Shop OBEX P1 Docs P2 Docs Learn Events
Limits and abilities of the Parallax servos.... — Parallax Forums

Limits and abilities of the Parallax servos....

GSUIndianaGSUIndiana Posts: 8
edited 2007-08-21 17:08 in Robotics
Heres hoping that I am not out of bounds!

Being my first post I am eager to impress,
so naturaly I am probably completely out
of line!.

So on to my question!

I· have ·4· Continuous· rotation·servo and
one standard servo,· I am curious of a few
things that I have been unable to discover
on my own.····· Can the continous rotation
servos be reverse enginnered to standard
servos and then back again, (depending on
the application) And also can I get the servo
to stop and hold a load steady?· Such as on
a robot arm?

Thanks for any help! GSU

Comments

  • Bruce BatesBruce Bates Posts: 3,045
    edited 2006-03-13 05:37
    GSU -

    Having never needed to do it, this is just a guess. I say probably not easily, or not at all without a good deal of work.

    Parallax sells the continuous rotation R/C servos for $6.95 (web special) and standard R/C servos for $12.00, so there hardly seems a need to reverse engineer them. I've seen the standard servos for even less when they're on special elsewhere. Just by way of example, here is a web site selling the Hitec HS-311 for $8.99:
    http://www.servocity.com/html/hs-311_standard.html

    As far as the holding power is concerned, so long as an R/C servo is being pulsed on a regular basis, it will hold its position. Shortly after the pulses cease, the ability to hold a position is lost.

    Regards,

    Bruce Bates

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    <!--StartFragment -->
  • Tom WalkerTom Walker Posts: 509
    edited 2006-03-13 14:43
    Just to clarify, Bruce was referring to an UNMODIFIED servo in "so long as an R/C servo is being pulsed on a regular basis, it will hold its position." A servo modified for continuous rotation will rotate at a continuous speed in the same situation...not hold its position.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Truly Understand the Fundamentals and the Path will be so much easier...
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2006-03-13 15:46
    Having modifed positional servos to continous rotation, I really believe it can't be done without certain additional parts being provided from th manufacturer.

    Why?

    You have to de-link the mechanical feedback system AND you have to remove the stopping block from the servo.
    These are done by a combination of simple removal and cutting away key areas on existing parts.

    Additionally, a lot of people remove the potentiometer and put in a pair of resistors. Again, it is not so easy to find an identical potentiometer.· The Parallax continuous rotation servo's pot is positioned 90 degrees from the axis it would have to be on for a sucessful modification to a positional servo.

    Without the feedback system, trying to use continuous rotation for positional uses requires setting up an external feedback system. And such a system usually needs to recalibrate each and every time you power up the device.

    Stepper motors really offer more choices in holding power while being able to do both rotation and position in software.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

    ······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan

    Post Edited (Kramer) : 3/13/2006 3:51:12 PM GMT
  • GSUIndianaGSUIndiana Posts: 8
    edited 2006-03-15 13:27
    Thank you! I appreciate the input!· So just recap the standard servos should hold position when the stop signal is applied.· The conitnous rotation are only designed to rotate not stop and hold position.· Ok, again thank you for the feedback.· I have no ideal what I want to do with this information but I am excited about the possibilities!·

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    From···· ·Northwest· ·· Indiana-

    Where·· the · introverts · stare
    at··· there·· ·shoes··· and·· the
    extroverts stare at your shoes!
  • Jon WilliamsJon Williams Posts: 6,491
    edited 2006-03-15 13:46
    C/R servos will stop at one point: when a "centering" command (1.5 ms pulse) is sent.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Jon Williams
    Applications Engineer, Parallax
  • Tom WalkerTom Walker Posts: 509
    edited 2006-03-15 14:08
    A simplistic view might be expressed thusly:

    You'll use the same commands (PULSOUT and PAUSE) for either type of servo.
    For the standard (unmodified) servo, these will dictate to what POSITION the servo will move. Values that indicate a pulse width of greater that 1.5 ms (on the BS2, this is represented by 750) will move the servo output various "distances" away from center one way, while pulse widths less than 1.5 ms will move the servo output various "distances" away from center the other way.

    In the case of a continuous rotation (modified) servo, using the "centering" value (again, a pulse width of 1.5 ms which is equal to 750 on the BS2), will cause the servo to stop. Values less than the "centering" value will cause the output shaft to rotate one way, and values greater than the "centering" value will cause it to rotate the other way. In either case, the SPEED will be dictated by the difference between the value used and the "centering" value...greater differences = greater speed.

    The "What's a Microcontroller?" text does a much better job of explaining this than my blurb...

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Truly Understand the Fundamentals and the Path will be so much easier...

    Post Edited (Tom Walker) : 3/15/2006 2:12:01 PM GMT
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2006-03-15 14:09
    I suppose a better way to say it is that ......
    C/R will stop at 1.5 ms pulse and rotate in one direction when going low, the other direction when going high.
    Where they stop is entirely dependent on when the command is issued and cannot really be easily repeated in software code.

    Positional servos never really 'stop' in a true sense, they hold a position. At the 1.5 ms pulse, it will hold at center. From there you go one or the other direction, but the pulse works with the internal feedback link to hold at a new position. If you were to force the rotor to move, it would move back.
    Where the hold is can easily be repeated by using the same code.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

    ······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
  • GSUIndianaGSUIndiana Posts: 8
    edited 2006-03-16 04:03
    wow! you guys put some thought into answering a new guy's question...... THANKS!!!!

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    From···· ·Northwest· ·· Indiana-

    Where·· the · introverts · stare
    at··· there·· ·shoes··· and·· the
    extroverts stare at your shoes!
  • XNORXNOR Posts: 74
    edited 2007-08-21 16:49
    Bruce Bates said...

    Parallax sells the continuous rotation R/C servos for $6.95 (web special) and standard R/C servos for $12.00,
    so there hardly seems a need to reverse engineer them.

    Hi,

    I was just curious why the continuous rotation servos have almost doubled in price ($12.95)

    www.parallax.com/detail.asp?product_id=900-00008

    Thanks
  • Ken GraceyKen Gracey Posts: 7,392
    edited 2007-08-21 17:08
    XNOR,

    In the past we had a massive (tens of thousands) excess of continuous rotation servos. Excess inventory·is a tremendous risk. Dated inventory has a decreasing inventory value, and accountants want to see it removed from the books quickly. So, we discounted them deeply and below our cost to get them out of our inventory as quickly as possible.

    After two years, Mission Accomplished. We're·now·managing normal inventory levels and selling servos for a small profit at the $12.95 price you see.

    These are based on Futaba S-148s, by the way.

    Sincerely,

    Ken·Gracey
    Parallax, Inc.
Sign In or Register to comment.