joystick problem

FIRST General Forum
1

Posted by Igor Likhotkin.

Engineer on team #419, Rambots, from U Mass Boston and NASA.

Posted on 2/12/2000 8:37 PM MST

Hi all!

My name is Igor. I am from team 419.

We seem to have a problem with one of our joysticks. When I run default program on our robot,
eventually one of the motors keeps getting some ‘juice’ even though nobody is touching the joysticks.
I tried swapping the joysticks which made the other motor act the same way. I tried looking at what is
happening to the input variable responsible for that joystick and found out that even though the joystick
sits in the middle, the variable value can be as high as 130-160 decimal. I tried calibrating the y-axis
with the wheel on the joystick, but, alas, after moving the joystick back and forth and then leaving it
alone it would still leak some ‘juice’ into the motor.

I assumed that since we ran the robot with the default code and since the recalibration of the joystick had
no effect on the performance, the joystick was bad.

Has anyone else had that problem? I really don’t know what to think at this point.

If anyone has any input, please reply.

Thank you.

Igor.

2

Posted by Quentin Lewis.

Engineer on team #42, P.A.R.T.S - Prececision Alvirne Robotics Technology Systems, from Alvirne, Hudson NH.

Posted on 2/12/2000 8:41 PM MST

In Reply to: joystick problem posted by Igor Likhotkin on 2/12/2000 8:37 PM MST:

If I recall, there is a calibration button on the motor controllers that allows you to set the middle and end travel of the joystick…that is, without changing your programming at all.

I heard about it at the INNOVATION FIRST meeting at the kickoff…and I’m sure it is documented somewhere in the Motor Controller specs.

I hope this helps…getting another joystick might also help.

-Quentin

: We seem to have a problem with one of our joysticks. When I run default program on our robot,
: eventually one of the motors keeps getting some ‘juice’ even though nobody is touching the joysticks.
: I tried swapping the joysticks which made the other motor act the same way. I tried looking at what is
: happening to the input variable responsible for that joystick and found out that even though the joystick
: sits in the middle, the variable value can be as high as 130-160 decimal. I tried calibrating the y-axis
: with the wheel on the joystick, but, alas, after moving the joystick back and forth and then leaving it
: alone it would still leak some ‘juice’ into the motor.

: I assumed that since we ran the robot with the default code and since the recalibration of the joystick had
: no effect on the performance, the joystick was bad.

3

Posted by Nate Smith.

Other on team #66, GM Powertrain/Willow Run HS, from Eastern Michigan University and GM Powertrain.

Posted on 2/12/2000 9:23 PM MST

In Reply to: Re: joystick problem posted by Quentin Lewis on 2/12/2000 8:41 PM MST:

: If I recall, there is a calibration button on the motor controllers that allows you to set the middle and end travel of the joystick…that is, without changing your programming at all.

Yep…you can recalibrate the speed controllers to work with the output that the controller is giving them…next to the LED on each speed controller, there’s an inset button that you have to use a nail, pen, etc. to press. Hold that button down, then move the joystick controlling that speed controller to its max and min positions, then let it go back to center and release the button on the speed controller. The LED should flash green if the calibration is successful.

4

Posted by Anton Abaya.

Engineer on team #419, Rambots, from BC High / UMass Boston and NASA, Mathsoft, Solidworks.

Posted on 2/12/2000 10:24 PM MST

In Reply to: Speed Controller Calibration posted by Nate Smith on 2/12/2000 9:23 PM MST:

hey all, Anton from team 419 too.

well, after much playing around with it, i finally remembered that when I was on team 97, we had the same problem for a while.

how’d we fix it? DUCK TAPE. I think the knob or wheel that calibrates it is a little tight on some joysticks and tends to uncalibrate itself naturally.

Just tape it or glue it permanently. Or, you can also open the joystick and fix the knob from the inside. It’s suggested that you tape it though from the outside to avoid tampering. (I know i tampered with it last year… hehe)

In any case, we’ve fixed the problem, and i hope this helps other teams.

: : If I recall, there is a calibration button on the motor controllers that allows you to set the middle and end travel of the joystick…that is, without changing your programming at all.

: Yep…you can recalibrate the speed controllers to work with the output that the controller is giving them…next to the LED on each speed controller, there’s an inset button that you have to use a nail, pen, etc. to press. Hold that button down, then move the joystick controlling that speed controller to its max and min positions, then let it go back to center and release the button on the speed controller. The LED should flash green if the calibration is successful.

5

Posted by Lora Knepper.

Student on team #69, HYPER (Helping Youth Pursue Engineering & Robotics), from Quincy Public Schools and The Gillette Company.

Posted on 2/13/2000 8:49 AM MST

In Reply to: it’s simple really! we had the same problem last year! posted by Anton Abaya on 2/12/2000 10:24 PM MST:

Though I know this post is a bit late, those joysticks have always been a problem for us. We’ve calibrated the joystics and the speed controllers for a temp fix. As a driver, I’ve gotten used to retrimming the joysticks on the fly, and even with tape (we tried both duct and electrical), it still is neccessary. I’m kind of curious if anyone has had the same problems, with a fix better than tape.

good luck,
Lora

6

Posted by Nate Smith.

Other on team #66, GM Powertrain/Willow Run HS, from Eastern Michigan University and GM Powertrain.

Posted on 2/13/2000 11:18 AM MST

In Reply to: Re: it’s simple really! we had the same problem last year! posted by Lora Knepper on 2/13/2000 8:49 AM MST:

: I’m kind of curious if anyone has had the same problems, with a fix better than tape.

Well, for various reasons, we have a joystick in pieces at our shop…I got to looking at it and figured out a way to lock them in place…take off the bottom cover(you’ll have to remove the rubber ‘feet’ to get to the screws), and center your trimmers, then glue them to the inside housing of the joystick. Haven’t tried actually doing this with our drive stick yet, but it should work…

Nate

7

Posted by Jerry Eckert.

Engineer from Looking for a team in Raleigh, NC sponsored by .

Posted on 2/13/2000 12:16 PM MST

In Reply to: All Too Common… posted by Nate Smith on 2/13/2000 11:18 AM MST:

: : I’m kind of curious if anyone has had the same problems, with a fix better than tape.

: Well, for various reasons, we have a joystick in pieces at our shop…I got to looking at it and figured out a way to lock them in place…take off the bottom cover(you’ll have to remove the rubber ‘feet’ to get to the screws), and center your trimmers, then glue them to the inside housing of the joystick. Haven’t tried actually doing this with our drive stick yet, but it should work…

Note that altering the joystick, including gluing the trim control, is a violation of rule C1.

8

Posted by Joe Johnson.   [PICTURE: SAME | NEW | HELP]

Engineer on team #47, Chief Delphi, from Pontiac Central High School and Delphi Automotive Systems.

Posted on 2/13/2000 12:21 PM MST

In Reply to: Re: All Too Common… posted by Jerry Eckert on 2/13/2000 12:16 PM MST:

Some how I got in my head that joysticks WERE allowed to be modified this year (in past year’s I know we could not, but this year, I KNOW I read somewhere that we could).

But… there is C1 staring me in the face…

Anybody know where I could have read this?

Joe J.

P.S. Who else among us regrets not being able to modify those 9 pin cords from the radio to the Robot Controller? .3 lbs just asking to be shortened to about .1 lbs – but there is that rule C1 again :frowning:

9

Posted by Nate Smith.

Other on team #66, GM Powertrain/Willow Run HS, from Eastern Michigan University and GM Powertrain.

Posted on 2/13/2000 12:49 PM MST

In Reply to: Didn’t I read somewhere… posted by Joe Johnson on 2/13/2000 12:21 PM MST:

: Some how I got in my head that joysticks WERE allowed to be modified this year (in past year’s I know we could not, but this year, I KNOW I read somewhere that we could).

: But… there is C1 staring me in the face…

: Anybody know where I could have read this?

Team Update 2 -> New Rule C29:
The joystick may be modified(rewired, disassembled, cut, etc.) in order to use the potentiometer and switches in a different manner. If by modifying the joystick you end up breaking it or destroying it, you will not be provided with a replacement.

: P.S. Who else among us regrets not being able to modify those 9 pin cords from the radio to the Robot Controller? .3 lbs just asking to be shortened to about .1 lbs – but there is that rule C1 again :frowning:

I know the feeling Joe…rather than it being a weight issue, it’s where to put all the extra cord that we don’t need…

Nate

10

Posted by Jerry Eckert.

Engineer from Looking for a team in Raleigh, NC sponsored by .

Posted on 2/13/2000 12:12 PM MST

In Reply to: Re: it’s simple really! we had the same problem last year! posted by Lora Knepper on 2/13/2000 8:49 AM MST:

: Though I know this post is a bit late, those joysticks have always been a problem for us. We’ve calibrated the joystics and the speed controllers for a temp fix. As a driver, I’ve gotten used to retrimming the joysticks on the fly, and even with tape (we tried both duct and electrical), it still is neccessary. I’m kind of curious if anyone has had the same problems, with a fix better than tape.

Tape is probably the best fix you will find, short of using a better joystick.

For those who have never disassembled one of the joysticks, there are two potentiometers inside the stick - one for the x-axis position and one for the y-axis. When the stick is moved the shaft of the potentiometer is rotated. The trim controls rotate the body of the potentiometer.

If the stick is moved vigorously or to the end of its rotation the entire potentiometer will move, which causes the value at the stick center position to change. If the trim knob is securely taped (not just using a thin strip of tape orthogonal to the rotational axis) the body of the potentiometer will move much less and the stick will tend to retain the trim setting.

Note that calibrating the speed controllers will NOT correct this problem. The speed controller calibration can correct an initial out-of-trim condition, but once the stick is moved the center values will change again and the control will once again require trimming. If you repeat the procedure enough times you may find a stable position, but in most cases it’s quicker, easier, and more reliable to tape the trim controls.

Jerry

11

Posted by Jerry Eckert.

Engineer from Looking for a team in Raleigh, NC sponsored by .

Posted on 2/13/2000 7:32 AM MST

In Reply to: joystick problem posted by Igor Likhotkin on 2/12/2000 8:37 PM MST:

: Hi all!

: My name is Igor. I am from team 419.

: We seem to have a problem with one of our joysticks. When I run default program on our robot,
: eventually one of the motors keeps getting some ‘juice’ even though nobody is touching the joysticks.
: I tried swapping the joysticks which made the other motor act the same way. I tried looking at what is
: happening to the input variable responsible for that joystick and found out that even though the joystick
: sits in the middle, the variable value can be as high as 130-160 decimal. I tried calibrating the y-axis
: with the wheel on the joystick, but, alas, after moving the joystick back and forth and then leaving it
: alone it would still leak some ‘juice’ into the motor.

: I assumed that since we ran the robot with the default code and since the recalibration of the joystick had
: no effect on the performance, the joystick was bad.

: Has anyone else had that problem? I really don’t know what to think at this point.

The trim controls on the joystick can easily be knocked out of adjustment simply by moving the stick. The first thing I’d try is using tape to secure the trim controls.

Jerry

12

Posted by Mike Gray.

Engineer on team HOT from Huron Valley Schools sponsored by GM Milford Proving Ground.

Posted on 2/13/2000 1:04 PM MST

In Reply to: joystick problem posted by Igor Likhotkin on 2/12/2000 8:37 PM MST:

Joysticks tend to drift as a function of temperature. The biggest change occurs in the first 6 minutes, and settles at 99% in about 30 minutes. (Actual times may vary):slight_smile:

Suggestions:

  1. Employ the DEADSTICK subroutine posted under White Papers at the InnovationFIRST website. It will help prevent ‘leakage’ and motor heating. You want all motors to be as cool as possible when the match starts.

  2. Keep the joysticks warm by keeping power applied. The potentiometers in the joysticks are used in current mode (as a rheostat)rather than voltage mode.

If you have trimmed the joysticks in the pits and kept them warm, then they should return to proper trim 2-3 minutes after you plug in at the player station.

  1. Connect an LED to an unused relay and program the relay to fire when the joystick is properly trimmed (126-128 A-D counts). Place the LED so as to be visible from the player station. This is more difficult than last year, as a relay module is now required. You could use the Dashboard to indicate a properly trimmed joystick too.
13

Posted by Rick Berube.

Engineer on team #121, Rhode Warriors, from Middletown H.S…

Posted on 2/24/2000 8:16 PM MST

In Reply to: joystick problem posted by Igor Likhotkin on 2/12/2000 8:37 PM MST:

Igor,

just implement a DeadZone about the center of your joysticks in software. We have done this for years. We simply implement a count of ~20 on either side of 127 as our deadzone (your limits will vary with your drivetrain).
Aside from that, we have our players trim the joysticks before each match by adjusting the trims forward (until we see movement) and then backwards (until we see movement in teh other direction). Then move teh trim as close to the center of these two positions as possible. Works every time!

Rick

14