Return to index: [Subject] [Thread] [Date] [Author]

Allowable Stresses In Wood

[Subject Prev][Subject Next][Thread Prev][Thread Next]
In addition to the references i posted last month on this issue, you might take
a look at the Wood Handbook published by the Forest Products Lab. Chapter 6
deals with lumber stress grades and derivation of design properties. It gives a
good overview of the development of "small-clear" design values and "in-grade."
It also provides some additional references for further study. Here's a link to
the pdf file on the FPL website for Chapter 6:

I researched your question about whether the size effect is different for new
growth versus old growth, and didn't come up with anything definitive. However,
note that a size effect in some form has been around for wood members since the
earliest NDS. The NDS Commentary section 4.3.2 provides some background on the
development of size factors for structural lumber.

Hope this information helps.

Buddy Showalter, P.E.

From: Michael Bryson <MBryson(--nospam--at)>
To: "'seaint(--nospam--at)'" <seaint(--nospam--at)>
Subject: RE: Allowable Stresses In Wood

I read through last month's archives, and there did not appear to be any
clear consensus on the issue. If anything the balance was in favour of size
effect being the cause of the reduced values.
However, reading through the NDS91 commentary, they explicitly say that the
extrapolation of test results from small to big was considered appropriate:
they did do tests on full-size structural members to confirm this. They also
extrapolated test results for different species and grades.
The testing they did in the late 70s - early 80s not only tested full-size
members but also the different species and grades. They did discover a
greater size effect on bending strengths for the lower grades. But could
this be a size effect of the new growth wood that was tested only? i.e. you
would not get the same size effect for the old growth wood. 

From: Frank Glazewski <frankgla(--nospam--at)>
Subject: Re: Allowable Stresses In Wood
To: seaint(--nospam--at)


My understanding is that the change is due to the fact that the older
design values were extrapolated from testing of smaller specimens (2x2,
etc.) These specimens were relatively defect free and the original
testing results perhaps did not accurately reflect performance of a
normal graded member (with normal knots, checks etc associated with the
grade) whereas the 'in-grade' testing program does. It sounds reasonable
that the decline in timber quality may also have something to do with
this change, but I'm not positive of that. (Try buying a 2 x 6 today
that doesn't have exhibit wane on diametrically opposing corners....)

For existing buildings, it seems reasonable that you could check the
framing members using the code that was in effect at the time of
construction. (Unless you are adding new loads to the structure, then I
would use the current code). Whenever possible, I always like to discuss
design decisions such as this with a knowledgeable building official
prior to doing the work.


Michael Bryson wrote:

>  I wonder if someone has some knowledge about the change in allowable
> bending stresses in wood?The fb values seem to have come down abruptly
> with the UBC94 code presumably because of large scale testing done.
> For example 2x12 Doug. Fir #2 went from 1250psi to 875psi. Now can
> this change be attributed to the size effect factor because the new
> tests were done on large scale specimens? Or because of new growth
> material i.e. all the older & stronger wood had been used up and we
> are now using newer & poorer second growth wood. This is relevant
> because when assessing existing buildings which fb do you use? Old or
> new values?

From: Mlcse(--nospam--at)
Subject: Re: Allowable Stresses In Wood
To: seaint(--nospam--at)

I would recommend using the current wood allowable design values for 
evaluating the existing wood structures.  I imagine you wouldn't be checking 
it, unless you were adding addtional load. 

My understanding is that there have been changes in testing methods over the 
years besides the "in-grade" testing and size effects done in the 1980's as 
well.  One particular area is "Tension Parallel To Grain" values which were 
decreased around 45% between the 1967 and 1970 UBC's.  Previously "Tension 
Parallel to Grain" design values were the same as for "Bending".  The change 
I believe was a result of the original testing program to determine allowable 
design values for "Machine Stress Rated Lumber" in the Late 1960's which 
indicated a problem in the test method for determing "Tension Parallel to 
Grain", thus the revised design values in the 1970 UBC.  Mr Showalter  with 
AFPA may be able to clarify this some more.

Michael Cochran S.E.

To: seaint(--nospam--at), seaint(--nospam--at)
From: Neil Moore <nmoore(--nospam--at)>
Subject: Re: Allowable Stresses In Wood

Also, don't use the older code with the new values.  Use the current code with
all it's adjustment factors and different formulas.  It is my understanding
that using the present allowable stress values and the current code, that
is very little difference in the final size of your member.

Neil Moore, S.E.
neil moore and associates
shingle springs, california

From: "vicpeng" <vicpeng(--nospam--at)>
To: <seaint(--nospam--at)>
Subject: Re: Allowable Stresses In Wood


Last time I had to assess existing timbers, in our case for assessing =
the structure to current code levels of snow etc, here in BC, I =
contacted the Canadian Wood Council and was informed that we should use =
current tabulated values because of the "better" testing methods.  =
Previous posts have identified why some of the tests have changed.

The alternative was to test the timbers in-situ for "true" values.  That =
had its own problems!

Thor A Tandy P.Eng
Victoria BC
e-mail: vicpeng(--nospam--at)

*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*   Questions to seaint-ad(--nospam--at) Remember, any email you 
*   send to the list is public domain and may be re-posted 
*   without your permission. Make sure you visit our web 
*   site at: