Affect of Adjusted Nodes on Results

[Throwawaynerd]

Hi all,

I am designing ultrasonic transducer in Abaqus. I am performing modal analysis and I encountered the following Warning message (560 nodes have been adjusted. The nodes have been identified in node set WarnNodeAdjust) during Data Check before submitting my job.

I checked the mesh quality and it seems quite well and I do not see any kind of analysis warning during mesh verification.

When I submit my job I see Numerical Singularity warnings during analysis. The warning reads :

Solver problem. Numerical singularity when processing node BOLT-1.70 D.O.F. 3 ratio = 15.4389E+09 .

Solver problem. Numerical singularity when processing node BOLT-1.55 D.O.F. 1 ratio = 66.9964E+09 .

Solver problem. Numerical singularity when processing node BOLT-1.55 D.O.F. 3 ratio = 45.9110E+09 .

Solver problem. Numerical singularity when processing node BOLT-1.44 D.O.F. 1 ratio = 174.920E+09 .

Solver problem. Numerical singularity when processing node BOLT-1.44 D.O.F. 2 ratio = 203.448E+09 .

Solver problem. Numerical singularity when processing node BOLT-1.44 D.O.F. 3 ratio = 117.101E+09 ."

The first three numerical singularity warnings disappear after the first increment while the last three warnings appear till the end of simulation.

When I modify the mesh, the Numerical Singularity appears at some other location of the model.

The analysis also gives me warning of Negative Eigen Value.

I tried to constrain the respective nodes at the respective D.O.F but still the warning appeared.

It has made me confused whether or not would this affect my simulation results?

P.S, These are warning not errors, the simulation does complete and give me results. However, some Eigen Modes are very close to each other and this make me think about the importance of warnings.

Please give me some insight to resolve this issue.
I have also attached pictures for your reference.

Thanks in anticipation.

 

[FEA way]

Not all warnings indicate something to worry about but it's good to read and understand them all. The one regarding adjusted nodes is purely informational but you can check if the tie constraint works properly. Open the results and highlight this node set. Visually check the connection - if there are no disconnected regions.

 

[Throwawaynerd]

There is a gap but that is because of electrodes. As in actual design, the bolt and electrode will be insulated but I did not consider insulation in the current design, and I think it won't affect the results much.
These warnings now appear only in the static general step where bolt load is applied. And this singularity message appears in the region where I created surfaces to apply prestressing load.
When the simulations switch to the harmonic analysis step, there are no warning messages related to numerical singularity and negative eigen values.

 

[Throwawaynerd]

I keep receiving this error in the Modal and Harmonic analysis of the design. "THE ELECTRIC POTENTIAL WILL NOT BE CONSTRAINED FOR CONTACT PAIR (General_Contact_Faces, General_Contact_Faces)."
I have same potential, i.e., zero, during Modal Analysis and different potential across PZT during Harmonic Analysis.

 

[FEA way]

Isn’t that just a warning, not an error ?

 

[Throwawaynerd]

Dear FEA Way, That's true, these are warning messages rather than errors.
But I am just wondering that would these warnings affect the results of simulation and can the results be trusted with these warnings?

Thanks,