Robot behaving strangely

[Kevin Phan]

After a quick search around chief delphi and a look in the technical forum on the first forum, I have not found a solution to what is happening to the robot. Recently, a side of the robot will drive without any command from the joystick. The weird thing here is that even using a blank project, the side will move. Now during testing, I changed which pwm for each side in the program.(i.e. pwm 2 for the left is now pwm 1 for the left) When doing this, the other side will now move without any command from the joystick. I probed the joystick output values and it verified that the joystick is not responsible for the issue. The robot worked before Tuesday night.

There's another issue I have when deploying code, is that the cRIO is not recognized. Yes, I have checked the laptop's ip, it is set-up correctly. Yes, I have tried switching cables. Only when I rewrite my code from scratch in a new project will it work.

Now I am at a dead end with this, I tried power cycling and reformatting the cRIO. I also checked the cables themselves and they all are wired correctly.

The vi we are trying to drive is with the tank drive vi. Temporarily it worked when we deleted the vi and bringing it back into the program. If it helps, the motors are y'ed together and they both show a solid green led light and the other side show the orange light. Again we used a blank project and in the project we are working in, we disabled all of our other code in tele-op except for the drive. There is also nothing filtering the joystick value to the tank vi.

[sanelss]

maybe a labview corruption of some sort? It doesn't sounds like it's a speed controller or electrical issue if the sides change when u change motor assignments. Try bypassing the tank drive vi. Set up 2 motors and just send the joystick value to all the motors and see how it behaves. you may need to re install labview or at least redo the tank drive vi if that is indeed the issue.

[Cecil]

Are all 3 lights on your Digital Sidecar brightly lit? Try replacing the y cables with individual PWM cables and see if it clears up. It may be a bad cable, but something makes me think its more than that.

[sanelss]

Originally Posted by Cecil

Are all 3 lights on your Digital Sidecar brightly lit? Try replacing the y cables with individual PWM cables and see if it clears up. It may be a bad cable, but something makes me think its more than that.

from what he said:

I changed which pwm for each side in the program.(i.e. pwm 2 for the left is now pwm 1 for the left) When doing this, the other side will now move

so a change in the software changes the behvarios so it sounds like software issue not hardware

[Cecil]

Originally Posted by sanelss

so a change in the software changes the behvarios so it sounds like software issue not hardware

Darn my short term memory

Are you hard wiring your robot to code/drive with? We have run into a problem before when other members were driving a robot from the past year. That driver station then connected to the robot we were programming and about drove it off the table.

If that is not the case, I can't think of a half decent reason it would do that, especially after changing the PWM ports.

[Alan Anderson]

Originally Posted by Kevin Phan

If it helps, the motors are y'ed together and they both show a solid green led light and the other side show the orange light.

Green means it's receiving a signal that tells it to drive forward. What kind of motor controller is it? Did you somehow miscalibrate them so that they think a neutral command is actually forward? Try resetting them to factory calibration.

[Kevin Phan]

Here's some more details to clear some things up. We are using four talons for our speed controller on the drive. We are directly connected to the cRIO throughout the entire time currently in the season. The talons have not been calibrated but early in the season we tested them with basic drive code they worked well. We made sure the cRIO has the right team address.(10.3.57.2)
Programming wise it is a straight wire from the axis of the joystick to the axis input to the tank drive.

[Jon Stratis]

Do you have a scope available to probe the PWM signal being sent to each motor controller?

[Kevin Phan]

Originally Posted by Jon Stratis

Do you have a scope available to probe the PWM signal being sent to each motor controller?

I do not think so. I looked in the wpi tank drive control and the motor values are dependent on the joystick values. We probed the joystick value and found that it was 0 or .007. Which is too low of a value to have the motors going full forward.

[Alan Anderson]

Originally Posted by Kevin Phan

There's another issue I have when deploying code, is that the cRIO is not recognized. Yes, I have checked the laptop's ip, it is set-up correctly. Yes, I have tried switching cables. Only when I rewrite my code from scratch in a new project will it work.

In the LabVIEW project, do you have the target IP address set to 10.3.57.2, or is it 10.03.57.2? I've found that having a leading zero can confuse the networking sometimes.

[Kevin Phan]

Originally Posted by Alan Anderson

In the LabVIEW project, do you have the target IP address set to 10.3.57.2, or is it 10.03.57.2? I've found that having a leading zero can confuse the networking sometimes.

Yes it is setup as 10.3.57.2.

Something else I forgot to add was that in the tele-op vi (where the drive code is), I gave a constant of zero to both sides and the side that moved without joystick command, moved. I moved the joystick axis and the other side did not move. I know that my coding is not wrong on the surface, but I don't know how in the world is this happening.

[Kevin Phan]

Tonight we created a new project and bit by bit reprogrammed the robot. I'm not sure what caused this phenomenon, but if it comes back...