Safety Factor When designing wood w/ NDS 2005 1

[hawkinjc]

I am designing hole covers for a construction site by spanning sawn lumber.

Per OSHA 1926.502(i)(2): "All other covers shall be capable of supporting, without failure, at least twice the weight of employees, equipment, and materials that may be imposed on the cover at any one time".

I cannot find anything in the NDS that deals with safety factors. I could always load test, but that seems to be too much work for what I am trying to do. Right now I am just doubling my design load and making sure it works with my allowables capacities from the NDS, but I feel that is being way to conservative.

Any suggestions? Thanks.

 

[MiketheEngineer]

If I remember correctly:

NDS (and lumber inspection bureaus)really do not
use "safety facors" per se. Values are based on probability curves.

Again - this has been years ago - I did my own math and was able to show that the SF is about 2:1. Remember ASD steel is about 1.67 to 1.

I may be all wrong - it was a number of years ago and things have changed.

 

[JAE]

I got into an expert witness case once and tried to find out the relationship between the allowable stresses provided in the NDS verses the ultimate "fall down" stresses. Didn't find much but after speaking with a number of individuals I got the impression that the "safety factor" was somewhere between 1.67 and 2.0.

 

[Splitrings]

NDS 2005 has both ASD and LRFD methods. The factor, Kr, used to adjust the listed ASD values for sawn lumber to values used in LRFD are listed in Appendix N in table N1. The factor, Kr, is 2.16/phi for all properties other than compression perp. to the grain and Emin. Phi for bending, as an example, is 0.85.

Another place to look is in the AASHTO LFRD Bridge Design Spec. I have happen to have the third edition. In chapter 8 they list values for visually graded lumber that are adjusted to the "lower 5th percentile of the ultimate stress".

 

[dhengr]

hawkinjc:
The exact assumed factor of safety has become a rather nebulous number with all that has been done with the building codes over the last 15 or 20 years. As mentioned above the FofS is now all tangled up in a bunch of probabilistic and statistical number crunching and various adjustment factors.
I would use NDS for design, double your assumed design loads to meet OSHA reqr’mts. and then design to max. allowable stresses, etc. per NDS. Doesn’t ANSI/ASCE 7 have something to say about construction loadings. I believe a 300# construction worker with a heavy wheel barrow load is a likely load; pallets full of solid conc. blocks on this temporary opening cover seem beyond what you should have to design to accommodate. Maybe a 5-600# tool box is reasonable, with point loads at its feet. Why would you load test, when you don’t really know the craziness of the potential loads? Ask the GC for loadings, he’s the one who has some control over this, and has experience with normal practice on his jobs, of this type. I would try to get him on record on this issue. You should not have to worry about ANY POSSIBLE loads, when only his people can control this. I also suspect that OSHA has something to say about signage, railings and fencing around openings, even when they are covered to at least prevent accidental falls.
The bigger issue with NDS may be the following: You say “sawn lumber”; 1x6's, etc. are really not graded for significant structural use, except as sheathing, etc., light uniform loading; 2x’s are graded for structural use as members on edge or in compression; planking and scaffolding, 2x’s & 3x’s, on the flat, are another grading group, separate from normal structural usage. Concentrated loads and impacts might be the biggest issue, maybe use plywood as the top layer, which tends to distribute loads and make sawn members and sheathing act more in a repetitive fashion, which we like. Proper bearings and attachment of these cover panels is an important issue, and I would think bending and shear stresses and the like can be right up at their limit; and I wouldn’t spec. a 2x8 because a 2x6 was 5 or 10% overstressed, since you have already doubled the loading (for OSHA) and NDS does already have a FofS.
You are allowed to use some engineering judgement, as any prudent engineer would do. You are not reqr’d. to know the FofS, +/- 2%, nor can it be determined, when you are dealing with an unknown potential loading, on material which is somewhat subjectively graded, which might see an unforseen impact loading; and when the carpenter might use a piece of junk with a knot on the tension face for a joist, because he is building it up-side-down, before installation. The NDS, in all its machinations, tries to do this for you, and it works most of the time. Finally, slightly conservative or not, these opening covers are probably not a real big percentage of the total construction cost, engineers are entitled to sleep peacefully at night too, not just the guys who make all the money.

 

[woodman88]

Hawkinjc, As I see it, OSHA has place the twice the design load as the safety factor. If you feel you are being conservative the only thing you can change is your calculation of the design loads.

Garth Dreger PE
AZ Phoenix area

 

[msquared48]

Doubling the expected loading automatically yields a FS of 2.0.

However, considering the NDS, we never, check that - very rarely - design fto a FS of 1.0. 1.5 is the normal accepted minimum. Hence there is a minimum of 1.5 FS already in the allowable stresses of the NDS.

Therefore, the combined FS is in the order of 2.5 to 3.0 or higher for the OSHA requirements you describe.

And we wonder why government toilet seats cost $1000.00.

Mike McCann
MMC Engineering