Use of FEA in showing FAR compliance 1

[jetmaker]

Can someone provide a reference to the FAR or AC or AMDT that indicates what is required to demonstrate FEA compliance with section 25.307 of the FARs?

Thanks,

jetmaker

 

[rb1957]

AC25-21 has some very general information and pointers to more general info; i suspect that there isn't much defining what is required to support an FEA.

maybe a few more details would help us help you ?

if the structure isn't primary then there shouldn't be much problem. if it is primary, then you need to be more careful. If it is carrying flight loads (from the basic airplane) then you need to account for these, as well as any loads due to your installation.

 

[Aerodesign]

jetmaker, you need to show traceability from your loads source to the post-processed output, through the stages of FEM creation (geometry sources from your CAD drawing tree), derivation of material allowables (material qualifcation tests, MIL, etc), properties within the FEM and how they were derived (fastener stiffnesses, equivalent stiffness for non-descreet items, etc, etc). The next stage is to show how you have applied the loads and verify they relate to the inputs (SPCFORCE balance, etc). You will need to show that the FEM gives balanced loads and does not utilise any automatic constraints (AUTOSPC). It is normal to run your FEM without boundary condisitons (like a normal modes SOL 103) to show rigid body modes and the first ten or so normal modes of vibration. There are several other checks that you should perform on the FEM, some more valuable than others (loading and model dependencies). The internal loads, stresses and deformations that come from the FE/FEA are then used as a basis for the rational stress analysis. It is very unlikely that the FAR can be shown directly from the FEM work, asubstantial work is required, using the fully validated and verified FEM as the source for the stress outputs.

In my experience the FAA , CAA (UK) and EASA all buy off on the FEM/FEA provided adequate v&v is included in the documentation. If holes (in the data) are apparent the whole MoC of the aircraft can be cast in doubt, invariably re-doing the FEM (or a part of it) has significant downstream costs in both time and money.

 

[jetmaker]

Thank you for the replies.

The analysis I am doing is on ECS and electrical support brackets for flight controls. The analysis covers the basic information that Aerodesign refers to. What I was eluding to is it appears that the DER who will sign off on the work wants checks made against theoretical analyses. Just wondering if there was a doc that said all FEMs need to be verified such that load balances are shown, sample checks against theory, output files included as hardcopy in report, etc.

Thanks.

jetmaker

 

[rb1957]

i suspect that the DER is more interested in your specific model (your post seemed to be referring to FEA generalities; surely everyone accepts that FEMs have zero force balance, the catch is where/how does your model react the loads, particularly SPCs and most importantly AUTOSPCs (which you don't control specifically).

i think MSC have a page on quality control checks for FEMs (i'll see if i can find it)

 

[rb1957]

actually its a NASA doc ...
go to ...

http://ntrs.nasa.gov/search.jsp

search for NASA-CR-4675,
most of what it says has already been said, but at least you have a proper reference

 

[Stache]

I ran a check of all the AC’s that may apply and came up with the ones below. I think AC 25.10 explains it best. Something you may want to consider is contacting the FAA MIDO office in your area as they will more than likely have the data or point you to it.

AC 20-107A Composite Aircraft Structure

AC 25-10 Guidance for installation of miscellaneous, nonrequired electrical equipment
This advisory circular (AC) sets forth a method of compliance with the requirements of Part 25 of the Federal Aviation Regulations (FAR) pertaining to installations of miscellaneous, nonrequired electrical equipment in transport category airplanes. It is for guidance purposes and provides an example of a method of compliance that has been found acceptable. It covers FAR 25.307.


AC 25.21 CERTIFICATION OF TRANSPORT AIRPLANE STRUCTURE
Background. Section 25.307 of the FAR was recodified from § 4b.202 of Civil Air Regulations (CAR) in 1964. During recodification, the text was revised for the purpose of technical clarification, but the intent of the requirement remained the same. Prior to adoption of part 4b of the CAR, a similar requirement existed in CAR 04, where § 04.3 required proof of compliance with the strength and deformation requirements of § 04.2.

The Civil Air Regulations (CAR) are on my web site at:

http://www.stacheair.com

under Repair Stations.

Stache

 

[YoungTurk]

25.307 states "Structural analysis may be used only if the structure conforms to that for which experience has shown this method to be reliable."

AC25-10 restates regarding 25.307 "Structural analysis was allowed only if the structure conformed to that which experience had shown the methods to be reliable."

According to the DERs I support, their FAA advisors interpret this to mean that (complex) FEA work must be correlated to test data in order to establish a method which can be considered reliable. The method may then be applied to similar structures.

Given the wide variety of FEA methods (different types of elements, constraints, element sizes, etc. may constitute different methods), the burden of showing correlation to test can be tedious.

I would be very interested in hearing experience of anyone obtaining structural approval via FEA without correlating test results.