Cob Re: cordwood homes

[M J Epko]

At 08:44 AM 12/1/97 -0800, you wrote:
>Cordwood in the wall would certainly reduce total thermal mass 
>compared to pure cob, and wood is not very insulative either 
>(although neither is cob...)

	About the same time that Shannon set up this list (undying gratitude to
Shannon), this stuff about cob appeared on the strawbale list:

*

Simon Randell wrote:
I recently conducted a thermal heat transfer and thermography test on a cob
cottage here in the UK (we have 40,000 structures in the SW). The result:

U'Value - 1.0 W/m2degC
density - 1860 kg/m3
test duration - 29 days 

Density obviously had a huge effect and so did the heating source. 
Moisture content unobtainable due to metals in the mix giving 
inaccurate reading. Anyway, not bad for 170 year old building!

*

Andrew Alcorn wrote:
In my experience, cob walls are typically 300 - 400mm thick for single
storey, and @600mm thick for the ground storey walls of double storey
buildings. Figures I have for earth walls give an R value of 0.7 W/m2degC.
Assuming that cob walls, having more straw and other fibre than your typical
adobe or rammed earth wall, have a greater R value, then maybe it was 300mm
walls that were measured?

*

Jim Lindley wrote:
The meaning of U-value of 1.0 W/m2degC is that given (for the SI system) a
one square meter wall surface area and a temperature difference of one degC
across the wall, the rate of energy flow would be 1.0 watt.  Or using the
U-value of 0.176 Btu/hr-sqft-degF. The heat flow would be 0.176 Btu/hr thru
a one sqft section of wall for every degree of temperature difference.
Note that this does not say anything about wall thickness and I suspect
that it should.  Obviously an R-value of 5.678 is not very high.

In the SI system U-values for materials are typically expressed in units
of per meter of thickness, i.e the units become W/m2degC/m or W/mdegC.  The
U-value of 1.0 W/mdegC would then be converted to the IP system as 0.5783
Btu/hrftF or 6.94 Btu/hrsqftF/in. This would give a conventional R-value of
0.144 which falls between values normally used for common brick and concrete.

*

I write now:
Rob Roy gives end-grain (as opposed to side-grain, like a regular log
house, which has about 50% better thermal resistance than end-grain)
R-per-inch values as follows:

White Cedar					1.00
Poplar, Hemlock				0.80
Douglas fir					0.76
Oak, Maple, dense hardwoods generally	0.60
Read heart cedar				0.60

	From there he goes on to calculate the total R of the stackwall matrix. In
any event, it appears that wood of any sort has equal or better thermal
resistance than cob & would likely serve to increase the total R of the
wall. A cob/stackwall hybrid would have significantly reduced mass
qualities though.

	The R-value of plain old *uncompacted* earth is said to be about 1 per
inch (per the University of Minnesota Underground Space Center). Depending
primarily on the quantity of straw in the cob, I'd think it could possibly
approach 1/" as well despite what was requoted above. In which case cob and
end-grain wood are roughly equal in this regard.



   ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      M J Epko        duckchow at ix.netcom.com
      almost Wyoming, north of Nebraska, USA
               by way of New Mexico
      (not soon enough) - for now, Minnesota
   ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      I saw the best minds of my generation
      destroyed by madness, starving hysterical naked. 
                                - Allen Ginsberg, Howl