Swing Bolt Closure Hinge Pin B&PVC Stress Analysis 2

[gvc99]

I'm re-evaluating two pressure vessels with swing bolt closures (pairs of parallel lugs welded equally spaced around the circumference of the vessel flange and closure head flange, hinge pins through the vessel lugs capturing swing bolts which swing up between the corresponding closure head lugs and are capped by washers and hexnuts bearing on the top surfaces of the upper lugs). The vessels are designed to Section VIII Division 1 of the B&PVC.

The vessels have been fabricated by a subsidiary, who neglected to calculate the bending stresses in the hinge pins due to the swing bolt loads. The vessels have been tested, stamped, and delivered to the customer.

I have used our company's evaluation method for hinge pin bending, which assumes that the pins are simply supported beams with a span equal to the width of the distance between the inside faces of the lugs, and loaded with a uniformly distributed load over the width of the swing bolt head -

|.|.|.|.|.|.|
....VVVVVVVV....
===========
^--------------^

Using this approach, the bending stress in the pin, 63000 psi, is over the allowable stress for the pin material, 25000 psi(SA193B7).

I've looked at two alternative loadings - point loading at the edges of the swing bolt head

|............|
V...........V
==========
^--------------^


and a "v-shaped distributed load" over the width of the bolt head -

......|..........|
......||.......||
......|||....|||
......VVVVVV
==========
^..................^

Using the company's evaluation method as a reference,
the alternative methods give bending stresses that are 55%
and 24% of the reference stress, respectively.

Only the point load case stress, 15000 psi, will meet code requirements.

The actual loading on pin probably falls somewhere between the extremes of the cases that I've evaluated. I recognize that using the point load case less conservative, but is it non-conservative? I don't believe there is a cut and dried answer.

Also, since the pins are made of a material normally used for bolting, should the more conservative factors of safety that form the basis for the allowable stress be used, or could a argument be made for using a higher allowable?

My recollection is that the design basis for the higher factors of safety on bolting materials is that they are heat treated to achieve their properties, and thus have yield and ultimate strengths fairly close together, which indicates that they may not be as ductile as materials used for shell and head construction.

I greatly appreciate any insights into this problem that forum members may have.

George Chapek, PE.

 

[prex]

Not sure to understand the setup. Particularly can't see how the swing bolt head is: would expect eyebolts, but this seems not being the case from your description: how is the bolt head bearing on the pin?
I assume that you are OK with shear stress in the pin: there is a limit specified in the code, but don't remember how much it is.
Concerning the bending in my opinion you could take an allowable equal to 1.5S, but I wouldn't feel comfortable by assuming a load distribution less safe than the uniform one. After all you are already not very conservative in taking the distance between the inside faces of the lugs (you could take the mid thickness distance), and you could be much more conservative by taking a single center point load.
However, depending on actual detailed geometry, the bending case could not be very representative: if there is a limited gap between the pin and the holes and also not much space between the lugs and the head bolt, failure of this assembly would (and should) occur by shearing the pin (assuming of course bolt heads and lugs are OK).

prex

http://www.xcalcs.com

Online tools for structural design

 

[gvc99]

Thanks for your insights prex.

I assume that using 1.5S for the allowable is based on beam limit design. This seems a better approach than fishing for a load distribution that solves the problem and then trying to justify why this distribution is ok.

I've links to sketches of the swing bolt lug and pin arrangement to clarify my first post.

George

 

[gvc99]

Oops! I linked to thumbnails instead of full images. Here are the full size pictures:

 

[prex]

Now it's clear, it is a common setup (though personally don't like the nuts subject to bending, the washer is not sufficiently thick).
I confirm that I wouldn't check the bending stress in the pin: checking the shear stress is sufficient.
With that geometry it's clear that the pin will fail in shear. When the pin deforms under bending, it is forced to stay straight by the bolt eye. Of course this could require some plastic deformation of the pin, but this is allowed in pressure vessel design for deformation controlled quantities like this one.

prex

http://www.xcalcs.com

Online tools for structural design

 

[gvc99]

prex,

Thanks again for your valuable insight.

George

 

[unclesyd]

We have a number of large filer housings using the swing bolting method for closure and our biggest problem was distortion of the brackets that hold the pins. This was caused by the pins trying to bend. Our solution required the replacement and reinforcement of the brackets that hold the pins. In the refurbishing of filter housings the clamp design was standardized where possible.

The above comments on the washers makes a very good point. On some of the older models we had to make a washer that actually kept the brackets in place. All the round washers were increased in thickness and were hardened.

 

[jte]

George-

Prex and UncleSyd have provided their usual good advice. One additional note that I'll throw in is that since you are now checking the bolts for shear, be sure to read General Note (c) in II-D Table 3 which I believe would apply to your situation and limits the allowable stress to 80% of that listed in the table.

jt

 

[gvc99]

Thanks jt