Control Engineering Basics

In the Ziegler–Nichols ultimate sensitivity method, the gain is raised under proportional control alone until a sustained oscillation appears, and from that gain Ku and the period of oscillation Tu the PID parameters are set to Kp = 0.6Ku, Ti = 0.5Tu and Td = 0.125Tu. Which of the following are the values for Ku = 8 and Tu = 2 s?

  1. AKp = 4, Ti = 1 s, Td = 0.25 s
  2. BKp = 4.8, Ti = 1 s, Td = 0.25 s
  3. CKp = 4.8, Ti = 4 s, Td = 16 s
Show answer and explanation

Answer: B. Kp = 4.8, Ti = 1 s, Td = 0.25 s

Kp = 0.6 × 8 = 4.8, Ti = 0.5 × 2 = 1 s and Td = 0.125 × 2 = 0.25 s. In parallel form, the integral gain is Ki = Kp/Ti = 4.8 s⁻¹ and the derivative gain is Kd = Kp·Td = 1.2 s. The ultimate sensitivity method is a practical method that works without a model of the plant, but the response it gives is rather oscillatory, with the amplitude shrinking to about a quarter each cycle. It is therefore used as a starting point, followed by adjustments such as lowering the gain while watching the overshoot.

  • A. Kp = 0.5Ku is the value for proportional control alone.
  • B. Correct. Substitute Ku and Tu into the given formulas.
  • C. This divides Tu by the coefficients. Ti and Td are found by multiplying Tu by the coefficients.
Practice this topic Back to topic

Related calculators

Books on Control Engineering Basics Ad

As an Amazon Associate I earn from qualifying purchases.

More Control Engineering Basics questions

  1. Which of the following correctly describes integral windup?
  2. Positional PI control is implemented on a microcontroller, with the duty cycle of a PWM signal (0 to 100%) as the manipulated variable. Which of the following is a correct countermeasure against integral windup?
  3. A digital PI controller with a sampling period Ts = 1 ms calculates the integral with the rectangular approximation Σe·Ts. With an integral gain Ki = 10 s⁻¹, which of the following is the increase in the integral term when an error e = 0.2 lasts for 50 samples?
  4. In digital control with a sampling period Ts = 1 ms, the zero-order hold of the D/A conversion acts as a time delay of about Ts/2. With a gain crossover frequency of 100 Hz, which of the following is closest to the phase lag caused by this delay?
  5. A continuous-time first-order system with its pole at s = −100 s⁻¹ is discretized with a sampling period Ts = 1 ms. Which of the following is closest to the discrete-time pole given by the relation z = e^(sTs)?

Comparisons in this topic