I remember reading in a code somewhere that the guardrail must resist the prescribed load "without failure or permanent set". Does anyone else remember reading this phrase as it relates to the design of building-type guardrails (e.g. OSHA, UBC, IBC, etc.)? If so, do you remember where you read it?
In my opinion, a guardrail is there purely to save someone's life - not to give you a feeling of confidence in its strength or stiffness. (If the engineer advises, and the Owner agrees that it should be designed to exceed the code minimums, then by all means that is fine, of course.)
If the phrase I remember is correct, then the intent of the code body that wrote the statement would probably be:
"We want the rail to stop a 200 pound force without breaking or deforming, but we really don't care if it exceeds some typical 'allowable stress' limit."
Designing the rail to stop a 200 pound force without breaking or permanently bending would probably mean letting the bending stress get pretty close to Fy, and the deflection could be most anything, as long as it stayed in the elastic range so it springs back once the load is gone.
The question then becomes what do you design the connections for?
I wonder if anyone would care to comment on this.
BTW - for what it's worth, I do design rails and connections using the allowable stress method, and the full 200 pound (or 50#/ft, if greater) load. I use the IBC 2000 allowable overstress of 1/3 (1607.7.1.3). I personally think this more than satisfies the 'without failure or permanent set' idea.
In my opinion, a guardrail is there purely to save someone's life - not to give you a feeling of confidence in its strength or stiffness. (If the engineer advises, and the Owner agrees that it should be designed to exceed the code minimums, then by all means that is fine, of course.)
If the phrase I remember is correct, then the intent of the code body that wrote the statement would probably be:
"We want the rail to stop a 200 pound force without breaking or deforming, but we really don't care if it exceeds some typical 'allowable stress' limit."
Designing the rail to stop a 200 pound force without breaking or permanently bending would probably mean letting the bending stress get pretty close to Fy, and the deflection could be most anything, as long as it stayed in the elastic range so it springs back once the load is gone.
The question then becomes what do you design the connections for?
I wonder if anyone would care to comment on this.
BTW - for what it's worth, I do design rails and connections using the allowable stress method, and the full 200 pound (or 50#/ft, if greater) load. I use the IBC 2000 allowable overstress of 1/3 (1607.7.1.3). I personally think this more than satisfies the 'without failure or permanent set' idea.
