Servo Percision

Jonathan Allison

What level of percision can you get from a standard servo?

Stepper motors can get down to "microsteppers" like 50, 000 steps per rotation, I definitly do not need that kind of percision.

A server turns 180 degrees, how many steps does it take to turn the 180?

allanlane5

The Servo control signal is 1000 uSec to 2000 uSec. The BS2 controls it in steps of 2 uSec. Thus I believe you get 500 steps for 180 degrees. Repeatability, gear-lash, etc. I can't help you with. Should be pretty good, though.

Jonathan Allison

Thanks, it should work for what I need it for then. I need to have the ability to move a wire guide lateraly approximatly .25 MM at a time.

My plan is to setup the servo to connect to the guide rod, sort of how a car windshield washer motor is setup, less the wiper arms, I only need lateral movement.



Thanks for the help.

·

Joe Fishback

Johnathan,

Besides working with Parallax robots, I work with industrial CNC machines.

Here are a couple notes that may help you out. First hobby servos depend on a electronic circuit that uses a variable resistor(pot) to give a feedback to where the servo arm is at in relation to the signal sent by the BS2. I do not believe you can place a hobby servo arm to the percision you need ( 0.25 mm equals about 0.010" ). Industrial servos use encoders, not pots for movement feedback to the servo circuit. There are number of good articlies on the WEB that explain how hobby servos work and how you can modify them do all kinds of things they are not suppose to do. But I believe Stepper motors would still be a better and·easier choice, not knowing what you are actually trying to do.

Joe Fishback- Fishback Designs

kelvin james

Hi Jonathon I saw your post about the wire winder idea and your concerns about the motor control. I think joe is right, that a stepper motor for the guide control for precision movement is the way to go. You get a precise increment especially in micro-stepping and the ablity for the controller to track the position without feedback. As long as the torqe of the motor is not exceeded (as in stalled ) the exact position of the motor is kept for monitoring. And on another note, it would best to get the 2 motors to "talk" to each other, as the position of the spooler motor should be monitored also. That way for precision control during winding, x movement of the spooler will equal x movement of the guide. As an example, for every complete revolution 360 degree of the spool, a signal is sent to the processor to increment the step on the guide by whatever step angle is necessary. Through trial and error or math you could determine the proper lay of the wire over the width of the spool. And you could pre-program random patterns of the wire depending on the postion of the spool by changing postional increments and speed of the guide with some simple math. You would need some type of shaft encoder on the spool motor to accomplish this. It might be more than what you figured to get it working properly, but it is the way i would approach it.
This is only my opinion, other people may differ.

kelvin

Jonathan Allison

Thats exactly how I invisioned it working, one rotation of the spool motor would mean one small step(s) for the guide motor. Obivously its gonna need some logic programmed into it, as the spools aren't necessarily all the same width. I'm not worried about that right now though lol Just need to get some sort of development environment setup for now.

I've started another thread on stepper motors, in paticular the one offered by parallax. Not sure if its gonna do the trick or not.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Johnny