[Strawbale]Measured heat flux through a real wall/ceiling

[Mark Bigland-Pritchard // Low Energy Design]

Jenik -
Your previous post made clear what I had negligently left unsaid about 
the role of convection currents in sb walls.  Thanks.  In response to 
the one below,:
Jeff Christian's view of the failed ORNL R-value test  (he backed up 
this view with CFD modelling) was that a convection cell the full height 
of the wall was established in the bales.  This was only possible 
because the plaster was inadequately bonded to the bales, leaving 
vertical air channels.
Somebody (I think it was Bruce King, from California, but I don't 
remember for certain) has suggested that all gaps between bales should 
be filled with straw-clay mix rather than loose straw.  The reason given 
was to reduce fire risk, but it could also help to cut down the 
possibility of convection.
As Dirk has hinted, though, bales laid flat will behave differently from 
bales on edge in this regard.  Convective flow across the bale wall will 
be much lower in the case of bales laid flat (at a reasonable bale 
density, anyway).
Mark

Jan Hollan wrote:

>>I remembered an email, I got years ago and found it in the forum. Would that
>>be a possibility to solve the problem of air circulation (as far as it
>>is...), to pack strawbales in wind-proof paper?:
>>    
>>
>
>In our two insulating envelopes in Hostetin, we separated bales by paper
>at relevant surfaces.
>
>However, it's possible that convection cells are just of the size of about
>one bale, so separating the bales may not help much and surely cannot
>solve the problem completely.
>
>The improvement achieved by separating the bales is also something what
>should be investigated.
>
>jenik
>
>
>____________________________________________________
>    European strawbale building discussion list
>
>Send all messages to:
>Strawbale at amper....muni.cz
>
>Archives, subscription options, etc:
>http://amper.ped.muni.cz/mailman/listinfo/strawbale
>____________________________________________________
>
>
>
>
>  
>