[GSBN] Jumbo Bale Question

[martin hammer]

John, thanks for your jumbo bale thoughts.  You have more experience with
them than anyone.  (btw, great winery building.)


On 5/6/09 5:14 PM, "John Glassford" <jacksflat at gmail.com> wrote:

>> As another point of comparison, the maximum aspect ratio for adobe
>> construction in the International Building Code (2109.8.4.2) is 10:1. Straw
>> bale codes to date have used a limit of 5.6 to 1.

> Which code mate? 

Every US SB code I'm aware of uses the 5.6 to 1 h/w ratio.  It emanates from
the Tucson / Pima County, Arizona code, which is (for practical purposes)
the first SB code.

>> At least in theory, it is the skins only that determine an SB wall's bearing
>> capacity, because the plaster skins, not the bales, carry the load (as the
>> stiffest element of the wall assembly).

> Not so in practice mate.  What happens is that the render creeps
> downwards around 7mm over 2.7m high.  We conducted creep tests at the
> UNSW and we only did the test using cement renders however lime and
> earth would do the same.  So in effect the render has moved away from
> the load bearing top plate and the bales are really doing the work.  I
> doubt that you could get two string bales to perform like the jumbos
> in fact I know you cannot.  What I mean is that a jumbo bale wall at
> 900 mm wide with a 1 metre top plate will support a much bigger load
> than a standard bale however maybe these jumbo bales need a test or
> two?

I'd like to understand this better.  Do you have documentation of your
render creep tests?

However, John, even if the render creeps away from the top plate (thus
eliminating direct transfer of roof load to render) it's likely the render
is still carrying most or all of the load.  It is transferred into the
render through the straw instead.  At least I think.

I suppose it's possible the load bypasses the render and goes straight
through the straw to the foundation (especially with dense jumbo bales).
But for this to happen the straw would compress immediately or over time,
whereas the stiffer renders would not compress, and there would presumably
be tearing at the straw-to-render bond.  (Maybe that happens within
acceptable limits?)  The dynamic depends on when the wall was loaded, when
the render was applied, how stiff the render is, and how dense and
pre-compressed the bales are.

I think this is an important question.  What really happens in load bearing
walls, and what factors affect what happens.  That is, what are the actual
load paths.  Although I generally still believe the renders carry the loads,
maybe this needs to be reinvestigated/rethought.  Especially with certain
wall systems, or roof-to-wall connections.

Martin