I have been using Allegro A5977 stepper drivers for a while now (a few years) and I had not too bad of a result meeting the EMI compliance targets even for CLASS B.

However I went by the seat of my pants when filtering the VBB (the Motor power voltage). I use a filter consisting of an electrolytic cap to the GND, 68uF SMT can right at the chip's VBB, then I put a BLM18KG221SN1D ferrite bead in series (220 Ohm at 100MHz, 50mOhm DC resistance 2A rated) and finish it off with another cap, this time a X7R 22uF chip capacitor to the GND.

This filter is at every driver. Then it runs to my 12V input where it meets the board's input filter which could vary depending on a board and POL DCDCs I use. But no matter what, there will be a 100uF electrolytic and 1uF film cap there.

I have my doubts as to the ferrite bead efficiency as the currents that could go into 1A peak will render the bead not very... resistive to the noise which is quite low in frequency in the first place and is certainly not in multi-MHz region.

How can a I design a board with stepper drivers and ensure that CLASS B FCC standards are met for EMI?


1 Answer 1


So, what do you guys use to make your steppers to behave to meet CLASS B?

I use a closed frame (faraday shield) with filters on the power to meet class B. If the motor cables need to go outside the frame, I use shielded cable.

One thing that may be an advantage to knock out 100Mhz+ signals is to short them out with capacitors. X2Y capacitors are multiterminal devices that cut the ESL in half and are effective in the GHz range. I also use them for low noise applications on analog electronics. They may be useful to dispense with the inductor an use only capacitors to short out high frequency noise that arises from high frequency sources.

  • \$\begingroup\$ Thanks for answering. My device is a small (let's say 8x4x4 inches) metal enclosure with everything inside including motors. There are openings for the moving parts and ventilation. I got no problems with 100MHz type signals, it is mostly low end of the spectrum. I always use copious decoupling for everything digital so my highs are usually pretty quiet. \$\endgroup\$
    – morzh
    Commented Jul 18, 2019 at 13:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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