Currently, I am working with a Four_Nine_QuadUP element and the Pressure Depend Multi Yield 02 material. While i am doing the gravity analysis of a single quadUp element, the top nodes (i.e., node 3, 4 and 7) have a "effective stress" different to zero, which obviously those nodes should have the stress as zero, and the bottom nodes (i.e., 1, 2 and 5) have a effective stress different to the expected one. Curiosily, if you sum the stress at node 1 and at node 3 up, you came up with the real effective stress that the bottom should have.
The stress is recorded at Gauss integration points, not at nodes. Four_Nine_QuadUP element employs 3x3 Gauss integration scheme. Please check a Finite Element textbook for the Gauss point locations. Also, in Four_Nine_QuadUP element, the order of the Gauss points is not the same as that of element nodes, e.g., Gauss point 2 (or material point 2 in recorder) is not near the corner of node 2. --Jinchi
