Motor noise on encoder ground, is it a problem?

[billhinge]

Hi

I've been modding my CG5 to run with a tracking belt and in doing so I had to reverse the motor polarity and the A & B encoder wires

When determining which were the A & B encoder signals I hooked up to a logic analyser which showed them quite easily

However, there appears to be noise on the encoder ground due to it sharing with the motor. Is this likely to be a problem and should I attempt to mitigate it with shielding and capacitors on the motor etc or is this just wasted effort ?

The encoder ground is the yellow ch4, orange ch 3 is the 5v line 

[billhinge]

I did a more detailed measurement today

Yellow ch 4 = encoder ground

green ch 5 and white ch 9 are encoder

ch 6 and ch 7 are motor 

The spikes on the encoder ground are of duration 20 ns, typically the encoder signal is of duration 20 ms or more ( i.e. factor of 1000 )

Also notice the bounce on the motor pwm channels (greatly magnified (ch 6 & ch7)

Everything seems to work so I'll assume the noise is of no consequence 

PS - the gap on the first pic is due to me switching motor direction part the way through the measurement 

[perfrej]

1) How did you measure encoder ground? In respect to what?

2) I wouldn't think capacitors should be used. What would you reference them to?

/per

[billhinge]

On the motor and encoder assembly there are 6 wires - The motor is fed from a good quality 13.8v psu 

2 motor wires

encoder  5v

encoder ground

encoder A & B

attach probe to each wire and measure (checked with logic analyser manufacturer that this is correct process and that the spikes are likely to be noise). As I said, there is a factor 1000 in the duration of spikes to the duration of the encoder pulse

The rationale for using capacitors is found here 

http://www.pololu.com/docs/0J15/9

and at the bottom here

http://learn.adafruit.com/adafruit-motor-shield-v2-for-arduino/faq