EIGENROOMLAB WORKSHEET 03ANSWER KEY · FOR INSTRUCTORSDC motor · PID tuning
What does P change?
Goal
Measure what a larger proportional gain buys and what it costs when the controller has no integral action.
Setup
Lab eigenroom.com/lab?lesson=proportional · short link eigenroom.com/proportional
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
- 120 rad/s
- Loop
- Closed loop
- Controller type
- PID
- Proportional Kp
- 0.16
- Integral Ki
- 0
- Derivative Kd
- 0
- Anti-windup
- On
- Load torque
- 0.04 N·m
- Sensor noise
- 0
- Controller period
- 20 ms
- Disturbance events
- None
- Recording
- 12 s
- APreset, exactly as the link opens it.
- BChange only: Proportional Kp → 0.24.
- CChange only: Proportional Kp → 1. (ext.)
1. Predict
Will a higher P gain reach the target faster without a cost?
More proportional action reacts harder to the present error. Saturation and overshoot can appear before the target settles.
2. Observe
| Quantity | Where to read it | A | B | C ext. |
|---|---|---|---|---|
| Requested voltage at t = 0 s (V) | Readout at 0 s | 19.2 V | 28.8 V | 120.0 V |
| Speed at t = 12 s (rad/s) | Readout at 12 s | 76.5 rad/s | 87.7 rad/s | 110.7 rad/s |
| Final error (rad/s) | Measured response | 43.48 rad/s | 32.26 rad/s | 9.35 rad/s |
| Control effort (V·s) | Measured response | 87.28 V·s | 97.01 V·s | 116.14 V·s |
Values computed by DC motor and cart-pole engine 1.5.0 at the settings above. A replay readout can differ in the last digit if the slider is not exactly at the stated time; values read from a plot are approximate.
3. Explain
Neither run reaches the target. Why does a P-only controller settle below it, and why is the gap smaller in run B?
With P only, the voltage is Kp × error, so the motor receives voltage only while an error remains. A larger Kp produces the needed voltage from a smaller error: 43.48 rad/s in run A, 32.26 rad/s in run B.
Compare the requested voltage at t = 0 s with the 12 V the motor can receive. What does the larger gain cost that the speed plot does not show?
Run B requests 28.8 V against 19.2 V in run A; both are clipped to 12 V at the start. The larger gain asks for more than the actuator can give and uses more control effort: 97.01 V·s against 87.28 V·s.
Extension (optional)
How large must Kp be to bring the final error below a tenth of the target? Try Kp = 1 and watch the voltage as well as the speed.
With Kp = 1 the final error falls to 9.35 rad/s, but the request at t = 0 s is 120.0 V and the control effort 116.14 V·s. A P-only loop shrinks the error only by asking for ever more voltage; it never removes it.