Parallax Standard Servo how accurate
Thirteeneven
Posts: 11
How accurate is the Standard servo from parallax we need a increments of .5 degrees will this servo do that
Thank you
Josh
Thank you
Josh
Comments
All servos will have a little bit of slop in the gear train at best. The servo mechanism as well is not that accurate. You can probably get premium servos with higher resolution and accuracy, but you'll have to look very carefully at the datasheet.
This is something that I had been meaning to test so since it was brought up I did a quick, somewhat crude measurement.
First I printed up a degree wheel 10" in diameter - the degree marks are are a little over 2 mm apart.
I set my BoeBot with Ping servo in the center with a wire attached to serve as a pointer.
Using my super nifty Servo Tester (yeah, that's a plug), I set the pointer so that it was at zero degrees when the pulse width is 1500 uS.
Slowly decreasing the pulse width to 1000 uS - the pointer is on exactly 310 degrees, for 50 degrees of travel. Close to 10 uS per degree.
Going back to 1500 uS (without overshooting) I end up at 0 degrees.
Continuing on to 2000 uS I end up at 53 degrees. I don't have the BoeBot centered perfectly.
Back down to 1500 uS, without overshooting, it is at .5 degrees. I need to decrease the pulse width one more microsecond to 1499 to get it at 0 again.
This particular servo has a measured 5 uS dead band.
If I program three pulse widths, 1494, 1500 and 1506, it will reliably move between the two outside positions but the middle one will vary by about 1/4 degree.
My conclusion is that you can command it to move in .5 degree steps but each new pulse must be different enough not to fall within the dead band and the accuracy will be +/- .5 degrees.
The repeatability appears excellent. Going back and forth between two settings It hits the same mark every time. The error occurs when you want to hit that mark coming from the other side of it. Then it can be .5 degrees off.
With some programming I think you could get .5 degree accuracy +/- 1/4 degree. You would need to take into account which direction you are approaching the target from and make sure to stay out of the dead band. It would no doubt be easier with a servo that has a tighter dead band.
Rich H
If you don't need the servo's full angular range of motion and you are adept at assembling accurate geartrains(Google anti-backlash gear), you can·add a stage of gear reduction and improve·any servo's accuracy by that much. That is, adding 2:1 gear reduction cuts your angular range of motion in half and doubles your angular accuracy. It also increases torque and reduces the negative effects of your servo load.
Nice work, W9GFO!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·"If you build it, they will come."
Also, the Parallax standard servo that I am using will work form 520 uS to 2420 uS, giving it a 190 degree range of motion.
Servo City has all kinds of servo accessories, including gears.
Rich H