This is why I would recommend an electric radiant ceiling. Will respond quickly and depending on ceiling height hot head may not be an issue. You have easy zoning and at your sq ft heat loads can be laid out in a way to minimize any hot spots. Will respond in minutes instead of hours.

...whereas a mini-split is a "set and forget" kind of backup, a high-efficiency method of putting a low-limit to the air temp, and doesn't need to be super responsive, and you get cooling out of the same system.

In a decently designed high-R building there's no need to micro-zone it to death, temperature differences between rooms will occur depending on the occupancy of the rooms and the heat loss out of the exterior walls, but those need not be huge deltas. Putting the interior head(s) in the highest loss or most-critical for comfort room for comfort is part of the overall design process. Passive solar homes usually have bigger gains and losses on the sunny side anyway, and will have significant temperature deltas with other rooms, and minimizing or putting a bound on those deltas is already part of the design process.

I am building a large vacation home at the Taos Ski Valley and I am using the radiant floor heating. I am researching the boiler situation. In a closed system in cold climates at 10,000 ft do you think in a 6,800 sq. foot house that is a vacation home, should I use 2 boilers or one large boiler? The plumber is recommending 14 zones. Do you think I could use distilled water and not use glycol or what is your recommendation? How many pumps would you need for 14 zones?

TreeTops: Without a room by room heat load calc and whole house heat load there's no meaningful answer to the question. Microzoning has a lot of associated problems, and might not be the way to go no matter what, but it depends. You might want to start your own thread- I won't be responding further to you on this thread, but might if you start another.

Not surprisingly, John Seigenthaler has already weighed in about using passive solar slab-mass as radiant floors (and has roughly the same as my take on it- on a cold clear night with a bright & sunny day it's easy to overheat the place.)

His proposed solutions are to use low(er) mass low-temp panel radiators  or radiant-wall to provide the supplemental heat with a reliably fast turn-off time, and leave the main mass alone.

Electric radiant ceiling...riiiight. Works great, like if you don't mind the trebled fuel bill.

As usual Dana and Ziggy are right. Making things simple is a lot of hard work.

Electric rates and hours of operation are not the same all over the world. In a lot of western states is cheaper to use electric then propane. An existing hydro plant would be kinder a gentler than hydrocarbons. There would also be off peak utilization to be considered.

Hydronic or air radiant in the interior walls (air heat with no drafts) makes sense to me. Heating the exterior surfaces (upper ceiling, exterior walls, lower floor) may be comfortable, but makes less sense energy wise.

If you have cheaper off-peak electric available, some type of water storage system makes sense. Preferably with a heat pump.

In just about any electricity market in US zone 5 or lower it's cheaper to heat with a mini-split than radiant electric or propane. There are a few places on Long Island where condensing propane could still beat a mini-split though (primarily due to power rates pushing 2x the national average, and steeply ascending residential rates with power use.)

I don't know of any market in the lower 48 on the gas-grid where radiant electric would be cheaper than condensing gas, even at off-peak, but they probably exist. If you can get 5 cent/kwh electricity that's ~68KBTU/$, and $1.30/therm gas in a 90% burner would match that. Gas is that cheap or cheaper in most US markets,and sub-5 cent off-peak power isn't exactly common. To meet or beat more typical buck-a-therm gas price levels you'd need < 4 cents/kwh pricing.

Existing hydro power isn't very cheap & available. anymore. In NM (where this house is going) coal fired thermal plants account for 49.8% the grid source of power, gas fired thermal is 41%, whereas hydro accounts for a scant 1% according to EIA data as of Y2009. (see: http://www.eia.gov/cneaf/electricity/st_profiles/sept04nm.xls ) A 90% efficient gas burner has about 1/3 the carbon footprint of the ~30% efficient grid sources in NM.

At an average COP of 3 a mini-split would about break-even on carbon footprint with a condensing gas burner on carbon footprint in that relatively temperate winter climate, and has a comparable operating cost at their (lower than US average) ~11-13 cent winter electricity prices (see: http://www.whatsmypower.com/locations/87043/rates/642 ) and ~ 95cents/therm gas.

I use electricity in one of my homes, it is of-peak and nearly as cheap as NG (unavailable). Naturally, the radiant system is hydronic driven by an electric boiler because I may want to change my mind (wood, pellets, solar, oil, propane, NG). Once you bury the wire you've decided for everyone, as long as the house stands.

Radiant ceilings are great, radiant electric ceilings need a good excuse.

OK, Dana and Ziggy, let us posit a cold night followed by an overcast day followed by a second cold night followed by a second overcast day. Is it possible you might want a means of charging the thermal mass directly? The problem here is your attempt to make passive solar something it is not. (Towit: pliant and predictable.) I say again: Pick a point in the day when you check the temperature of your slab, and sundown makes more sense to me than sunrise. Experiment to the point where you know how much heat to add; embrace the amplitude. Among the folks who have actually built and lived in passive solar houses, the aux heat of choice is a masonry heater, which is significantly heavier than drywall. Their advantage perhaps is not knowing who John Seigenthaler is.

let us posit a cold night followed by an overcast day followed by a second cold night followed by a second overcast day.

Ha ha. I think you are describing my climate. Our goal here with passive solar is to absorb as much as possible during the shoulder seasons (generally Spring and Fall). The fact we have a mild climate makes that strategy more workable. During the Winter, you have to be in an active heating mode, and in Summer, in defense mode.

If insolation to the passive absorbers tapers off, either during a cloudy period, or at night, the active heating system simply steps up to fill the gap.

A masonry heater has a lot of thermal mass, but is rare used as the primary mass system for the passive solar ( which is usually slabs, sometimes the south facing interior walls of the heavily glazed south wall. ) With some empirical tweaking it's possible to figure out how much heat you'd need to put into the massive stove to have it's emittance decay roughly with the rise of the passive solar end, but they're fairly independent systems. I'm sure it's possible to make them work very well with passive solar with cognizant-human control

John Seigenthaler is a hydronic heating designer who is quite well versed in how to do the math on the mass & temp of the heating emitters, and selecting the appropriate controls. (And he knows the difference between mass and thermal mass.) His primary advantage is good instincts for the basic physics, and the ability to both calculate & measure actual performance as a system. He doesn't design masonry heaters, but he has vast experience with high-mass radiant floors, as well as active & passive solar. If you want the systems to work simply and reliably without a lot of human input such as measuring the temp of the slab and scouring the 48 hour forecasts to figure out how much pre-charge you need) his advice is relevant. Not everyone is heating-system-obsessed, most just want to be comfortable. A low/mid mass hydronic heating using separate emitters is a "set and forget" backup system that can handle all weather variations without screwing up the passive solar design the way using a high-mass slab for both does.

That's some lovely hair splitting on mass/thermal mass, Dana. So overheating a slab is worse than overheating a massive fireplace because -- what? -- the fireplace can be rolled outside? Or are we to assume that if human cognizant control works for one there is no reason to think it won't work for the other?
As for simple, reliable automation of aux heat, show me a house so equipped that gets 75 percent or more of its heat through its windows. Even in predictably sunny climes like New Mexico, snow/no snow would make a big enough difference in many house orientations that someone should be paying attention to forecasts. Set and regret?
Automation is possible with a radiant slab. Sort of. I am installing tekmar 511 thermostats that can programmed by slab temp and time of day. Set the clock for midnight to, say, 9 a.m., and all that remains is finding the Goldilocks setpoint that splits to the difference between warm enough in the morning and cool enough in the evening.
On an average day.
If such a thing exists.
My view is that only cognizant humans need apply.

There's no "hair-splitting" on mass & thermal mass- it's a fundamentally different characteristic of the material. A hundred pounds of water has 9 times the thermal mass of a hundred pounds of iron or steel, and 5 times that of a hundred pounds of concrete. But they all have the same mass.

The primary thermal mass in most passive solar houses are place where the get direct sun, which is what distinquishes it from the other thermal mass in the house (including massive stoves.)

My bad. So overheating "primary thermal mass" is worse than overheating "other mass in the house (including massive stoves)" because -- what? -- the stove can be rolled outside? Or is an injudicious addition of heat injudicious regardless of how it is done?

Right over his head Dana. Why you do it...I can not fathom.

If you can't get ICF's simplified - yet functional - description, you are in the wrong room, unless you sell or are otherwise emotionally committed to primitive technology. In which case, fancy slab sensors are contrary to the primitive credo.

So, where does "primary thermal mass" stop and "other thermal mass" begin? For example, my main floor is all one slab, but not all of it can be seen by the sun. And, the parts that are seen by the sun vary during the year, of course. I bought a box of children's sidewalk chalk at WM and am having fun mapping out the solar areas to see how closely they match with predicted. I can also change a section from one classification to another by dropping a shade or even by moving a piece of furniture or a throw rug at any one time.

I certainly haven't read the entire body of Siegenthaler's work on passive solar and radiant heating, but one thing did jump out at me about the article that Dana1 provided. Well, actually, a couple things. The first is that you don't want to arrive in the morning with a fully heat-charged home, precisely so there is "capacity" for the expected solar contribution you hope is arriving later that day. The second is that heat loss from your overall structure continues, particularly on a cold clear day. Once you turn off the energy to the radiator (slab) something has to replace that heat input. The solar input usually starts more gently, controlled by the transition from East to South glazing. The interior is cooling and the slab is cooling as well, so the solar impact isn't quite as dramatic as one might expect. I am still unconvinced by the argument that radiant heat in a passive solar home needs to utilize some other radiator than the slab. I still think it is a perfect marriage. It also allows us to chill the loops during the Summer, thereby defeating much of the unwanted solar input in that season.

Dana is simply calling the primary receiver of solar energy the "primary thermal mass". I don't know who brought up "secondary" thermal mass... it's not really a relevant term.

Using a masonry heater in any fluctuating internal heat situation is a matter of management. If they could achieve the same type of lifestyle with a low mass heater they'd have better results. the only advantage of a masonry heater it is tames a very spikey heat source (wood, which wants to burn fast and dump tons of heat all at once) with thermal mass, making a short spike into a long emittance. Masonry heaters are thus used only because the passive solar people want wood and do not realize that a wood boiler setup would serve their needs better (though, perhaps with a bit more maintenance). IN many cases cheaper too, a lot of those masonry heaters are really expensive.

If they had a wood boiler and tank of water with low mass heating they would not be required to plan ahead and the system could simply adapt to the reality in realtime. that's an improvement by any measure other than the "look at my pretty masonry heater" measure.

That said, in a passive solar situation a floor slab is likely only a few degrees above room temperature in most cases if it is heated. I do not accept that those few degrees are an insurmountable problem in most cases. Passive solar homes have overshoots when they are picking up solar... otherwise, they are not passive solar homes, you have to pick up more heat than you need, at a temperature above room temp to have heat for use later... That's life, otherwise all you're doing is killing your daytime heat demand. with a simple floor sensor you can avoid undershoot on the back end, and you might lose a few BTUs but collector efficiency does not drop so fast that just a couple of degrees means that much.

in terms of storage cap to a PREDETERMINED storage temp, it does, but then, that's not passive solar. the reality of passive solar monolithic slabs is that their bottom halves are cold at the end of the day without floor sensing, so the storage cap is not fully utilized in any case, so there has got to be significant wiggle room in the realized storage capacity vs theoretical maximums we are nowhere near.

for suspended floors that argument may be different, but in that case your slab mass is likely much lower as well and so you only lose a relatively small amount of energy early on.

"That said, in a passive solar situation a floor slab is likely only a few degrees above room temperature in most cases if it is heated. I do not accept that those few degrees are an insurmountable problem in most cases."

Insurmountable, no, but it's an unnecessary complication to the design & control, and the advantages of few in sunny NM.

toddm: All the thermal mass in a house participates in all heating/cooling aspects, but not all is directly involved with the passive solar aspects of a passive solar design. The (thermally) massive components that get direct sun exposure are the most relevant (and best left unheated by other sources.) But yes, overheating any huge thermal mass to meet the overnight heat load can lead to overheating on a sunny day, which is why Seigenthaler (and I) consider a lower mass backup a better solution. (The wood boiler/buffer tank + lower mass radiation solution works too.)