INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Jobs

FEA Application for an Aircraft Seat Frame

FEA Application for an Aircraft Seat Frame

(OP)
I thought that other users of CADRE might find this usefull so I decided to provide an example of a project I am working on.  My experience with FEA is limited so I welcome any comments or constructive critisism.

My first project with CADRE Pro is going to be an eighty inch long side facing divan (couch) that carries up to four passengers.  The couch frame consists of 1"x1"x0.035" wall steel square tubing that is mitered and welded together to create a box like frame. The seat belt bolts and floor attachments bolts mount directly to the steel tubes through bushings that have been welded into the frame.

I started modeling the frame using AutoCAD line elements connecting the frame to the attachments the floor and passenger seat belts.  The trick here was a little planning to setup different layers to assign for each type of element that would be included in the model. I used seperate elements for the tubes, bolts, and rigid links that I will use to introduce the passenger loads into the frame. The line elements were broken at all intersections of the frame and where ever an attachment would be added such as at seat belt attachments, floor attachments, and passenger bracing points.  

Since the bolts used to attach the seat belts to the frame are all on one side of the tube and can rotate about the axis of the bolt, I decided to add 0.5" long line elements that extend from the divan frame to the normal location for the connection so that I can use pinned connections for the links to the passenger CG.

Since modeling a passenger and simulating seat belts is a little to complex for my capabilites, I decided to simulate the passenger and restrain system loads by simply creating more lines linking the seat belt attachments on the outboard side of the frame to a point above the frame representing the prescribed CG of the passenger.  Since the first loading condition is a forward loading condition the restrained passenger of a side facing seat would tend to apply more pressure on their forward thigh so I also added a line from the CG to a point on the seat frame directly forward of the CG.

Once the AutoCAD model was complete I exported as a DXF file and imported it into CADRE PRO.  The line elements in autoCAD were imported flawlessly as beam elements and nodes were aded at all the line end connections.  The layers were automatically imported into the element library but they still needed section and material properties.  I used the basic shape editor to generate the properties for type of element and copied the data into the elements that had been imported.  I also changed the default colors to give each element its own unique color.  I used standard beam elements for the tubes and bolts but used pinned beam elements for the links to the passenger CG.

Given that the mass of the seat cushions, belts, and underseat stowage were not accounted for in the FEA model I decided to forgo usin the dead weight loads and simply apply equal loads at each of the passenger CG's equalling the sum of all the masses for the passenegers, stowage, cushions, and frames multiplied by the specified load factor (9g).

I then set the floor attachment bolt locations as fixed boundry conditions and ran the static analysis.  The results wil need to be checked but I think that it gives me a reasonably good simulation using a fairly quick approach.

Steve



RE: FEA Application for an Aircraft Seat Frame

sjbjcl,

I only used Autocad once to see if it worked, and it did.  Other programs I tried did not work as advertised concerning dxf files.  At some point you will probably try creating models with Cadre.  Cadre's 3 point method of identifying nodes is actually a very reliable way of getting the member in with the orientation right.  It takes some getting used to.

As a structural engineer (mostly industrial type structures) I almost always use line elements for my analysis.  This requires that you be aware of end conditions and framing issues.  Are intersecting members fully rigid and are pinned connections really pinned.  Tubes and pipes can be difficult because the cross section doesn't run continuous through the connection.  This effects the analysis and your results.  For critical designs or where design standards don't give any guidance a FEA may be required.

I started my career without the use of computers and can say that structural analysis now is almost fun.  

-Mike

RE: FEA Application for an Aircraft Seat Frame

(OP)
Hi Mike,

Thanks for your input. Having worked on AutoCAD for a long time (release 2), I felt a little more comfortable generating the lines in the CAD package but needless to say I will give CADRE's 3 point method a try.  My analysis work has also been using hand calculations but I have used other FEA packages in the past albeit in limited frequency and duration.

I did put some thought to the effects from the end conditions and how they would affect the results and I know that I will likely need to refine the model and account for the issues you pointed out.  In my first post I also for got to mention that the idea for this excersize is to use this model only to derive the reaction loads at the floor attachment points.  Our industry standards would require us to verify the FEA results via structural test before we could use it as a basis for approval of the divan structure.  

During the course of the day today I also made changes to the way I distribute the load into the frame. I decided to apply loads for the lower frame and underseat stowage areas divded onto four nodes equally spaced about their CG logically positioned relative to the direction of load.  This should provide a better distribution of the loads within the lower frame and reduce the loads at the seat belt attachments.  In addition there are armrest cabinets attached at each end of the divan.  I decided to use traditional static analysis to derive the attachment loads for these cabinets and add them to as point loads at the bolt locations for the armrest on the FEA model.

I am left with a few new questions like:

1- How to adjust the end conditions of certain joints rather than just using a fixed or pinned boundry (K factor?).

2- I understand that the nodes for plate elements need to coincide with those on beam elements in order for them to be joined but is there a way to add nodes on a beam element after the fact?

3- Is there a way for me to upload either images of the FEA model or the FEA files themeselves?

Steve

RE: FEA Application for an Aircraft Seat Frame

I have used Cadre to model an aircraft seat/installation.  It is going through Transport Canada certification at the moment.  I used it to determine all the attachment loads as it is a very statically indeterminate structure.

Nigel Waterhouse B Eng (Hon's)
Can-Am Aerospace,LLC, Canadian Aircraft Certification Centre
www.canamaero.com

RE: FEA Application for an Aircraft Seat Frame

(OP)
Hi Nigel,

Thanks for your post.  Did you use a similar approach to simulate the passenger loads?

Steve

RE: FEA Application for an Aircraft Seat Frame

sjbjcl,

I haven't used CADRE for about a year.  Now I use Multiframe which is directed more towards building type structures and handles multiple load cases well, which it necessary for engineers designing to building codes.  I believe it is based on the stiffness matrix approach and doesn't include plate elements.

Concerning question#1:  I model structures based on the general guidelines in the AISC design spec.  Clip angles and framing angles allow rotation at the joint so they would be modeled as pinned.  Rigid connections are reinforced and/or gusseted per the design spec.  All of this is fairly specific to the industry and would be of much help to you.

Question#2:  I think you can add nodes by splitting a member and then changing the coordinates as required.

Question#3:  There are sites on the internet that allow you to upload files for others to access.  I thought there was one mentioned in faq at eng-tips but I can't find it now.

Regards,
-Mike

RE: FEA Application for an Aircraft Seat Frame

(OP)
Hi Mike,

When it comes to boundry conditions I usually try to evaluate the joints to determine if the joint is fixed, pinned, or if it is somewhere in between.  For example the aircraft OEM's usually define a spring rate for airframe joints if it has certain level of flexibility. CADRE has the ability to allow us to vary the boundry conditions of the beam elements be either fixing or releasing the boundry conditions on a node. I did not see anything that allowed an intermediate setting for the boundry condition (ie K factor).  I wonder how we could achieve this otherwise?

Thanks for the tip on the divide feature in CADRE.  It takes some additional work at add them after the fact but this feature in CADRE it is very functional and easy to use.  I also did the same by dividing the beams in AutoCAD before I did the import.  Both take a fair bit of time and planning to make sure the plates line up properly.

Steve

RE: FEA Application for an Aircraft Seat Frame

Steve,

In my last post under #1 I meant to say that what I do and the spec I use would NOT be much help to you.  I'm sure buildings and airplanes are very different from the design standpoint.

On the CADRE site I see that both Lite and Pro versions have "Elastic (spring) boundary conditions" and a "spring" element, but I have not personnaly used either.  Depending on the implimentation of these capabilities they could be very useful.

-Mike

RE: FEA Application for an Aircraft Seat Frame

(OP)
Mike,

Thanks for the heads up on the spring elements.  I suppose I could use short spring elements at the attchment points.  I will read up these.

Steve

RE: FEA Application for an Aircraft Seat Frame

(OP)
Mike,

It just discovered that the pinned beam elements allow you to define the a fraction between 0 & 1 to define an intermediate level of end fixity.  I think that this is more what I was looking for...  Check out the Pinned beam properties in the CADRE help menu.  It explains it very well.

Steve

RE: FEA Application for an Aircraft Seat Frame

(OP)
I am working on changing the way I represent the attachments to the aircraft.  I have the following criteria:

1- Reaction loads cannot be influenced by the stiffness of the divan frame.  (e.g. Assume that the divan frame is rigid.)

2- I need to account for variations of the support stiffness at each attachment.  (e.g. Assume that the airframe side of the attachment is stiffer in the longitudinal axis than the lateral and vertical axis)

I have seen that this has been done using Nastran usingh individual rod elements in three axes (X, Y, Z). They reported constraint forces for each rod as the reaction loads.  I don't have access Nastran so I posted a quiry in the Nastran forum to see if they can provide me a definition of the rod element properties so I cna better understand how they did it.

Any Ideas?

Steve

RE: FEA Application for an Aircraft Seat Frame

(OP)
After trying out a couple scenarios in CADRE Pro 5.0, I think that one solution is to add three short (1" long) "axial springs" to represent the X, Y, Z directional constraints of each attachment point to airframe.  Then set the boundry conditions at the unsupported end of the spring to be a "spring" with the spring constant adjusted to represent the relative stifness of the airframe in the direction of the load to be reacted and free in all other directions.  

I think that this will allow us to see the effect on the load distribution for an uneven airframe structure stiffness without having the stiffness of the divan frame affect the reaction loads.

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources


Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close