Unity check for pin connection? 2

[Spoonful]

Hi All,

When designing the pin connection, we have tensile and shear stress and also bearing stress.

Assume all individual stress are within its limit, say Actual/allowable = about 60% each.

DO YOU also do a unity check?

eg

tensile actual/ tensile allowable + shear actual/shear allowable + bearing actual/ bearing allowable <= 1?

If so the design will now fail.

Thank you.

 

[Guastavino]

Can you post a quick sketch or describe the detail and application?

 

[Spoonful]

Please see attached.

 

[gwynn]

It depends on what the application is, and if the pin is going to see significant rotations and how many cycles of load and/or rotation (fatigue/wear considerations). CSA-S6 gives the interaction equation for bending and shear stress in a pin as Mf/Mr+(Vf/Vr)^3 < 1, with Mr based on elastic section properties. AREMA has different specifications which can be more stringent (and that I don't know off the top of my head). In general, for a pin with low rotation demand, I would satisfy both requirements.

If it sees high rotation demand, neither of those apply. If it is for a below the hook lifting device (a common area to see pins), neither of those apply. You do need to give a bit more information on the application in this instance.

 

[KootK]

Your sketch leads me to believe that you're checking the connected parent material rather than the pin itself. In that case no, a combined stress check is not applicable. The three failure modes are treated independently.

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.

 

[dhengr]

Spoonful:
Generally, I agree with KootK’s last post, the various stresses are checked separately, except... you should check combined stresses where appropriate, and that combined stress should be kept within some bounds, at your judgement, maybe a unity check of some sort. Examples: If the end distance to the pin hole is small, you can get a splitting shear failure on only one shear face, at the centerline axis of the pin hole, induced by combined shear, tension, bending and bearing stress which are max. at the bearing contact area of the pin & hole. At the sides, where you show sigma, you can get high tension and bending stresses combining; with tension being essentially σ = T(material area in tension, net area); and bending stresses being in tension at the hole edge, and compression at the pin plate outer edges. Again, combined stresses should be less than Fy(some FoS). Take a look at ASME BTH-1, “Design of Below-the-Hook Lifting Devices,” they have a pretty good section on design around pins and pin plates. Look at the commentary sections of all codes too, they often give you considerable insight on your problem, not just a bunch of formulas to grind through. Look also at some good Strength of Materials or Theory of Elasticity texts for the design of eye bars and pin plates. In the same texts take a look at Hertz contact stresses, these are the bearing stresses btwn. the pin and the pin holes.

 

[rb1957]

is "unity" check an expression you structural people use ? sounds odd to someone not in your field (i'd call it a margin, or MS, check)

to the OP i think you're combining incorrectly. bearing is bearing ... it is how the load travels from the pin to the parent material; you are checking that the hole doesn't elongate. once the load is in the parent material, then it is creating stresses ... you can do a simple net section tension check. if you've got tension and shear stresses I'd prefer a principal (or von mises) stress, summing tesnion stress/tension allow + shear stress/shear allow is very conservative. If the shear stress you're talking about is in the pin, then it has nothing to do with the parent material.

another day in paradise, or is paradise one day closer ?

 

[KootK]

is "unity" check an expression you structural people use ?

It is. It's analogous to a Von Mises check and employed where failure modes are more complex and macroscopic than straight up material yielding etc.

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.

 

[dhengr]

Rb1957:
You and I would have known the ‘unity check’ as interaction formulae for combined stresses when we went to school. You would have seen one when we studied column design, (fa/Fa) + (fb/Fb) ≤ 1, where (axial stress/allowable axial stress if only axial stress existed) + (bending normal stress/allowable bending stress if only bending stress existed). There were other variations on the theme for other stress conditions and combinations. I’m pretty sure I’ve seen this approach for combined stresses in areo ref. books too. At least, at that time, we kinda understood what we were doing (what we were addressing) when we used these interaction formulae. I dare say, that half the population using FEA software these days, don’t have the vaguest idea what Von Mises stresses are, or what to do with them. They just see a red area in the pretty output, and then don’t know what to do.

We mostly used factors of safety in Civil/Structural work, not margins of safety, they are slightly different, I’m sure you know. MS is pretty much an Areo approach to stress limits. Because of the cross discipline work and research I’ve done over the years, I’ve used MS for a long time.

 

[Spoonful]

Hi,

Thanks for all the useful replies,

I don't recall ASME BTH do a unity check for tensile, shear and bearing, (so doesn't AISC). I now think maybe I am being over conservative.

Rotation is not a concern here, there are actually one row of 3 holes, I am dividing the load by the number of holes(pin).


rb1957:

regarding Von Mises stresses. at the pin hole edge side, your tensile stress is your σx and bearing is your σy and shear is your τxy. Correct?

dhengr
The example case your have described, when you said "Again, combined stresses should be less than Fy(some FoS)" do yo meaning simply added all three stresses? or Von Mises?


Thanks again for the helps.

 

[rb1957]

I think your "factor of safety" is like our "reserve factor" (>1, allowable/applied).

I would not use bearing stress in a von mises calc.

bearing is one failure mode, parent material (ie the stresses in the plate from the load applied, away from the load) are another.

another day in paradise, or is paradise one day closer ?

 

[Spoonful]

So in that case, Von Mises here, would only have tensile stress σx and shear τxy ?

I trying to understand why should bearing stress not to be considered here when combine all stresses?

 

[KootK]

I trying to understand why should bearing stress not to be considered here when combine all stresses?

Strictly speaking, you are correct. In general, all stress should be considered in a Von Mises check, including the compression stresses implied by a bearing condition. The usual condition, however, is that bearing stresses occur where other stress are low and thus can be ignored or dealt with in a simplified fashion. Additionally, in many members, bearing compression stresses will wind up being trivial compared to flexural compression/tension forces. There are, of course, always exceptions to general rules like this.



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.

 

[rb1957]

for me, the bearing stress is very local to the hole; and the parent material stresses are far from the hole. ie, two different stresses at two different locations

another day in paradise, or is paradise one day closer ?

 

[dozer]

Spoonful, you want to be very careful trying to take average stresses from three different failure modes and cram them into the equation for von Mises stress. I think of it as von Mises stresses are "real world" stresses that should be arrived at by taking into account stress concentrations, Poisson's effect, eccentricity, etc. The stresses you're talking about are nominal or average stresses that we work-a-day engineers use in conjunction with AISC or something like it, so we won't spend forever designing one connection in a structure that has a thousand connections.

You should either do a detailed analysis where you figure out the "exact" state of stress in a component or do a code check in accordance with some accepted specification like AISC 360, but you shouldn't try and mix the two.

 

[Shu Jiang PEng SE]

You do not need to check shear and tension combination. they are two types of failure modes which will not occur simultaneously . However, you need to check another failure mode: block shear strength for the limit state of block shear rupture along a shear failure path and a perpendicular tension failure path.
You also need to check bearing stress.