Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Modeling Plates+Fasteners with Holes 3

Status
Not open for further replies.

VN1981

Aerospace
Sep 29, 2015
186
Hi,
I am trying to model a Lap Joint with Holes connected by a Bushing element and coupling elements.

02_ab_w86cks.png


The wire represents the fasteners and is assigned Bushing connector elements with Elastic stiffness (or flexibility) defined. The fastener element is connected to respective plates (top & bottom) via Distributed coupling.

A tensile load is applied to the top plate while the corresponding opposite end is fixed on the bottom plate as shown below:

03_wlxewk.png


However, the analysis gets aborted when I run it. The message files says

"
***NOTE: THE SOLUTION APPEARS TO BE DIVERGING. CONVERGENCE IS JUDGED UNLIKELY.


***ERROR: TOO MANY ATTEMPTS MADE FOR THIS INCREMENT"

I have attached zip file containing the job analysis message file. Also there were several warnings of singularities. I suspect some how there is Rigid body motion happening!

ALso, I am inserting images of my Bushing Connector & Distributed Coupling Settings entry in Abaqus CAE

04_rq4boy.png
05_hkvv7c.png


Note: 1) The above problem setup runs fine in Nastran i.e. using RBE3 MPC with CBUSH (Sring/DOF element) for fasteners. I thought I could replicate the same modeling method in Abaqus.

2) I did a much more simpler Lap Joint model but without holes and used Fastener connection in Abaqus with everything else remaining the same (except for manually created couplings) and analysis ran fine with no errors and expected outputs.

I would really appreciate help with the above issue.

Thanks...
 
Replies continue below

Recommended for you

Can some experienced person suggest what is the best way to model the above setup? Rivet along with some kind of coupling representation in Abaqus? I searched the net but unable to get any pointers.
 
OK, after some more Googling...I found this slide from an old Abaqus presentation (open source. Hosted on Univ of Sydney website).

06_maoioy.png


So, I have used the same method, except that I am using a Bushing connector element instead of beam. I don't Beam allows me to enter flexibility values for different directions like Bushing does. Any suggestions on replacing Beam with something that allows stiffness/flexibility values to be entered. I guess, I can create a Spring element for each direction, but that would drive up the computation costs, although for my simple model it shouldn't matter. Anyways, the intent is to find a single 6 DOF Spring element if I need to do the above in large models.
 
Last try here. I can't be the only one who has faced this issue. I have hit a wall here. Would really appreciate help from folks here if they have experience with the above.

Thx
 
Hello,
I have succeeded a long time ago with a beam and two rigid couplings. The two things I had difficulty with/forgot to do at first was applying the beam section to the line (I know...) and getting the couplings to connect to the end of the beams and not a reference point hanging in the air.

Have you connected your nodes with a ridig coupling to see if there is something else that is causing your convergence issues?

***WARNING: SOLVER PROBLEM. NUMERICAL SINGULARITY WHEN PROCESSING NODE
ASSEMBLY.1 D.O.F. 5 RATIO = 1.85444E+15.


***WARNING: SOLVER PROBLEM. NUMERICAL SINGULARITY WHEN PROCESSING NODE
ASSEMBLY.1 D.O.F. 6 RATIO = 3.94648E+15.

***WARNING: THE INCREMENTAL ROTATION AT THE FIRST NODE OF THE CONNECTOR
ELEMENT NUMBER 1 (ASSEMBLY) IS VERY LARGE. THIS MAY LEAD TO
NONCOVERGENCE OR SUBSEQUENT SYSTEM ERROR MESSAGES.

Where is that? Try checking the "job diagnostics". It could be that your beams/bushings are rotating around the lengthwise axis, or would that be DOF 4 (anyone?)?


Good luck! Hope you get it working.
 
I guess you may try to define the springs (six DOF) between the top and bottom ref points in stead of bush elements. Based on my experience, the connector elements (like bush) with too many DOF free sometimes is hard to converge for static problems.
 
@stefCon.....I am new with Abaqus. Can you help please help me.. how to preload the bolt in a simple model (a beam and two rigid couplings).
 
VishnuVP: It was a long time since I did that. I mostly used solid elements. Perhaps the load called "bolt preload" works for beams too?

Otherwise you can do it by writing in the input file. Maybe this will help you: Link
 
bolt_qr7h0s.png

I don't have any idea how to preload in this model. Normal bolt preload function cannot apply to this model. If you have any suggestion please share it
 
Use a Hinge-Connector instead of the MPC. In the connector you can add a load and later also a boundary condition if necessary. Output is also available for connectors.
 
VishnuVP, I was trying to model a riveted connection not a bolted one. So no issue of pre-load!
 
Do your cbushes have stiffness in all six dof? If not, change any 0.0's to 10.0 and rerun.
 
I think change connector type to "Basic" and chose "Cartesian" and "Cardan" will suit for you,and i often use this way to solve this problem.
 
Hi,

Dear VN1981, I have the same problem like you. I want to to enter flexibility values for different directions such as Bushing. Could you find any solution for this?

Thank you
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor