Gravity fed in-floor heating for Off-Grid home
I’m in the process of designing an off-grid home in North Bay, Ontario. with a 100% gravity fed heating system. I have no idea whether my system will work or if it even makes sense.
The water from the 400 gal tank is heated by the wood fireplace or gas water tank and via convective flow moves into the attic tank. Water from the attic tank flows down through the in-floor (or rads) back into the 400 gal tank. Since the attic tank is 10ft higher than the infloor/rads I’m sure I’ll get enough flow for this part. It’s the heating side I’m not sure about.
I was thinking I may need to add a pump to get the water flowing or to have enough flow. Since solar power is limited in the winter months I want the system to use little power and not use pumps if possible. Any suggestion are greatly appreciated.
Replies
Hot water ( lower density) rises, Cold water sinks. Without mechanical assistance, the heated water in the attic tank will stay there.
Run the numbers. The pressure difference required to move the water is the density difference between the hot and cold water. It is not much.
Plan on using circulators. The new ECM circulators do not use much electricity.
This is an Open system. When the value is open at the bottom of the attic tank the water is going to flow out. If it doesn't work I can always add the pump after.
https://www.finehomebuilding.com/forum/hydronic-heat-and-gravity-loop
Some previous discussion.
To start with, you need to know how many BTUs you need to distribute to the house in your design case. Have you done a manual J to understand the heating system size you need, and the heating requirements of each of the rooms?
House is just under 2000 sqft and very energy efficient. Heating needs are 7500 BTU/hr. Just need to get the heat distributed around the house.
https://www.engineeringtoolbox.com/gravity-heating-systems-d_189.html
Some more light reading
In-floor heating will also increase the challenge of gravity fed systems. a nice, tall radiator, with a hot and cold side would help things move along, while a long, flat, circuitious in-floor plumbing will not be as helpful.
You would want the cold (relatively) return back to the heating source, not directly to the storage tank.
the attic would be the place for an expansion tank, not a hot water storage tank. you will have to address keeping the system full.
It's not a pressurized system so no expansion tank is needed.
Attached is a system using a wet-back on a fireplace to heat water in a tank on the 2nd floor. Rocket Stoves seem to heat water very well without the use of pumps so I'm trying to incorporate some of that design.
Expansion tank is needed because you will heat the water. when it heats, it takes up more room.
You can use an expansion tank that is open to the air, but it still has to be in the system.
Since the tank in the attic is open, would it be considered the expansion tank? Or are you saying there needs to be another open tank?
an open to air tank, precludes the need for an expansion tank, which is only required in closed systems.
What's not obvious in my diagram is that the 400 gal tank is 2 ft higher than the fireplace. As in a wet-back fireplace heater when the water is heated it has to go someplace so it's going to rise up through the system and into the attic tank (as long sufficiently heated - Rocket Stoves do this very well) and cold will be drawn in at the bottom. This type of system was used in Europe a century ago so it should work for me if it's designed correctly.
I know I could stick in a couple of pumps, but I'm trying to avoid that if I can.
https://www.oldhouseweb.com/how-to-advice/gravity-hot-water-heating.shtml
Here is another discussion of gravity driven hot water heat distribution.
Your design has a few basic flaws.
There is no way for the hot water to get to that attic tank. Assuming you somehow get the tank full of hot water, as soon as you open the valves, it will empty and put hot water into your large storage tank, and in the process, empty all the distribution pipes. (Fill them with air)
The only reason a wetback heater circulates water into a tank is that there is a continuous path for water to circulate, With a cold (relatively) return from the storage tank to the heat source. The water in this system does not have to be pressurized over that provided by the atmosphere and elevation, but it needs to fill the pipes. air in one section will prevent circulation in the whole circuit.
Your design would require a pretty hefty pump to fill the attic tank. It would have to overcome the elevation head pressure between the level of your heat sources and the attic.
https://www.engineeringtoolbox.com/pressure-head-water-d_1354.html
for a 15 foot elevation, you need to make up for 6.5 PSI. (Just gravity, will need more to overcome friction losses)
Density reality:
The density of water at 180 degrees f is 60.586 lbs/square foot per foot of elevation
at 160 degrees it is 61.006
at 40 degrees f, the density is 62.423.
Between 40 and 180 degrees, the density of water change is 1.837 lbs/square foot per foot of elevation.
This is 0.013 pounds per square inch. for a gravity-driven system to work, this density change is the only thing to move the water.
(between 160 and 180 degrees the density change is 0.003 PSI per foot of elevation.)
The only way this can move water is if there is a return path, and the entire circuit is full of water. If you have 15 feet of elevation, the density change will add to give you 0.045 pounds per square inch of pressure difference to push the water around, and circulate heat.
The 6.5 psi of head pressure is very real, but if the pipes are full of water, the 0.045 psi of pressure will move water through the system. The 6.5 psi will spread throughout the system up and down, and similar elevations will have similar pressure.
it is worse. if the system is not full of water, the only thing that heating water will do is make it expand.
Thanks for the link.
The design I posted leaves out much of the detail. The full design has a shut-off value to stop flow when the attic tank is low so all the water doesn't drain out of the system. There's also a fill value connected to the pressurized house water to fill the attic tank.
If there was a separate pipe running from the attic tank back to the heat source would that work?
In gravity systems, the motive force is provided by the density difference between the hot and cold water. In the diagram from your original post, that heat source is the water heater and wood stove. If the 400g storage tank is hot enough, it will provide some motive force as well.
The tank in the attic will not provide any motive force since it is higher than were you want the heat. It is acting only as an expansion tank. Expansion tanks are normally (Single) piped on the cold side, with an overflow. Unless you are piping everything with corrosion resistant materials, the system should be pressurized, and closed from atmosphere.
If you really want a pure gravity feed system:
-Run the heat loss calculations to determine how much heat is required.
- Layout of the heat emitters to put the heat where required
- Layout the piping layout for minimum flow loses. (Large diameter pipes, few elbows. Where elbows are required, use large radius)
-Calculate the flow rate and temperatures required to match the design conditions.
-Calculate the pressure drop through the system at the design conditions.
-Calculate the temperature difference required to provide that pressure.
It can be done, with attention to a lot of small details.
You have to run the numbers to see if it can be done in your situation.
Large dia pipes are expensive, Circulator pumps are realatively cheap. ECM pumps are more expensive, but stingy on electricity.
Thanks for all the input. I see I have a lot more work to do.
For most of the winter months we don't get enough sun to charge the batteries and running a generator uses fuel and the fuel prices keep going up.
An ECM pump doesn't use many watts, however 1 or 2 circulating pumps running 8 - 10 hours per day will still use close to 1kW / day. That's 15% of my total daily energy usage calculation. I need heat circulating on it's own or I have to increase my battery capacity and batteries are already the most expensive part of the system.
in my experience with a Grudfos ECM pump with a heat loss of 6 Kw average power consumption was 12 watts per hour when only a couple of zones were open it could be as low as 5 watts,power consumption was negligible
If the numbers work out so that the system will circulate by gravity feed, that would be great. The calculations will tell.
I have a TACO VT2218 circulator in my home. When it is running, it is idling along at ~ 8W, with an occational burst to purge cold water when a zone first calls for heat.
8W x 24hr = 192 Wh
Your system has many heat sources, many heat destinations, and many loops.
You may want to simplify each loop, and understand the dynamics and operating conditions. Temperatures of each heat loop source and sink, and what the conditions are.
You solar can easily heat the water so hot that you need a tempering valve to prevent scalding from the domestic hot water.
The large tank heated up by solar should be great and should heat the house and domestic hot water in mild condtions, but when it is cold and long nights, you want to make sure your water heater and fireplace loop has a way to heat without heating the whole large tank of water.
so a smaller storage tank to moderate heat from the fireplace is a good idea, but for a gravity flow system, it should be in the basement, so gravity will work. (if the tank is in a loop with your in-floor and rads, but is in the attic, gravity will want the hot water to stay at the highest point. If the smaller tank is below the elevation of the heaters, the density difference of hot/cold water in the pipes will help things move along.
You will want to understand all the combinations of temperatures over the heating seasons, and interactions, including how heat from solar moves to the large tank,and from there to the secondary tank.
Controls will be an interesting discussion. low electric power, low fluid drag. Where to put automatic operated valves, and where to put manual valves
I think this is an interesting design discussion. Thanks for posting here.
has anyone near you built something like this?
seems they would be glad to help you with what works and does nor.
designing a first of the kind rig might be risky