Old Timber Design Values 2

[EngineeringEric]

Hello, I feel like I recently came across a post regarding this but i am unable to locate it.

I have a building that i am working on which was built in the early 19th century and this isn't uncommon from one my firm's large clients. We often analyze existing structures for current loads, residential. Often the structures are abnormal rough hand scrapped lumber. sometimes great quality workmanship and sometimes not.

I was wondering if anybody has any information about the first or early timber property design manuals? old design values are really what i am looking for. I would feel confident with our analysis if I knew how old timber compared to modern. Most of what i see is mixed-red oak and sometimes mixed pine, poplar and cedar finishes.

I know deflection is still an issue but with these structures deflection isn't an issue, it is nearly always bending stress or shear governed. The client doesn't care about occupant comfort (bouncy) or finishes as much as keeping it all original, it is my job to see if original is sufficient for loads.

 

[BAretired]

Early in the 19th century? Do you mean early in the 20th century?

BA

 

[wannabeSE]

I think the Carnegie Pocket Companions on Sliderule Era's website have some wood properties.

http://www.slideruleera.net/miscellaneous.html

 

[JAE]

Three quick thoughts -
1. My older mentors (started their careers in the 1927-1947 range) would almost always use Fb' = 1000 psi. Of course they didn't call it F'b but simply the maximum stress that they'd let wood experience (service level loads).

2. If you need to use something to correlate your engineering judgement here ... i.e. what another reasonable engineer would do ... and need to tag to something published, the only thing I can see you could do under the building code would be to do a load test on the floors. This is very expensive (for a residential use especially) but that would at least give you something to depend on as a rational basis for telling a homeowner of a 1800's era house it is "safe".

3. Take a wood sample and test it for flexure and shear. I'm not sure of any ASTM tests for this - and when I ask testing labs about testing samples of wood I've gotten nowhere. There are labs that can take samples and tell you the species...but I'd be afraid that older wood might be more dried out or rotted and a direct species-to-NDS table look up would be meaningless.



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[JAE]

I guess my last item - the fact that older wood might not still carry the same physical properties as it did when first used in the structure, would negate any use of older design manuals or guides.



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[TomDOT]

JAE, I would go further and posit it extremely likely the wood has significant differences in physical properties today compared to installation over 100 years ago. Absent rot and other damage, I would generally expect significantly reduced flexibility and increased hardness. I've heard numerous stories of being unable to drive a nail into old wood (absent predrilling) due to the hardness.

For testing wood properties, I suppose one should look for a lab which performs ASTM D4761

http://www.astm.org/Standards/D4761.htm

A quick google came up with these:

http://www.wpcinfo.org/codes/documents/WMEL_IAS_Scope.pdf

http://www.appliedtesting.com/tests/mechanical-properties-of-lumber

 

[CANPRO]

Eric, see attached. Hopefully of some use.

 

[jayrod12]

The earliest stuff I could find in my office was a book titled "wood structural design data" published by National Lumber Association 2nd edition from 1939. For Dense D.Fir structural grade they had a working stress of like 1700 psi. For select structural it was in the range of 2100psi. With a modulus of 1e6.

 

[JAE]

TomDOT - thanks for that reference. That is probably the one to use for a testing lab and samples taken from a site.
The next challenge would be to assert a safety factor as the test, I'm sure, only provides an ultimate failure load, not an allowable.
I've always had a hard time finding out what NDS uses for wood safety factors. I guess we could go LRFD in the wood design but even then, you
need to also determine a 90% confidence line for values from the tests.



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[woodman88]

Attached is some lumber values from 1922 from the NDS web site.

Garth Dreger PE - AZ Phoenix area
As EOR's we should take the responsibility to design our structures to support the components we allow in our design per that industry standards.

 

[EngineeringEric]

I have been out all day. so i apologize for not contributing earlier.

First, thank you all.
Second i do mean early 19th (ie 1830 one-eight for this particular project!) I know there were not design manuals for the period so i was mainly looking for information as old as i could go back to see how things have changed (property specific). Testing this wood isn't really an option but i have done that before on different projects.

I think the minimum req't is take 3 samples (I've used 12 so not sure if 3 is truly allowed). test to failure. then make sure you use an ultimate value that is 95% confident it will work. then apply an Omega (factor) of 1.67 for F'b. I think it was 1.67 but now i have to many #'s floating around. Verify the above as now i lost confidence...

I am concerned with the increase of hardness but i am not sure how that would result in a decrease in bending stress... but that is a good valid point.

I will check out the documents that you guys all gave. Thank you so much for thinking about this one for me!

 

[jayrod12]

Just act like a contractor "It's been standing for almost 200 years, It's got to work right?" May the AHJ will accept that

 

[msquared48]

You could always look at this book too from 1916, a predecessor of the Douglas Fir Use Book, first published in 1930.

http://www.amazon.com/G-P-M-Goss-Structural-Handbook-Pacific/dp/B00HMNJWA2

Mike McCann, PE, SE (WA)

 

[EngineeringEric]

if it wasn't for the darn beetles that ate 10% of the depth and some width i probably would have just told the owner that "there is no way one can evaluate it to code and show it good, but it has lasted this long..."

 

[msquared48]

EngineeringEic:

Conjecture here: I think the reduction in bending values, from dealing with older wood materials myself, some as hard as concrete, involves the loss in pliability of the material and increase in overall rigidity, increasing the tendency for a brittle failure of the material. The material also has a greater tendency to crack across the grain, decreasing notches that increases tension stresses.



Mike McCann, PE, SE (WA)

 

[EngineeringEric]

Mike, valid point. i have seem brittle timber failures in the past where a bending failure looks more like a steel tensile. The basis of more brittle creating a lack of ability for the natural grains and fibers to transfer from some into others is fairly rational. In conclusion, it seems that the best anyone can say is the contractor action per JayRod and if we try to prove it good we most likely will have to make so many assumptions it is useless... now somehow i have to put that into a nice report.

 

[samdamon]

Eric- An article that may be of some use was published in the ASCE Practice Periodical on Structural Design and Construction in Feb 1996 called "Investigating and repairing wood bowstring trusses". Its still available from the ASCE online. The article includes a discussion about how allowable bending and tension values for wood were both considered equal prior to the 60's, and how this contributed (along with other factors) to underdesign of portions of wood trusses. I wont paraphrase the whole text but it would be good to read as background before using wood allowables published in old reference manuals.