Need Opinion on Stress Analysis Procedure for Interior Monument Honeycomb Laminates

[VN1981]

Hello,
In some of my previous aircraft interior monuments analysis which are constructed from Nomex or similar Honeycomb Laminates, I've modeled the laminates using smeared or equivalent PSHELL formulation (Ref: FEMCI NASA document). So Nastran will spit out stress values for the entire laminate instead of each individual layer if I had defined laminate using PCOMP/PSHELL formulation. I've used Panel long beam bending allowables obtained from long bending test to check against the Von Mises output from FE for the laminate. Similarly, one can extract the shear running forces from Nastran and convert them to shear stresses and compare against short beam shear allowables which are for core shear failures. Is the procedure is accurate? If no, can a more accurate method of check be suggested i..e how to use panel allowables from long beam & short beam test against FE results? Thanks

 

[SWComposites]

How are your long beam bending allowables defined?
- facesheet stress?
- smeared stress?
- bending moment?

 

[VN1981]

THey seem to be defined for the entire panel i.e. smeared stress as well as bending moment. Strength was calculated from Ultimate Bending Moment.

 

[VN1981]

Hi SWComposites,
Any input on my original question based on my response about how strength is defined for the laminates?

Thx

 

[SWComposites]

smeared stress as well as bending moment.? which one? you need to be very specific about how the allowables are defined. If the are based on a smeared stress, then exactly what equation was used to calculate them?

Are these allowables from your own testing? or from a vendor or customer?

 

[VN1981]

The allowables are obtained from vendor in the following format

Short Beam Shear Test
In the short beam shear test, core compression failures were observed in the first samples. Load plates were added to the loading fixture and holding fixture to prevent this type of failure.

The following equation converts the load from test results to core shear stress:
σ = V/hb, where V = P/2, σ = core shear stress, V = shear load on the panel, h = height of the core and b = width of the panel. The shear load is half the applied load as external load is applied at the center between two supports.

Long Beam Bending Test
The bending moment allowable is quoted in both inch/lbs and lbs/sq. in. The following equiation translates the moment in in/lbs to lbs/sq. in
M = PL/8b
σb = M/tfh, where tf = thickness of facing sheet

Fs is in psi

 

[MOHR1951]

What exactly is "smeared stress",don't mean to sound uninformed,but haven't ran across this term before ?

 

[stressebookllc]

You can learn more about how the sandwich panels are tested and allowables developed here:

http://www.stressebook.com/sandwich-panel-long-beam-and-short-beam-tests/

How the flexure stress equation is derived here:

http://www.stressebook.com/sandwich-panel-flexure-stress-equation-derivation/

And a bit more about the concepts and test coupons here:

http://www.stressebook.com/sandwich-panel-flexure-and-core-shear/

The flexure allowable is at the top of the panel in compression on the facesheet in the middle.
So I suggest you use minimum principal stress from your hand calc or FEM results on the panels to compare against the allowable stress.


Aerospace Stress Analysis and FEA Courses

http://www.stressebook.com

Stressing Stresslessly!

 

[MOHR1951]

Still waiting and hoping to hear from someone with a definition of what"smeared stress" is ?

 

[VN1981]

Thanks for the response folks.

MOHR1951, per my understanding, a honeycomb panel is made of two facesheets and a core. The facesheets themselves may be a laminate made of different plies. In smeared stress, the stress distribution is obtained for the entire laminate as opposed to stresses in individual plies/layers. As you can see, some of the panel allowables are provided for the entire panel rather than individual layers. Having a smeared output makes it easier to make direct comparisons.

 

[MOHR1951]

Thanks for the explanation,VN 1981.

 

[autothesis]

Hello VN1981.
I am in need of using pshell elements for modeling laminates. I am curious how did your analysis worked out. I did a quick study of results between plate modeling and laminate modeling and the results are not comparable. I compared shear stresses as well as vonmises and principal stresses. Do let me know where is your analysis standing and I will update you what my research findings are.

 

[MOHR1951]

Have checked for the term "smeared stress",in the two texts on composite design I have,and all of my structural analysis design texts ,I have related to airframes,nobody even speaks this language.Who came up with or coined this and when ? We need to all be on the same wave length to communicate properly.Stessebookllc,is this some of your terminology,or do you have anything to add.I viewed your links earlier,but did not see the term used.Thanks in advance for your reply.

 

[ESPcomposites]

It is just a term engineers use for convenience. You won't see it used much in technical documentation because it does not sound professional. For documentation, something like "average through-thickness stress" may be used instead (also more descriptive). Also, many composite books of the academic variety focus on the actual ply stresses (not smeared stresses). But in engineering practice, smeared stresses are often more valuable than individual ply stresses (due to methods/allowables development and convenience). In other words, practicing engineers do things a little different than the objectives of many composite text books. Text books are usually better for the understanding of mechanics/principles, but less so for actual design/analysis.

Brian

http://www.espcomposites.com

 

[MOHR1951]

Brian,will you pls. let me know when your text becomes available.It may give me a better feel for this idea of smeared stress,thanks in advance.
george_hobbs@msn.com

 

[SWComposites]

smeared stress = laminate stress = homogeneous stress = P/A. =not lamina / ply stress.

 

[SWComposites]

Or in the case of sandwich structure it ignoes the fact that face sheets and core are different materials and = P/A + Mc/I

 

[MOHR1951]

So,just treat the material as a homogeneous item,no difference in materials,yes,that sure makes things simple from a strength of materials stand point.But is it safe to make such simplifications ?

 

[SWComposites]

No, probably not in general. I did not say it was a good thing to do. But it can work in some cases IF you have test data that exactly matches your structure and loading, and the same calc method is used to get the allowable stress as to get the applied stress.

 

[VN1981]

Hi autothesis,
Sorry for the late response.

Please refer to this document on how to convert PCOMP to PSHELL in Nastran.

http://www.aersys.com/AERSYS-7001.pdf

I have tried the above and I was not able to obtain comparable results to approved hand calculated test cases.

Your query does not specify about the nature of the laminate panels i.e. if they are honeycomb sandwich panels or just plain composite laminates.

If it is the former, then I recommend following the procedure listed in this link:

https://femci.gsfc.nasa.gov/hcplate/honeycomb_pshell.html

I can vouch that FEMCI NASA procedure works correctly and I was able to get comparable & accurate results to an approved document containing hand calculations for simple test cases. Of course, please consult with your FAA DER or any other certifying agency approved signatory about the validity of using the above approach for your project/work.

Hope the above helps you.