Recently I've been working on a simulation of concrete 3D printing, and has successfully simulated a layer-by-layer 3D printing process using model change. However, when I proceeded to try simulations of uneven load during the 3D printing process(which is dividing a layer into like 4 parts and activating them one by one in a clockwise order, then move on to the next layer.), there was an error message even before Step-1 finished, saying that the global mass matrix is singular. So I did as it hinted and set *dynamic, singular mass=warning, so I could finish Step-1, where all element sets except the 1st set are deactivated until their activation step.

However, Step-2(where the 2nd element set gets reactivated) went straight from U1 to U5, there were zero pivot points located between the 2 adjacent element sets, and elements distorting excessively for unknown reasons.

The most blizzard part is that when I consulted one of the senior students who had already successfully finished his similar model, and he sent me his FUNCTIONAL model(.inp), there were still the same error messages that I encountered when I tried to run his model. Could it be that we ran the same model on different versions of Abaqus that led to different outcomes? Mine was Abaqus 2021 and his was 2023.

Here are some of my error messages and zero pivot points.

https://tsunamicoldness.blogspot.com/2024/07/errors-and-zero-pivot-points.html

The entire model was functional until the same layer was set to be activated in 4 adjacent steps.