[GSBN] using loose straw insulation in roof

Sat Oct 28 00:41:49 UTC 2017

Hi all,

Good discussion and advice on this.

The PAKSBAB (Pakistan) SB buildings use 12² (30 cm) of clay-coated straw for
insulation above the ceiling (photos and drawing attached) (second photo and
drawing courtesy of Darcey Donovan Messner). Andy Mueller and I used a
lesser depth of the same ³loose straw-clay² for our SB building in Haiti
(photos attached).

Both make a noticeable difference in the building's thermal performance in
their respective climates. In both cultures thermal insulation of any kind
is rarely used. But as happy as I am to use this inexpensive natural
insulation in these and similar contexts, I worry about the flammability of
the ³loose straw clay². Even if the material resists ignition in the first
months or year after installation, I¹ve seen evidence over time of clay
³dusting², thus diminishing whatever protection it offered the straw
initially.

I remember reading (in a TLS article?) Kelly Lerner describe a straw
insulation fire that started (electrical source, as I remember) in the attic
of a SB house in China. It was not part of the prescribed design, but I
think it was reasoned that if it¹s good in the walls, why not the attic? I
don¹t remember if it was loose or baled straw, but either way it is an
indicator of the fire risk of straw as ceiling, attic or roof insulation.
This event gave SB a bad name in the region that took time to overcome.

That said, there are many factors at play. I can imagine reasonably
fire-safe ways to use straw (with or without clay) as ceiling, attic or roof
insulation. (Leaving the structural issues out of it.)

I found the following statement in TLS17 (1997):  "Robert Laporte, timber
framer and straw-clay builder, commonly uses straw-clay stuffed loose
between rafters as insulation, with the clay discouraging pests."  I don¹t
know if Robert still practices this 20 years later. His partner Paula
Baker-Laporte who is on this list, might be able to say more.

Also, I¹ve attached an article from TLS38 (2002) ³Bales in My Roof . . . A
Costly Decision?² that speaks to some of the identified issues.

Lastly, to the extent that it's relevant, and as far as I know, Rikki is
still working is Spain (and maybe neighboring countries).

Martin Hammer

From:  Gsbn <gsbn-bounces at sustainablesources.com> on behalf of John
Swearingen <jswearingen at skillful-means.com>
Reply-To:  GSBN <GSBN at SustainableSources.com>
Date:  Friday, October 27, 2017 at 6:57 AM
To:  GSBN <GSBN at SustainableSources.com>
Subject:  Re: [GSBN] using loose straw insulation in roof

Quite a long time ago we built a house using bales for roof insulation, with
site built trusses.The biggest consideration about the insulation value came
with the numerous large gaps between bales and trusses that needed to be
filled. The building was near a major farmland highway with large trucks
passing a few yards away, and the bales did an excellent job of dampening
the low frequency noise of the trucks. I would consider them again is a
similar situation, but not for insulation; the cost and difficulty of doing
a good job is too great, and other good alternatives are available.

This year we have again been up close and personal with fire, this time
dealing with melting levels of heat, generating tornado winds in
concentrations of timber buildings feeding off of each other. We have four
projects that came within yards of the worst fires, but none of them were
touched by flames, so we don't have any forensic information to offer.  The
closest endorsement we've had came from firefighters who came to defend our
house and we showed them our strawbale cabin with a grassy green roof.  They
immediately freaked, then looked, pondered, said "no problem" and moved on
to check out the wood structures.  Nothing like thick plaster and few inches
of wet dirt to slow down a hot fire.

Since we're designing a home in the same fire-prone region of Napa, we've
been looking at the weak links that would make a good plastered house
vulnerable. In an intense fire, assuming the plaster and roof are good, the
weak link is heat transfer (and breakage) through windows igniting the
interior, and we're thinking of mounding sprinklers on the exterior within
the window cavity, primarily for cooling. We've done this in urban
situations when a building has been inside the required setback in lieu of a
one-hour fire wall This would be a slightly different use from preventing
the spread of fire--primarily to cool the glazing area long enough for a hot
wildfire to pass. When firemen show up before and advancing fire, they do a
lot of wetting. We've installed large garden sprinklers on our roofs in the
forest, and the firemen are big fans of those. Water is their only friend.

Any thoughts on this?

John "I'm All Wet" Swearingen

On Thu, Oct 26, 2017 at 7:13 PM David Arkin, AIA <david at arkintilt.com>
wrote:
> Hi World:
> 
> Having just taught a few workshops where we measured the density of bales and
> discussed the possible reasons for a minimum density, this is a timely
> question.  We actually had some bales that didn¹t pass the minimum 6.5
> lbs./cu. ft. specified in the International Residential Code (IRC).  Reasons
> for a minimum density include structural integrity, insulation properties, and
> fire safety.  The use of loose fill that doesn¹t achieve a minimum density is
> not supported in this code, but that doesn¹t rule out the possibility that
> some version might be developed that could achieve similar properties to loose
> fill cellulose and other materials.
> 
> For insulation, the IRC and also the IBC specify a maximum flame spread index
> of 25 and a maximum smoke developed index of 450.  An ASTM E84-98 Standard
> Test Method for Surface Burning Characteristics test of a Straw Bale in May of
> 2000 yielded a flame spread index of 10 and a smoke developed index of 350,
> thus verifying straw bale as a viable insulation material.  Note that this was
> a test of surface burning of a bale, not loose straw.
> 
> Curiously, the IBC notes that Cellulose loose-fill insulation only needs to
> meet the smoke developed index not exceeding 450 criteria, provided it
> complies with CPSC 16 CFR Part 1200 and CPSC 16 CFR Part 1404, and further
> stating that Œeach package of such insulation material shall be clearly
> labeled in compliance with' [these].
> 
> I¹ve heard some folks in the UK have been experimenting with a chopped straw
> insulation infill, but the exact parameters of the processing and the
> resulting fire resistance I don¹t have information on.
> 
> That all said, I¹ve never done it (put straw in a roof / ceiling), nor do I
> recommend it, but neither can I say with certainty that it shouldn¹t be done.
> 
> David A.
> 
>> On Oct 26, 2017, at 4:57 PM, Bill Christensen <lists at sustainablesources.com>
>> wrote:
>> 
>> In the ASTM tests that Bruce King coordinated back in 2007, one of the walls
>> was clay plastered.
>> 
>> Yes, the clay cracked and the straw eventually burned (though the clay
>> plastered wall did better than cement, we believe because it spent the first
>> 20 min or so "firing" as in a kiln, and the chemical changes from that
>> absorbed a lot of the heat).
>> 
>> Loosely packed straw might not fare so well once oxygen can get to it.
>> 
>> We had a problem because the clay plastered wall had a crack on the exterior,
>> which allowed more oxygen in between a couple bales.  Though we had stuffed
>> and cobbed those spaces maniacally, the fire followed the path back through
>> the wall.  We had to cut the test off at 1 hour because of this - otherwise
>> the clay plastered wall probably would have out-performed the cement
>> plastered one. 
>> 
>> The guys at the testing facility told us that "Everything burns...
>> eventually.   Everything".
>> 
>> I have to agree with the earlier comments in this thread - loose straw as
>> ceiling insulation is just asking for problems, whether hit with clay slip or
>> not.  
>> 
>> On 10/25/17 4:56 PM, Bohdan Dorniak wrote:
>>> Hi All
>>> Graeme  I¹d be worried about your statement about a ³coating of clay²  in
>>> our fire test of rendered strawbales by CSIRO - done in 2002  the earth
>>> rendered bale (render about 50mm thick  Frank Thomas did the rendering)
>>> started to show cracking in the render.
>>> Wondering whether eventually they would burn?
>>> Other considerations is the weight factor and stronger ceiling structures
>>> (if using thicker coats of render).
>>> That¹s my 2 bob¹s worth.
>>> Bohdan
>>>  
>>> From: Gsbn [mailto:gsbn-bounces at sustainablesources.com] On Behalf Of Misha
>>> Rauchwerger
>>> Sent: Thursday, 26 October 2017 8:04 AM
>>> To: Global Straw Building Network
>>> Subject: Re: [GSBN] using loose straw insulation in roof
>>>  
>>> I believe another issue is whether the attic space is vented or not, as well
>>> as the kind of roofing.  After the Oakland fires, I remember this being
>>> debated furiously, because the codes require attic venting, and going with
>>> something like polyisocyanurate rigid insulation without vents was
>>> controversial.  Obviously the fire danger will increase if air can mix with
>>> flammable insulation fuel (frieze block vents and ridge vents create a
>>> perfect means for fully combusting the attic materials) . If it is encased
>>> and covered with earth on the outside and inside (plaster on the ceiling),
>>> as in a living roof, we can decrease the flammability factor.
>>>  
>>> The Lobo fire that swept through our neighborhood recently came right to our
>>> friend's straw bale house with only superficial plaster damage on one
>>> corner.  Of course there are far too many factors to say it was saved
>>> because of being straw bale or not.
>>>  
>>> Then again, if you get a fire like just swept through Napa and Sonoma
>>> Counties, all bets are off...
>>>  
>>> Misha 
>>>  
>>> On Wed, Oct 25, 2017 at 1:38 PM, Graeme North <graeme at ecodesign.co.nz>
>>> wrote:
>>> This list is a fantastic resource for learned info and is greatly
>>> appreciated I can tell you.
>>>  
>>> The fire thing is scary - there is no requirement in NZ for single dwellings
>>> to have fire resistant ceiling materials, but we do need to have smoke
>>> alarms near bedrooms and escape routes.
>>> The really scary thing is that the fire people here advise that a dry house
>>> with an open attic with light timber rafters or trusses for roof framing
>>> will burn from end to end in 60 secs.
>>> Goodness knows what adding an accelerant like straw (without a good clay
>>> coating) might do. I would advise against it.
>>>  
>>> I would use straw coated with clay  - pretty much a LEM mix - if my
>>> structure allowed it to happen easily and I was considering it for my own
>>> house but the work required and extra weight decided against it
>>> and again recently I considered it on another design but the considerably
>>> increased extra depth of the roof structure to accommodate it, along with
>>> the extra cost, work, and weight, decided against it again so I went with
>>> wool insulation in both cases.
>>>  
>>> I am also not convinced that cellulose/paper insulation with borax retains
>>> it fire rating over time - I have seen examples of old (only a few years)
>>> cellulose insulation that would not ignite when new,  ignite readily with a
>>> match and it seems to keep burning quite happily.
>>>  
>>>  
>>> Cheers
>>>  
>>> Graeme
>>>      
>>>  
>>>  
>>>  
>>>  
>>>  
>>>  
>>>> On 26/10/2017, at 5:36 AM, Bruce EBNet <bruce at ecobuildnetwork.org> wrote:
>>>>  
>>>>  
>>>> Don¹t you love it when someone else chimes into these discussions ahead of
>>>> you, and says everything you wanted to say?
>>>>  
>>>> I sure do.  Thank you Derek and John;  what they said.
>>>>  
>>>> Bruce King
>>>> (415) 987-7271 <tel:%28415%29%20987-7271>
>>>> bruce at ecobuildnetwork.org
>>>>  
>>>>  
>>>>  
>>>>  
>>>>> On Oct 25, 2017, at 7:56 AM, John Straube <jfstraube at uwaterloo.ca> wrote:
>>>>>  
>>>>> I would echo Derek¹s concerns exactly. Loose fluffy straw burns very
>>>>> quickly and you may as well say you lost the house.
>>>>> Adding clay slips will increase the fire resistance to the point is
>>>>> acceptable, as will dense bales with some sort of clay slip top, but once
>>>>> you do that, you have a heavy and labor intensive roof insulation.
>>>>> Cellulose with 20%+ borate treatment is inexpensive, gives good R-value,
>>>>> is widely available, and is very good at fire resistance.
>>>>> 
>>>>> 
>>>>> On Oct 25, 2017, at 10:00 AM, Derek Roff <derek at unm.edu> wrote:
>>>>> 
>>>>> I have a different concern about using straw packed at low-densities in
>>>>> the roof.  I think that the fire risk increase is much more of a problem
>>>>> than the decrease in insulation value.  If you have seen flakes of straw
>>>>> or loose straw burn, you will be aware that they are much more flammable
>>>>> than standard bales.  Straw flakes are probably about half the density of
>>>>> a building bale, and loose straw might cut the density in half again.
>>>>> 
>>>>> Losing the roof in a fire usually means losing the whole house¹s integrity
>>>>> and value.  While enclosing the flakes or loose straw for the roof in
>>>>> plywood, for example, would help somewhat with fire resistance, I
>>>>> recommend against design choices that depend on a few things going right
>>>>> to avoid a catastrophic failure in a fire.
>>>>> 
>>>>> An additional consideration that has been mentioned here before, is that
>>>>> while bales may be relatively inexpensive, placing them in the roof
>>>>> requires increasing the size and number of the roof¹s structural elements,
>>>>> which will likely raise costs more than the amount saved by using bales
>>>>> instead of other insulation materials in the roof.  Using lower density
>>>>> straw diminishes that problem to some extent, but straw is still likely to
>>>>> be significantly heavier than other kinds of roof insulation, for an
>>>>> equivalent insulation value.
>>>>> 
>>>>> Derek
>>>>> 
>>>>> Derek Roff
>>>>> derek at unm.edu
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> On Oct 25, 2017, at 4:34 AM, Rikki Nitzkin <rikkinitzkin at gmail.com> wrote:
>>>>> 
>>>>> Hi!
>>>>> 
>>>>> I don¹t know if you all remember, but not long ago I asked about the
>>>>> MAXIMUM density of a SB before it begins loosing thermal propertiesŠ most
>>>>> people agreed that we should not worry about a bale being too dense.
>>>>> 
>>>>> Now I ask about the opposite question: is there a minimum density?
>>>>> 
>>>>> The other day we were talking about using flakes of straw to insulate a
>>>>> roof. One of the builders insisted that is was important that the
>>>>> insulation cavity was filled with highly compressed straw, and another
>>>>> said that as long as the cavity was properly filled (leaving no big holes
>>>>> for air to circulate) that the density of the infill is not important, as
>>>>> the straw (loose or dense, but enclosed in the insulation cavity) impedes
>>>>> the circulation of air, and that is what insulates.
>>>>> 
>>>>> Can any of you technicians refer me to studies about insulation properties
>>>>> and how they relate to density? or at least clarify my doubt: Is it
>>>>> important to highly compact the insulation in the roof? and whyŠ so I can
>>>>> explain it better.
>>>>> 
>>>>> thanks!
>>>>> Rikki
>>>>> 
>>>>> ______
>> 
>> -- 
>> Bill Christensen
>> http://SustainableSources.com <http://sustainablesources.com/>
>> http://LinkedIn.com/in/billc108
>> _______________________________________________
>> Gsbn mailing list
>> Gsbn at sustainablesources.com
>> http://sustainablesources.com/mailman/listinfo.cgi/gsbn
> 
> 
> 
> *  *  *  *  *
> Arkin Tilt Architects
> Ecological Planning & Design
> 1101 8th St. #180, Berkeley, CA  94710
> 510/528-9830 ext. 2# <tel:(510)%20528-9830>
> www.arkintilt.com <http://www.arkintilt.com>
> 
> David Arkin, AIA, Architect
> LEED Accredited Professional
> CA #C22459/NV #5030
> 
> Director, California Straw Building Association
> www.strawbuilding.org <http://www.strawbuilding.org>
> CASBA is a project of the Tides Center
> 
> "There is no way to peace. Peace is the way."
> ‹ A. J. Muste 
> 
> 
> 
> 
> 
> 
> _______________________________________________
> Gsbn mailing list
> Gsbn at sustainablesources.com
> http://sustainablesources.com/mailman/listinfo.cgi/gsbn
_______________________________________________ Gsbn mailing list
Gsbn at sustainablesources.com
http://sustainablesources.com/mailman/listinfo.cgi/gsbn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sustainablesources.com/pipermail/gsbn/attachments/20171027/d051b408/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ClayCoatedStraw_AtticInsulation_Pakistan.jpg
Type: image/jpeg
Size: 380120 bytes
Desc: not available
URL: <http://lists.sustainablesources.com/pipermail/gsbn/attachments/20171027/d051b408/attachment.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ReedMatClg_Pakistan.jpg
Type: image/jpeg
Size: 333649 bytes
Desc: not available
URL: <http://lists.sustainablesources.com/pipermail/gsbn/attachments/20171027/d051b408/attachment-0001.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: HouseF1_Section_11_17_08.pdf
Type: application/pdf
Size: 118544 bytes
Desc: not available
URL: <http://lists.sustainablesources.com/pipermail/gsbn/attachments/20171027/d051b408/attachment.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ClayCoatedStraw_AtticInsulation_Haiti.jpg
Type: image/jpeg
Size: 410555 bytes
Desc: not available
URL: <http://lists.sustainablesources.com/pipermail/gsbn/attachments/20171027/d051b408/attachment-0002.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: BambooClg_Haiti.jpg
Type: image/jpeg
Size: 503333 bytes
Desc: not available
URL: <http://lists.sustainablesources.com/pipermail/gsbn/attachments/20171027/d051b408/attachment-0003.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TLS38_p20-21_2002.pdf
Type: application/pdf
Size: 469260 bytes
Desc: not available
URL: <http://lists.sustainablesources.com/pipermail/gsbn/attachments/20171027/d051b408/attachment-0001.pdf>