Shop OBEX P1 Docs P2 Docs Learn Events
Continuous Rotation Servo Question — Parallax Forums

Continuous Rotation Servo Question

chaosgkchaosgk Posts: 322
edited 2009-04-12 06:44 in General Discussion
Can someone tell me a little more about the continuous rotation servo from parallax.·
A project I will be working on in the near future (when my CNC and relay board is done) is going to be a RC Predator UAV with aerial photography.·
I already have experience with building and flying RC aircraft and using servos and all of that. What I want to do is build a mount for the onboard cameras, the idea is to have a pod in the front that is self contained with the cameras and battery pack. I want the pod to be able to track the same spot no matter what position is in the sky. I'm thinking of using heading hold gyros to control the two servo's for the X and Y axis of the camera mount with the continuous rotation servos connected to them.· This should allow the camera to rotate 360 degrees and hopefully 180 degrees for the Y axis.· How exactly do the continuous rotation servo's work. I know on regular servo's, the pot tells the control board where it is and where zero is.· Is this the same for the continuous ones.· I have a 9 channel transmitter and I figure it is going to take 5 channels for the camera system alone,
(not the channels used, just a list of what I picture)

1. X axis rotation
2. X axis gyro enable
3. Y axis rotation
4. Y axis gyro enable
5. Snap picture.

I'd like to be able to use knobs or sliders on the radio to aim the camera at an object or location, then flip on the gyro switches and have the heading hold lock the camera in place.· Using heading hold gyro's, they should be able to sense the changes in pitch and roll and compensate for minor changes in direction.

What do you all think?

·

Comments

  • James LongJames Long Posts: 1,181
    edited 2009-04-06 18:41
    Chaos,

    A continuous rotation servo is like a motor, it has no reference to any point in it's rotation. It can not be used in the manner of which you want. They basically take a servo, and disconnect it feedback pot. This causes the servo to "hunt" it position it is attempting to reach. This is what makes it continue to turn.

    What you are looking for is a >90 degree rotation servo. I think Servo City has them, but you would need to browse around there. There are other servos (not R/C) which are capable of doing this. Dynamixel is one manufacturer of a servo which will serve the purpose you are attempting.

    I hope this helps.

    James L

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    James L
    Partner/Designer

    Lil Brother SMT Assembly Services
  • chaosgkchaosgk Posts: 322
    edited 2009-04-06 21:30
    Actually all of the servos I use in my planes now are about 140 degree total travel. I am looking for something for the x axis that will go beyone that. I can probably use a standard servo with a belt drive system to get higher rotation out of it on the camera side. This may end up being an upgrade option later and just aim manually.
    The other question, how does a servo know which direction to move, is it based on pulses, with a heading hold gyro attached, wouldn't it be sendin out pulses constantly telling it what direction to move, if a servo gets the pulse that says stay in one spot, wouldn't it stay where it is until the gyro says to move one direction or another?
  • James LongJames Long Posts: 1,181
    edited 2009-04-07 02:00
    Chaos,

    Yes servos use a frequency bandwidth to know where they need to go. If you need them to go left the frequency goes up, right and it goes down. It is a little more complicated than that, but the basis is simple.

    Well you have the basic idea about the movement.

    Basically the "on" time of the frequency tells the servo where to go. If the on time is 1ms the servo will go to 0 deg. If it is 2ms it will go to maximum (90,100,140, etc). If the pulse on is 1.5 ms it will center. You tell the servo where to go, by adjusting the "on" time of the frequency.

    This makes it easy, for the receiver of a RC plane doesn't have to keep telling the servo to move a little. It just tells the servo where to move, and the servo goes to that point. If the position needs to be changed, the receiver tells the servo the new location.

    I think this will explain enough of how it works.

    James L

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    James L
    Partner/Designer

    Lil Brother SMT Assembly Services
  • mctriviamctrivia Posts: 3,772
    edited 2009-04-07 02:46
    for my camera terit I am using a 720 degree serve rated at 10lb loads for pan and 180 degree servo also rated for 10lb loads for tilt. Cost $500 for the 2 of them though.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Need to make your prop design easier or secure? Get a PropMod has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.
  • chaosgkchaosgk Posts: 322
    edited 2009-04-08 05:37
    I did some reading and figured out how they worked and understand the ms pulses now and that makes sense why they need the pot to know where to find center at. I may end up using a belt and pully system and get 300 degrees rotation out of it depending on the ratio and just use a regular servo and linkage for the tilt part of it. This is way on the back burner anyway, at least till after the relay board project is finished and the CNC project and after everything has gone boom on the 4th of July.
    Plus the UAV itself still needs to get drawn up and built, but of all of my current hobbies, making something built from scratch fly seems like one of the easier ones. I made a combat plane last year out of some scrap rain gutter downspout, pink foam for a wing core with carbon fiber arrows for support and balsa sheets for the tail secition. I love that plane, cause there's really nothing in it, it flies great and is designed specifically to crash so if it does, we just laugh and I replace a part and put it in the sky again.
  • James LongJames Long Posts: 1,181
    edited 2009-04-08 06:54
    chaosgk said...
    I did some reading and figured out how they worked and understand the ms pulses now and that makes sense why they need the pot to know where to find center at. I may end up using a belt and pully system and get 300 degrees rotation out of it depending on the ratio and just use a regular servo and linkage for the tilt part of it. This is way on the back burner anyway, at least till after the relay board project is finished and the CNC project and after everything has gone boom on the 4th of July.
    Plus the UAV itself still needs to get drawn up and built, but of all of my current hobbies, making something built from scratch fly seems like one of the easier ones. I made a combat plane last year out of some scrap rain gutter downspout, pink foam for a wing core with carbon fiber arrows for support and balsa sheets for the tail secition. I love that plane, cause there's really nothing in it, it flies great and is designed specifically to crash so if it does, we just laugh and I replace a part and put it in the sky again.

    Sounds like a real cool project.

    I was working on one a year or two ago, but dropped the project. The complexity of the project got to be enormous, so I decided to put it on the "way out back, down south back burner". The project was way more than a simple RC to UAV type project, and probably so much over my head I couldn't justify it.

    With all the Kalman filters being written, it may get started back up after the two projects I'm working on. It will be a minimum of a year before they are done and totally debugged though.

    Just for giggles, you may want to use a linear servo for the tilt. Here is an address of one I've been watching for a long time. One version does use the pulse method of control. http://www.firgelli.com/

    They are pricey, but really cool.

    James L

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    James L
    Partner/Designer

    Lil Brother SMT Assembly Services
  • chaosgkchaosgk Posts: 322
    edited 2009-04-12 06:44
    Getting good rotation out of standard servo's isn't a big problem. A friend of mine and I built a RC duck retrieval boat last year and we have ducted fans mounted on pivots on the floats. Below the surface is a couple of gears with a servo attached to a larger one for a 3:1 ratio I think. Out of normal travel on the servo's we get almost 360 rotation, close to 180 degrees each way.
    We didn't have any rain for the last half of the summer of 08 so every place we hunt was too dry to even use it so it hasn't even been out yet.

    I'll do some posts when I start the UAV project but I need to light up the sky first unless my A.D.D kicks in one night and I decide to start drawing up some plans. I would like to have it functional by the end of August for our club funfly.
Sign In or Register to comment.