Giant servo question....

krokikroki Posts: 21
edited 2005-02-07 - 19:11:56 in Robotics
Hi!

I have two Giant scale, very strong (343 oz/in) servos, and i'm wondering, if i could use them, a GPS and a Stamp to control a 1/5 scale car to go on a specific route?

I searched the archives for servo projects, but every one of them uses modified servos.
I have a BS2sx and a BS2p24, and i don't want to modify the servos, because if anything goes wrong, then if a servo as strong like this goes over the "end" then it can brake many things.

in a few words:
Two srtong servos, a radio reciever, a 1/5 scale car, a BS2sx/p24, and a controller.

Please help me!

Many thanx

p.s.: i will use the Stamp to watch the voltage too:-)

Comments

  • MatthewMatthew Posts: 200
    edited 2005-02-07 - 14:52:10
    How complex is the route? Are you going to use sensors to moniter obsturctions?

    Are you trying to race your car without controlling it on a specific track [noparse];)[/noparse] ?
  • allanlane5allanlane5 Posts: 3,815
    edited 2005-02-07 - 14:55:08
    Servo's were created to be used at the end of an RF link to move a control linkage through an arc. As such, they are optimized to allow you to select a position (zero to 180 degrees, typically) of the output shaft.

    A 'Modified' servo is modified so that it no longer allows you to select a position. Instead, the servo can run continuously (like a wheel) and you select the speed of rotation. Any time a servo runs a driving wheel, it MUST be a 'modified' servo. The nice thing about a 'modified' servo is that a BS2 can control them very easily, and the price is good too (compared to every other speed controlled motor approach).

    So, if you are going to put them in your car, what is your car going to use for moving forward? Does it already have electric motors?

    When 2 servo's are used in a robot, typically differential speed is used for steering. This wouldn't work very well in a model car. What you need is one powerful servo to drive the driving wheels, and one un-modified servo to steer.
  • krokikroki Posts: 21
    edited 2005-02-07 - 15:08:17
    allanlane5:

    I am a nitro modeller, and i use nitrous engine for driving the car forward, and the servos are for pulling the gas on the carb. and the brakes, and for steering the model.
    So i need to get the servos working and positionable.
  • LarryLarry Posts: 212
    edited 2005-02-07 - 15:16:14
    Take a look here http://www.stoneflyers.com/gps_guided_truck.htm
    for someone's attempt at this problem. There are a couple problems with this approach. The heading· and bearing only update once per second in most GPS units. so most of the time is spent guessing where you are going.Also, since the Stamp spends most of its' time in a serin statement waiting for new GPS info, it only goes through its; loop once per second!

    I now use one BS2 Stamp to read the GPS, and a BS2p to do everything else. The BS2p·Stamp uses the polled interupt feature to detect when the BS2 is ready to send info. You may want to use a compass or rate gyro to keep the vehicle moving in the right direction between updates.

    As to the Servo being big, Our robot used a power window motor big enough to break your hand if it were in the wrong place.

    'Larry

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
  • krokikroki Posts: 21
    edited 2005-02-07 - 15:27:27
    Good idea Larry!

    I have a 2P and a 2sx, so it's possible to do it, just the controlling is the question for me! Thanx for the link.
    btw: We do shows with the models, so it's enough to go slowly, because it's just for the showsmile.gif
    And we are thinking about a "following act", and a Stamp is ideal for it smile.gif
  • Paul BakerPaul Baker Posts: 6,351
    edited 2005-02-07 - 17:49:44
    If you use an optical encoder on the two rear wheels, you can get rotational rate for each and determine basic direction and speed, then you could use the GPS coordinate update as a synchronization. Remember though, the GPS system has a deliberate deviation built into it to prevent foreign countries from using it for missile guidance. So your GPS coord will have a built it 9-15 ft deviation, and this is if you have a 5 satellites locked and with good signal, if the receiver is under heavy cloud cover, under trees or near a building the accuracy will be even less. There are methods of using 2 receivers spaced approximately 15 ft apart that permits you to triangulate a more accurate measurement. You can also time average, but this requires a very long stationary reading, which is of no use to you.

    Post Edited (Paul Baker) : 2/7/2005 5:53:31 PM GMT
  • Beau SchwabeBeau Schwabe Posts: 6,400
    edited 2005-02-07 - 19:11:56
    I wouldn't say "Modified", but I have "Harvested" old servo electronics who's gears have striped and used them in a "Modified"
    cordless screw driver for a very effective powerful servo.

    Anyway since you already have your servo's you don't need to hack a cordless screw driver.

    Years ago I made a large scale (20 inch wheel base) remote car powered by a gas weed eater. One of the biggest challenges
    was steering. I decided to make a rack and pinion steering frame from stock aluminum. Then I needed to convert the
    rotational servo (screw driver) into a linear servo. To do this I used a semi large spring (2 inch diameter about 5 inches
    long with 10 turns) I welded washers (1/4 inch ID) on each end of the spring (used silver solder and a torch -- heat both
    spring and washer until glowing, and then spot wick the silver solder until it flows between spring and washer) On one end
    of the spring I connected the screwdriver via a 1/4 inch bolt (also welded). On the other end of the spring I used a 1/4 inch
    steel dowel bent in the shape of an "L" . The short leg of the "L" stuck out past the spring coil and rode along an aluminum
    guide track and wala a linear servo adapter that I could use to move the rack and pinion steering. (See images below)

    Note: The spring was an added benefit to absorb steering shock on rough terrain.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    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) : 2/8/2005 4:13:02 AM GMT
    630 x 412 - 36K
    631 x 413 - 48K


    Beau Schwabe -- Submicron Forensic Engineer
    www.Kit-Start.com - bschwabe@Kit-Start.com ෴෴ www.BScircuitDesigns.com - icbeau@bscircuitdesigns.com ෴෴

Sign In or Register to comment.