3
\$\begingroup\$

I'm building a remote tuning device for a magnetic loop antenna. It will use a stepper motor to turn the shaft on the mag loop's variable capacitor. The electronics will consist primarily of a microcontroller and a stepper motor controller. Control and power will be provided through a shielded twisted pair cable. One twisted pair will carry a half-duplex RS-485 signal at a low baud rate. The others will carry DC power and ground, somewhere around 7 – 12 volts depending on which motor I choose. I have a draft schematic here.

The transmitter must be on during tuning, so this device will be next to a source of around 5 – 10 watts of RF at 7 – 30 MHz. There is no RF circuitry inside the box and I would like to keep out stray RF as much as possible. I have never done a project like this before so I have a number of questions about making it robust to RFI.

  • Should I build this inside a metal box? And what connections should exist between the chassis, the control cable shield and the DC supply ground inside that cable? Should I ground the shield at one end or both?

  • What kind of protection do I need on the power and control lines? I have TVSes on the power supply, as well as both RS-485 lines (using a SM712, designed specifically for this purpose). I put a ferrite bead on the power line as well (but I'm unsure how to choose a specific part) and I have plenty of bypass capacitors. There are also 10 ohm resistors in series with the RS-485 lines to prevent damage due to overcurrent for whatever reason. (I realize this creates a RC low-pass filter with the line capacitance, but I believe this won't be a problem with the low data rates I'm using.) Is there anything else I'm missing?

  • I'm using a very basic RS-485 transceiver chip; should I switch to an optically or galvanically isolated part? Am I correct in thinking this would be pointless without an isolated power supply? In my current design the sender and receiver share a power supply and will not have a ground offset.

  • What PCB layout rules do I need to follow, besides "use a ground plane" and "keep traces short"?

Any other advice on making this work well would be greatly appreciated! Thank you.

\$\endgroup\$
  • \$\begingroup\$ Thanks for sharing the tuner schematic. I think we need a complete system diagram to answer most of your questions concerning RFI. \$\endgroup\$ – Stefan Wyss May 21 at 4:23
  • \$\begingroup\$ Any cables will become energy-storage structures, thus dampening will be your friend. I've used this mindset on PCBs of 6" size, with MCU and AnalogSignalProcessing experiencing loss-of-data or undergoing-Reset, in the presence of 400MHz EMI testing. The solution was the insertion of 1Kohm resistors next to the AnalogSignalProcessing IC, which no longer had measurement errors. Thus the 6" traces were highly dampened, and even if instantaneous energy levels were high, the 1KOhm was useful to block RF current flows into the ASP. \$\endgroup\$ – analogsystemsrf May 21 at 5:31

Your Answer

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

Browse other questions tagged or ask your own question.