1. This forum section is a read-only archive which contains old newsgroup posts. If you wish to post a query, please do so in one of our main forum sections (here). This way you will get a faster, better response from the members on Motherboard Point.

Stepper motor interfaces

Discussion in 'Embedded' started by Mr_D, Feb 8, 2006.

  1. Mr_D

    Mr_D Guest

    Greetings,
    I'm looking at an under-hood stepper motor interface application,
    however I don't know the full specs of the system yet. In particular,
    I'm assuming there are 2 basic approaches one invlovling the micro
    provide the actual stepper pulses, and another where a dedicated chip
    recieves commands from the micro, and off-loads the micro of the step
    pulses.

    Can someone point me to a good stable source for such stepper interface
    chips, and perhaps tell me a little about their experience with them?

    TIA,

    -D
     
    Mr_D, Feb 8, 2006
    #1
    1. Advertisements

  2. Mr_D

    linnix Guest

    The stepper pulses are relatively slow. You should be able to
    do it with one micro.
     
    linnix, Feb 8, 2006
    #2
    1. Advertisements

  3. There are several considerations.

    If you have a unipolar motor, and want to do full stepping, using
    a very simple micro and n-channel MOSFETS provides a workable system.

    If you have a bipolar motor, you need a bridge to drive it, and
    such a config argues more for a dedicated driver chip.

    If you want to do microstepping, that argues even more for a dedicated
    driver chip.

    http://www.trinamic.com/ - Have not used these, but they look very
    good.
     
    Bryan Hackney, Feb 8, 2006
    #3
  4. Mr_D

    Ico Guest

    Depends a lot on the exact type of motor you're driving. Unipolar or
    bipolar, what current, what voltage, etc.

    I've created several projects with the 3967 from Allegro:

    http://www.allegromicro.com/sf/3967/

    This is a driver chip for bipolar motors, full H-bridge, current
    regulators and microstepping. Driving the chip is easy : one pin for the
    clock for each (micro)step and one pin for the direction. It needs a few
    resistors and capacitors for the PWM current regulators. Be sure to
    provide proper thermal connections to your PCB for cooling. (I know what
    I'm talking about, I fried a few)
     
    Ico, Feb 8, 2006
    #4
  5. That's true, but the software and testing effort needed to get optimum
    performance from a particular motor can be expensive. You need to
    have your program control acceleration curves, watch out for resonance
    frequencies and a number of other functions. You still have do the
    tests and make the decisions with a motor controller, but you don't
    have to write and debug the code that implements the functions.


    If you have the micro do the pulses, you are adding a somewhat
    demanding real-time task to the chores your micro must perform.Mark Borgerson
     
    Mark Borgerson, Feb 8, 2006
    #5
  6. I have a guess that with sufficiently small microstepping, the torque-
    frequency curve flattens out, and the resonant problems disappear.
    Smoothly commutated brushless DC motors (BLDC) have no problems with
    resonance. The smoother the microstep, the more the stepper motor
    resembles a 3 phase BLDC.

    This feature alone is a strong argument for microstepping - one
    that had not occured to me before.

    Aside : brushless DC motors should really be called brushless
    AC motors.

    [...]
     
    Bryan Hackney, Feb 10, 2006
    #6
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.