Cob: Cold Temps

[Shannon C. Dealy]

On Mon, 12 Aug 2002, baco at pacinfo wrote:

[snip]
> I am not an engineer and have not constructed anything of cob. My interest
[snip]
> The specific problem that I saw was a lofted structure. The main support for
> the loft area had less support on one end than I would have chosen to use.
> It was also not tied in as well as it might have been and was missing a
> gusset that it was notched for. Maybe it will hang there for a thousand
> years, I don't know. My problem is that the builder does not know either and
> almost makes a study of not talking engineering.
[snip]

Did the builder actually say they didn't know if it was strong enough to
hold together, or did they simply not have any engineering data to provide
you with as proof that it was strong enough?  There is a big difference.
Sometimes it is enough to simply know your material, and if you aren't
pushing it anywhere near it's limits, engineering data isn't really
necessary.  If I wanted to build a platform to keep my 2000 lb honda off
the ground, I could simply bolt a solid row of six foot long oak 4x4's
across a couple of steel I-Beams lying on the ground four feet apart, and
no one who is even slightly familiar with the materials would question
whether or not it could support the weight.  Sure you could get all sorts
of engineering data, but what would be the point?  Don't mis-understand, I
am as hard core on science and engineering as anyone you will find, but
the actual "need" for the data and computations is very much application
specific.  In the case that you described, it may be that the builder
knows the materials well enough to know they can easily handle the
loads involved, where you being less familiar with them do not have the
knowledge to make such an assessment.  Of course the person may in fact be
wrong in their assessment of the material, but this also happens in
engineered structures as well.  Many modern structures absolutely
require engineering at some point, in part because with modern buildings
we try to use the absolute minimum of materials in order to cut costs, so
we push those materials to their limits, where for many older/traditional
building techniques, the materials were chosen because they were cheap and
plentiful, so they can be (and were) over-built, using far more material
than a carefully engineered modern structure would require to accomplish
the same thing.

I could easily design (without any engineering data) a cob house that so
long as the ground under it didn't collapse would withstand any earthquake
in recorded history without any danger (from the structure) of injury to
the occupants, but who wants to spend all the time required to build eight
foot thick cob walls on a one story 100 square foot house :-)  Of course
someone like John who knows alot more about earthquake design than I do
might decide that even without data on cob, four foot thick walls were
sufficient, and whatever he decided I'm sure would be more than adequate.
Ultimately this is why engineering data is needed, so that we know from
an experience independent perspective, how far we can actually push the
materials without having to overbuild the structure, but until the data is
available, a combination of adequate experience/knowledge of the
materials, and overbuilding can be a safe and reasonable alternative
for many circumstances.

FWIW.

Shannon C. Dealy      |               DeaTech Research Inc.
dealy at deatech.com     |          - Custom Software Development -
                      |    Embedded Systems, Real-time, Device Drivers
Phone: (800) 467-5820 | Networking, Scientific & Engineering Applications
   or: (541) 451-5177 |                  www.deatech.com