Silvaire wrote:
The use of water to heat houses, particularly when circulated in the floor is a bit strange to me.
To me it makes perfect sense. But you’ve got to think of the perspective of heat transport, not heat containment.
Most houses have a single central furnace somewhere. The heat from that furnace needs to be transported to the various rooms in the house. That’s done by circulating some sort of medium that can absorb and release heat.
Water has a heat capacity of 4200 J/kgK and is a liquid at 1000 kg/m3, while air has a heat capacity of 993 J/kgK and is a gas at 1 kg/m3. So in order to transport a certain amount of heat (energy) from one place to another, given the same temperature differentials allowed, I need to move 4000 times the volume of air vs. water. That’s the difference between a small-diameter water pipe (standard here is 22 mm for main piping, 15 mm for branches) and quite some significant ducting for air. The water pipes are also easier to insulate (smaller circumference) and require a lower flow rate (less noise). And since I need to move significantly less water, both in terms of volume and in terms of weight/mass, I can do with a smaller pump.
And it gets better. When you use ducted air, the air that comes out of the duct needs to be at slightly above room temperature, maybe 30C?. But radiator water is commonly heated to about 60C. Which makes it even more efficient as a transport medium. (Water that goes into a heated floor is at a lower temperature though or you’ll burn your feet.)
Now I’m not an American plumber, so I have a question: Where does your furnace gets its air from? Is this lukewarm air from the rooms that gets ducted back to the furnace to be re-heated (closed-circuit), or does the furnace use cold outside air? What happens to the lukewarm air from the rooms? Is it just expelled into the atmosphere or is there some sort of energy win-back? With our water systems, once the water has heated the radiators or the floor, it gets piped back to the furnace via insulated pipes for re-heating. So it’s a closed circuit and virtually no energy/heat is lost in a place where you don’t want to lose it.
Silvaire wrote:
The mass being heated increases the time constant for changing the interior air temperature to the point where you have to hold the house at a given temperature continuously,
The amount of water that’s circulating in the system is relatively small. I have not done the calculation but I would be surprised if it’s more than, say, 100 liters or so. So the heat that’s permanently contained in there is tiny compared to the heat that’s transported. So there’s no need to run the system 24/7. In my house, from turning up the thermostat to feeling the temperature rise in the room takes minutes at most. (Heated floors are an exception and need to be kept warm 24/7, but that’s not because of the water but the surrounding stonework/concrete/whatever that can’t be heated up too quickly or it’ll crack.)
I’d have to look up the furnace fan power, but surely the waste heat from the fan goes into the air you are trying to heat with the furnace anyway? Same with the aerodynamic loss of circulating air in ducts (which is most of the fan power). It heats the air, as does the energy loss when the air ejects into the room and stops. This is a surely a problem when you are using an air fan to circulate cooling air, but it doesn’t seem to be an issue if the intent is to heat and circulate the air, other than the (locally 4X higher) price for electric energy.
I’ve worked on an air cooling system where unavoidably half the heat to be removed from the air by a heat exchanger was created the fan circulating the air 
Luckily the machinery being cooled was 98+% efficient so the fan loss was manageable. We didn’t vary the fan speed, which would have been one approach to increase efficiency of the cooling system at lower system power levels.
Answering your question, yes, in a typical US home heating forced air system the air circulates (closed loop) in the house, picked up from the rooms and reheated.
I would think a recuperator on the furnace exhaust would be beneficial, but again I’m no expert given our minuscule use of room heating over the year.
Air duct size versus water pipe size might be an issue depending on the house. Our heating ducts are in a well insulated attic, plenty of room for them, and are also individually fiberglass insulated. Presumably losses there heat the room ceilings.
Re liquid heated rooms, as you’ve noted, the mass being heated by water must include the entire floor, unless you have above floor radiators in the rooms. That’s why I said “particularly when circulated in the floor” in my first sentence. Presumably the 24/7 loss is to the dirt underneath, limited by the thermal diffusivity of dirt or any insulation present under the heated floor. I’ve never known anybody in e.g. Germany to regulate the temperature in their modern house on a daily cycle as would be the norm here, but maybe some do. It’s always struck me as strange but OTOH the houses I’m thinking of are almost always occupied by somebody during a given day.
Silvaire wrote:
but surely the waste heat from the fan goes into the air you are trying to heat anyway?
I was thinking more in terms of noise. I agree that any energy wasted in the fan/pump is turned into heat so not really lost in that respect.
And yes, my house has standard radiators in each room. Kind-of old-fashioned but they work just fine.
Most of the issues are to do with retrofitting. You can retrofit a water based heating system but one which uses air ducts is much more work.
I’ve seen air ducts done – for aircon; no other practical way to do that – and it is a huge job in an existing house. You basically have to move out while they are doing it 
Nothing is perfect… water systems can develop leaks.
Underfloor heating is otherwise not new to the UK. I stayed in Scotland in 1969 where they had underfloor electric heating, in concrete floors.
Interesting that nobody has illuminated the matter of heat pumps. I think it is basically a load of hot air (BS).
@BackPacker, yes noise is an issue with forced air. Our house’s system in particular would not be acceptable to us if we ran it a lot. Others are better, but none are as silent as water heated floors.
I had never considered that water filled radiators would be uncommon in the USA, but if you favour circulating air rather than water, perhaps it’s because air conditioning is much more common there. Radiators don’t work well for cooling whereas circulating air can do both. All the institutional buildings I’ve worked in have never-ending issues with air ducts: they whistle, heat or cool unevenly, take up masses of space, and can spread legionnaires’ disease.
I have a nearly new ultra efficient gas boiler that I got installed as I thought it was the right thing to do. My fault for not checking… But the idea that it wasn’t has come as an unpleasant surprise.
I’ve never needed to do repairs on air ducts in any house I’ve owned, but I’ve fixed a lot of corroded water pipes. As with air cooled engines, the lack of liquid is a maintenance benefit.
I once lived in a small US house that had direct, resistive electrical heating. I still own the house but don’t pay the electric bill Electrically heated ceiling panels were a fashion in the 70s and 80s, super cheap to install, no maintenance, but its practical only if you hardly use the heat – the monthly cost is outrageous otherwise.
Personally I’ve had far more issues with plumbing for tap water than plumbing for heating – water for the latter is recirculated and has anti corrosion additives added, so the pipes don’t fur and corrode to the same extent.
When I worked in Galashiels Academy in 1966-68, it had underfloor heating in concrete floors. I found it good, but my boss wore wooden sandals as it harmed his feet.
The 1970 building I worked in in in Inverness had water radiators. We often had to open windows as there was no effective control.
In the 90s, the Council had Consultants look at energy saving. Lights were fitted with sensors to switch them off if no movement was sensed. With a class busy writing, they went off every few minutes, and were turned back on. (Fluorescent tubes, which use most electricity when turned on.)
And the windows often were open to keep the temperature down.
Staff raising this were told the experts hadn’t suggested radiator thermostats, just the light controls.
When I worked in Galashiels Academy in 1966-68, it had underfloor heating in concrete floors. I found it good, but my boss wore wooden sandals as it harmed his feet.
My parents used to have a dog that loved to lie on the warm bit of floor over the hot water pipes. I know what form of heating she would have chosen.
