|
|
re: strawbale houses, superinsulate verses thermal mass
27 dec 1999
robert w. tom wrote:
>>>...strawbales are insulation, not mass per se.
>> us r50 or so, if there's no thick layer of plaster running from one
>> side of the wall to the other to reduce their thermal resistance...
>you might want to check some of test results gleaned from tests
>done on wall *assemblies* , as carried out by dr. christian at
>oak ridge national lab...
jeff christian's group also checked out sip wall assemblies, which have
a lot less thermal bridging and air infiltration than stud walls and
may work better for people with more money than time. i built a strawbale
wall with no thermal bridges a few years ago, and can't imagine building
a whole house that way. a little mass-production would help, eg some ties
and t-strips to allow fastening interior stucco wire and vinyl siding,
and some way to avoid a) voids in the insulation and b) having to age and
tension the walls while they squish down. a rigid frame infilled with bales
solves b) but not a). encasing each bale in concrete all round solves both
but adds thermal bridging.
>...earlier canadian studies of the "quebec syle" of mortared straw
>bale construction (cmhc 1986) where the bales are set in a
>structural grid of concrete (complete with massive thermal bridges
>around each and every bale)...
well, that's canadians for ya...
>...found that there was an overall reduction of about 30% in the
>effective thermal resistance due to conduction losses.
still pretty good insulation, but lots of work... they might use
all-vermiculite concrete (astm c332-66) instead. here's a recipe
from prof richard hill's 1977 stickwood furnace report:
proportions: 8 parts vermiculite (about $2/ft^2 in bags or
cheaper in bulk from horticultural suppliers)
1.5 parts portland cement,
3.5 parts water.
1. mix equal volumes of sifted #1 and #5 vermiculite (#5 is the finest)
or, if available, use #3.
2. place half the water into the mixer.
3. add all the cement and mix to a thin, soup-like consistency.
4. add the vermiculte. the mix will become dry as it absorbs water.
5. add additional water and 2 oz. of air entraining agent (available
at any ready-mix concrete plant) until the mix becomes workable (the
consistency of shaving cream is about right. this must be done slowly
and carefully. the mixture can become too soupy if care is not taken.
prof hill says this can raise the us r-value from 0.2 to 2 ft^2-f-h/btu-in
and lower the density from 120 to 15 lb/ft^3. much easier schlepping...
compressive strength is directly proportional to density.
with stiff enough inner and outer walls and a good tensile connection,
we might just sprinkle vermiculite on top of each bale.
nick
|
|
