Bolted Joint Stiffness - Analytical Approaches for Special Case

[SpaceEngineer4]

Hi everyone. I know that in the aerospace industry, NASA-STD-5020 is a standard that most companies rely on. However, I have a question regarding estimating joint stiffness factors for cases where there is not enough material to develop the full frustum suggested by Shigley's approach. Does anyone have experience in looking at bolted joints where this situation is true, for example if close to an edge or on a scalloped flange? I'm curious what the best way to handle this is or if this is usually neglected (and if so, what is the rationale?). I understand that FEA is probably the best option, but its usually not something we have enough time to do on each bolted joint meeting this condition, especially for linear dynamic analyses like random vibration and shock where using preloaded joints isn't a feasible option.

Thanks in advance for your thoughts.
SE4

 

[SWComposites]

Joint stiffness in what direction? shear? thru-thickness tensile?

Your choices probably are a) run some tests, b) bludgeon it with a 3D FEM

 

[MintJulep]

The unfortunate reality at many companies in many industries is "ignore it and hope for the best".

Given your citation of NASA-STD-5020, you handle and that you've taken the time to ask the question it's clear that approach isn't a viable option for you.

You state that you have many joints meeting this condition. A best path forward is to not do that. I understand that might not be an option right now for the project at hand.

Since you have multiple joints it's probably worth your time in the long run to develop parametric model to quickly generate inputs for FEA and to define a standard approach to boundary conditions and loading.

Make the effort as efficient as you can.

Update edit: Perhaps you could incrementally build a catalog of joints pre-solved for a unit load. Then just multiply by the actual load to have an immediate answer in the future.

 

[rb1957]

going by your handle, I'd've thought you would have better references than Shigley. I'd've thought spacecraft would have their own body of knowledge to draw on, since they are a highly specialised field. Does Lockheed SM88 have anything to offer ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[sdm919]

The European Space industry has a document called "Threaded fasteners handbook" with ID number ECSS-E-HB-32-23A (the copy I am looking at is dated 16 April 2010). I think it is freely available on their website and should be easy to find.

Section 7.6.2 of that document is called "Compression Zone Configurations" and discusses different cases that may be of use to you. Basically, the cone would terminate and become a cylinder when it intersects an edge. If one edge is closer than the other, use the smaller one and assume both sides are the same.

I cannot say if it is right or wrong, but some people simply use a cylinder with the diameter equal to the washer (or head or nut) diameter. Also, I believe there are several assumptions in these methods that try to account for the wide range of uncertain parameters involved and be on the conservative (safe) side.

 

[Stress_Eng]

This is one of those occasions where the method tries to simplify reality. The cone is used to simplify the compression zone that, in general, seems to look more like a barrel. As with the cone, the barrel emanates from the underside of the washer and bolt head. Another observation would be, by being close to an edge on one side and a large distance to an edge on the other, why would the presence of the edge alter the compression zone on the opposite side? It is obviously a non-symmetric condition, when compared to the standard case. It begs the question, is it a comparison of compression zone volume?

 

[SpaceEngineer4]

I am familiar with many of the other classical aerospace textbook references like Bruhn and Niu, but Shigley's is the only such textbook I have on my bookshelf so I tend to default to that. I've seen some qualitative suggestions from Sandia National Labs as well, but never anything concrete. The Shigley's method for calculating joint stiffness is also in-line with NASA-TM-106943. I don't have easy access to Lockheed SM88, as I don't believe that is a publicly available document. Regardless, my question is not only applicable to aerospace, I'd be equally as interested to hear how those in other industries handle such a problem (with the understanding that responses from those in other industries will be limited in this forum). Thanks for your feedback everyone.

SE4