Need a book? Engineering books recommendations...

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

RE: Acceptable Level of Overstress

[Subject Prev][Subject Next][Thread Prev][Thread Next]


Stan makes a good point here assuming that either a failure occurs or a visual deflection brings the design engineer into court because the owner finds it unacceptable and the finger is pointed at the overstress allowed in the calculations. However, this does assume that damage occurs.

The other issues that are relevant are; what materials are you referring to and what method of design is used. There is much more room for error in working stress design than in  allowable stress design. However, in wood, you need to consider the long term effect of creep on the member and while wood probably offers the widest range of safety factors, the builder can always install a beam crown side down and enhance the error if the overstress is in bending.

I am much more conservative than most. I won’t exceed 100% of the allowable stress of a member or connection – in fact, I try and keep it below 85% and I look at issues like deflection whether it is a beam, column or shearwall. Maybe this is why I don’t do much in the way of panelized roofing and tilt-up design – commercial and industrial buildings generally count every dollar and I would rather keep my practice on the conservative side. I think in the long run it does the owner (who may not be the developer or builder) more in performance and compensation for what the builder may lack in ability than a more stringent code.

I received a private e-mail about cantilevered columns and SEAOC’s position paper on the use of an R=2.2 for only where the columns exist. However, the code is the law and SEAOC’s position is arguable in a court of law. A talented expert can argue that the code is specific as to the full-compliance method of design and that the base shear is taken for the lowest R value in the direction of the force. Now you and I both know that nearly every engineer violates this law in lieu of SEAOC’s position. But SEAOC created the code by the nature of their work. Their follow-up position is not law, it is only an opinion. If damage occurs and if any of it can be interpreted to be deflection related that may have been mitigated by using an R of 2.2 for the entire direction in all walls, the engineer of record will get nailed.

So in short – Stan is correct – it is a matter of what counts if litigation occurs. If not, then the limit on overstress is only a moot point.




-----Original Message-----
From: Gerard Madden, SE [mailto:gmadden(--nospam--at)]
Thursday, December 04, 2003 1:33 PM
To: seaint(--nospam--at)
Subject: RE: Acceptable Level of Overstress


I guess what I am saying is I am not concerned about a 6% overstress, but the plan checker is. I didn’t realize it was written in stone somewhere about the level of overstress. I know the 5% rule is there for DSA projects.


Thanks again,



-----Original Message-----
From: Suresh Acharya [mailto:struct(--nospam--at)]
Thursday, December 04, 2003 1:48 PM
To: seaint(--nospam--at)
Subject: Re: Acceptable Level of Overstress


 I  would not worry about slight overstressing if:
    -it is a roof  beam and the tributrary area is substantial.
    -it is continuous beam and the controlling mode is flexure because there a chance of redistribution of moments. /*The beam would not know whether you used ASD or LRFD method!! */

Suresh Acharya, SE

Gerard Madden, SE wrote:

What is the general practice on acceptable level of overstress when checking an existing member for capacity?


I have an existing wood beam that is 6% overstressed with new loads. I am prepared to deem that “Acceptable”. I was always taught that 10% or less overstressed is okay. I have a plan checker wanting to cut the limit at 5%. Is this somewhere in the code???





Gerard Madden, SE, PE

California Licensed Structural and Civil Engineer


Madden Engineering

2183 Eaton Drive

Lodi, CA 95242

T: 209.368.9955

F: 209.368.9966

E: gmadden(--nospam--at)