Methods to analyse a joint in composites

[Statiker57]

Does anyone know of a good method or program to allow me to calculate the MS (or RF) for a bolted joint in a composite structure? I have a multi bolt joint between two sheets that I need to check for all possible failure modes (eg bearing, bearing/bypass, shear-out etc etc) considering the bolt loads and bypass loads. I want to avoid reverting to a FEM if possible. I've worked with some OEMs in the past that had their own in-house methods and programs but I can't use these for this Project so need something that is in the public domain. Thanks.

 

[ESPcomposites]

That is an involved topic. But to help you get started, consider this image from "Practical Analysis of Aircraft Composites". There are two distinct aspects: (1) determine the loading for the isolated element and (2) failure analysis for various checks.

To determine the loading, you can use a program like eDistribution (link below). The fastener and bearing checks are relatively straightforward. However, the bearing-bypass failure is inherently semi-empirical. You can use a public domain tool such as BJSFM, but you will need complimentary test data for it to be useful (and the required test data is not often made public). That is the reason you don't see a "complete" public domain tool for the bearing-bypass failure mode.

Brian

http://www.espcomposites.com

 

[rb1957]

Brian,

Is your book available ? (Amazon says "out of stock")

cheers

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

 

[ESPcomposites]

rb1957, can you contact me directly about that (via the website)? I will see what the issue is.

Brian

http://www.espcomposites.com

 

[SWComposites]

A3,

All of the publicly available methods like BJSFM require test data for empirical factors. You won’t find much publicly available test data. There is a little bit in CMH-17 rev G. And even with FEM methods you will need joint test data. Don’t even think about using lamina properties and text book failure criteria for joints.

 

[Statiker57]

Hi Brian
I saw in one of your posts from a few years ago that you were developing an "eBJSFM" program. Is that available from you?

 

[Sparweb]

A3tdi,
Regarding the program, have you not tried to click on the link in ESPcomposites' signature, follow the page links and download the program yourself? Are you missing something obvious, or am I?

http://www.sparweb.ca

 

[crackman]

A3tdi
I would also recommed the programs by Hart-Smith, A4EI, A4EJ, A4EK. Some of the best and lots of documentation You will still need some data for your material and joint config if it is not covered by the existing datasets.

 

[ESPcomposites]

A4EJ is an option to determine the fastener loading if you perform a nonlinear analysis. This option is noted in the top half of the figure above. Be aware that many times one will opt for a linear analysis. This is for analytical simplicity, less test data is required (you need a nonlinear load-deflection curve for the fastener connection), because composites have less load redistribution than metals, there may not be a significant difference between linear and nonlinear, etc. However, a nonlinear load redistribution analysis may still be performed (and sometimes is).

However, for the failure analysis of the bearing-bypass mode (bottom half of the figure), you will probably find A4EJ to be of limited use. It is for uniaxial loading only and requires calibration. I don't think its main intention was a failure analysis of the bearing-bypass mode. Rather, it is probably better thought of as the fastener distribution analysis analog to the Hart-Smith's bonded joint analytical solutions (which determine the stress/strain it the bondline over the length of the bondline). In other words, we are referring to the fact that the load distribution (fasteners) and stress/strain distribution (bondline) is nonuniform. For the actual bearing-bypass failure mode, most prefer BJSFM because you can allow for a more general state of loading (testing is required to calibrate it). It is also a more physically consistent approach.



Brian

http://www.espcomposites.com