Hydronic Pipes in the floor: Material and Heating options

tel jetson wrote:pex and copper: when I was looking into it, prices and embodied energy were comparable.

Mike Cantrell wrote:

tel jetson wrote:pex and copper: when I was looking into it, prices and embodied energy were comparable.

Embodied energy might be, but price is definitely not.

tel jetson wrote:

Mike Cantrell wrote:

tel jetson wrote:pex and copper: when I was looking into it, prices and embodied energy were comparable.

Embodied energy might be, but price is definitely not.

it's been a couple of years since I looked. the two used to be much closer. I don't know if copper has gotten more expensive, or pex has gotten cheaper.

tel jetson wrote:RMH wrap: at first blush, it seems like a pressure release valve would make this safe, but those really only work to relieve the pressure in a large vessel like a water heater, not in a tube. if it were my project, I think I would heat an open tank of water with the RMH and run your hydronic tubing through that water for heat exchange. that way, your water tank will never be above boiling, so steam explosions are not a risk.

tel jetson wrote:
pex and copper: when I was looking into it, prices and embodied energy were comparable. pex is cross-linked polyethylene, and I believe it's recyclable. most sources seemed to think the pex was a bit easier to work with than copper. copper is mined, which isn't a friendly scenario. pex is likely made from petroleum, which also isn't great. I would guess that copper conducts heat a bit better, but I don't actually know if that's an advantage in this application.

R Scott wrote:The poorer heat transfer of PEX is actually a good thing in the floor tubes. It allows you to have a longer run in the circuit and have more even heat. Copper will dump the heat too fast so the start of the loop is HOT while the end is COLD. You can have about 100' per run with 1/2" PEX before your heat loss and backpressure get too high. Run parallel loops if you need more than that.

Around here, PEX is a little cheaper up front but installs in WAY less time. You need a special tool to make connections, or buy the stupid expensive sharkbite connectors. Spend the money for the crimp tool if your plan is complicated, it will pay for itself in about a dozen connectors or less. But you can do a simple heat system in fewer connections than that.

It also is much more robust against mechanical damage like pounding it into a floor, although it still needs to be pressurized (or filled and capped) while packing the dirt to prevent collapse.

+1 on an open storage tank, or picking up heat from the bench area of the RMH. RMH + water is DANGEROUS. BOOM SQUISH as Paul, Ernie, and Erica refer to it. The problem is the RMH is SO hot the water flashes to steam and a valve simply can't react fast enough. We are talking EXPLOSION of steam here, same speed and energy as gunpowder.

Rob Irish wrote:
That makes sense come to think of it. But where I'm scratching my head is how then to heat up this open tank? Do you mean like it would sit on top of the big metal drum?

Rob Irish wrote:
I'm not sure if this sort of concept would work, but could a water tank be built inside the mass section of the RMH so that it gets warmed up just by the auxiliary heat?

tel jetson wrote:
that's one option, and probably a good one. a water jacket partially or completely surrounding the barrel could work, too, and I would guess that it would increase the draft of the unit by cooling the exhaust a lot more than air would. you might also be able to use a gas hot water tank as the down-draft portion instead of a barrel. route the exhaust downward through the flue. never tried any of these, but I will at some point.

Rob Irish wrote:
Wow, I didn't realise I would be able to have that long a pipe (100') in one loop. What would be the spacing for 1/2" Pex?

Rob Irish wrote:
We want to avoid sudden explosions of boiling water while relaxing if possible, so will definitely not try and be too innovative with the heating of water with the RMH

Just so I'm clear, is this the kind of method that is really unsafe? He seems to be using the heat itself to migrate the water in a loop i.e. no external pump to push the water around, just heat pressure:

R Scott wrote:
Normally the spacing is 8-12 inches between runs. You can do a little more than that in deep earth floors but you will start to see uneven temps if you go too far apart for the mass..

Brian Knight wrote:
Cool project you got going. Great diagram..Sketchup? Tend to agree that avoiding high temps is critical and challenging. Tej's suggestion of only running tubes through the bench portion is a good one.

The water jacket is just to create more surface area? Is that a certain product? I think for thermosiphoning with the two tanks one has to be completely below the other one. Large diameter tubing (3/4 thin wall copper as minimum) and avoiding 90 degree bends is critical.

It you attempt something like this it would be smart to put in a temp sensor at the closest part of the loop to the rmh vent along with the open tank advice. Pex does have temp restrictions too.

Not sure if your project is for new or existing but have to point out that a home where you can actually feel the heat or (toasy toes) from radiant floors has a very inefficient building envelope. There is a lot more science and repeated success of maintaining comfort and minimal fuel use with high performance building envelopes. You may want to consider focusing more in that area.

tel jetson wrote:you can buy or build manifolds that allow multiple parallel runs of pex. you wouldn't want to bury those in the floor, but you could frame them into either the floor or the wall so that they're still accessible.

Rob Irish wrote:Does a manifold system require more pressure or a more powerful pump?

Rob Irish wrote:Are all manifolds pretty much the same or is there a certain type of manifold which is ideal for this sort of system?

Rob Irish wrote:And more probably stupid questions: Say on one wall you have a manifold, on the other side of the floor where you need to turn the pipe around with 2 right angle turns, would that be done with another manifold connected the same way? Or would you bring those pipes out of the ground, and into another frame or into the wall and then turn them and loop them back to the one manifold?

R Scott wrote:Those bend supports are great for holding tight bends, but you don't even need that for most turns. Normally the pipe gets zip-tied to the re-mesh for concrete floors. For an earthen floor, you could use landscape staples, or the plastic pegs, or sandbags (removed after you have packed around them) to hold it down as you build up the floor. To hold a fairly tight bend, you could tie it with twine to hold the loop. They definitely help where you come up out of the floor to the manifold.

There is a science to how you run the pipes, it isn't a straight back and forth zig zag. You skip runs and then loop back to get them--it helps distribute heat more evenly and reduces the number of tight bends. Youtube and the library are your friends. They also show how to unroll pipe without kinking it.

Manifolds can run from simple (basically a stack of Tee's) to insane (flow meters and adjusting valves and shutoff valves and pressure meters and ...) Running parallel circuits may mean you need a bigger volume pump, but probably not, as the reduced backpressure usually evens things out. Usually the problem is finding a small enough pump for a DIY system. The big pump manufacturer's (Taco, Grundfos, etc) have pump sizing tools on their websites. If you want a solar/DC pump, take the numbers from their tool and then find the DC pump with acceptable specs. www.builditsolar.com has a lot of info on DC pumps.

Zero joints in the floor. Pressure/leak test every run before you cover them up (pipes do get pinholes in shipping and install). Have them pressurized while putting in the floor.

Brian Knight wrote:Great advice above. Was thinking more about your system. One of my main concerns is creating an overly complicated plumbing system to distribute heat from a source that is already going to be distributing well enough on its own. How much useful benefit will there be to heating the floors in an already heated room? Now if youre talking about a different level or distant room that's different..

DC pumps powered by PV are in a world all themselves. DC pumps last a very long time and the PV is running it outside of the grid power. The minute you hook up a grid tied AC pump to any heating distribution system you are introducing some serious concerns.

AC pumps tend to break down a lot and can represent a surprising amount of energy use. Its possible that an AC pump in this type of system would use as much or more power than a mini-split heat pump. The mini split would heat (and cool) the area almost instantly and could be dialed back instantly unlike a slowly responding radiant floor.

So that leaves you with DC pumps powered by PV or a completely passive and thermosiphoning system. DC with PV is a problem because it wont work very well when you need it the most (cold, cloudy dark).

A system that circulates itself would obviously be the ticket but presents its own challenges. The main thing is that the RMH would need to be below the floor area to be heated (warm water rises, cold water drops). Again, back to my original big picture problem that a RMH will probably be heating the upstairs just fine depending on the building envelope.

tel jetson wrote:
I think you could do it many ways. you could do an out-and-back run (or any multiple of that) and have the supply and return manifolds right next to or on top of each other. you could also do just a single run (or any odd number of runs) and have your supply and return manifolds on opposite sides of the room. unless you find or make a really fancy contraption, though, I would guess that you will need two or more manifolds. more than two would be if you wanted to have more than one heating zone so you could heat only specific parts instead of the whole system.

R Scott wrote: Pressure/leak test every run before you cover them up (pipes do get pinholes in shipping and install). Have them pressurized while putting in the floor.

R Scott wrote:Normally we did it dry. We had a home-built pressure tester that included a pressure gauge from home depot from the irrigation/sprinkler aisle, a hose Tee (with no valves, just a cap on the other side), a tire valvestem, and an air pump.

Or you can buy this already done: http://www.homedepot.com/p/DANCO-0-15-psi-1-10-Increment-Gas-Test-Gauge-94352/100180536#.UgueftKkqss

You can use sharkbite adapters and plugs to temporarily close off the pipes.

Pump it up to 15 psi and leave it overnight. If it isn't 15 psi by morning, then you get to troubleshoot. Soapy water on every connection along the mainfold, that is probably the leak. If it isn't there, then you may want to fill with water to find the leak--but it can make fixing it messier.

Rufus Laggren wrote:Sorry if I missed it but what temperature do you plan on running through your radiant floor piping? What flow do you plan? Unless you wish to constantly monitor and modify your temperature/flow you need to have some idea of what you're shooting for and try to build to get somewhere in that ballpark by default. But regardless, you still need a way to control the heat into the radiant pipe. Otherwise you get either hot or cold feet.

"Open" tanks generally need topping up regularly, more so if temperatures are high; you may want to design in a fail safe so that a low water condition doesn't destroy things or at least "fails gracefully".

What effect will this heat have on your floor? Cracks due to thermal changes or moisture changes? Do you want insulation below the radiant piping to reduce the heat you send to China?

www.heatinghelp.com is a "clubhouse" of professional heating contractors where a lot of this stuff is discussed in more depth than any sane person needs - at least in the context of "standard" construction practice. A year or so ago they were still answering questions from polite and intelligent "home owners".

Like you said, steep/high learning curve. But on the other hand usually radiant fluid temps are low - 80F. to 110F IIRC which makes them relatively benign, at least in the floor piping.

Nice project. Best luck.


Rufus

Rufus Laggren wrote:> warm water...

That's what radiant floor heating uses - at least compared with radiators which will use water anywhere from 140F to 170F when things get going in winter. Just remember that w/out insulation the heat is going to China as well as up into your room; may not matter if you don't mind burning a few extra BTU but the floor will maintain it's temp longer after the burn if you use insulation.

> just the right temp...

Yes that's always the hoped for scenario. It pretty much implies some sort of controls because the chance of you building your heat exchangers just right so it works perfectly right out of the box... If you can figure out how to control your water temps between about 85F and 110F you have solved the main design problem. After that it's just a matter of specific construction details.

FWIW vertical tanks (like a standard water heater) have two temperatures in them - top and bottom. IIRC there can be as much as 25F difference between the two. If you heat the tank by circulating water from and back into the tank near the top somewhat slowly (so as not to create a large mixing action in the tank) you can heat the tank water from the top down and may be able to maintain the thermocline (imaginary boundary in the tank where the water below and the water above differ significantly in temperature) while doing it. The point is that you then have a choice of drawing "cool" (from the bottom) or "hot" (from the top) water from the tank to mix and circulate in your floor. Be warned - I really don't know if this idea has merit; I never have seen it done this way. However heating system regularly utilize the difference between top and bottom temps to provide thermo-syphon flow; the two temperatures in the tank _may_ provide a somewhat passive way to obtain the particular temperature you want to circulate.


Rufus

Rufus Laggren wrote:
What effect will this heat have on your floor? Cracks due to thermal changes or moisture changes? Do you want insulation below the radiant piping to reduce the heat you send to China?

Rob Irish wrote:Which means I think for this project, we're going to need to look at using some more modern tech to do the job. Which I think involves:

1. A standard hot water heating unit.
2. A unit of some kind which measures the temperature of the water in the pipes, and brings in more hot water and cycles the water to maintain a constant temperature.
3. Since we currently have AC, and not DC, (solar coming down the track), a AC pump of some kind which connects to this 2nd unit, i.e. they talk to each other.

Rob Irish wrote:

Rufus Laggren wrote:

What is a good insulation for a cob floor?

So currently we have sand, then rocks, then cob. Where would insulation go?

Where we are renovating, we told a neighbour we will have clay floors and he said that will be too cold. I'm guessing this cold can be reduced with insulation?

Any information greatly appreciated.

Here is the profile of my earthen floor with radiant heat:

1/2" troweled earthen floor layer - fine finish, oiled etc.
1/2" troweled earthen floor layer
1/4 to 3/4" troweled earth floor leveling layer - checked with laser while applying
~2" clay-rich crusher fines placed moist and tamped with plate compactor
1/2" PEX tubing 9" on center tied to 6x6 concrete reinforcing mesh with rebar ties
~2" clay-rich crusher fines placed moist and tamped with plate compactor
Landscape fabric (fibrous, breathable stuff)
9" of 1/2"pumice for insulation (14" thick adjacent to foundation for about 24")
Compacted native ground / fill

This floor certainly isn't cold like your neighbor suggests. You could use 1.5" or 2" expanded polystyrene or the like to insulate the floor instead of pumice. We used pumice because it is natural, we needed material/fill to build up the floor/site anyway, there are pumice operations reasonably close. The thermal break the insulation provides is important IMO. Some people advocate no insulation for natural cooling. This really comes down to your local climate. In most temperate climates where you can count on cooler evenings in the summer - a well insulated building (floor included) with decent thermal mass will get you far.

Another note: Don't underestimate how much sand needs to go in your floor. If you use a similar mix of clay to sand as normal cob, adobe, earthen plaster, etc. you will get cracks.