[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GSBN:fire test -- your opinions?

Hello all,

I've been reading with great interest all these comments and ideas and insights. Thanks to all for contributing them and this message is not intended to halt to this process. I just have some contextual comments and some thoughts of my own about this and about the thinking that Bruce and I have done in regard to what we're trying to achieve with minimal resources given the full range of tests we would love to be able to do.

First, I think it is important to clarify a few things about the test. Overall, it would be nice if we could do one test, see how it goes, and then decide on how we would build the second wall and then test it. But because of related expenses and timing, including the desire to have these results end up in Bruce's new book this fall, coupled with the amount of time required for these test walls to cure and dry before the test (a month) and our desire to have adequate curing and time between coats of plaster, we need to figure out what we're going to do and build both test walls at the same time and test them on consecutive days in July.

The tests we are planning to do are according to the ASTM E-119 Full Scale Fire Test standard and we plan to seek a two hour fire rating for the assemblies we test (for those who might want to know we're using the newest version - E119-05a). This test includes a hose stream test at the end of the specified duration of fire testing. We could test longer though we would have to pay more and the longer the test the longer and stronger the hose stream test. We have chosen to conduct the test for non-loadbearing specimens for a variety of reasons including cost, simplicity and timing. We recognize that we are still going to want to have the wall under some compression because if makes for a better wall and so we will be planning to work this out in the design of the walls. The frames in which we are building these panels in are, as we understand it, roughly 11' x 11' and about 16" deep so the unheated side of the wall will overhang the frame and need to be sealed up when we plaster it. We plan to have the bales placed in the frames such that the plaster finishes flush with the surface of the frame that will be placed against the test furnace. Thus the plaster will be supported at the bottom by the test frame and not merely hanging on the bales.

The criteria for a successful test result is that the wall has withstood the fire endurance test without passage of flame or gases hot enough to ignite cotton waste (on the unheated side), for a time equal to that for which classification is desired. The transmission of heat through the wall during the fire endurance test shall not have been such as to raise the temperature on its unexposed surface more than 250°F (139°C) above its initial temperature. Also, that the wall has withstood the fire and hose stream test without passage of water from the hose stream. The assembly shall be considered to have failed the hose stream test if an opening develops that permits the projection of water from the stream beyond the unexposed surface during the time of the hose stream test. The hose stream test for a rating 2 hours and over and less than 4 hours requires a water pressure at the base of the nozzle of 30 psi (207 kPa) for 2-1/2 minutes.

The walls need to have the same finish on both sides of the wall or we would be required to test both sides for a succesful test of the whole assembly. Although it isn't clearly delineated in the standard, because the requirement for a full scale test is a minimum area of 10ft x 10 ft, I believe that we do not have the option of creating a panel with different finishes on different parts of the test wall and conforming to the requirements for a recognized test result. On the other hand, it might be worthwhile to think about doing this, even though it would be spending a chunk of money that would advance our knowledge but not create a strictly usable test result. We could choose to build a wall that had, say even four different combinations of plaster and orientation and test it if we could agree on what the best single configuration would be for the other test wall - something that would give us a successful two hour test we can rely on that is the most useful to those who need this test result.

The issue comes down to choices made in the realm of risk and uncertainty. As Bruce put it, the uncertainty regarding whether the specific wall assemblies we choose will pass both the fire and hose stream tests may be small but the risk if they do fail is high - the cost of the tests and the lack of a usable test result. So, for instance, if we were to test earthen plaster on one of these panels and it didn't pass the hose stream test we would not have a successful test for earthen plastered bale walls even if it survived the two hour fire endurance part of the test.

It's important to consider that what we test will likely be what is required when someone uses these test results to obtain code approval for a building. We would like the results to be as unrestricted as possible and if we were certain that earth plastered walls would pass we'd be more inclined to be doing them now. We're fairly certain that a lime-cement plaster with no mesh will get us that successful test in either bale orientation. As Bruce and I talked about this we found it hard to imagine that even if the plaster on the heated side of the wall failed at the onset of the hose stream test, because of the limited amount of charring seen in previous tests, there will almost certainly be some amount of uncharred straw and the other layer of plaster on the unheated side of the wall still in place to impede the hose stream from projecting beyond the unheated surface.

As for the orientation questions, I think its important to be precise in our thinking here. It was suggested that the test results related to insulation and orientation suggest that it doesn't make any difference which way the bales are placed for this test. But in fact, the r-value is not the same per inch - on-edge provides a higher insulating value per inch, which is why it makes little difference overall whether bales are flat or on-edge. So, when we think about fire resistance in these two orientations, the behavior may or may not be the same. We actually don't know. And the bigger issue is that this is the sort of thing that code officials are likely to cite if we test one way and not the other.

This is where the idea of doing one test panel that is our "sure thing" and one that is a combination of finishes and orientations may make some sense. Test with cement-lime plaster with bales on-edge let's say as the sure thing, and then do a combination wall panel that incorporates bales flat with lime plaster, bales flat with earthen plaster, bales on-edge with lime and bales on-edge with earthen plaster. We end up with our successful two hour fire test and we get a heck of a lot of information that we can point to and say here is a test that was done at the same time in the same lab and (in the best case scenario) these all also passed the fire and hose stream tests. If we had good documentation, which we hope to have, this could work to our advantage. I do need to say that I have specific concerns about lime plaster because of how much time it takes to cure and harden raising questions about whether it would be able to pass the hose stream test.

Perhaps I've opened a can of worms here and it would be much simpler to just do fairly conservative tests that will give us the most useful information and results. But I do agree with Bruce's baseline assumption that the places that are most likely to require the test results are also more likely to require a lime-cement kind of finish. I also want to just mention that the results of John Straube's moisture testing with regard to the dramatic increase in water vapor permeability of cement stucco with the addition of more lime relieves some of my concern about the use of stucco on bales. It is worth downloading and reading this from Bruce's website www.ecobuildnetwork.org.

Got to run, but I wanted to make sure we have as much as possible of a whole and realistic picture as we discuss the various possibilities.

David Eisenberg

-----Original Message-----
From: John Swearingen jswearingen@...
To: 'GSBN' GSBN@...
Sent: Wed, 26 Apr 2006 11:32:06 -0700
Subject: RE: GSBN:fire test -- your opinions?

I agree with Bill that I wouldn't expect much performance difference between flat and edgy bales as far as heat transmission goes--wouldn't that closely
correspond to the R-value, which is similar.

So the only reason for testing bales on edge would be to see if the strings would pop, and we know that with some certainty. What we don't know is what will happen after they pop, in a situation where the plaster is compromised.
Once a bale even begins to disgorge, the conflagration spreads like
wildflowers in a prairie. In an assembly using mesh, disgorgement would be prevented as long as the mesh holds. (Obviously metal mesh has an advantage
over plastic or natural fiber here).

Marty mentioned that in the dark of the night he's worried that a plaster
failure would allow disgorgement and this could lead to a failure of the
entire assembly which would result in frowns on the faces of inspectors and
could ruin the value of the test for just about everybody.   He might be
right, and so I'd suggest doing a test with some kind of standard metal
meshâ?¦or doing the first test w/o mesh, and if the scenario above occurs, use
mesh for the second test.

Of all the performance variables that could be tested--such as plaster,
mesh, orientation--the one that I would predict would give the greatest
difference, potentially, would be mesh, followed by earth plaster.

John "My Strings are Melting" Swearingen

John Swearingen
design and construction
HYPERLINK "www.skillful-means.com"www.skillful-means.com

-----Original Message-----
"<a target="_blank" href="mailto:GSBN@..."mailto:GSBN@...";>mailto:GSBN@..."mailto:GSBN@...] On
Behalf Of billc_lists@...
Sent: Wednesday, April 26, 2006 12:57 AM
Subject: Re: GSBN:fire test -- your opinions?

Hi all,

Interesting points re: edge vs flat, cement vs cement-lime vs lime vs
earthen plasters.

I think it's clear that order to really get the full data, more than two
tests will need to be done. Typically in a scientific procedure you'd change just one variable at a time, for instance bales on edge vs bales flat, with
all other variables (plaster, etc) remaining the same. Then you'd vary
another constraint, such as the plaster types, so that if you'd already done
edge vs flat with one type of render you'd be able to predict how the
performance would vary (or not) using different plasters as a result of bale
orientation.  Otherwise you have trouble saying exactly what caused any
difference you may have observed.

I personally don't think that we'll see a huge difference between edge vs flat, though I can certainly see the reasoning behind doing such tests to be
sure.  I don't think molten strings are going to make much difference,
especially in a test wall which is built within a four sided frame - it's not likely that the straw is going to go anywhere. The question is more a matter of how heat is transferred through the bale - do the hollow straws allow it to move faster through flat bales? Do vertical straws cause heat
to move to the top more quickly, causing failure there?

Regarding plasters, curiosity got the best of me and I pulled some numbers from the SB Registry to see just what people are using on their walls. For
exterior walls, they can choose from:

cement plaster
earth or clay plaster
lime plaster
wood siding over plastered bales
wood siding over unplastered bales
other - Describe if other

or any combination of the above.  Of those who reported their exterior
finish, the resulting numbers are:


Cement only: 235

Earth only: 72

Earth/lime: 34

Lime only: 51

with a scattering of other strange combinations (perhaps different
treatments for different walls).  For interior walls, they can choose

cement plaster
earth or clay plaster
lime plaster
gypsum plaster
other - Describe if other

or any combination.  The resulting numbers are:


Cement only: 116

Cement plus other: 39 (includes cement &amp; gypsum, cement &amp; sheetrock, cement
&amp; wood, etc)

Earth only: 94

Earth &amp; Gypsum: 13

Earth &amp; lime: 13

earth &amp; other: 15

and again, a few who did other, more bizarre combinations.

So though these numbers are by no means definitive, it looks to me that
cement beats earth for exteriors in general by about 2 to 1, but they're
pretty close to an even match for interiors.

Which then begs the question:  Are we testing for fire resistance of
interior walls or exterior, or both?

On the flat vs edge orientation, I see 316 flat vs 84 on edge.

I don't know if those numbers will help anyone make a decision, but I
figured I'd throw them out there.  There's more reported use of earth
plaster than I expected there to be.

Bill Christensen
"<a  target="_blank" href="http://sustainablesources.com/contact/"http://sustainablesources.com/contact/";>http://sustainablesources.com/contact/"http://sustainablesources.com/contact/</a>>

Green Building Professionals Directory: <HYPERLINK
"<a  target="_blank" href="http://directory.sustainablesources.com"http://directory.sustainablesources.com";>http://directory.sustainablesources.com"http://directory.sustainablesources.com</a>>
Sustainable Building Calendar: <HYPERLINK
"<a  target="_blank" href="http://SustainableSources.com/calendar/"http://SustainableSources.com/calend";>http://SustainableSources.com/calendar/"http://SustainableSources.com/calend</a>

Green Real Estate: <HYPERLINK
"<a  target="_blank" href="http://SustainableSources.com/realestate/"http://SustainableSources.com/real";>http://SustainableSources.com/realestate/"http://SustainableSources.com/real</a>
Straw Bale Registry: <HYPERLINK
"<a  target="_blank" href="http://sbregistry.sustainablesources.com/"http://sbregistry.sustainablesources.com/";>http://sbregistry.sustainablesources.com/"http://sbregistry.sustainablesources.com/</a>>

Books/videos/software: <HYPERLINK
"<a  target="_blank" href="http://bookstore.sustainablesources.com/"http://bookstore.sustainablesources.com/";>http://bookstore.sustainablesources.com/"http://bookstore.sustainablesources.com/</a>>
GSBN is an invitation-only forum of key individuals and representatives of regional straw construction organizations. The costs of operating this list are underwritten by The Last Straw Journal in exchange for use of the GSBN
as an advisory board and technical editing arm.

For instructions on joining, leaving, or otherwise using the GSBN list, send
email to GSBN@...HELP in the SUBJECT line.

No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.1.385 / Virus Database: 268.4.6/323 - Release Date: 4/24/2006

No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.385 / Virus Database: 268.4.6/324 - Release Date: 4/25/2006

--- StripMime Report -- processed MIME parts ---
 text/plain (text body -- kept)
GSBN is an invitation-only forum of key individuals and representatives of regional straw construction organizations. The costs of operating this list are underwritten by The Last Straw Journal in exchange for use of the GSBN as an
advisory board and technical editing arm.

For instructions on joining, leaving, or otherwise using the GSBN list, send
email to GSBN@...HELP in the SUBJECT line.

GSBN is an invitation-only forum of key individuals and representatives of regional straw construction organizations. The costs of operating this list are underwritten by The Last Straw Journal in exchange for use of the GSBN as an advisory board and technical editing arm.

For instructions on joining, leaving, or otherwise using the GSBN list, send email to GSBN@...HELP in the SUBJECT line.