[Strawbale]Testing loadbearing straw bale structures ++ Need HELP for conception ++ Questions at the end
chrismowatt at i12...
Thu Sep 9 02:26:17 CEST 2004
Dirk Scharmer wrote:
"Our main intention is to proove, that unplastered straw bale walls are
able to resist vertical and lateral loads coming up in two-storey
buildings (one-storey consumes to much space in small germany).
Dietmar Lorenz sent me the recent ebnetsbresearch documentation, but
unfortunately I was not able to get suitable information out of it/ to
apply the results to our german conditions yet."
"So after 1,5 hour writing I would be very happy to get some feedbacks
from experts (not only)."
After 1.5 hours writing Dirk deserves a bit of feedback, so, in the absence
of any responses from experts, here goes (from a not only expert)....
Like Dirk I have been studying the available structural test results and
have come to the conclusion they are all inadequate for any ambitious build.
This is not because the tests were badly conducted, but because there are so
many variables. The number of variables is so great that you are unlikely to
find two sections of wall on the same build that exhibit the same structural
characteristics let alone be able to define a test with universal
applicability. Variables may include type of straw (wheat, rice etc), straw
length, straw moisture content, straw density, type of binding (metal,
sisal, twine, polypropylene etc.), number of bindings (2,3), bale size, bale
orientation (flat, on edge), construction bond (running bond, etc.), wall
even-ness and plumb, pinning (material, method and placement - if any), mesh
(type, gauge and placement (if any), render (materials, material mix,
thickness, consistency, setting conditions, curing conditions, drying
conditions, quality of bond, skill of application, number of coats, method
of application - hand, pump, spray), wall layout (unsupported wall length),
number and placement of penetrations, bracings (if any), buttresses (if
any), strappings (if any), top plate and bottom plate design etc.,etc. etc..
Little wonder that structural engineers throw up their hands in horror,
commercial builders turn away and architects opt for safe single-storey
The only option seems to be to select your preferred material and
construction method, build a sample and test it -which is exactly what Dirk
"We want to test an element of 2m width and
3m height with high precompressed bales. In the first step we'll work
with small jumbo bales (0,85mx0,5mx2m). That means in every course there
is only one bale (no running bond). This not only our test specimen but
also our wall element on the building site."
This sounds a reasonable test specimen, but why not build a test building, a
garage or a workshop perhaps, and conduct your tests on something that that
will serve a purpose? Can the University of Applied Sciences in Magdeburg
not conduct their tests on your build site? I would want the tests to
conducted in conditions that are as close as possible to the real conditions
of your build.
"Here are reasons for our election:
1. Based on the consideration, that our common 2-string bales should not
loaded above 15-19KN/m2 (and this is only possible by excellent work of
the baler, which is difficult to guarantee) and based on the calculation
that a typical german 2-storey- building with strawbales in the roof
shows a maximum load at the root point of more than 35-45KN/m2 we'll
first test the small jumbo bales with a much higher density."
Jumbo bales are more dense and structurally stronger than 2 string bales.
This makes constructing the wall prior to plastering a lot easier. The wall
will be more stable and will go up quicker. However, the straw in these
jumbo bales will contribute very little to the additional strength of the
plastered wall. Most of any additional strength will be derived from the
fact that the wall will be much thicker and so able to resist out-of-plane
lateral loads much better. An alternative approach, if you need more
strength, would be to have more thinner walls built using two string bales.
This approach would maximise the amount of render used and it is from the
render that rendered straw bale walls derive most of their strength. You
could in theory have a two skin cavity wall in your first storey,
alternatively you could subdivide your ground floor into smaller rooms, so
you have more (thinner) ground floor walls supporting the weight of your
first floor than you would using jumbo bales.
I don't have the ebnet Bruce King summary of structural test to hand, but I
thought that rendered two string bales are entirely adequate for supporting
a two-storey load-bearing wall. Proving this to your code officials may be
more difficult, but here I the UK they are happy to discuss the options.
2. According to the wet climate in germany, we want to have the option
to build very fast with prefab elements (but it should be also possible
to build the elements on the building site). These prefab elements
minimizes the time exposed to possible rain to 2 days. After fixing 6 to
8 elements in normal rectangular floor plan, you can immediately
continue with the ceiling,.... Building the exact labour intensive
openings comes after the completion of the roof. We think more bigger
prefabed wall elements need to much effort to give enough stability for
hanging on cranes clamp.
I'm not convinced by the prefab idea. If you are thinking of rendering the
prefabricated wall sections before assembly then the sections will be very
heavy and you will lose strength at the intersection of the prefab sections.
I would forget the prefab idea and just have a good temporary roof and
protection for the walls. Building the openings after assembly sounds a good
idea to me, although cutting holes through .85m of straw may be hard work.
3. The reduction of the static requirement to an 2m-width element gives
hopefully more freedoom in designing openings in the walls, especially
for the southsided walls for passive solar winnings. In the output there
should be more openings possible, and you don't need to consider about
the distances between doors, windows, enough undisturbed wall length."
I'm not sure the jumbo bales will give you more freedom in designing
openings. Remember, with thicker walls and denser bales comes increased
weight and greater structural demands.
"4. We think prooving the stiffing Qualitys of earth or lime plaster
would nearly be impossible under german/ european building codes. We
prefer earth plaster, so in this case, it would be impossible to make a
safe definition of the mixture, the execution and the resulting
stiffness of it. So we decided to try our work with unplastered bale
walls, hopefully with sufficient results. If not this 2x3m wall element
gives us the chance to add wodden tie rods on each side to stiffen the
I assume from this that your reason for choosing to test unplastered bales
is based on the conservative notion that the structural strength of the
straw is all you can absolutely rely on and that the German code officials
will more readily accept the results. This has always seemed a strange
notion to me, as it is only true for the period of the build during which
the walls remain unplastered. As soon as the plaster goes on the wall
becomes a much stronger structural panel sandwich. The residual strength of
the straw bales become almost irrelevant as it is the render that carries
all the structural loads.
Have you spoken to the code officials to confirm that your understanding is
correct? It would be sad to go to a lot of trouble to tests that have little
bearing on the structural strength of the finished wall in order to satisfy
imagined code requirements that are not real.
5. The thickness of 0,85m is very high, so reducing the utilisation of
the thick jumbo bales on the 2m elements gives more space. Between the
thick 2m-elements we can you use 2-string bales (0,35m) on edge.
Am I correct to understand that you'll be using the jumbo bales more like
pillars and will be infilling with 2 string bales on edge?
Besides that we'll test small 2-string bales, too but we're not
expecting good results, probably they'll work only by using plaster.
We'll work together with a small institut at the University of Applied
Sciences in Magdeburg.
So after 1,5 hour writing I would be very happy to get some feedbacks
from experts (not only).
What do you think about our test layout/ our concept?
Will it be possible to proove precompressed unplastered bale walls for
loadbearing use (vertical a n d horizontal loads, no earthquake)?
Do you think our choosen way, not to build bale walls in running bond
with openings included is a possible solution?
"Possible, yes, but I'd talk to code officials first, test on rendered two
string bales used in a useful non-code building first. This gives you
something to show the code officials, somewhere to keep your tools and
building materials and something to learn on.
"How do you calculate the resistance of bale walls against lateral loads
Employ a structural engineer!
More information about the Strawbale