PushOver: problem with steel structure

[LuCanic]

Hi there;

I am running pushover analysis on a ten story steel structure with concentric brace frame V-inverted (chevron-type). To model the instability of compressed brace I introduced an eccentricity at the middle of the brace. The value of this eccentricity is approximately 11 mm (for all brace). The analysis, however, does not converge, it crashes.
It 'possible that the problem is very small eccentricities?
(The length of the span is of 6 meters)
Putting larger eccentricity analysis after some interactions going to convergence.

kind regards.

[huffte]

What type of element did you use to model the braces, LuCanic? They will likely need to be inelastic frame elements with moment releases at joint ends only (as opposed to inelastic truss elements) to prevent rotational instabilities at the middle node where you introduced the eccentricity.

[LuCanic]

Hi,
I used InfrmFBPH for the braces, and i used a "link" element at the ends of the braces with low rotational stiffness to simulate the cinematic hinge.

[huffte]

I have seen similar models work - models with eccentricities built into the braces at midpoint. The models I have seen work employed infrmFB elements with releases at the ends (not at the midpoint node). So you may wish to try such a strategy, though there is no guarantee that this is the problem. Best of luck LuCanic.

[Behnam]

Dear LuCanic
based on experimental formula,L(brace)/300, the 11 mm value is absolutely reasonable.
1.did you do an eigenvalue analysis to check if the model is ok?
2.how did you define the imperfection(eccentricities),naturally you should define the node using "structural node"?
2.please fined attached one of my models
http://www.fileconvoy.com/dfl.php?id=g3 ... 1b03e6b8a9
BR
Behnam

[LuCanic]

Hi Behnam,
Thank you for your reply.
1) I have carried out the analysis of eigenvalues, and the model is ok
2) I have used structural nodes for eccentricity
3) Your model is similar to mine, I'm studying to understand the differences.
I want to ask you one other thing.
On the same model I'm doing time-history dynamic analysis. I noticed that the brace reach a strength value of Buckling (Nb,RD - Eurocodice) greater of the one from which I calculated eccentricity. The isteretic cycle is bigger than I expect.
Do You have done this type of analysis?

[Behnam]

Hi LuCanic
yeah,i have done time-history analysis.
1.are you verifying any special experiment?if so,there could some other reasons apart from eccentricity(i'd rather call it "initial imperfection"!)regarding imperfection pls take a look at 7/28 pdf paging in the following paper)
2.what damping are you using?i suggest using tangential stiffness damping with 2%.(pls find attached the paper regarding this point on it's 23/28 pdf range page)
3.i strongly suggest thoroughly read this paper since it has been done via SeismoStruct and is about steel concentric braced frames.
Please let me know
BR
Behnam
http://www.fileconvoy.com/dfl.php?id=gf ... 7930ec2378
PS:
"The isteretic cycle is bigger than I expect."
have you tweaked the strain hardening parameter,because it directly effects the buckling strength value.

[LuCanic]

Hi Behnam,
I will study the paper as soon as i can.
I will check the strain hardening parameter to chang the the buckling strength value.

Kind regards,
LuCanic