Controller for 5-wire 'dumb' servo.

So, I’ve been tinkering around (as I do) with some old toys that have these proprietary little servos in them. They have 5 wires (2 motor, 3 pot) and no controller board. The servos work, and I’ve been able to play around with them a little bit with an hbridge (L293D) and pot. I’d like to be able to attach these guys with the standard 3-wire connector like you find on most RC servos. So, I’d like to make a little control board for each one, but I have no idea where to start. I’ve been internetting for weeks, and I’ve only found schematics for generating PWM, but not consuming it and adding logic to control the servo. I can control them with my arduino, but I’d like to find more of a “hardware” solution, as it might be nice to hook them up to one of those lovely 16-channel servo shields. So, any information would be most helpful.

Relevant info: The servo consists of a 3v motor and a 5k pot.

http://www.willingtons.com/mymac/servo_driver.htm

There is quite a bit of info on constructing your own servo. Here are some sites.

https://www.sparkfun.com/products/retired/9014 (Has a schematic)

http://www.openservo.com/

With these pages, you can see that there are a few ICs that you can use to control a DC motor with a pot for positioning.

I think you’d get far better control and precision if you kept them as-is rather than reducing their capability by making them into standard R/C servos. After all, you’ll have to write servo control code anyway, why hide it behind a pulse-width interface?

Just a thought!

Ross Robotics:
There is quite a bit of info on constructing your own servo. Here are some sites.

https://www.sparkfun.com/products/retired/9014 (Has a schematic)

http://www.openservo.com/

I like this open servo stuff. I’ll read through it tonight! Thanks for the info!

lyndon:
I think you’d get far better control and precision if you kept them as-is rather than reducing their capability by making them into standard R/C servos. After all, you’ll have to write servo control code anyway, why hide it behind a pulse-width interface?

Just a thought!

There is total merit to that. The toys currently run off of PPM8 signals, and I’m trying to get them back into working order. If I can have the servos understand PWM, the code that I do write will be more of a pass-through. Or at least, so it seems.