Understanding the output of rotational joystick

Hi! First timer here for asking questions on a forum, so please let me know if this is not a great question.

I am looking to override a currently installed joystick with my own automation controls, but the current joystick is a mystery to me. There is no part number on it. The machine manufacturer is Japanese, so the joystick is likely made in Japan.

There is a button and a spin-able rotating action available to the user. The button is very simple (a simple switch controlled with two wires). The spin controls the Z direction motor on the machine. The joystick’s X and Y motion is not electrical, rather there is a mechanical “foot” that makes contact with the base. I’m not concerned with this motion currently.

The joystick has 6 wires (30 gauge) and a ground coming out of it. Two are for the button, leaving the other 4 wires to control the Z motor. Shorting two of these wires gives a slight ozone smell…no good, so I backed off (the machine still works!).

In the service manual they don’t specify exactly how it all works but in the troubleshooting section (see attached image) it says to check if the output voltage between some of the pins shows an 0.8V waveform when the joystick is rotated. It also notes that the frequency differs depending on the rotational speed of the joystick. It says this waveform will be seen when reading the output voltage between pins 6 and 9 and pins 7 and 9 (pins 8 and 9 gave me the ozone I believe). These readings are coming from a board that seems to be directly connected to the joystick.

My questions are:
What basic principle might this joystick be run using (like is it similar to a potentiometer, or is there something similar that I could look up to understand it better?)
What tools do I need to safely provide a similar input to what is being provided from the joystick?
Is there a similar product I could learn from that has available specifications?

Thank you for the help!


image

Hi danmiller55,

Welcome to the Tech Forum.

Questions:

  • Does your rotating head have a spring mechanism that returns it to a common point when you release it, or does it continuously rotate and stay at whatever position you leave it?

  • What is the effect on your system when it is turned? I assume it raises or lowers something as the head is rotated, but does it raise or lower something until the head returns to the center position, or does the height of that which is being raised/lowered depend on the final position the head is left in?

  • Do you have access to an oscilloscope to look at the actual wave-form of the outputs of the joystick (that might prove very useful)?

If I had to guess, based on the info you have provided, the rotational information comes from a fixed magnet in the head and a magnetic sensor which detects the angular position of the rotating head and outputs a sinusoidal signal (or possibly two signals, one 90° offset from the other) based on the angle of the head, similar to the function of these.

1 Like

Hi David, thanks for getting back to me.

The head does not have a spring mechanism to return it to a common point, but continuously rotates and stays in position when you let go.

When the head is turned the system raises or lowers (based on the direction of the turn). There is no return to center position. It seems to depend on the final position, although it may not be a 1:1 or linear relationship, its hard to tell. It seems like spinning it fast or slow doesn’t affect the speed of the raise very much (i.e. you can spin it fast and a lot and it still just raises “some”)

I don’t have an oscilloscope unfortunetly. It’s just me and my multimeter here.

Additionally, I realized it might be helpful to understand a bit about the circuitry between the joystick and the motor and where the measurement listed in the original post was meant to be taken. The attached screenshot is the best circuit diagram I have. You can see the motor is connected to an U/D switch board, and then the switch board is connected to a tracking board.

The tracking board has a lot of circuitry, but the U/D motor is very simple (see attached picture). There doesn’t seem to be any resistors or switches or anything affecting the signal as it is passed through. Of the 6 pins that come into the U/D board, 4 are from this joystick.

The troubleshooting guide says to check pins on the tracking board (J5 - the input wires on the tracking board). I’m assuming there is some action after this point to adjust the signal, but the signal read at J5 should be the same as from the U/D SW board, right? In which case it should be the same signal as is coming from the joystick directly?

Thanks again for your help! If I can get an oscilloscope I’ll mention that later.

Thanks


It would seem that the lines going into the U/D Switch Board just pass through, as you surmise. You can test this by using the continuity test mode on your multimeter. Just check for continuity from leads on the input connector to each lead on the output connector. If it beeps, you have a direct pass-through connection.

So, what are the buttons on the U/D Switch Board for? Are they also to raise or lower the U/D motor, or for something else? If the former, it would seem to me it would be much easier to tap into those signals, as they would just be closing a contact, whereas, based on your previous information, you may have to produce a sine wave to mimic the output of the joystick, which would be far more challenging.

1 Like

TLDR:

  1. The U/D buttons are for a separate motor
  2. Continuity can be seen between all the pins from J2 to J1 connectors
  3. One of the 4 relevant wires seems to be shared with a pin used by the joystick button

Sorry, yeah, I should have been more clear about that. Those U/D buttons go to a separate part of the machine to raise and lower a separate motor.

J2 on the board above includes 6 wires, 2 for the joystick button, and the rest for the Joystick U/D function. I found continuity for the 2 joystick button wires in the J1 connector (pins 5 and 8).

On the J1 connector pins 1-4 are reserved for the U/D buttons (not relevant to this effort). The joystick button wires connect to pins 5 and 8. That leaves only 3 wires. I did the continuity check and saw that the U/D joystick function J2 pins 3, 4, 5, and 6 connect directly to J1 pins 6, 7, 8, and 9. The #4 pin on J2 is connected to #8 on J1 (slightly mixing up the order), but that #8 pin is also connected to the J2 pin #1 (one of the joystick button wires).

That J2 pin 4 to J1 pin 8 connection is the wide path visible on the board it seems. Not sure if this is useful info, but thought I’d point it out.

From your description, Pin 8 on J1 is a “common” pin for the joystick switch and the rotational signals. Probably the ground reference, but possibly a power supply voltage (whatever voltage is used). The wide trace reinforces that theory, I believe.

Based on what you’ve determined and the note about the waveform on pins 6 and 7 of J1 to pin 9, I would guess that pin 9 is either a power supply or possibly a ground reference for the device inside the joystick which generates the wave forms. I’m pretty certain that to imitate the rotational signals, you’ll need to be able to produce two synchronized out of phase sine waves on pins 6 and 7 of J1 of the proper amplitude relative to pin 9. The phase difference would allow the receiver to determine the direction of rotation. That’s not something easily tackled, even by an experienced electronics design person.

To confirm that this theory is correct, you would need to use an oscilloscope to view both signals simultaneously. Anything else would just be guessing, as I am doing here.

1 Like

Ugh, yeah I can see you’re probably right. Thanks for the help, I might need to go a different direction, like driving the motor directly from its input wires (there are only two of those and its a DC motor…Shouldn’t be nearly as tricky.)

Thank you so much for the very detailed response and help!