Spark Max RF Interference

Here’s a quick story that seems pretty unbelievable…

To anyone experiencing seemingly unsolvable problems with the Spark Maxes (ignore the grammar), try moving your motor controller further away from the voltage regulator. I know, I know. It seems crazy. However, at least for us, we spent over a week trying to figure out what we were doing wrong, what crimps were wrong, talking to tech support, etc., and we just discovered that it’s a simple case of radio or electromagnetic interference.

Try it. If it works, let me know. If not, I’m afraid you might have a more complicated problem.

What was the problem you were experiencing?
Are you saying that EMI from the regulator was preventing the Spark MAX for working correctly? (Or vice versa) This seems very odd to me.

2 Likes

Continuing the discussion from Spark Max RF Interference:

Hi @bsimmons,

We were experiencing problems with one of four REV SparkMAX controllers and its connected NEO motor:

  • intermittent LED blinking pattern Orange/Magenta, documented as “brushless encoder error”
  • intermittent problem with the NEO motor not starting without “assistance”, similar in behavior to this thread
  • “mysterious” modification of SparkMAX configuration parameters (e.g., follower mode values)

Simply “shielding” (with @jonah4975’s hand) or moving the SparkMAX away from the CTRE Voltage Regulator Module caused our problems to “disappear”. We were able to consistently repeat this behavior (that is, clearing up the problem or introducing the problem) by moving or shielding the SparkMAX.

As @jonah4975 mentioned, we spent lots of time checking and rechecking Anderson Power Pole crimps, our Java code, the CAN bus configurations of all devices, etc., etc. This is the only thing that consistently remediated the problem for us. This is a “lab robot” so the electronics are pretty tightly packed in the center of the mounting board. So it’s possible a different device was causing RF interference, but the VRM seems to have been the culprit.

I’m going to put the VRM inside a Faraday Cage this evening to see if that mitigates the issue.

Check your phase connections (Controller to Motor) and make sure they are strong connections, rev did say that if you dont have a good connections, you could have this issue. But I do see you have checked them, but I’d check the crimps on the power poles.

I have heard of problems with the VRM giving off RF interference that intermittently interferes with the radio. I don’t know what causes it because sometimes they can be next to each other without a problem. Even if the VRM were giving off RF interference, I don’t see why that would affect the SPARK MAX; but it’s not inconceivable.

7 Likes

The symptoms you describe are consistent with a piece of equipment with low immunity to radiated EMI. Please try enclosing the SPARKMAX with aluminum foil that is connected to one of the black terminals at the PDP. This forms a Faraday shield around the SPARKMAX. You can also try placing a metal plate between the VRM and the SPARKMAX with the plate connected to one of the black terminals at the PDP. While the VRM may be emitting some EMI, there are not a lot of reports of problems with any of the other components such as the radio, Roborio, PDP, PCM, cameras or other co-processors. This makes it more likely that it is the SPARKMAX that is particularly sensitive to EMI. EMI immunity testing is done with a setup like this.

We absolutely encountered PCM faults when the OMP radio was placed near it, but have not had any SparkMax issues as you describe.

I wrote this analysis way back in 2015 when the VRM was first introduced. Even back then, I found that the VRM could affect radio performance. paper: 2015 Cross The Road Electronics Voltage Regulation Module (EVM) Characterization

1 Like

I do not recommend doing this.

In addition to possibly violating rule R42, it can also be very dangerous to have exposed electrical connections around devices on the robot.

@jonah4975 and @dklann, you mentioned talking to tech support. Could you please DM me the email address you used so that I can find the email in our support inbox? I couldn’t find your team number or any support tickets in our system that resembles what you’re seeing. I’m not saying you didn’t reach out, I’m just trying to tie this thread to any existing support tickets.

I would like to take some further troubleshooting steps for each of these symptoms, as they all have likely causes that should be relatively easy to isolate. Let’s either continue the existing support thread or start a new one through our support@revrobotics.com email.

1 Like

I would also construe this as in violation of R38 unless you then had an insulator around that cage.

Sorry that I did not make clear that this was only meant to be a test to see if shielding made a difference. One will find rolls of aluminum foil laying around at EMC test labs for just this purpose.