1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Stopping time for a cycloidial dial table

  1. May 28, 2015 #1
    I am working on a Risk Assessment for a dial table, the manufacturer states that the indexer will stop in 0.86 sec, worst case. Since the indexer is cam driven the speed varies throughout its motion. That being said I expect stop times to be highest at 90 and 270deg, while lowest at 0 and 180deg. My thinking is to multiply the SIN(r)*θ. Am I on the right path, or off on a tangent.
     
  2. jcsd
  3. May 28, 2015 #2

    Baluncore

    User Avatar
    Science Advisor

    Welcome to PF.

    Please give a diagram of the cam mechanism. What is the reference direction relative to the cam?

    You cannot take the Sin(r) of a dimensioned radius, only of a dimensionless angle. Sin(θ)*r would make more dimensional sense.
     
  4. May 28, 2015 #3

    berkeman

    User Avatar

    Staff: Mentor

    What kind of risks? Risks that the workpiece will be ruined for some reason, or risks that the operator can be hurt?

    When you say the indexer will stop in 0.86s worst case, that's 0.86s from what?
     
  5. May 31, 2015 #4
    It sounds like the manufacturer quoted you the time to disable power to the driving motor and engage the motor brake. If no brake, then stoppage occurs due to cam motion friction between the mechanical components.

    If they did their engineering sizing calculations properly, the motor is matched to the specified rotating payload mass & inertia on the dial table. So it follows that the brake is properly matched also. On ESTOP, cut power to motor and clamp on the brake...rotation stops in 0.86 seconds. Seems reasonable. But the cost to do that is likely damage to the mechanical components if the rotational mass payload is spinning at max speed.
     
  6. Jun 4, 2015 #5
    Thank you all for prompt reply.

    Baluncore,
    The move profile is sinusoidal, this type of indexer is common in the manufacturing industry, it give a nice slow start/stop at the pitch, this system has two positions. Thanks for correcting my math.

    Berkeman.

    Risks in my business is prioritized to people, tooling, part. I am assuming the .86 represents from max velocity, since the profile is sinusoidal, ithink the velocity varies throughout the index cycle, at this point I assume at 90 deg intervals.

    Tigerdawg.

    The stop time is based on power disconnection from the motor driving the thing, the time to disable is the system response time. This includes the actuation of the stop signal, fieldbus (Ethernet) transport time, processor input update time, logic execution time, I/O faults within the system, output update time, and contactor opening time (power removal). Under normal stopping conditions the drive (VFD) supplies regenerative braking, when the dial is in dwell. Under the operator protection scenario, the contactor rules.

    I am reasonably sure the calculations are correct, this is a common system, and the technology has been around for a long time. Braking is common, and works just like the brakes on your car, no damage unless you hit something.

    All,

    I am using worst case .86 seconds for stop time, but my intellectual curiosity always overpower me and I have to know more. Interestingly enough I find the response time of a contactor with an integral diode, is faster that one with an external diode. The diodes are used to 'snub' counter EMF. Does anyone have an explanation? Maybe some additional inductance, or slower field collapse?
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Stopping time for a cycloidial dial table
  1. Thermodynamic tables (Replies: 1)

  2. Thermodynamics Tables (Replies: 5)

Loading...