I have a set of shafts that need to be rotated at a high frequency (~1000 RPM) but not at a constant speed. That is to say, the angular velocity of the shaft needs to be a periodic function of the shaft's angular position.

This bizarre predicament has arisen in the context of operating an unsteady, transonic wind tunnel. We have a set of vanes whose angular positions determine the Mach number of the flow through the tunnel's test section. For the research we hope to conduct, the test section Mach number will have to oscillate in a specific pattern as a function of time. Technically, there are several ways we could modify the tunnel's design to accomplish this. We are looking into all of them, but in theory, having a motor whose angular velocity is precisely controlled as a function of its angular position would be the most ideal solution to our problem.

I'm no expert in electrical engineering. I understand that some motors, like servomotors, are made to rotate at any desired speed within their operating range, and that their speed can be precisely controlled by a closed-loop feedback system. Does this mean that I might be able to program a servomotor to rotate in such a way that its angular velocity is constantly changing as it rotates, even at average rotation speeds of up to 1000 RPM?

If so, what type of equipment might I need to do this? Is this what a Variable-Frequency Drive (VDF) is for?

If a servomotor is not appropriate for this task, what type of motor might I need?

  • $\begingroup$ Hi Jack, I believe that adding a position feedback and a control loop to your motor can do the trick (within the limits of your desired velocity profile!), although the control loop will not be a simple one. Your desire is to make the motor run at non-constant speed with constant load, when most want the motor to run at constant speed with a variable load... similar after all. I'm not aware of an already-made solution, so I'm not posting an answer, but if you feel like delegating, I may know some people who'd be interested in working on this. $\endgroup$ Jun 29, 2017 at 20:36
  • $\begingroup$ Above I'm assuming that you'd want to be able to change the profile... if you don't, perhaps making a non-circular gear implementing your profile would be a good way. $\endgroup$ Jun 29, 2017 at 20:40
  • $\begingroup$ cJ - thanks for the response. What type of equipment might I need to go about setting up a control loop like this? $\endgroup$
    – plam
    Jun 30, 2017 at 16:58
  • $\begingroup$ non-circular sprockets are actually another option that we've been looking into. They present some mechanical issues though (namely variable tension in the chain), so we're exploring other options as well. $\endgroup$
    – plam
    Jun 30, 2017 at 17:00
  • 1
    $\begingroup$ If you're already using electric motors, they're probably already synchronous motors, and you'd already have the power electronics, all of this which you could keep. You'd have to add in position sensing (optical/magnetic) somewhere in the chain (you may already have speed+torque sensors on the turbine shaft), and it's possible that at your speed, which is not so high (though it'd depend on the profile irregularity), you may get away with a simple MCU (eg. 32-bit / 48 MHz) to fine-control the switching with a profile that could be learned and mostly hard-coded in LUTs. $\endgroup$ Jul 1, 2017 at 16:12


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy