FEA resources for Laminate Filled Hole Analysis 1

[Burner2k]

While I wait for ESP Composites book to come out, in the meanwhile are there any good resources/material on performing a composite laminate filled hole analysis, especially using Nastran?

Edit: I was going through some theoretical material and I do not understand completely the need to transfer the bypass stress along the draping angle. Is it because that it is assumed that 0 deg ply will carry most of the loads and hence the need to resolve the loads?

Any good material which will address the physics part is also appreciated.

Thx

 

[SWComposites]

Are you refering to a bearing/bypass interaction analysis, or local stresses around the hole due to fastener contact? The former is usually done with empirical interaction curves, the latter is almost impossible to get useful results.

 

[ESPcomposites]

I agree with SWComposites. You are not going to find anything "magical" in my book because there is simply no well established failure criterion for this problem. Instead, you have to consider either a fully empirical or a semi-empirical approach.

For the semi-empirical approach, you can use an analytical solution to assist/improve/expand the results from test data (within reason). For tension, one can analytically combine the effects of bearing and bypass (such as BJSFM). However, note that for the bearing component, the actual distribution is not easily determined. Instead, you can attempt to analytically capture the influence from each component and then calibrate the solution with test data. At that point, you may be able to use the analysis method (within reason) to address certain geometric parameters or load cases that were not tested.

Note that for compression, you may have to use a fully empirical approach. I won't go into details, but the failure mechanisms are different and the compressive bearing pressure does not allow you to use the resulting local compressive stresses in a meaningful manner.

Another thing to consider is that a lot of the "pure" analytical solutions that work for ductile metals do not carry over to composites (although it has been attempted many times). So you may have to think about composites in a different manner that requires more testing and semi-empirical approaches. This is largely due to the effect of notch sensitivity for composites. They are neither brittle nor ductile and require test data to accurately determine the effect of "pseudo-plastic" behavior. Unfortunately, there is no failure criterion that can accurately determine this (at least not one that is well established).

Brian

http://www.espcomposites.com

 

[Burner2k]

Yes,
I am interested in learning bearing/bypass interaction analysis. I would like to get a better understanding of the above for composite laminates at least some basic physics part of it.

I am reading through the procedure to be followed for extracting FEM results from 2D elements (shell) around a fastener representation. I have requested a training session with the vendor to get a better understanding of the procedure, but I was just wondering if any folks who is already familiar with the process might chip in.

 

[ESPcomposites]

Take a step backwards. There are two aspects to the analysis. First, you develop the bearing and bypass loads for the isolated element (fastener location). Second, you analyze the isolated element for the bearing-byapss interaction. And then there is the physical behavior of why there is a bearing-bypass interaction, which is a different topic. Learning how to extract the bearing and bypass loads from a FEM is not going to help you with with the physical understanding (its just a procedure). You might want to make your question more specific. Are you interested more in the physics of bearing-bypass, how to determine the bearing/bypass loads, or how to analyze the isolated element?

Brian

http://www.espcomposites.com

 

[Burner2k]

Hi ESPComposites,
Thanks for your patient replies.

I would like to start with Why i.e. physics of bearing-bypass interaction (I have a decent understanding of this phenomenon in metallic structures) and then proceed to understand the How i.e. analysis part.

I am in no hurry or rush. I can spend some time going through any material which are referred.

 

[SWComposites]

Track down a copy of "Mechanically-Fastened Joints for Advanced Composites - Phenomenological Considerations and Simple Analyses" by L. J. Hart-Smith

and

"Stress Analysis and Strength Prediction of Mechanically Fastened Joints in FRP: A Review" Camanho and Matthews

also

McCarthy, M. "BOJCAS: Bolted Joints in Composite Aircraft Structures"

and there are hundreds of other papers and reports related to bolted joints in composites, some more useful than others.

 

[ESPcomposites]

That is one of the better H-S papers for fastened joints and does a good job at explaining the factors that influence the bearing-bypass failure mode (tension only). But you need to be aware that the "C factor" is a function of many variables and must be determined via test. In effect, this is an application of the PSC (point stress criterion).

For the Camanho and Matthews paper, be aware that the use of ply based criteria is inappropriate for fastened joints since the damage mechanisms are much more complex than that. And the use of progressive damage analysis is still predominately a research topic and has not been as successful as hoped for aircraft applications (that I am aware of). Same can be said for the BOJCAS paper. I would be interested if any companies are using a 3D physics based failure criterion, combined with damage progression, to do final analysis.

Some others that I think are decent. There are quite a few by Hart-Smith that can are useful:

Nelson, W.D., Bunin, B.L. and Hart-Smith, L.J., “Critical Joints in Large Composite Aircraft Structure”, NASA Contractor Report 3710, Contract NAS1-16857, August 1983.

Hart-Smith, L.J., “Bolted Joint Analysis for Composite Structures – Current Empirical Methods and Future Scientific Prospects”, Joining and Repair of Composite Structures, ASTM STP 1455, K.T. Kedward and H.Kim, Ed.(s), ASTM International, West Conshohoken, PA, 2004.

Garbo, S.P. and Ogonowski, J.M., “Effects of Variances and Manufacturing Tolerances on the Design Strength and Life of Mechanically Fastened Composite Joints,” Vol 1, 2 and 3, AFWAL-TR-81-3041, April 1981.

Brian

http://www.espcomposites.com

 

[Burner2k]

Thanks folks for all the replies. Will have to go through the references one by one.

Meanwhile, any body have an idea of equivalent bearing-bypass stress calculation? I want to get some background on the caclulation. Just learnt that a some chunk of our work will involve the above (from FEA).

I have attached link to a company white paper which describes the process albeit not in great detail! The paper is about analytical approach for fatigue analysis of a laminate filled hole.

[link ]

https://www.quest-global.com/wp-con...led-Hole-Locations-An-Analytical-Approach.pdf

[/url]

 

[SWComposites]

Aircraft composite bolted joints properly sized for static loads are typically not critical for fatigue, so why worry about it?

The equivalent stress approach does not work. Have tried to correlate that and similar things and they dont work, unless one applies a very conservative factors. You need test data.

 

[Burner2k]

Awesome SWComposites. These are the types of insights which makes this community such an asset!

The work we are contracted to do is mainly Certification calculations for a major OEM. The laminate joints may not be critical in fatigue but I guess for certifying purposes, a positive Margin needs to be shown and I think equivalent stress approach is recommended by our client.

I would still like to learn the background of the approach. I tried searching on net but was unable to get good references.

 

[Burner2k]

So, I have access to one of the client’s composite laminate filled hole static stress analysis document. The following steps are to be followed:

1) Calculate Gross Running In-Plane Loads based on pitch corrections. The In-Plane loads are defined with respect to Laminate Global (or Reference) Coordinate System.
2) Calculate Net Running In-Plane Loads based on the running loads from Step 1
3) Transform Net Running loads from Step 2 in to a Coordinate system specific to Bearing Load.
4) Add the running loads due to bearing to the loads from Step 3
5) Transform the running loads from Step 4 in to their Main Coordinate System
6) Running loads from Step 6 are multiplied by Factor Kt
7) Running loads from Step 6 are transferred to Laminate Global or Reference Coordinate System
8) A CLT calculation is performed and ply stresses & strains are calculated
9) A Failure Criteria may be used to determine Failure Indices

The document does not provide detailed explanations. Currently, I have no idea what Main Coordinate System, mentioned in Step 5 means. I don’t think it is the orthotropic coordinate because it is defined separately.

Is the procedure similar in programs like BJSFM etc, which use point stress criterion to identify failure regions?

 

[SWComposites]

Um, why don't you ask the client?

I am very familiar with several OEM's methods for bolted joint analysis, unfortunately not this one. Issues I see:
Step 2 makes no sense; typically gross area stresses/strains are used.
Step 4 is very vague; seems to assume bearing loads are only critical for bypass, which is not correct
Step 5 - I have no idea what Main Coord System is (only guess would be directly of max principal tension stress)
Step 6 - using same Kt for far field bypass as for bearing induced bypass is incorrect
Steps 8, 9 - the result here is similar to what is calculated by BJSFM - local peak stresses or strains near the hole

Again, you need to ask the client for details and some example problems worked out in detail.

 

[Burner2k]

SWComposites,
Currently, I am kinda too low in the project hierarchy ladder to have direct interaction with our client. My boss who is a delivery kinda person just insists on "following" the steps and deliver rather than to try understand what they mean. The calculations are performed on a client supplied highly validated tool...so its like a black box...just feed in the required parameters and it throws out a detailed MS summary for various failure modes.

Anyways, those are my problems and I need to find a way to resolve it. May be as the project proceeds further, I'll get a chance to interact with client and clarify my questions.

You are correct that the overall procedure seems similar to BJSFM. I am currently going through some old BJSFM documents to familiarize myself with the overall procedure.

Thanks for your replies anyways. Appreciate them.

 

[SWComposites]

Brilliant. Aircraft analysis by plugging random numbers into a black box and outputting more random numbers, probably to 12 decimal places.

 

[Aerodesign]

Did anyone mention ISAMI?
40,000 flight cases, 12 thermal flight cases, 3 material conditions, about 15 failure modes, at least 3 loops of load = 25 TB of results data.
Stress engineers become data controllers.
Welcome to the new world.