Design of Lifting Lugs (Ricker, AISC EJ - 3Q 1991)

[Illbay]

thread507-446784

I take particular note of the referenced thread (now closed - why do they close them so quickly? This one is about a year old and didn't receive nearly the attention it might have otherwise) and wish to make a couple of observations.

First observation: the OP states that "if you check shear failure using his equation of Pu=2*(0.4)*Fy*e*t and then apply a factor of safety of 5 as he recommends this becomes the controlling case with a Pu=100K so Pa=20K, he is designing the lug for a tension load of 20.4K so that would mean the lug doesn't pass."

I think the fallacy of this observation lies in the difference between how things were done in 1991 - in basically the ASD era of steel design if you will - and how we approach them now. The OP is assuming (as I did at first) the "limit states" approach where you come up with various limit-state capacities, then choose the most critical and apply the "safety factor" or "strength reduction factor" accordingly (depending on if you're using ASD or LRFD respectively).

But Ricker is doing things the OLD "allowable stress" way, where you have capacities that are each treated as separate problems, rather than using the integrated approach we now use in steel design.

Note that on page 154, Ricker states "In allowable stress design it is difficult to establish a base for an overall factor of safety because some allowable stresses are based on ultimate stresses and some on yield stresses and the relationship of these stresses is not uniform as we change steel types. For ASTM A36 steel based on a bending stress, a factor of 1.8 multiplied by the actual weight will produce a factored design weight which can be used in subsequent design calculations. The resulting factor of safety is approximately 5." [emphasis mine]. In other words, Ricker is referring to the load factor divided by the allowable stress factor = 1.8/0.4 = 4.5. (Not sure why he didn't use a "load factor" of 2.0 here which would give a F.S. of exactly 5, but there's a lot of seat-of-the-pants engineering going on in this paper).

So in Ricker's method, the F.S. of 5 applies only to the "tensile load capacity" (2*a*t*Fu) not the shear capacity.

Second observation: I know this isn't a topic that a lot of structural engineers deal with, but it is an important one in construction engineering or in any application where you're lifting loads (I work as an in-house engineer at a mine, and I'm asked to rate lifting lugs all the time). It's disappointing that among all the literature out there, only the Ricker paper (1991) and an earlier paper in the AISC Engineering Journal by Tolbert and Hackett (1974) treat this subject (the former is a practical but non-rigorous treatment, and the latter a formal investigation of limited scope).

It would seem to be a great topic for someone's thesis or dissertation, coming up with a practical methodology that is more in line with current limit states design philosophy, supported by physical and nonlinear FEM modeling.

"No one is completely useless. He can always serve as a bad example." --My Dad ca. 1975

 

[Teguci]

Are you using ASME BTH-1 "Design of Below the Hook Lifting Devices"? -

https://www.asme.org/codes-standards/find-codes-standards/bth-1-design-hook-lifting-devices

 

[Craig_H]

First, I agree that the design of lifting lugs is sometimes poorly carried out by structural engineers. In my experience, lugs are usually poorly dimensioned for the installation of lifting tackle, and lugs are oversized/underrated (the engineered ones, that is - millwright specials are a whole different story). As Teguci has mentioned, ASME BTH-1 provides quite good guidance on lifting lug design, and it is the reference used by many lifting contractors in the design of their custom equipment. ASME BTH-1, however, provides ASD methods of analysis, so that does not do much to help you in your search for a limit states design method.

Seeing as the Euros are ahead of us in adoption of LSD principles, their codes can be useful starting points for LSD analysis. EN 13001-3-1 and EN 1993-1-8 both provide guidance for the design of pinned connections. It is possible to re-derive these codified expressions, and modify them to meet the slightly different load and resistance factors of your local area. It would also be possible to adapt BTH-1 analysis methods into LSD equations. Choice of load factors will also be an interesting exercise.

At the end of the day, I wonder what the efficacy is of reinventing the wheel here. While it could be a fun exercise to reinvent BTH-1 to LSD principles, is it really worth the effort and added risk to you personally?

 

[canwesteng]

BTH-1 is the gold standard. There would be a mittful of problems re-deriving equations using FEM (bearing stress between shackle and lug, stress distribution through the shear plane), and BTH-1 provides design that are economical in terms of both fabrication hours and engineering hours. There is also very little to gain going to LFRD, since you have only live load on the lug anyway.

 

[skeletron]

BTH-1 as a basis with Ricker for insight and independent checks from AISC/CISC based on approximations. Unless you do it by FEM (probably don't sink THAT much time into a lug) size it for ease of construction, fabrication, and rigging.

I agree that it would be ideal to have a tried-and-true fully dedicated guide to designing a lug, but I don't see it as a major time investment with regards to a lot of research programs.