re: thermal mass for multi-story house.
28 jan 2005
>where i live, southern califonia, near the ocean, the weather is very
>mild and pleasant. except for a few chilly winter nights there are no
>large temperature fluctuations...
sounds like you don't much space heating or cooling at all.
>nonetheless, i woudl like to obtain the benefits of both high
>insulation and thermal mass.
and airtightness, altho i tried to help one californian figure out
how to solar heat a new house with the windows open :-)
>land prices are so high that one cannot afford to build a home with
>the usual "earthship look"...
>herein lies my problem: an earthship provides large amounts of thermal
>mass, because it has extremely thick earth walls and an earth floor;
and not much insulation.
>...while i can build a concrete slab for thermal mass on the bottom floor,
>i don't think i can do so for upper floors due to the heavy weight involved.
a slab can help cool the house, with ceiling fans. a low-e massy ceiling
can help warm it, with thermosyphoning air heaters. you might look into
"the barra system," which is popular in europe, or put some fin-tube pipe
under a low-e ceiling, with a stratified heat storage tank on the ground.
this can also preheat water for showers.
a whole-house fan with a smart controller may also be useful, letting you
bring in cooler outdoor air when the house needs cooling and warmer dry
outdoor air when the house needs heat.
>...i plan to instead compromise by using eps tridipanels. though i do not
>have any experience with tridipanels, they sound like an ideal way to
>quickly erect a building, while obtaining thermal mass on the inside...
indoor thermal mass is good, but these panels look heavy and expensive.
>however, i wonder if this is enough thermal mass.
it wants to have lots of surface, with short internal and long external
time constants rci and rce. a thermal capacitance c btu/f with a ft^2 of
surface in slow-moving air has an internal time constant rci = c/(1.5a)
in hours. with a g btu/h-f house conductance, rce = c/g hours.
for instance, a 30'x40'x8' house with an r40 attic and r30 walls and g=128
btu/h-f and 2400 ft^2 of 1" drywall (2 layers) with 2400 btu/f plus 2400
btu/f of possessions with a = 4800 ft^2 would have rci = 4800/(1.5x4800)
= 0.67 hours (40 minutes) and rce = 4800/128 = 37.5 h.
>for a residential home, i would build a three-story house consisting of
>a finished basement, a ground floor, and a second story. for the basement,
>i can pour concrete to my heart's content so that i have the house resting
>of a thick thick slab of thermal mass.
concrete also has internal and airfilm thermal resistance...
>i might even be able to build a concrete slab onto the ground floor...
or tuck some 4" pvc water pipes between the basement ceiling joists.
>i don't know. though, if i can provide a concrete floor for the second
>story. it seems that the expense of reinforcing the house to support
>the extra weight is prohibitively costly...
water stores about 6x more heat per pound than masonry. you might
"pug" the floors with epdm or poly film water ducts between joists.