In the Advanced Simulation help, section Mesh mating conditions it is written:
Free coincident mesh mating conditions With a Free Coincident condition, the software aligns the mesh on both the source and target face but does not connect the meshes. With Free Coincident, this results in duplicate nodes at the interface between the source and target faces. This is useful, for example, for setting up surface-to-surface contact problems. See Surface-to-Surface Contact overview for more information."
In the Nastran NX help, section Surface Contact for SOL 101 paragraph Understanding Source Regions and Target Regions I can read:
"In general, of the two contact regions you use for the pair, choose the one with the finer mesh for the source region. When the source and target regions have different mesh densities, more elements on the source region will mean that more contact elements are created, which will produce a more accurate solution."
So it seams that for a better solution it is desirable that source and target regions have different mesh densities.
Why then is it suggested to create a free coincident mesh on the two surfaces???
The guidelines you quote are somewhat independent of one another. For example, I've often heard that ABAQUS surface-surface contact works best (converges faster, etc.) when the contacting faces have coincident meshes. Also for any contact between curved surfaces there's potential for undesired interference simply due to differences in the meshes. If the meshes are identical, there will be no interference, therefore no induced contact due to the mesh dissimilarities.
The NX Nastran guidelines really apply to situations where the meshes are different. It is simply stating that when this is the case, it is best to choose the finer mesh as the source for the contact. That is because contact elements are formed from the source element faces and projected to the target element faces along the normal to the individual source contact faces. Following this guideline assures the formation of the maximum number of possible contact elements. Having more contact elements increases the accuracy of the contact. The NX Nastran guideline is not suggesting that it is best to have dissimilar meshes between the source and target faces.
Dear Gisseppe, As an add-on to perfect Mark comments, the use of "free coincident mesh mating" feature is when you plan to define explicit CGAP elements node-to-node contact. This is the classical 2-node GAP contact where the user mesh explicitly the contact elements in the preprocessor, prescribing the gap stiffness (compression or tension), gap distance, and orientation.
I use a lot CGAP elements when dealing with contact between Shell elements, is simply an practical, avoid problems with shell OFFSET (currently a bug problem in NX NASTRAN 7.1, already solved in NX NASTRAN V8.0), but requires both 2-nodes of CGAP element to be aligned in the direction of contact force, works very bad when transverse motion is developed between contact nodes. In this case modern contact surface-to-surface feature is welcome.
Thanks for your kind answer.
I'm not going to use CGAP elements because I want to use the Surface-to-surface contact by adding an offset to both the surfaces (pin fitted into a hole with interference).
Should I use the free coincident mesh mating also in this case?
Not neccessary, but I suggest always to use the same mesh density in both SOURCE & TARGET regions.
Also please caution, there is a bug in NX NASTRAN 7.1 regarding the use of OFFSET with Shell elements and contact surface-to-surface, the bug is solved in NX NASTRAN 8.0, I reported this error when considering contact between Shell elements. Here you are the error:
2161330 When the shell offset feature is used for shell elements like CQUAD4 and CQUAD8, valid results are only produced if all of the shell elements that have contact defined have the same offset value or if the ORDER of the shell elements that have contact defined does not change when the shell faces are sorted.