Deflected - Bent beam due to installation - Analysis / Repair Method

[RFreund]

Background ->

This is a temporary shoring installation situation. It is a square excavation sheeted and braced with walers. The walers have kicker/braces at each corner, basically 4 perimeter wales with corner braces, all welded. Due to the sheeting being slightly slanted the hole gets smaller as you go down. The contractor installing the bracing was pushing at the center of the waler and bent/curved the waler (due to the restraint at the corners). It has deflected 5" in the weak direction. The welds are not broken. Attached is a photo (remember the deflection is vertical not horizontal)

Is there a way to analyze or decide if the beam is acceptable or if it is acceptable if pushed back into place?
I've back calculated the stress due to the deflection assuming a curved shape similar to what a uniform load would produce and the stress is above the allowable. However I'm not sure this is an appropriate method. Any good suggestions for repair?

Thanks

EIT
www.HowToEngineer.com

 

[SlideRuleEra]

Leave it alone. Anyway, with load applied you probably won't able to move it with reasonable force. If it does move, you may get movement at other spots, too... causing more serious problems. If the member is judged not adequate for the load, install another complete independent ring adjacent to the existing ring.

As for analyzing, I suspect the true loads are not really known - just estimates.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[RFreund]

That was my first thought as well until calculating the stress from the assumed deflected shape would be about 80ksi. However, Due to the fact that there is no load in the vertical direction, I'm not sure that it is critical.
As with any earth retention problem, true loads remain a mystery. The soils though are a very stiff which I suppose is another reason that the system should perform acceptably.

EIT

http://www.HowToEngineer.com

 

[SlideRuleEra]

RFreund - From looking at the photo, I surely believe your calculated 80ksi is a valid number. That member's span to depth ratio looks way too much - maybe 12:1, from corner brace to corner brace. The ring is likely undersized, even if it was not damaged. What was the calculated stress in the undamaged member? Hard to accept that a 5 inch vertical defection would change the calculated stress significantly.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[RFreund]

Interaction between strong axis bending, axial, and axial x 5" deflection weak axis moment calcs out ok.

Maybe have them weld it to the sheeting a closer intervals.

EIT

http://www.HowToEngineer.com

 

[RFreund]

Sorry jumped the gun there.

Again the 5" is in the vertical direction so I back calculated. finding what uniform load causes 5" deflection then finding the moment associated with that uniform load and finally the stress from that moment.

The loads are technically zero for the soil when you include the cohesion, however I changed the input and added surcharge, etc and it is satisfactory. Spans are 16' center span with 5.5' each side. No nearby existing structures.

EIT

http://www.HowToEngineer.com

 

[SlideRuleEra]

Did the member DEFLECT 5" (vertical, weak axis), due to the horizontal soil load? Or was the member BENT (permanently) 5" (vertically, weak axis) when contractor installing the bracing was pushing at the center of the wale?

Bending a wale (permanently) due to mishandling is common. Not at all good, but not as bad as deflection because of load.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[RFreund]

Sliderule - "Or was the member BENT (permanently) 5" (vertically, weak axis) when contractor installing the bracing was pushing at the center of the wale?" Correct

"Bending a wale (permanently) due to mishandling is common. Not at all good, but not as bad as deflection because of load."
That's what I had figured. I see it from time to time but we don't get many calls about it. Maybe the project engineer on site is new or wanted reassurance.

EIT

http://www.HowToEngineer.com

 

[TehMightyEngineer]

You could almost consider this a curved beam that was rolled the easy way (albeit with a very tight radius). You'll get some torsional effects from the displaced centerline of the beam relative to the supports and you would want to verify that the residual stresses from the bending (mishandling in this case) don't invalidate the local buckling and lateral-torsional buckling equations. Otherwise, given the temporary nature of it and the fact that any failure mode should be highly ductile, I would expect this turns out fine after analysis.

Maine EIT, Civil/Structural.

 

[RFreund]

Good points. I am unsure how to calculate the residual stresses portion, any suggestions?

EIT

http://www.HowToEngineer.com

 

[TehMightyEngineer]

See figure 1 of this paper:

http://www.nordicsteel2009.se/pdf/120.pdf

Based on my (limited) understanding residual stresses will only effect your buckling strength calculations. And then only if they exceed the residual stresses that a straight hot-rolled steel beam normally has. This is the reason for the 0.7Fy term in the lateral-torsional buckling equations for steel. Roughly if you keep your residual stresses below the 70% you should probably be acceptable.

You'll also have reductions in notch-toughness due to the tight radius of your bend. This probably isn't an issue unless you have major impact or fatigue issues which seems unlikely for a temporary beam.

Maine EIT, Civil/Structural.

 

[RFreund]

Thanks!

EIT

http://www.HowToEngineer.com

 

[dhengr]

Rfreund:
I would be inclined to leave it alone also, for fear of doing more damage than good. Yes it has yielded, and been buckled, but you say no cracking or breaking of the main members or the welds. When (and if) it is loaded by the soil and sheeting will this tend to try to straighten the buckled member or make the buckle grow worse? This is a kinda subjective question, but how much has the new shape really changed the member section properties? The flg. has moved laterally by 4 or 5" and the web has bent and buckled a bit in the process, but you haven’t decreased the actual depth of the member by 20-30% have you? The buckling is actually fairly concentrated isn’t it, and if so, plastic deformation took place and regular deflection calcs., and regular stresses and loads causing this go out the window. As long as the anticipated loading doesn’t make the buckle grow, the member should be able to be loaded to yield again without significant problems. As SlideRuleEra suggested, these kinds of dings are fairly common in this kind of work. They just can’t seem to miss em with backhoe buckets and the like.

 

[RFreund]

Good points, thanks DH.

EIT

http://www.HowToEngineer.com