Servo Movement At Power Up Before Initialization

The likes of a cam, bell crank, or four bar linkage can often be used to limit the physical travel extent independently from that of the driving motor.

However, if you reduce the length of the servo horn (radius), you may be able to reduce the length of travel at the animal to within safe operational limits.

Do you have a photo / sketch that shows the actual servo to animal set up ?

I totally agree, it's not always easy seeing the limitations of hobby servos. I've been bitten by the stalled servo smoking scenario. Despite something working in principle, when it comes to in practice, someone or something might forcibly prevent a servo turning, perhaps it might accidentally hit another piece of mechanism unexpectedly, and then it's stalled and overheating damage can occur. 

Maybe someone (not me, no time : ( should write a blog (or do a video) on "Why hobby servo's stink, and how to make them stink less through mechanical, electronic and software means". 

I hope the show goes well! Looking forward to hearing how it goes.

Equipment

Servo
     SG92R
     SG90
     MG90S

Nano (knock off)
     ATmega328P (old bootloader)
     ATmega168P

Drawing represents a servo with a horn attached. At the lower end of the servo case is were the control cable connection is shown. The servo rotates counter-clockwise along the arch from 0 to 180 degrees (except when noted).

Servo motors SG92R & SG90 can only be move by hand along the arc. Physical stops in the servo prevent it from being positioned outside the arc. The MG90S servo has no physical stops. It can be moved by hand outside the arc.

If no default servo position is provided in void_setup (servoA.write(#); the servo horn will move to 90 when powered on. This happens for all three servos. The MG90S servo horn will move to 90 but if the horn is between 181-360 it may move counter-clockwise or clockwise to reach 90. If the horn is >200 the servo will move counter-clockwise. <200 the horn moves clockwise to reach 90.

The code used for testing is provided.

#include <Arduino.h>
#include <Servo.h>

uint8_t ServoA = 3;
int buttonPin = 2;  // Button to trigger animation

Servo servoA;  //initialize servo

//debounce
int counter = 0;
int buttonState = 0;
int lastButtonState = 0;

int currentButtonState = 0;
unsigned long lastDebounceTime = 0;
unsigned long debounceDelay = 50;

void setup() {
  pinMode(buttonPin, INPUT_PULLUP);
  //servoA.attach(ServoA, 544, 2400);
  //servoA.attach(ServoA, 544, 2400);
  //servoA.attach(ServoA, 544, 2400);
  //servoA.write(0);
}

void loop() {
  currentButtonState = digitalRead(buttonPin);
  if (currentButtonState != lastButtonState) {
    lastDebounceTime = millis();
  }
  if ((millis() - lastDebounceTime) > debounceDelay) {
    if (currentButtonState != buttonState) {
      buttonState = currentButtonState;
      if (buttonState == LOW) {
        servoA.write(0);
        delay(1000);
        servoA.write(180);
        delay(1000);
        servoA.write(0);
      }
    }
  }
  lastButtonState = currentButtonState;
}

Pushing a button the servo moves from 0 position to 180 and then back to 0 position. The servo is conditioned using the min/max default values servoA.attach(ServoA, 544, 2400); provided in the library. The button push is wrapped in a debounce routine.

Lesson Learn:

Set a default position for the servo in void_setup. This simple step will prevent the servo from moving to 90 position, especially if the position is outside the range of movement.

Start by using servo library default min and max values servoA.attach(ServoA, 544, 2400); I've found that tweaking these values may improve the length of travel of the horn. On the MG90S & SG92R increasing the maximum value to 2450 increased the distance closer to 180. Decreasing the minimum value on the SG90 did not get it closer to 0. Posts on the internet suggests manufacture will provide these values. I had no success finding those specification sheets.

I had a similar problem with a cheapo ( but 200W) audio amp powering up with the speakers, causing the speakers to pop loudly. I had to use the control power relay thought to keep the speakers disconnected from the amp until the amp was fully powered up.  

oh those pops are a stab to the heart of audio buffs.