2
\$\begingroup\$

I'm pretty new to oscilloscopes so I apologize in advance. I'm experiencing significant signal distortion on the rotary encoder output from a linear actuator and this is what the oscilloscope is looking like. (Apologies, I don't have the images of the circuit or part numbers at the moment).

It's strange that the noise spikes are constrained to the high state only. What could be causing these types of oscillations? Is it probably just noise?

If so, what would you recommend to fix it, perhaps a schmitt trigger?

I'd appreciate any help/guidance

\$\endgroup\$
8
  • \$\begingroup\$ It looks like that should not happen. What if you leave the encoder state high without movement? Maybe that's the noise of the motor running? How are you measuring with the scope, what is the setup and wiring? \$\endgroup\$ Commented Sep 9 at 7:23
  • \$\begingroup\$ Looks like 26 spikes on the high. And the perfect alignment and exact tops of each would make me very curious. It's not random. What is the exact model of the rotary encoder? I'd be looking for root-cause that is not poisson-that-sums-to-gaussian noise. That's highly correlated-looking. So the rest of the circuit operating this would also be a nice-to-see. Knowing the scope may also be important, if it is a cheap one. \$\endgroup\$ Commented Sep 9 at 8:25
  • \$\begingroup\$ You have to find what that is. What's the time base? Maybe it pwm making it's way into your digital signal? \$\endgroup\$ Commented Sep 9 at 8:43
  • \$\begingroup\$ Alex, a very warm welcome to the site. Please can you edit your question (don't add new info in comments) and add (a) the encoder manf and part number, (b) a schematic of your circuit between the encoder, this measurement and whatever else is connected. Don't forget to show the power supply etc. Thanks. \$\endgroup\$ Commented Sep 9 at 10:55
  • \$\begingroup\$ Isn't that just wiper contact noise? Unless it's an optical device, then perhaps just noise injected from an external source, onto a higher-impedance node (not actively pulled low, by a transistor, say). \$\endgroup\$ Commented Sep 9 at 11:15

2 Answers 2

Reset to default
6
\$\begingroup\$

If it's a mechanical encoder or an optical encoder with open collector or open drain outputs then the output impedance might be 10kΩ when high and 0.1Ω when low, so this kind of waveform is quite plausible when there is noise present and wires running to the encoder.

One common source of noise is switching power supplies, especially those without a ground when used with a grounded oscilloscope.

As per comment @td127 below, lower value pullup resistors can help with this (as low as 1kΩ if power consumption is not a great concern), and for mechanical encoders, many manufacturers recommend a small capacitor to ground on each of A and B pins, something like 10nF-100nF.

\$\endgroup\$
1
  • 2
    \$\begingroup\$ Agreed this is likely the issue. OP asked for recommendations to fix: making the pullup resistors (much) smaller would reduce the coupling significantly, if this were in fact the issue. \$\endgroup\$ Commented Sep 9 at 20:53
1
\$\begingroup\$

Taking into account you don´t have the part number of the encoder you are using at the moment, I provide this image from a bourns rotary encoder datasheet I personally use and it works perfectly to filter noise. As you notice it use resistors and caps.

enter image description here

Hope this helps.

\$\endgroup\$

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.