Old Thread "Railing Posts Anchored with Non-Shrink Grout" 2

[MacGruber22]

Has anyone become more enlightened regarding this thread?

http://www.eng-tips.com/viewthread.cfm?qid=415064

I am struggling to prove that an existing railing embed connection works.


One in the hand is worth two in the bush.

 

[XR250]

That response made my brain hurt :>
Let me re-phrase: When you design guardrail assembles, what factor of safety do you use?

 

[Ron]

I use SF=1.0, but I use ASD for design....so there's more than a 1.0 margin of safety in there. For unwelded aluminum, it works out to about 2.0+. For steel, it is about 1.67. Not a whole lot different for LRFD, I just prefer ASD.

 

[MacGruber22]

No worries, Ron. All good information.

KootK - the lug procedure is the way to go. Much more representative of the failure planes, I think. The only issue I do not like is that the bearing stress is not uniform as assumed in the lug breakout capacity (not to say that is was by rigging ACI anchoring equations!). I still worry about added tension from the moment about the btm edge of slab. The change in moment about that point is 18% when comparing a uniform bearing stress versus the "real" non-uniform, which in my case, gets me 5% under capacity. How have you rat

ionalized that? Maybe your calcs never got that close.

One in the hand is worth two in the bush.

 

[XR250]

I use SF=1.0, but I use ASD for design....so there's more than a 1.0 margin of safety in there. For unwelded aluminum, it works out to about 2.0+. For steel, it is about 1.67. Not a whole lot different for LRFD, I just prefer ASD.

I prefer ASD as well. This still does not make sense to me.
You are essentially testing the ultimate capacity of the assembly using 200 plf. If we design the system using wood, aluminum or steel, the factor of safety is built into the material allowables and ranges from 1.67 to 2.0 +/-, yet we still use the same 200 plf. load.

It seems you should either be testing the rail at 334 - 400 plf or we should be able to factor that load down when we design the system under ASD.

 

[Shotzie]

@XR250, if you were to test the rail to the 334-400 plf range wouldn't you expect the rail to yield, assuming that it was designed to 100% of the design load? I would expect you to load test it to the specified code load as this is the actual intended load for it to resist. We build in our factors of safety in the off chance that it does get overloaded, there will be plastic deformation and the rail will prevent collapse. It wouldn't be practical to test the rail to the increased load as you'd likely have to replace it afterwards.

 

[Ingenuity]

Miami-Dade County has this requirement from Link:

IBC 2012, ASTM and AC test requirments:

Keep in mind that "test load" sometimes refers to the requirement for the manufacturer product testing in a laboratory-type test, not necessarily a field-type load test.

When I did field-load testing to a plastic/vinyl handrail system about 20 years ago, the AHJ required we test to 2 x 200 lb.f. But the max test load in this instance was moot, the system failed at less than 200 lb.f!!!

 

[Ron]

XR250....I finally understand your question! Sorry.

There are some items for which load tests are carried to 125-200 percent of design. Pile load tests for instance. Guardrails are not one of those items. The ASTM Standard for testing guardrails does not require an additional safety factor nor does the building code or life safety code.

I agree with the points Shotzie noted.

 

[KootK]

@MacGruber22: I clearly failed to communicate my zero depth lug intent effectively in my previous post. Let me know if this doesn't clear it up. My lever arm is admittedly aggressive. It's based on the deformation picture that I have in my head which is a me-specific thing. For many arrangements, lowering the center of compression will increase the size of the resisting shear frustum. It should offset some.

Sometimes, to remain solutions focused, I think that one has to do something other than make all of their choices conservatively. I'll probably take a beating for saying so in present company but I really don't care about guardrails very much. I believe the code intent for guardrails really boils down to "stout". If I can parse out a defensible evaluation technique that gets me in the ballpark of "stout", I'm happy with that. Functionally, doing otherwise means that my client heads down the road and gives his business to a crappier engineer who will produce crappier engineering. Thus, my semi-crappy engineering makes the public safer than some other guy's fully-crappy engineering. And I can stay in business. I know, that probably sounds more than a little arrogant. I'm cool with that. This is the place for the unadulterated truth to be told after all.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[Ron]

Ingenuity....your post points out the need for a consensus on guardrail/handrail requirements, which does not currently exist. The references in the Miami-Dade requirements to Florida Building Code requirements (FBC 1618) apply only to the HVHZ areas which are Dade and Broward Counties ( 2 counties out of 67 counties in Florida).....another pet peave of mine with the Florida Building Code. The code has provisions in the HVHZ sections that only apply to those sections, but should apply to all sections of Florida....many of them have nothing to do with wind loads (HVHZ means "High Velocity Hurricane Zone")

Also....Ingenuity...did you live in or near Ponte Vedra Beach at one time?

 

[NorthCivil]

KootK said it right

Especially in the world of glass guards. For every hour I have spent agonizing over stress distributions around holes in point supported glass guards, 3 other engineers in my city have rubber stamped a glass guard that would never pass a concept review. (for half the design fee too).

Sometimes, depending on the probability and consequence of failure of course, you are doing the world a favor by signing off on a guard that can resist 80% of its code required resistance, if everyone else in town is signing off on 50% or less.

Cue the debate on Locke vs Mill vs Kants theories of ethics

 

[Ingenuity]

but should apply to all sections of Florida....many of them have nothing to do with wind loads

Interesting. I 'survived' 200+ mph typhoon back in 1997 when I lived on a tropical island in the Pacific - NOT many guardrails/handrail failures during that event compared to other structural failures, like 5" thick concrete roofs peeling/folding back from overhang corners of homes, prestressed concrete light/power poles snapped like toothpicks etc.

Also....Ingenuity...did you live in or near Ponte Vedra Beach at one time?

No, never visited/lived in any part of Florida...and sorry to admit that I had to google' where Ponte Vedra Beach is located. :-(

 

[XR250]

It wouldn't be practical to test the rail to the increased load as you'd likely have to replace it afterwards.

Makes sense.

 

[MacGruber22]

KootK - I should have asked what you meant by zero length. My bad. Consequently, in my case using a lever arm equal to a linear bearing stress distribution (with full-depth lug) results in about the same capacity as your generous full-depth lever arm with smaller frustum area.

One in the hand is worth two in the bush.