1

I'm trying to control a 3-axis stepper motor motion system using a 3-axis Hall-effect sensor joystick (the APEM HF Series joystick, to be specific).

My problem is that whenever I move the joystick in the x-direction, I pick up a signal like I have moved the joystick in the y-direction even though I have not. And when I move the joystick in the y-direction, I pick up a signal like I have moved the joystick in the x-direction even though I have not. Motion in the z-direction, which is achieved by twisting the top half of the joystick handle, fortunately does not interfere with either the x- or y-axes, and vice versa.

You can see this in the image I have attached, where the blue line shows x-joystick displacement and green shows y-joystick displacement. I created this graph by pushing the joystick +x, -x, +y, -y, +x, -x, +y, -y. You can see those minor peaks under each of the major peaks, which is the problem I'm dealing with:

enter image description here

For some context, I have the joystick hooked up to an Arduino Uno (5 V, ground, A0, A1, A2). The joystick has two outputs for buttons and a secondary output for each axis, but I am not using those outputs.

Is anyone familiar with this problem?

ocrdu
  • 8,705
  • 21
  • 30
  • 42
JNB
  • 41
  • 2
  • 1
    Just to be clear, you're absolutely sure you're moving the joystick perfectly on one axis at a time? One thing I'm noticing is that the correlation between the two is not consistent. The first x+ move causes y+, but the second x+ causes y-. If it were a magnetic issue, I would expect both x+ moves to result in the same y signal. I think the most likely culprit is a mechanical problem like backlash. – Drew Nov 17 '22 at 02:56
  • Now that I think about it, you are probably right. The joystick is very sensitive. I initially thought it might be some sort of electromagnetic crosstalk issue, but the way you put it makes a lot of sense. – JNB Nov 18 '22 at 23:56

1 Answers1

0

I think this is caused by the joystick being very sensitive to small movements, just as Drew mentioned. I initially got wrapped up into thinking this might be an electromagnetic crosstalk issue, but that's probably not the case.

JNB
  • 41
  • 2