Hi All,

I have had a search through the discussion forum and found one message thread on the sliding mesh problem. However this was from 4 years ago and things have moved on since then I am sure. I am looking to implement a slding mesh technique into a structured multi-block code (and so a sliding plane problem I suppose is more acurate). I am just starting to think about this problem so any information I can find will be useful undoubtedly. I have found previously that this forum is always a good place to start! Are there any papers that can be recommended on the sliding mesh problem (background and approach) for structured grids??? Many thanks in advance, Ryan.

Sliding interfaces is a feature of commercial codes and I assume you will not find much in detail in the literature. To direct your thoughts, the following may help.

A sliding interface between the regions allows the cells on either side of the interface to slide relative to each other in small increments during a transient analyses. The relative movement and connectivity between the two regions can be done by regular attachment or by arbitrary attachment.

Regular coupling: With regular coupling/attachment the cell faces across the interface always have a regular (one-to-one) connectivity. The relative mesh movement between two re-attachments can be made smaller by defining a number of intermediate movement steps. During the intermediate movement the cell faces stay attached while the cells connected to the interface become distorted. The advantage of the 'regular method' above the 'arbitrary method' is that the regular method is faster. The method is however limited to movement of constant increments in a local cylindrical coordinate system.

Arbitrary coupling: With arbitrary coupling the movement increments can be constant or user-defined and movement is not limited to certain coordinate systems. Arbitrary coupled meshes across a sliding interfaces are also allowed to get disconnected from each other. Coupling should take place during the solution phase as well as updating the connectivity of the cells. The arbitrary method allows more flexibility in terms of the number and shape of the cells across the interface compared to the REGULAR coupling method.

Thanks very much for your input Onno, it is much appreciated. As you say, sliding interfaces are indeed a feature of most commercial packages. Since my post I have found a few papers but none give detailed discussion of techniques, as you thought. Your description has certainly made some aspects clearer. I will collect all these ideas and techniques and then think about it further, and then make a start at implementing some option. Thanks again, Ryan.