Shop OBEX P1 Docs P2 Docs Learn Events
Faster Servo — Parallax Forums

Faster Servo

GHJN14GHJN14 Posts: 2
edited 2009-01-07 17:33 in Robotics
How can you make a continuous rotation servo faster? yeah.gif

Comments

  • Robert SchwartzRobert Schwartz Posts: 141
    edited 2005-02-11 04:06
    More volts [noparse]:)[/noparse] Unforunately that will shorten its life as part of the process. You could attach an output gear and use gear "reduction" to make it go faster (use a larger gear to spin a smaller gear).

    Post Edited (Robert Schwartz) : 2/11/2005 4:16:46 AM GMT
  • MatthewMatthew Posts: 200
    edited 2005-02-11 06:21
    Or you could just buy a new servo which is faster than your current one. If it isn't a continuous rotating servo, you would have to modify it slightly.
  • edited 2005-02-11 14:16
    Beware of more volts on the Parallax Continuous Rotation servos. 5 and 6 V are okay, but 9 V lets the magic smoke out or a capacitor inside the servo. Then it becomes a paper weight.
  • Beau SchwabeBeau Schwabe Posts: 6,566
    edited 2005-02-12 07:16
    Here is a servo modification that I have done to drive larger motors such as a cordless screw driver.
    Need a quick and dirty ESC (Electronic Speed Controller) ?

    With this mod you can drive larger motors requiring more voltage and more current while still using the
    original hobby servo electronics.

    I usually replace the POT with a multi turn trim pot that I can set and fix.... basically send a continuous
    1.5uS signal and turn the trim pot until the motor stops in either direction and never touch the trim again.



    I have added another image.... "hpd_sm.gif" is an older modification schematic very similar to the
    "ServoMotorModification.jpg" schematic but without the use of a dual opto-isolator. If you use the
    older version you want to make sure that the drive output from the servo electronics pulls the
    signals down to ground when the motor is resting, otherwise you will let the smoke out of the H-Bridge.
    The opto isolator version takes care of this problem and should work for any servo.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe - Mask Designer III

    National Semiconductor Corporation
    (Communication Interface Division)
    500 Pinnacle Court, Suite 525
    Mail Stop GA1
    Norcross,GA 30071

    Post Edited (Beau Schwabe) : 2/13/2005 6:55:36 AM GMT
    403 x 305 - 28K
    1044 x 742 - 21K
    695 x 567 - 103K
  • Paul BakerPaul Baker Posts: 6,351
    edited 2005-02-13 05:53
    That info is a gem, thanks Beau.
  • edited 2007-07-11 05:12
    Dear Beau et all,

    I was hoping to swap a big motor (actually, a SPAL linear actuator) into a circuit that had previously controlled a servo. Those actuators come with potentiometers for position feedback, so I was able to pop open a dead Hitech servo (dead motor, not drive) and wire the Hitech pot connections to the linear actuator's pot. I breadboarded your opto-isolated circuit -- wasn't able to get it running at first with I made a few modifications and made up a board in Eagle (free) . It's single sided and pretty big -- only a couple of between-pad routings -- so most people should be able to etch one using press n' peel or another method. Works like a charm. Note that the eagle parts show the TIP31's and TIP32's laying flat on the board -- they should be standing up. They aren't getting warm at about .5 amps continuous, but it would be easy to add heatsinks.

    I'm attaching the board and schematic files, as well as a printable pdf (for etching) -- just be sure when you print it that it's to scale: The opto-isolators and transistors should have .1 between their pins. I'm not proud of this board -- it was a one-night-hack -- but it worked well for me. The downside of it being one-sided is that it's tough to route all signals. As a result, when building it you have to add a few wires to jumper signals across the board (poor man's second layer!). There should be three jumper wires added to the top of the board: You'll see six green pads with no connections. Just add wires to connect the pairs of pads that are parallel horizontally.

    http://www.trossenrobotics.com/store/c/3103-Linear-Actuators.aspx
    http://www.cadsoft.de/freeware.htm
    http://www.techniks.com/
  • WhitWhit Posts: 4,191
    edited 2007-07-11 13:50
    Thanks Beau - That's really interesting.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Whit+


    "We keep moving forward, opening new doors, and doing new things, because we're curious and curiosity keeps leading us down new paths." - Walt Disney

    Post Edited (Whit) : 7/11/2007 2:27:54 PM GMT
  • D FaustD Faust Posts: 608
    edited 2007-07-11 14:29
    Beau,

    Your schematic shows bipolar transistors, right?· Wouldn't MOSFETs be better suited for the job?(because of a lower resistance)

    Could you replace the right portion of the schematic for the modifiation with an H-bridge IC? (A and·D would be connected to the + side and C and D to the ground side)

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    D Faust
  • Beau SchwabeBeau Schwabe Posts: 6,566
    edited 2007-07-11 15:00
    Wow... this is an old post coming back to life.


    D Faust,

    "Your schematic shows bipolar transistors, right?..." That is correct

    "Wouldn't MOSFETs be better suited for the job?..." Sure, I think this was a case where I just used what I happened to have at hand, but MOSFETS would work fine.
    If you use the Opto-Isolator version you can avoid the headache of creating a "high-side driver" and just replace the bipolars with discrete MOSFETS or use an H-Bridge IC.

    http://forums.parallax.com/attachment.php?attachmentid=37088


    "...(A and·D would be connected to the + side and C and D to the ground side)..."·· - This depends on the H-Bridge IC configuration, some are pretty basic, while others
    might have a little bit of built-in logic.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.

    Post Edited (Beau Schwabe (Parallax)) : 7/11/2007 3:05:40 PM GMT
  • D FaustD Faust Posts: 608
    edited 2007-07-11 17:47
    Thank you for the clarification.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    D Faust
  • SRLMSRLM Posts: 5,045
    edited 2009-01-07 06:38
    @ Beau

    You linked to the same thread... I think you meant to put your reply on the other page.
  • GWJaxGWJax Posts: 267
    edited 2009-01-07 08:05
    Here is a good Hbridge circuit that I use for high current using a LM298N chip.

    Of course I use the Opto's to control it to protect the stamp.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    If a robot has a screw then it must be romoved and hacked into..
    658 x 433 - 29K
  • Beau SchwabeBeau Schwabe Posts: 6,566
    edited 2009-01-07 17:33
    Oops!!!

    I meant to post the link here... http://forums.parallax.com/showthread.php?p=774960



    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.

    Post Edited (Beau Schwabe (Parallax)) : 1/7/2009 5:39:02 PM GMT
Sign In or Register to comment.