C shaped material lifter - Unbraced Length

[PEVT]

Take a look at the attached sketch and tell me what you think the unbraced length should be as I evaluate the allowable stress on the section.

I am designing a C shaped lifter to replace one with very poor weld details, pipe sections welded with partial pen welds in a 90 degree knee joint at each of the two corners.

The load is very small at 350lb. I would like to cut this from a plate to avoid welds in the corners, but am challenged by the allowable stresses due to lateral torsional buckling. When I check the allowable stress for the very small sections I need to fit the geometric constraints at the site, with an apparent Lb = 24"+12"+24"=60", the allowable stresses are too low to be useful (1-2ksi), and a closed section (pipe) seems to be my only choice, but the weld details for joing pipes at corners just seem like bad options.

I am designing in accordance with ASME BTH-1-2014, Design of Below-The-Hook Lifting Devices. In this document Lb is defined as: the distance between cross-sections braced against twist or lateral displacement of the compression flange; for beams not braced against twist or lateral displacement, the greater of the maximum distance between supports or the distance between the two points of applied load that are farthest apart.

 

[dhengr]

PEVT:
It would be nice to see a few photos of the exiting lifting device, and the details in the curved section of the device. It would also be helpful to understand what you are lifting, its shape, dimensions, etc. and those “geometric constraints at the site” which you elude to. Certainly, you can’t be serious about the sharp square corners, btwn. the vert. leg and the horiz. legs, at the back of your new lifting device. I’ve always paid much more attention to these details and the special stress problems associated with a curved beam than the buckling of the entire device, but I’ve also been careful about the proportioning of my beam elements. A square or round shape or tubes of these shapes are much less susceptible to torsional or ‘lateral torsional buckling’ problems, than your plate section with the large dim. standing vertically. In any case, you do want generous inside radii or the tension stress just explode. Also, watch out for radial stresses which tend to compound the normal bending stresses. Rather than cutting a hook out of a plate, you might be better off heating and bending a more squat piece of steel bar stock. You would end up with the steel grain better oriented. And, of course, they forge hooks and rings, etc. for this very reason. Take a hard look at the design of curved beams, the Winkler-Bach Formula, and the like. Seely and Smith, in their “Advanced Mechanics of Materials,” pub. by Wiley & Sons, has a good section on ‘Curved Flexural Members,’ as do a number of other good Theory of Elasticity and Strength of Materials texts.

 

[oldestguy]

My choice would be pipe and maybe a little larger size than you are thinking. I'd saw 45 degree cuts so that the joint is pretty well visible for welding, using beveled joints for full depth welding except at inside corners.. That inside corner might take two or three passes. If this is for carrying pipes, I'd size it so the loaded position is still a little tip pointed up some.

 

[PEVT]

See attached photo of an example unit. The geometric constraint is the unit needs to fit inside a 2.375" diameter hole, with the long end buried all the way up to the handle. The unit lifts rolls of paper, so the long part slides into the center of the rolls. You can see in the photo the welds which I really do not want to duplicate. Thanks

 

[EdStainless]

We have a few of these devices.
The ones that have worked best are rather complicated to build.
We use heavier pipe for the top and vertical pieces than the main horizontal one.
We mill the joints so that the tubes fit square against each other, not a 45deg miter.
The welds are all full penetration.
The horizontal tube actually fits over a pin that extends about 40% of the distance. We machine a steel bar to exactly fit inside of the tube.
The bore a hole pin size through the vertical and weld the pin in on both sides. (the vert tube should extend a bit further than the joint)
Then slide the tube over it (with it's end cut to exactly fit against the vertical) and then we weld that.
If there is any ID clearance we will add a gusset full length to the bottom of the horizontal tube to stiffen it.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[cal91]

If the section is a pipe, LTB is not an issue. Am I missing something?

 

[dhengr]

PEVT:

You know that the 2" pipe you are using for the lower horiz. leg works now, in terms of length, round fit, and moment (stresses) at the back, near the vert. leg. That is, M = (350lbs.)(some FoS)(24") = 8.4"kips. This moment, give or take a few inches in lever arm is much less significant, less difficult to deal with in a good design, your problem, than all the stress amplification and stress raiser b.s. which happens back at that joint. Make that lever arm 27 or 28" long, a 17% lever arm increase, so you can get a generous radius, 3 or 4", in the two corners at the vert leg. Have a fab./machine shop heat bend a 3" dia. +/- high strength stl. bar with those 3 or 4" radii, 12" high and with 3" of additional stick-out on each end. This looks something like a flat “C” shape, not a “C”, rather a “[“ with radii. Your 3" bar is first machined so that the 3" of additional stick-out on each end will lightly press fit into your horiz. pipe sections. The machining also included a 45-50̊ bevel from the 3" dia. to the new smaller dia. Cut a bevel on the end of the pipes too. Press the three parts together and weld the groove full, that is, a CJP weld. Have the shop grind smooth the welds on the tension sides. Put a snug slip washer (no cross pipe welding allowed) on the lower horiz. leg at the weld joint, so you can’t shove the pipe in far enough to damage your paper roll. Alternatively, make the stick-out on the ends of the 3" dia. bar 12" long, but same slip-fit machining at the ends. This gets the CJP welds away from the corners and at a lower stress level. You could also just heat bend a full 12" dia. “C” shape on the 3" dia. bar with a slightly longer lever arm. If you could load your machine with something akin to a forklift, you would only need the lower horiz. leg. Grab it by the end with a clamp or fitting on the forklift and insert it into the machine.

 

[PEVT]

Thank you all.

I would like comments on your thoughts regarding the Lb, unbraced length, that you think would apply.

cal91, The reason I would like input on this is because during the analysis for shapes, other than closed or sections in weak axis bending for example, the Lb is really important. Imagine I was just to cut this from a plate and stiffen the compression edges for example...you can imagine the plate deforming out of plane at some point...which I would like to understand better.

The most interesting design challenges are :
Keeping the weight at or less than the existing unit (42lb) which is by far the biggest challenge and,
Making a corner connection with weld details that are not prone to fatigue.

dhenger, I really like the bent bar concept and will look into that further, see weight issue below.

Attached is a partial photo of the unit I originally designed and fabricated. The workers at the plant have to physically pull and move the unit high up in the air, so weight is super critical, hence my counterweighted unit worked fine at a different machine but was not doable where they had to manipulate it above their heads...

Hence here I am, trying to make something light, safe and durable.

 

[PEVT]

Here is a drawing of the lifter that was designed and fabricated, with a counterweight, that is just too heavy....

 

[cal91]

Oh okay. I thought you were only considering pipe. Keeping in mind that LTB is torsional rotation of a cross section from a strong axis "I" to a weaker axis "I" to minimize the potential energy. I think that the beam not being in a straight line has the same effect as it would for a straight line beam. At every point across the beam it will torsionally rotate in such a way that the point of load application is at a minimum height.

Also, this seems like a situation where you'd want to be more conservative, we're not trying to design aircraft here. So I would treat the unbraced length as the entire length along the beam from the point of load application to the point of support. Cb = 1.0.

 

[cal91]

If you go with pipe maybe put a sandwich plate (thickness = pipe wall thickness) at each 45 cut so you can just have fillet welds?