Torsional constant for Complex Mullion sections

[sutki1]

I am looking for an efficient and quick way to calculate torsional constant,J for mullion sections which are unique are highly complex (

http://www.vistawall.com./details/fastnav/fastnav_frames.htm

u can download from this website. I am dealing with a variety of sections.
It is time consuming and inefficient to hand calc the J(torsional constant). A worksheet wont work as each section is unique.

The values of area, moment of inertia, radius of gyration can be calculated for these complex sections in matter of seconds using AutoCad, it is the J, "torsional constant" I am have problems with.

What I really am looking for is a program that calculates J for non conventioanal sections, or if AUtoCad has a LISP developed that covers J.

 

[GregLocock]

FEA is the fastest way, and to be honest it is the only really practical accurate way if the section is thin walled and multiply reentrant, for example. The assumptions that classic beam theory makes overestimate J for these sections.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[chicopee]

MF Spotts "Design of Machine Elements" has a section to determine the torsional J constant of composite sections. You probably can develop a similar methology on a spread sheet by using his formula in the general case and then applying the formula to a composite section.

 

[sutki1]

Which Finite Element program can import a section from AutoCad and calculate the section properties?

 

[eromlignod]

Are you talking about the polar moment of inertia? If so, then it is just

J = Ix + Iy

Where Ix and Iy are the moments of inertia about the x and y axes through the centroid.

Don
Kansas City

 

[GregLocock]

Don't think so. Not for thin walled sections. Plane sections don't remain plane, for a start.





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[SilenceJes]

I downloaded a demo version of Section Maker (just do a search for section maker and it will show up).

To use the program, you have to region a shape in AutoCAD and then explode the shape and save the AutoCAD file as a DXF file.

Then open Section Maker and do file - import - then you will be allowed to change the material of the shape to Steel, Aluminum, etc - then move the shape to the centroid and it will have a table listing the J value (along with the Ix, Ix, Sx, etc values of the shape).

 

[Qshake]

As a structural you learn not to confuse the polar moment of inertia with the torsional constant.

It's likely that the mullions are cold formed elements and if so, I would defer to the Cold Form Design Handbook or texts by Wei Wen Yu on the matter. unfortunately most cold form design is about calculating section properties.

Regards,
Qshake

Eng-Tips Forums:Real Solutions for Real Problems Really Quick.

 

[SilenceJes]

I am looking for a vay to verify the torsional constant obtained from the demo version of Section Maker I downloaded.

I have Mechanical AutoCAD 2006 which has a Finite Element program in it - will that give me the Torsional Constant of complex mullions? If so, how do I get it to do this?

-SilenceJes

 

[GregLocock]

Create a beam with that section, fix one end, apply a torque to the other.

??????????????????????????????????????

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[btrueblood]

No, AutoCad 2006 is a 2D program, and its FEA is similarly only for plane stress (2D).

You need a 3D FEA package, for reasons that Greg gave (you have open/reentrant/non-circular sections which will warp when twisted). AutoCad Inventor Professional comes with a bundled "lite" version of Ansys FEA, for instance, if you wanted to stay with Autocad (but you'll have to learn how to use Inventor - not hard, and you can import the 2d section from Autocad directly and extrude it to form a beam). Similar products exist for ProE and Solidworks (SW can also import autocad files). And NEi offers a good price for their stand-alone FEA modeller.

 

[sutki1]

The best way to verify is to use a steel handbook. The handbook lists the torsional properties for a variety of sections. That is how I would go about doing it. I did download the program but one problem i am encountering is that almost all of the Aluminum sections for mullions I deal with are closed formed. When I import the dwf file it does not recognise the thickness but reads the outer and the inner polyline as different sections.
In other words how do I use the AuotCad equivalent command for "SUBTRACT".

 

[arto]

look in roark:

http://virtual2002.tau.ac.il/users/

http://www/34085/Roark's Formulas for Stress and Strain (7th Edition)/2542x_10.pdf