1
\$\begingroup\$

I am trying to connect a encoder to a motor controller which runs very close to the motor that it controls (30+ amps). Due to the risk of high EMI of the motor inducing on the encoder signals, we send these signals differentially (RS485) over an SF/UTP cable. Currently, the shielding of this wire is connected directly to the ground line of the encoder. The figure below shows the current wiring:

SF/UTP cable connections

I am not sure if this would be the proper way of wiring these signals under these circumstances. I am a bit worried that the groundloop causes ground noise and thus ruins the proper grounding that the encoder requires (it's powered by 5V and can take up to 1V of fluctuation worst-case, but better grounding is absolutely preferred). I've seen solutions where one side is connected with a cap, however this is not ideal in my case because I do not really have any room for this.

My question in a nutshell: would the groundloop in this case possibly cause issues? Note that the currents to the ground plane are limited to <250mA. Or are there other solutions that are more suitable for my case? Connecting the shielding to the frame for example.

\$\endgroup\$
  • 1
    \$\begingroup\$ How can you be sure that groundloop currents are limited. In my experience, they are out of easy control of the designer. \$\endgroup\$ – Andy aka Jul 5 '17 at 16:52
  • \$\begingroup\$ This might not have been entirely clear, but the <250mA that I mentioned was just the converter IC's current limit. I wasn't sure if this would limit groundloop currents, but apparently it does not. \$\endgroup\$ – Timmeh Jul 6 '17 at 7:12
  • \$\begingroup\$ Are you worried about the shield acting as an antenna to conduct noise into your digital GND? \$\endgroup\$ – Krunal Desai Jul 6 '17 at 8:20
  • \$\begingroup\$ Not necessarily, since right now it is connected on both ends of my GND (it is terminated in this case and won't act as antenna). However, I am worried by the noise caused on my GND due to the ground loop, and I am looking for the solution that would minimize that noise. Options I could come up with: one-sided grounding, two-sided grounding (current solution) and two-sided grounding with one side having a capacitor. Not sure which one is preferable, however. \$\endgroup\$ – Timmeh Jul 6 '17 at 9:19
  • \$\begingroup\$ Dies anyone specifically recommend any of the options? \$\endgroup\$ – Timmeh Jul 10 '17 at 7:54
-1
\$\begingroup\$

All of your GND connections should be referenced together in order to ensure they are all at the same potential. This includes the encoder ground line, the casing of the encoder, as well as the shielding connections.

\$\endgroup\$
  • \$\begingroup\$ The problem that I'm facing is that this cannot be the case entirely. The SF/UTP cables are about 20cm long and I cannot connect shielding to the ground line in the cable other than at the ends. A solution could be to use shielding as ground plane (elimitating the ground loop), however I believe this will induce a lot of motor EMI on the ground line. \$\endgroup\$ – Timmeh Jul 6 '17 at 7:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.