EIGENROOMLAB WORKSHEET 19DC motor · Design challenges
Design model versus actual motor
NameSectionDate
Goal
Measure how a controller tuned for the nominal motor behaves when the real inertia or friction differs.
Setup
Lab eigenroom.com/lab?lesson=model-mismatch · short link eigenroom.com/model-mismatch
Open it on a phone or a computer; there is nothing to log in to or install. The link loads run A's settings:
- Target speed
- 95 rad/s
- Loop
- Closed loop
- Controller type
- PID
- Proportional Kp
- 0.22
- Integral Ki
- 0.6
- Derivative Kd
- 0.008
- Anti-windup
- On
- Load torque
- 0.04 N·m
- Sensor noise
- 0
- Controller period
- 20 ms
- Actual inertia
- 1×
- Actual friction
- 1×
- Disturbance events
- None
- Recording
- 12 s
- APreset, exactly as the link opens it.
- BChange only: Actual inertia → 2×.
- CChange only: Actual friction → 2×.
Pin each run before you change a setting, so you can compare runs at the same replay time.
1. Predict
Will a controller tuned for the nominal motor behave the same if inertia doubles?
Write your prediction and a reason before you run anything.
2. Observe
Read each value in the lab and write it in the table.
| Quantity | Where to read it | A | B | C |
|---|---|---|---|---|
| Rise time, 10–90% of target (s) | Measured response | |||
| Overshoot vs target (%) | Measured response | |||
| Settling, ±2% of target (s) | Measured response | |||
| Control effort (V·s) | Measured response | |||
| Phase margin (°) | Frequency workbench |
3. Explain
Doubling the actual inertia: what happens to rise and settling time, and what does the frequency workbench report?
Doubling friction instead reduced the overshoot. Does that make the controller robust? What would you test before claiming it?
Extension (optional)
Try an actual inertia of 0.5×. Does a mismatch always make the response worse?