[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GSBN:Re Thermal Properties of Straw Bale Walls

Nehemiah -

Good explanations.

You talked about the straw-stuffed gaps at the top and sides of a wall as
being places where convective losses could occur.  So I'll add the thermal
importance of stuffing vertical joints/gaps between bales for the same
reason of limiting convective losses.  (This would also be important between
bales and "posts" that go mostly or all the way through the thickness of the
wall.  I've seen I-joists or steel trusses used this way.)

Straw-clay seems the best material for this purpose, although too much clay
might increase conductive heat loss in the stuffed area.  Horizontal joints
between bales don't seem to be an issue because the weight of the bales
appear to cause the surfaces to lock in well enough to limit air movement
between them (although the French dipped bales might seal that joint even
better, and bales on-edge probably nestle together better than laid-flat).

Then there's always the question of what material is between the bottom of
the plates, and what the insulative qualities of the roof bearing assembly
are.  And then there's the ceiling/roof, and the windows/doors, and the
amount of infiltration throughout, and . . . . .

Martin Hammer

> John,
> I did not see a response from Andrew.  Was that off list?  I am always
> interested in what new or other information people have on the thermal
> properties of straw bale construction.  Care to share his input?
> Also, your Q about how compaction affects R-value is a potent question.  If
> hot box testing wasn't so expensive and time consuming, or if there were
> funders lined up to pay for it, I'd already have an answer for you.  There
> are a number of confounding factors, so until someone has done the actual
> research, we can throw around lots of theories.
> For example, it is air that creates the insulation value of almost
> everything used for wall insulation.  ,,,not the spun glass, not the solid
> portions of the foam, not the cellulose, not the straw.   Therefore, if
> bales are compacted too much, one would expect the insulation value to go
> down.  But, what is "too much?"  If bales are too loose, then the air can
> circulate in the air pockets and research HAS shown that this can lead to
> convective currents that lead in turn, to a dramatic drop in R-value.  That
> was one of the causes (we think) for the relatively low R-values in the ATI
> lab tests in Fresno, CA.  Once we stacked the bales in the hot box wall
> opening, and compressed them as they'd be in a building wall, we had a six
> inch gap at the top.  We filled it with straw as tightly as we could, but
> we are not match for either a baler or truckers' strap tightening levers,
> so we KNOW that the top (where the greatest amount of heat exchange would
> naturally occur anyway) was much looser than the rest of the wall.  Ditto
> the sides, though those gaps were significantly smaller (so perhaps, harder
> to compact straw into).
> Further, though in theory greater compaction - after the optimal point -
> will lead to a decreasing R-value, no tests have yet shown that to be the
> case.  Perhaps we just haven't found the optimal compression force yet.
> Perhaps the theory is wrong.
> Lastly, I would question your assertion that jumbo bales are "naturally
> compacted a lot more than the smaller bales."  It is mechanically more
> difficult to compact a larger bale to the same density as a smaller one.  I
> am not saying that the machinery isn't designed to do so - perhaps it is.
> But, from a pure physics point of view, it is not "natural" as you said.
> One way to verify whether the compaction is greater or not is to measure
> the water content and density.  Rice straw bales in California (the ones we
> tested) are typically at least 8 pounds per cubic foot at a moisture
> content of about 6%.  Do you have similar data on the Aussie jumbo bales?
> If you want to get an accurate reading of the moisture content (more
> accurate than a moisture meter stuck a random depth into the bales), let me
> know and I will send you (offline) a description of how we did it.  The
> density (#/cf or kG/cM) is pretty easy, assuming you can weigh a
> representative sample of the bales.
> Hope this helps.
> Thanks,
> Nehemiah Stone
> stoneandstraw@...