SBL2360 quadrature encoder input logic levels

1 year 2 months ago #29534724 by mpthompson
With an SBL2360, I'm attempting to connect optical quadrature encoders to ENC1A/B and ENC2A/B inputs for closed loop position/velocity control.

The main symptom I'm experiencing is that on the ENC1 input, I'm observing miss encoder tick counts which is hurting closed loop control.

With an oscilloscope, I'm observing the ENC1A/B signal doesn't produce a 0 to +5v voltage swing, but rather a 0 to +3v swing. The ENC2A/B signals on the other hand are producing the correct 0 to +5v voltage swing. My suspicion is that the lower voltage swing on the ENC1A/B input along with some noise on the signals is causing ticks to be missed.

If the encoder is disconnected from the ENC1A/B inputs, the encoder itself is producing the expected 0 to +5v swing on the A/B signals so I don't believe the fault lies on that side of things. When the signals are connected to the Roboteq ENC1A/B inputs the signals are pulled down from +5v to +3v.

Is there something in the configuration of the SBL2360 that is causing the ENC1A/B inputs to have some other load on them that could account for the bad voltage swing? I know these pins can be purposed to other uses (ie. pulse input, analog input or digital inputs), but I don't see anything in my configuration that would cause problems such problems (these other uses are disabled).

Any help diagnosing this would be very useful.


Please Log in or Create an account to join the conversation.

1 year 2 months ago #29534726 by mpthompson
I'll answer my own question.

Turns out my encoders had very weak internal pull-up to 5v. Apparently the encoder output was not enough to properly drive the ENC1A/B inputs on the SBL2360. The ENC2A/B inputs probably have some minor circuitry differences that caused the slightly different behavior I observed.

By adding a 1K pull-up resistor to +5v at the encoder inputs on the SBL2360 I was then able to see a proper 0 to +5v swing on all the encoder signals. This pull-up resistor also cleaned up much of the noise observed on the encoder lines. With this change, the motors are now running much better under closed loop velocity position control.

Sorry for the distraction.

Please Log in or Create an account to join the conversation.

Moderators: tonysantoni
Time to create page: 0.102 seconds