dispBeamColumn and BeamWithHing

Forum for OpenSees users to post questions, comments, etc. on the use of the OpenSees interpreter, OpenSees.exe

Moderators: silvia, selimgunay, Moderators

Post Reply
saeedeh
Posts: 9
Joined: Mon Mar 30, 2015 1:58 am

dispBeamColumn and BeamWithHing

Post by saeedeh »

hi
i have 2D concrete farme model with BeamwithHing element for beams and dispBeamColumn element for columns + rotational spring for modeling Kslip in columns,
when i push the frame it can not converged, is there a problem with having 2 different element type for beams and columns? i checked every thing it seemed there is not any special problem with model

p.s: i have a same model with dispBeamColumn as beam element and it doesn't have problem in push load

i will really appreciate it if you help ,
thanks :)
fmk
Site Admin
Posts: 5884
Joined: Fri Jun 11, 2004 2:33 pm
Location: UC Berkeley
Contact:

Re: dispBeamColumn and BeamWithHing

Post by fmk »

mixing and matching is not a problem.
have you a rigid constraint in the floor? if not, make he hinge ends elastic;
if it works
check force-deformation in the hinges elastically to see the magnitude of the deformations your nonlinear hinges are being pushed to.
if it fails
change roaional springs in columns to elastic and check as above
saeedeh
Posts: 9
Joined: Mon Mar 30, 2015 1:58 am

Re: dispBeamColumn and BeamWithHing

Post by saeedeh »

thanks a lot for answering
i have rigid element as beam column joint (elasticBeamColumn element with large E) to ignoring joint effect in frame
i omitted rotational spring in columns but it doesn't work again!
i used beamWithHinges element as :

element beamWithHinges $eleTag $iNode $jNode $secTagI $Lpi $secTagJ $Lpj $E $A $Iz $transfTag
fmk
Site Admin
Posts: 5884
Joined: Fri Jun 11, 2004 2:33 pm
Location: UC Berkeley
Contact:

Re: dispBeamColumn and BeamWithHing

Post by fmk »

a rigid element with large E can effect the ability of the nonlinear program to converge as it effects the condition number of the matrix and also ability of algorithm to converge .. these large numbers should not be picked willy-nilly but some thought should go into what the numbers should be (as nothing is actually rigid) .. if 0 joint rotation is what is wanted, remove the joint totally and only have 1 node at location.
Post Reply