Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Control systems

  1. Sep 17, 2011 #1
    can someone please give me some ideas of control system projects which have someinterface with computer?
  2. jcsd
  3. Sep 17, 2011 #2
    Last edited by a moderator: Apr 26, 2017
  4. Sep 17, 2011 #3
    I've found that slow systems and nonlinear systems can be best dealt with in a computer. A good example is an oven controller.
    An oven is typically so slow, that analog components are not suited to the long time constants. In addition, people generally desire the oven to settle at temperature promptly.

    Through utilizing a PID controller, one can get an oven tuned in, but it will either be slow or overshoot (or likely both). Through the magic of programming, you can change the nature of the control algorithm as the oven heats.

    Start with open loop control and a maximum call for heat until the oven reaches a specific setpoint.

    Switch in a PID loop with a reduced D coefficeint and a preset integrator value (such that you are targeting a sever overshoot). Continue this until you are quite close to the target setpoint and then...

    Switch over to a PID loop with sufficeint D to ensure little overshoot and again preset the integrator.

    Not an easy task in analag, but great for a computer. And, it has a practical application in that you can bake your brownies faster :D

    - Mike
  5. Sep 17, 2011 #4
    Good suggestion Mike_in_Plano (and I've been to Plano). I designed a similar controller to keep a tank of molten wax between 170 and 190 deg. The controller could not overshoot 190 because above 190 deg. the wax would deteriorate.

    My design was the reverse of yours. I added D to P without any I, and compared the sum to the reference to prevent overshoot regardless of how much or how little wax was in the tank. At the time D + P crossed the reference, I was added in which not only maintained the temperature at the reference but because it was out of phase with D, it prevented oscillation.

    Clarification: I added the derivative of the measured temperature to the temperature and compared that sum to the reference. This presumes there is a lag between applying or suspending heat and that change being measured. By adding the derivative, the heating element will be shut off earlier if the temperature is rising faster compensating for the time lag. The integral of the set point minus the temperature was added only after the temperature plus derivative reached the set point in order to avoid lengthening the settling time.
    Last edited: Sep 18, 2011
  6. Sep 19, 2011 #5
    The inverted pendulum is always good, or the levitating magnetic balls.
  7. Sep 20, 2011 #6
    levitating magnetic balls look like a good chalenging project but i need a project which have some kind of interface with computer
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook