I am doing the analysis on roof bowstring trusses from 1943. The attached note I think means the wood was to have 900 psi bending strength and 1200 psi axial but I am not sure. It looks like there may be an "f" after the 1200 psi. The tensile strength of wood is commonly listed as much lower than the parallel to grain compressive strength in NDS. If the 1200 psi represents axial, was it common back then to use that for both tension and compression?
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations TugboatEng on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Old wood stress values 3
- Thread starter haynewp
- Start date
My guess would be fc = 900 psi and fb = 1200 psi.
Not much reference to wood tensile values in older documents that I've seen. Most wood construction from the 1940's tended to avoid tension in wood as the connections weren't as developed or understood. I've seen numerous old wood trusses that had compression diagonals but steel rod or pipe tensile members.
Check out Eng-Tips Forum's Policies here:
faq731-376
Not much reference to wood tensile values in older documents that I've seen. Most wood construction from the 1940's tended to avoid tension in wood as the connections weren't as developed or understood. I've seen numerous old wood trusses that had compression diagonals but steel rod or pipe tensile members.
Check out Eng-Tips Forum's Policies here:
faq731-376
- Thread starter
- #3
I found an old listing for wood properties and it does appear that the allowable bending stress is higher than the allowable compression parallel to grain. I was comparing values using NDS for southern pine which varies as to the one that is higher. Unfortunately, these trusses are all wood members.
-
1
- #4
SlideRuleEra
Structural
haynewp - From the 1942 Southern Pine Manual, allowable tension stress was considered equal to allowable bending stress:
Also, the nominal lumber sizes specified in 1943 were larger than today.
![[idea]](/data/assets/smilies/idea.gif "[idea] [idea]")
![[r2d2]](/data/assets/smilies/r2d2.gif "[r2d2] [r2d2]")
Also, the nominal lumber sizes specified in 1943 were larger than today.
It was very common for those old bowstring trusses to use tensile stresses up to 1500 psi.
I've seen many failures of these bottom chords, typically after many years of service.
Typically failures were due to knots or other defects, excessive slope of grain (> 1:12), or compression wood fractures that were undetected with visual grading.
Also, if war-time construction, many trusses were built with unseasoned wood, which may have excessive splits or checks due to restraint of the gusset plates and connections. Also the stresses may have been pushed a bit due to war-time urgency and limited expected life-span.
Be careful with these trusses!
I've seen many failures of these bottom chords, typically after many years of service.
Typically failures were due to knots or other defects, excessive slope of grain (> 1:12), or compression wood fractures that were undetected with visual grading.
Also, if war-time construction, many trusses were built with unseasoned wood, which may have excessive splits or checks due to restraint of the gusset plates and connections. Also the stresses may have been pushed a bit due to war-time urgency and limited expected life-span.
Be careful with these trusses!
- Thread starter
- #6
from aitc:
From at least as early as the 1930’s until the late 1960’s the tensile strength of lumber was believed to be equal to or
exceeding that of its bending strength. The following statement from the 1935 Wood Handbook reflects this belief, “The
tensile strength of wood is greater than the modulus of rupture as obtained from bending tests. Hence stresses in the
body of a tension member fully equal to those given for fiber stress in bending…are justified.” (Wood Handbook, 1935,
p. 105). Limited full scale testing in tension parallel to grain resulted in failure of the wood specimens due to crushing in
the test machine grips prior to tension failure.
This belief was held until the 1960’s when improved grips for tension testing were developed to determine the tensile
strength of machine graded lumber. With improved grips, the testing of full scale specimens indicated that the tensile
strength was significantly lower than previously assumed. By 1970 virtually all of the lumber grading agencies had
reduced their published design values for tension parallel to grain to reflect this new knowledge.
From at least as early as the 1930’s until the late 1960’s the tensile strength of lumber was believed to be equal to or
exceeding that of its bending strength. The following statement from the 1935 Wood Handbook reflects this belief, “The
tensile strength of wood is greater than the modulus of rupture as obtained from bending tests. Hence stresses in the
body of a tension member fully equal to those given for fiber stress in bending…are justified.” (Wood Handbook, 1935,
p. 105). Limited full scale testing in tension parallel to grain resulted in failure of the wood specimens due to crushing in
the test machine grips prior to tension failure.
This belief was held until the 1960’s when improved grips for tension testing were developed to determine the tensile
strength of machine graded lumber. With improved grips, the testing of full scale specimens indicated that the tensile
strength was significantly lower than previously assumed. By 1970 virtually all of the lumber grading agencies had
reduced their published design values for tension parallel to grain to reflect this new knowledge.
Good article on this topic: "investigating and repairing wood bowstring trusses", by Richard J. Kristie and Arne P. Johnson, February 1996 "Practice periodical on structural design and construction", published by ASCE. The change in wood tension allowables as well as other factors to consider(unbalanced snow loading for example) are discussed.
- Thread starter
- #9
The attached from the manual that SRE found has both a 900 psi stress grade (No 2 medium grade) and a 1200 psi stress grade (No 1 grade) in bending listed. So now I am thinking maybe it was understood back then that the chords were made from the higher stress No 1 grade and the webs from the No 2 medium, instead of the 900 psi being the required compressive strength and the 1200 psi being the required bending strength. Or maybe each grade was common for members of a certain dimension and that was the understood intent. I am not sure what to use now.
-
1
- #10
SlideRuleEra
Structural
hawnewp said:
Is it similar to the sample roof bowstring truss, with TECO split ring connectors, shown on pages 112 & 113 of the 1942 Southern Pine Manual?
Scan of those two pages attached.
- Thread starter
- #11
SlideRuleEra
Structural
haynewp - You are welcome. Also, note that both top and bottom cords are built-up members. Although the "repetitive member factor" did not exist in 1943... those members are repetitive. I would give that information, together with larger dimension nominal size lumber, due consideration for the analysis.
- Thread starter
- #13
I don't know for sure this is Southern Pine since the original drawings did not say, but based on my research into what was commonly being used at the time and that diagram and values from the 1942 Southern Pine Manual I feel a lot more confident that it is Southern Pine. It kind of looks like Pine too (attached). I am trying to figure out what the best comparison of today's NDS values would be to use for the analysis given the fc and fb listed on the original drawings and figuring this is So Pine. There's not an exact match for current fc and fb based on what the original drawing listed. I understand that wood used to be of better general quality than today, but the testing has also been advanced to give more accurate values (hence the whole tension value issue mentioned above).
racookpe1978
Nuclear
Old school indeed! Square nuts.
Change those tiny "washers" to a larger square black-painted plate (1/8 thick, 3/16 or 1/4 thick at critical joints) so your clamping force works over a larger area. By not pulling the washer and nut through the wood at each bolted connection, you'll regain some of the strength "lost" by the lower predicted wood strength in the new calc. Also, the clunky black painted plate connections will be an architectural detail attracting interest to a needed structural feature. Will hide the stains around the washers as well.
Change those tiny "washers" to a larger square black-painted plate (1/8 thick, 3/16 or 1/4 thick at critical joints) so your clamping force works over a larger area. By not pulling the washer and nut through the wood at each bolted connection, you'll regain some of the strength "lost" by the lower predicted wood strength in the new calc. Also, the clunky black painted plate connections will be an architectural detail attracting interest to a needed structural feature. Will hide the stains around the washers as well.
-
1
- #15
SlideRuleEra
Structural
haynewp - From the photo, I'm 99%+ sure that is southern yellow pine, and high quality, too.
As for the design value, take a look at the current southern pine, visually graded, design values. NDS values come directly from this data. IMHO, it is then time to apply engineering judgement to select a reasonable set to use for the calcs. Here is what, and why, I would select:
Assuming the 2" to 4" thick, 2" and wider range, Southern Pine, No.2, Dense - Not because that is what the lumber is, but because that designation has values are a good match for the most dependable 1943 values. The most dependable values, in my opinion, are Fb = 1200 psi and E = 1,600,000 psi. Then use the other modern values for that grade. There is so much variability in wood, that pinpointing an exact value for any design value is futile.
For example: In 1967 tests by the Forest Products Lab, measured failure of #1 SYP for tension parallel to grain varied from 1570 psi to 12,640 psi, with an average of 5480 psi.
For #2, the range was 1000 psi to 10,430 psi, average 3400 psi.
Making a well thought out and reasonable choice of design values is as good as it gets.
As for the design value, take a look at the current southern pine, visually graded, design values. NDS values come directly from this data. IMHO, it is then time to apply engineering judgement to select a reasonable set to use for the calcs. Here is what, and why, I would select:
Assuming the 2" to 4" thick, 2" and wider range, Southern Pine, No.2, Dense - Not because that is what the lumber is, but because that designation has values are a good match for the most dependable 1943 values. The most dependable values, in my opinion, are Fb = 1200 psi and E = 1,600,000 psi. Then use the other modern values for that grade. There is so much variability in wood, that pinpointing an exact value for any design value is futile.
For example: In 1967 tests by the Forest Products Lab, measured failure of #1 SYP for tension parallel to grain varied from 1570 psi to 12,640 psi, with an average of 5480 psi.
For #2, the range was 1000 psi to 10,430 psi, average 3400 psi.
Making a well thought out and reasonable choice of design values is as good as it gets.
- Thread starter
- #16
Thanks. I am going to assume the same No 2 Dense for the 4x8 bottom chord members (mine are 4X8's instead of 4X6's that are shown on the 1942 diagram).
Completely different project but I attached some photos from another building I went in recently that I thought some of you might find interesting. These trusses are from 1879. They were a mix of steel rods and large wood members. They used notches instead of steel fasteners to make the large member to member connections. There was a steel spike every once in a while in some of the smaller framing. Some of these trusses were huge. The structure was multi-story built on foundations and walls originally from the 1830's.
Completely different project but I attached some photos from another building I went in recently that I thought some of you might find interesting. These trusses are from 1879. They were a mix of steel rods and large wood members. They used notches instead of steel fasteners to make the large member to member connections. There was a steel spike every once in a while in some of the smaller framing. Some of these trusses were huge. The structure was multi-story built on foundations and walls originally from the 1830's.
Some things don't look right in those photos:
In the first one, there is a double notch heel joint and the member framing into it is framing in at a much steeper angle and doesn't look like it has matching joint notches. To the right, there appears to be a wedge block that is not bearing on the member supposed to rest on it.
Last photo appears to show beams with mortise and tenon joints that are pulling out of the supporting member.
Looks like there have been some major alterations, or something has shifted.
In the first one, there is a double notch heel joint and the member framing into it is framing in at a much steeper angle and doesn't look like it has matching joint notches. To the right, there appears to be a wedge block that is not bearing on the member supposed to rest on it.
Last photo appears to show beams with mortise and tenon joints that are pulling out of the supporting member.
Looks like there have been some major alterations, or something has shifted.
- Status
Similar threads
