Motor noise interfering with ultrasonic sensors

[Andy Grove]

Hi,

I've just finished an overhaul of my classic AVC entry. I'm using ultrasonic sensors to detect obstacles and I've just confirmed that I'm getting incorrect readings once the motors start. I assume this is because of some form of sonic interference from high frequency noise being generated from the motors.

Does anyone have any tips on what I can do to work around this? So far my only idea is to try moving the sensors further away from the motors. 

Thanks,

Andy.

[Minuteman]

Is it just the motors, or is it vibrations in general. I mean, maybe you have to isolate the sensors from any vibration in the car, not necessarily the motors...? That's just a guess, since I have very little experience with ultrasonics.

[jesse brockmann]

Waiting to last minute?  :). What ultrasonics you using?  The more likely cause I would think would be electrical interference.  Are you using one battery for everything?  It's best to have a separate battery for you control electronics.  If not lots of caps, etc to filter.  Brushed or brushless motor matters a lot as well.  Brushed motors generate a lot of emf noise.

JesseJay

--
You received this message because you are subscribed to the Google Groups "diyrovers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to diyrovers+unsubscribe@googlegroups.com.
To post to this group, send email to diyr...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/diyrovers/557ad722-e9c1-4686-b1d2-f32b5e832952%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Jon Watte]

I second mechanical insulation -- something as simple as using got glue may add enough viscosity to dampen chassis transmission.

Add dense foam panels/baffles on the paths between the motors and the sensors, to Shor's direct resisted ultrasound.
Additionally, add panels/baffles to block done going out from motor and bouncing back.

For electronics, rather than two batteries, you can use a Y splitter, and put a filter between the motor controllers and logic. A DC DC converter is often a decent filter; a linear regulator often has excellent rejection (80 dB) but has problems at high correct. Or use a cap-inductor-cap topology. Big caps and big inductor needed though --- 1000 if, 470 uH, and up (rated for your expected currents plus 30% margin.)

Finally -- the Doppler effect of moving (especially fast) may get in the way. Of you keep the car on stilts and run the motors, do you still observe this problem?

Jw

To view this discussion on the web visit https://groups.google.com/d/msgid/diyrovers/CANqFZg2g-9U%3D4BiQFn%2Bf-oYy9dLRNKm5fHCHOkoMQj6nC4EBMQ%40mail.gmail.com.

[Joep Suijs]

Hi Andy,

Had this issue with strong (high current) cheap (yellow plastic) geared motors. A 100n capacitor on the motor contacts significantly reduced the issue. I've seen motors with two capacitors too, one from each motor contact to the housing. You might try that too.

Joep

Op zaterdag 3 september 2016 22:45:10 UTC+2 schreef Andy Grove:

[Andy Grove]

Thanks for all the tips! I tried adding capacitors and although it may have helped a bit, it didn't solve the issue. I then experimented with distance between motors and sensors and found that they worked fine at a certain distance away so I printed some new mounts for the sensors and now it's working much better.

Here's the video from the latest test run https://www.youtube.com/watch?v=NxSX1803Q98

@JesseJay If it wasn't for the last minute, nothing would get done ;-) Well, not by me at least.