Parallax Continuos Rotation Servo-Angle

GeorgeL

Is there any formula, etc to find the "angle" of the servo mentioned in the topic?

The situation is I have a servo fixated to a table, and it has, for example, a PING))) sensor. The starting position relative to the design is 0 degrees. When I have for example

Pulsout 12, 740

how much does the servo angle change.

W9GFO

If it is a continuous rotation servo then there is no way to find angle without an encoder of some form.

For a regular servo, every 10 uS of pulse width is roughly 1 degree of movement. In your example 740 is the same as 1,480 uS pulse width - which is 20 Us different from the center setting of 1,500 uS, which would then be about 2 degrees.

Rich H

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Servo Boss, a 12 channel servo tester kit from Gadget Gangster.

Mike Green

Regular servos, even though there is a correspondence between pulse width and position, are not accurate with regards to positioning. You'd have to calibrate your program with a test program where you can adjust the pulse width (like using DEBUGIN to set it). Once you have known points (like center and +/- 90 degrees) you can calculate the pulse width for any angle (keeping in mind that it's not very accurate).

GeorgeL

So I see that a non-contious servo is the best/most accurate solution?

W9GFO

GeorgeL said...
So I see that a non-contious servo is the best/most accurate solution?

Between a continuos rotation and a standard servo, yes. For more accuracy you could use a high quality digital servo. Look for one with the lowest dead band.

For the most accuracy you would need an encoder.

Rich H

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Servo Boss, a 12 channel servo tester kit from Gadget Gangster.

NWUpgrades

Have you considered using a stepper motor? VERY accurate and can be programmed for half steps.