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.

 

[MacGruber22]

bump

One in the hand is worth two in the bush.

 

[TehMightyEngineer]

Pull test? Easy enough to perform.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[Ron]

As TME notes, a lateral load test. We do them routinely as Florida has a requirement that balcony rails be tested periodically because of the idiots on spring break.

If it is close by calculation, load test it. If not close....provide a fix. If aluminum, the fix is more difficult. If steel, not so much.

If you are proving the depth of fixity for an embed, check the attachment....

 

[MacGruber22]

Thanks for the sample calc, Ron. I have reaction results from my calcs that are almost the same as using yours. The issue is the breakout at the edge. Currently, I have rigged ACI anchoring provisions. I also tried assuming a tension stress distribution to simulate a sort of zippering of failure planes. Both models resulted in under-capacity, but testing may be plausible.

One in the hand is worth two in the bush.

 

[KootK]

I've had success treating it as a zero depth shear lug per ACI 349 provisions. I take the lever arm at the full embedment depth.

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.

 

[MacGruber22]

KootK - interesting. I investigate this.

One in the hand is worth two in the bush.

 

[racookpe1978]

I understand the practical, definitive advantages of doing a load test on the railing.

50 lbs per linear foot, nominal 200 lbs per post if the posts are at their usual 4 foot interval.

But, where do you find the 200 lb test device? Surely not a bathroom scale turned sideways between the railing and a nearby wall.

 

[Ron]

racookepe1978.....here is our test setup for lateral load testing guardrails....

 

[KootK]

Curious Ron: state wide, what kind of failure rates do you see and what is the most common mode of failure?

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]

KootK....actual failure rate is fairly low. Usual failure mode is either a weld failure, end anchor pullout, or failure to recover deflection (though that's not a code requirement).

 

[XR250]

Ron,

How much load do you apply and for what duration?

 

[Ron]

XR250....200 lbf lateral and/or vertical....it is immediate load and release. There are no criteria in the code (IBC and derivatives) for any duration or deflection limits/recovery. The applicable ASTM Standard (E985) does not have deflection criteria either, except for Public Assembly buildings. Otherwise essentially the same as building code.

Adding to my comment previously of a low failure rate....that is because we are usually working on higher end construction I think. Hotels in Florida are required to be load tested; however, we do little of that because of our fee. There are some out there who will do this test for $200. We can't do that.

 

[wannabeSE]

Ron,
The embedment depth calculations based on concrete compressive strength are not clear to me. I include shear with the moment and calculate the elastic stress as V/A + M/S with A = diameter x embedment depth and S = diameter x (embedment depth)² / 6.

 

[Ingenuity]

Ron:

In your horizontal load test setup you are using the column/wall on the LHS as a reaction to the ram/jack force, through the horizontal SHS?

Is that a 5-ton or 10-ton capacity ram/jack?

 

[Ron]

Ingenuity. It is a 10-ton ram; however, the gauge is calibrated for lower loading up to 500 lbf. That way we can control the loading to much smaller increments.

 

[Ingenuity]

Ron: Yes, a low-end calibrated gauge is important when a 200 lb.f load on a 10-ton cylinder (that is capable of 10,000 psi) is only at a pressure of less than 100 psi.

With a single-acting cylinder like the one you use (that uses a spring for the return) the calibration should consist of system-calibration (gauge, cylinder and pump) against a NIST-traceable load cell, not just a gauge calibration to a master gauge. At low levels of load the spring in a 10-ton ram provides considerable 'internal' resistance, resulting in less externally applied load/force, if using ram effective area x gauge pressure (which I don't recommend).

I usually use an external small-capacity calibrated load cell in these instances, for redundancy.

Sorry for the hijack, MacGruber22.

 

[Ron]

Ingenuity...same here.This is "system calibrated". We have several rams but we don't mix and match without separate system calibration. All of our equipment is calibrated with NIST traceability or other recognized standards if NIST traceability is not available...load cells, gauges, manometers, moisture meters, sound testing, etc.

 

[XR250]

200 lbf lateral and/or vertical....it is immediate load and release. There are no criteria in the code (IBC and derivatives) for any duration or deflection limits/recovery.

So it seems you are testing the ultimate capacity of the guardrail. Does that mean when we design them we should be designing without a factor of safety? Does not appear correct.

 

[Ron]

How they are designed is not the impetus for the test loads. The test loads are code mandates. They are obviously required to resist those. In design, it is incumbent upon us as designers to use our engineering judgment with regard to safety factors, load factors, load combinations or other criteria we deem necessary for safe and efficient design, all the while meeting at least the minimum criteria required by code.