Noisy Sensor when ClearPath intergrated Servo Motor Runs.

[house231]

Hello and thank you for reading this. I am a hobbyist so please bear with me .

I am making a linear actuator. Pretty simple setup. The servo (NEMA 34) needs a PWM pin of my Mega (pin5), and three digital pins (6,7,8). The servo then is grounded to the Mega GND pin. I then run a linear pot for positional measurement, 5V, GND and A0 of the Mega. As long as the servo is powered down the pot readings are perfectly stationary (+/_ 0%). But when the servo is powered up and especially moves the reading are all over the place (+/- 5%). The pot wires are really long (10 ft) but are shielded but not grounded. The motor supposively uses optoisolators in the interface. The Mega is being powered by PC USB and the Servo by an dedicated power supply from the servo manufacture. I spoke to the company and said the servo utilizes optoisolators so it might be on the AC/DC side of things. I have no idea what that means. From reading the forum a couple of ideas would be: 1. to put capacitors on the sensor lines somehow, 2. average the sensor readings, 3. shield everything better, 4. beg for help. I would at least like to understand where is the noise coming from. Thank you again!

The servo is a Clearpath intergrated Servo Motor CPM-MCVC-3446P-RLN > ClearPath IP-5 power supply .

[Paul_KD7HB]

Would be of infinite help to provide a schematic of your project, and identify where on the schematic you made the trace you pictured. Also identify what was happening when the short spikes occur and when the tall spikes occur.
Identify your linear potentiometer. Some are wire-wound and will show contact bounce when moving. Some are cheap plastic and have poor contact pressure when moving. Was your pot designed to work properly during movement?
Having shielded wire that is not grounded on ONE end is just a big antenna.
Paul

[projecteinsteinvr]

Thanks for the help Paul. For others that might be looking up this thread it turned out that one end grounding of my sensor that ran off a foil and wire shielded cable worked perfectly. Sometimes a "antenna loop" or "ground loop" will happen as Paul pointed out but other times it will correct the problem like in my case. I don't fully understand it but a few Google articles talk about how the sensor signal runs in the shielded cable as to if one end grounding will work. It worked for me perfectly!!!