A good margin, a bad run
Why can the frequency plot look acceptable while the motor overshoots?
Try it in the current lab
Compare the preset at 120 rad/s with a smaller target, holding controller gains fixed. Report when requested and applied voltage separate.
Open Bode and Control effort. Pin the run, reduce target until voltage no longer saturates, and compare requested with applied voltage.
The lab recalculates after a setting changes. Pin a run before changing one variable; the response and control-effort plots share the same replay time. PID experiments also show the P/I/D terms; estimator experiments show measured and estimated states.
What the system is doing
The frequency workbench analyzes a nominal linear small-signal loop. A ±12 V actuator limit changes the large-signal time response, so a phase margin cannot certify this saturated run.
Connect the design method
The sampled nominal Bode and Nyquist plots use a linear motor loop. A large target can demand more than ±12 V; then requested and applied voltage part company. Compare a smaller target at identical gains. Treat phase margin as a local design indicator, not proof for the saturated run.
phase margin = 180° + ∠L at |L| = 1