The accompanying picture is of a structural linear stress analysis under a temperature/differential thermal contraction load.
One stainless steel plank [green in attached jpg] is sandwiched between two other planks of a differing steel [gray]. These two materials have slightly different mean coef. of thermal contraction, so if both ends' surfaces are (independently) tied together such that all three planks' ends move as one in the X-direction, the more shrunken steel (green) will be held in tension under the less shrunk steel planks's ends. (Of course, the gray steel planks will reciprocally be under compression via the green's relatively tugging both its ends together.)
I know how to tie everything together at the far end surfaces using general constraints in Tx, Ry, & Rz.
(Also, I'm using a sliding/no separation contact for both touching surfaces and 'anchoring' the whole assembly to one node's fully fixed general constraint so the whole thing doesn't go spinning off into space.)
What I'm after is how do I keep the pictured near ends' surfaces tied together (with regard to Tx, Ry, & Rz) while still allowing them to displace (as one) in the X direction under their individual pieces thermal differential contractions' loading?
FWIW, my model's default temperature is 70F, while its load temperature is -320F. (i.e., the model represents chilling the assembly down to -250F.)
Thanks!
