i created a few motion simulations with nx and i have the consistent problem, that solving the mechanism often leads to a "Recurdyne Solver Lockup" error message (End of msg-file is always "WARNING : CURRENT STEPSIZE IS TOO SMALL"). This happens in both Kinematics and Dynamics environments in NX 126.96.36.199.
I understand from the documentation that this is mostly because driving the joints quickly leads to instant movement in the joints and huge accelerations. Lowering the speed does not always seem like a feasible solution and often does not solve the problem of the error message appearing.
I have attached a simple example model where this happens:
Do you have additional information that help me understand when this error occurs?
Other question: How comes a 100 seconds solution does not animate in 100 seconds but much shorter?
I have come across your post regarding the solver error in recuDyn simulation. Currently, i am facing a same problem in which solver gives a errror message saying "Current step size is too small". I know it's been a long time for you. But i was wondering if you have found a sloution for this error. It would be of great help for me.
The solver is decreasing automatically the step size to converge and use the error tolerance. The default parameters with Initial Step size 1E-6, max step size 0.01 and error tolerance 0.001 already are sufficient in most cases.
This error "Current step size is too small" is often due when there is a mechanism lock-up. It's a Mechanism Design issue.
The solver message (.msg) gives you the maximum driver position the mechanism can go before the lock-up. You can also animate to that time to understand why a lock-up issue.
BTW, I took a look at the sample sent earlier and that is exactly what happens. For a 100 sec solution, the solver locks at 68.5 sec. In this case, the lock-up comes from slab3 that constraints the movement of slab5 - it's too short here.
Then, by simply editing the sketch, I am actually able to get the full cycle by increasing the slab3 length to 250.
Or you may change to get the pivot point of that slab3 closer to slab5, with same relative angle.
I can send back my data if needed - I use NX10.