Composite Analysis Software

[ESPcomposites]

What programs can analyze composites for:

- Laminated Plate Theory (I know of several including "The Laminator")

- Panel Buckling (ESDU has some fortran programs)

- Cutout analysis (Obviously FEM, but think simpler and easier to use. SQ5 in fortan)

- Bolted joints (BJSFM, A4EJ in fortran)

- Fastener distribution in a joint

- Radius opening failure analysis

I am developing composite analysis software in Excel, to be used in the public domain. I find the Excel interface to be more user friendly than company internal software, but not sure what is already available to the public. And fortran is not such a friendly interface. I have a few projects done, but don't want to go overboard. Perhaps it has all been done before. Also, I am not sure if this type of analysis software is even needed or wanted by the community. Large companies already have their own (though often times a bit clunky).

Brian

http://www.espcomposites.com

 

[ESPcomposites]

Is there just little interest in composite software? I would really like to continue developing a suite of tools, but maybe there is no value in that.

Perhaps everyone is just using company tools and third party software is not needed? Is there anything that would be needed?

Brian

http://www.espcomposites.com

 

[SWComposites]

There are some public domain laminate plate theory codes, not much for the other topics. As you say, most big companies have their proprietary codes. Would public, free codes be useful? probably, particularly for folks in smaller aircraft companies.

 

[ESPcomposites]

The software at large companies is good for final sizing, but I have found it be cumbersome for preliminary sizing where several iterations or parametric studies are required. I thought a better user interface would be beneficial, even in those environments.

I suppose as educational software, it could be useful. But if the need is really minimal, perhaps there is no reason to develop a suite of software that is comparable to the big companies.

Brian

http://www.espcomposites.com

 

[berkshire]

ESPcomposites
Brian,
Check out this months Professional boat builder
( October November page 8)
they have an article about Simulayt which is a plug in for Solidworks.
It does a lot of the things that your software does, at I believe, a much greater cost.
B.E.

 

[rybose]

The commercial package HyperSizer does many of those things including CLT, stiffened and unstiffened panel buckling, and BJSFM.

http://www.hypersizer.com

 

[ESPcomposites]

rybose,

Thanks. I worked with them some years back, but that was an effort to develop specific meshing algorithms. Some of my peers have used it for preliminary analysis, but final production analysis is still done with the specific tool for the individual detail. But I will have a deeper look since it is currently available to me.

Brian

http://www.espcomposites.com

 

[cpinz]

Brian,

Simulayt, mentioned by Berkshire, has the same capabilities of Laminate Modeler in Patran, that is, it's a pre-post processor for draping,creating plies, choosing failure theory fo results and so on. It needs a solver , like patran or cosmos (fo solidworks); then I believe it'is of little help for those who need to do quick calculation or parametric studies (..and have no money for so expensive tools)

I agree with SW about the target: small companies could be the right choice for your codes, specially those companies involved in small aircrafts field or simple structures made with composites.

I have seen your website and the collection of your codes: really great job !!
What I would suggest is the implementation of these codes, or part of them, in a new code where some typical structures can be easily and quickly analyzed. For example: wing box,beams like spars with diffeent sections,conical shape structures loaded in torsion and/or bending.
Although they clearly start with airplanes in mind, they could be useful in other fields where such structures are easy to find( wind turbines blades,civil structures where composite are used).

Cpinz

 

[ESPcomposites]

Cpinz,

Thanks for the feedback. Currently, I am focusing on "detail" methods. There seems there is a lack of commercial software that can do detail analysis. Even when considering something as good as Hypermesh, the detail analysis is not as flexible as a dedicated detail model. I have been looking into it further and know a few people who use it for preliminary sizing.

However, I do see your point about bringing the software to the structure level. Simple shapes like beams and cylinders, etc. should be trivial, but something like a wing box would be beyond my current scope. I think Hypermesh would be a better option for something like that. Compared to other software I have looked at, I think I am (or will be) able to provide better detail level analysis solutions, which would be applicable to final sizing.

The problem there is that most small companies probably want a "push button" solution. My software is more geared towards a seasoned analyst, but they usually work for a large company with their own software solutions. My original intent was provide tools that would be used for training, which I think I have accomplished. But I started to realize I can go a lot farther as well. However, I am not sure what direction I will take it from here.

Brian

http://www.espcomposites.com

 

[cpinz]

Brian,

I saw that you use the english system only,in your codes, is it right?
Maybe be a good ( and simple) way to increase the number of user could be implement the choice between English and SI system. Here in Europe it would be appreciated very much !!

Other tip: what's about a code concerning specifically the buckling of sandwich panels (in particular with honeycomb cores). If I well understood , "eStability" considers only monolithic laminates.

Cpinz

 

[ESPcomposites]

Cpinz,

Another good point about the units. I debated about this initially and decided to keep it simple for the time being. However, that can be changed with relative ease. There is a temporary workaround though. Although I specify units, they are often only as guidance for the type of input. In reality, if you use a consistent set of units, most problems should still work (i.e. similiar to how a FEM might expect units).

My knowledge of sandwich structure is only average, but it is something that I am actively working to improve. My plan for eStability is to add the ability to do curved panels, panels with cutouts (including ring doublers). But I can also address a sandwich structure at the same time since that would be beneficial. Fortunately, the approach I have chosen (i.e. FEM based solution) will allow for this. I should also be able to address the transverse shear deformation affect that is more significant in sandwich structure. After I complete the tool set (2 programs left), then I will start on improvements.



Brian

http://www.espcomposites.com