follow-up points:

I have never built a system that needed an air scoop/ burping device of any kind. A simple air vent and a flushing tee with shutoff valve is all that is ever needed. These are placed back at the boiler and not at the manifold. Remember that these systems are closed loop. Once the air is out, all that special air handling equipment is just sitting there doing nothing, forever. The only thing you need to do is get it burped enough to start running. Then it will clear itself completely over time by placing a single air vent in a good location, or an open expansion tank up high. Bubbles are the first problem, and they are relatively easy to deal with. Entrained air is the next problem and air vents do not do anything about that. Oxygen in the system, absorbed into the water will get used up in a slight bit of corrosion initially and then the corrosion stops and you have a perfect closed loop system that will run for many years without rust. Oxygen barrier tubing almost completely prevents oxygen from permeating into the system and a good additive is best to prevent minor corrosion issues.

In my house, I put the expansion tank up in the attic area so it would accumulate the last of the free air in the system and provide a place to fill the whole system. It is an open tank with no bladder. In fact, it is a sandblasting tank of about 6 gallons. I did not want, or have any reason to fill my system with a fill regulator.

As far as pressure testing, I may not have been clear. ALL systems must be placed under pressure before the pour and held there during the pour. Code says a100 PSI hydrostatic test. That isn't necessary for function though. A 60 PSI air test is fine. Sitting overnight is my general rule. 99% of all leaks are at the manifold, above grade, but be sure you find any that occur and fix them before the pour. Do not leave the tubing exposed outside for any longer than is necessary. I've had squirrels eat holes in it overnight.

As far as tube spacing goes, 1' is a good general number because it heats fairly well. And when tied to rebar is fairly easy to walk around in during the pour. A 9" spacing heats better, but gets more abuse during the pour. The rebar spacing should always the same as the tube spacing so the tube can be tied alongside the bar. Never run the tube along by itself in a concrete pour. Do not worry about the position of the the as far as being close to the surface or deep down. Consistency is better and it all heats well anyway.

I have drilled many holes in my radiant slab to bolt things down, to glue in vertical rebars to build a glass block wall, to add more plate hold down bolts and to install thresholds. You just have to study photos and draw a heat map on the floor to match the photos. It's scary, but completely doable.

Back to thermostats. A conventional setback thermostat is as good as you can get for comfort, timing and efficiency, with a few minor exceptions. A NEST thermostat is generally a waste of money and causes a loss of comfort, in most applications. For slab sensing a setback thermostat with remote sensor ability is the way to go.

Being in the business of installing these systems means sometimes work must be done in a more conventional and consistent way. That's fine. But on your own system you can afford to be more creative and fine tune the design more to your own needs and the house design.

With dry gravel under the slab you get excellent support for the slab. There is no heat loss straight down under the middle of the house. There is conduction, to a certain degree that increases the mass being heated. That is not a disadvantage or an energy loss. Remember, I made my slab 8" thick and it works very well with all that mass and no insulation under the slab. I do recommend insulation around the perimeter, say, two feet in from the edge under the slab and slab edge insulation And, of course in the walls. I also alway hold the heat back from the edge of the slab by about 1' from the inside of the walls. This reduces slab edge losses too.

When positioning a slab sensor tube in the slab, the exact location is not important as long as it's between some heating tubes and not off by itself, but it is only detecting an average temp at it's particular location. There is no "proper" reading that represents true comfort as slabs are different temperatures all over their surface. What we want is a reading that we can measure for a relative and consistent number. That's all. So the exact temperature is not important. Consistency is. I have a thermostat lying on the floor in one location that I use to monitor a representative spot. I know where it should read to be pretty comfortable in the house. Comfort is a fleeting and different situation for everybody and every situation in every room. Radiant rooms can and are very comfortable with quite cool air temps, for instance. So looking at a wall thermometer often has limited meaning. Reading the temperature of a radiant return line in the slab, with a floor sensor, is pretty much meaningless. A representative area that has had a chance to settle in over hours and is similar to the area you want to spend some time in, is much more accurate as a measure of whether you'll feel comfortable or not. And that spot is mainly valuable because you can set your floor thermostat to watch it for you.

I overwhelmingly use 1/2" PEX (5/8" OD) in new installations. But in my own home with an 8" slab I used 3/4" PEX (7/8" OD) throughout. I used number three rebar (3/8") on 12" centers in both directions and I added extra bars wherever needed. All tied with 6" or 8' rebar ties. Using bulk tie wire leaves a lot of sharp ends that can puncture. The 3/4" PEX is extremely rugged and allows a lot longer loop if needed. Longer loops are not always helpful, but sometimes. A stronger tube is less likely to get damaged during the pour and a thick slab gets a lot of stomping around during the pour.

So much of this endeavor is simply common sense, and trial and error. It takes into account mass, various temperatures, energy, the nature of water, lifestyle and time. Often, as in using thermostats, the thing you expect, is not what happens. An example of this is setting a thermostat to a certain temperature and expecting to have the house arrive there and stay there. Not gonna happen. Why not? That is where the fun begins. Another one is "average" floor temp. Slab vs room sensing and why. Insulation where and why. High vs low mass. Predicted and proper recovery rates (this is a big one). What the best energy source is. Weeding out inaccurate or poor advice. Understanding that there are too many variables to just expect a one-size-fits-all approach will be best for you.

You being in Nevada and me being in Wisconsin adds a lot of variables. Nobody I know around here would skip insulating under the slab (in fact many people will insulate the sides of the slab because concrete will radiate out further than I would think). So design is heavily dependent on climate. It gets cold here (-20F not unusual) and can stay cold for weeks.

 

[ovrszd]

I'm insulating the footings 4' deep and under the entire slab.

Friend of mine has been in his shop 3 years. He insulated as I described above. He says the grass stays green about a foot out around his shop. That's an example of how far concrete will radiate heat.

 

[Raspy]

teejk,

Yep, climate has a lot to do with design. I only said I would not insulate under the middle of the slab, but of course you must insulate on the edge and in under it for the first couple of feet too. Especially in in colder climates. Also, there is no need to heat right up to the edge of the slab inside. Hold the heat back and let it fade out near the edge.

Loosing enough heat out through the edge to keep a lawn green in very cold weather is terrible. Nobody would advocate that! Heat loss in that situation is through conduction, not radiation.