I carried out a small scale practical experiment to check the computer model. I used a 900x150 pin kerb balanced on a stack of concrete paving blocks 400mm high. When it hit the ground (not a hole) it continued to rotate by only about 50mm from the stone blocks. That's nowhere near what my model had predicted. I went back to the model and found the problem: When a block is pivoting about its end, the moment of inertia is 4 times larger than when it rotates about its middle. I had accounted for this, but had forgotten about the principle "conservation of momentum". What I should have done is to also divide the angular speed by 4 to maintain the same angular momentum. When I made this correction, the model now behaves very similarly to the pin kerb. However, this upsets all the figures I gave in my previous message. So please ignore them.
I've got a fairly busy day today so I'll have to wait until tonight to do any more, but "I'll be back".