Voice Controlled Robot: Bad Servo Modification?

[michealcollins]

Hey ya'll, still working on the VRbot (Admin's Instructible). I got it working correctly, but I had not known you could modify the servos for continuous 360 rotation, so it had limited movement.

To make this modification, I followed various online tutorials, where each one had slightly different procedures dependent on servo type. Mine are HS-311's.

After taking the cover off and unsoldering the control circuit from the motor, I disconnected the servos potentiometer, soldered in two 2k ohm resistors in it's place, chopped the "stopper" thingy on the restricting gear, and buttoned everything back up.

This VR bot requires two servos: controlling left, right, forward, and "dance" motion commands. Since my modification, Every command registers as "left". I get very full, fluid motion in the left direction, but for EVERY command.  Meaning, both servos (left and right) are rotating clockwise. Before the modification, the servos at least switched to the appropriate directions depending on my vocal command.

Can anyone tell me what I may have done wrong in my modification?

[Gertlex]

You probably have the resistor values wrong, but I've no idea what proper values or wiring are, off-hand.

Does the rotational speed vary with different positions sent to the servo?

When I did my servo modifications, I didn't replace the potentiometer, I just glued it in place.

[michealcollins]

Well according to the various mod videos, each one said you could either glue the pot in place, or add the resistors. I figured to increase my "servo hack" experience, I did both (even though adding the resistors cuts the pot out anyway lol).

As far as the rotational speed...I attached a screenshot of the program code. The "speed" values given to each command are definitely different.

One of the videos specifies 2.2k resistors...but I wonder if the value is dependant on the servo. Because the control circuit of my servo definitely looked slightly different than his.
 
Servo modification tutorial (360 degree of rotation)

Anyway, thanks Gertlex! You are always helpful!

Mike

[Soeren]

Hi,

One of the videos specifies 2.2k resistors...but I wonder if the value is dependant on the servo.

They aren't. They only have to be of identical value to make a mid voltage reference.
Anything from 1k to 5..10k will work. Lower values will work as well, but will increase the current and higher values will make it more susceptible to noise (including from its motor).

Are both servos acting precisely the same?
What if you swap them around?
Do you have access to a 'scope?

You could build a cheap servo tester from a single 555, some R's and C's and a pot and check the servos manually - that way you get around the possibility of programming errors.

[michealcollins]

well I know for a fact there are no program errors. the bot commands worked perfectly before the modification. it turned left, right, forward, and did a little dance thing where it moved right to left.

But when I soldered in those resistors, it made the servos only turn clockwise on any given command.

Both servos are standard HS-311's made for 90 degree rotation (the cheapest ones).

I think what needs to be done is I need to reattach the modified pots and see if my super glue job was enough to keep it from working...

[Gertlex]

Test method for you:
-re-attach the pot in place of the resistors (doesn't have to be soldering). 
-Probably don't need to put the servo back together, you can just have the electronics + motor + potentiometer on a table top.
-Connect servo to your bot and give it whatever command to just sit there (i.e. servos not moving, ideally).
-Turning the potentiometer should vary the speed at which the motor turns.

Curious if that works... Haven't used hobby servos in a couple years, really.

(98% sure nothing bad can happen doing this =D )

[michealcollins]

Oh i already snipped the pot switch thingy and super glued it in it's most center position (the other modification technique). I'm thinking that when I reattach the pot wires, things should work....hopefully lol

[Gertlex]

Probably not going to have a centered servo that way. You need to center it while the servo is powered on

[Admin]

Gertlex is mostly correct, but I want to chime in . . .

1) micheal, reattach the pot wiring like it was before.

2) Then, using your microcontroller, send a 1.5ms pulse (the neutral zero position command).

3) rotate the pot until the servo stops spinning

4) unsolder the pot

5) using a voltage meter, measure the resistance on the pot

6) then solder a resistor using that exact same value you measured

But it's easier to just glue the pot in place . . .