geometry imperfection and RESTART card after Riks analysis

Hi all,
 
I am having some dificulties in understanding some issues.
 
1. I am having a beam model, with thickness variable along its length. The beam 
is subjected to an offset load (e.g., 0.3% of the beam length). the question 
is, if I want my beam to exhibit large displacements (elastica), should I 
introduce the imperfections from a *Buckle analysis? Or it's just enough that I 
have the offset load. I tried both cases, and the results of maximum 
displacement on the beam are almost the same.
 
2. I am interested in finding solutions of exact values of load. The issue here 
is to use *RESTART after a Riks analysis, but in another inp file. I have used 
in my former Riks LPF=1 to get the behavior of the beam after I apply one 
certain value of a load. To find an exact solution of displacement under a 
certain load, I have to Restart the Riks analysis with a non-Riks one. I guess 
it can be a Static General one (with NLG on???!!) and I have to stipulate the 
magnitude of the load. The question is: which value I should continue with? As 
I have already done a Riks with my final Load and LPF=1.
 
How can I assess this problem?
 
I would appreciate any help from you, for either of the questions.
 
 
 
Regards,
 
Narcis

                
---------------------------------
Do you Yahoo!?
vote.yahoo.com - Register online to vote today!

[Non-text portions of this message have been removed]






------------------------ Yahoo! Groups Sponsor --------------------~--> 
Make a clean sweep of pop-up ads. Yahoo! Companion Toolbar.
Now with Pop-Up Blocker. Get it for free!
http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/PMYolB/TM
--------------------------------------------------------------------~->