Stabilizing Lift

Okay, so I posted something a couple days ago about having trouble with our arm lifting crooked. The image is a picture of our gearbox (three motors on each side of the arm). It is a 7:1 gear ratio. The bottom of the c-channel is attached to the 84 tooth gear and the top is attached to a shaft using lock plates. Currently, when we run our motors, the lift is going extremely crooked. We have looked at the internal gearing and they are all the same. We added integrated encoders to all of our lift motors. Our next step was going to use PID control to stabilize our left. Is anyone else doing the same? If so, can I see the code, or just sample code? I’m honestly not sure where to start with this, any other suggestions would also be helpful
IMG_4288.JPG

Are you setting the speed to 127 to all motors? Are you using a high-speed motor for one side and regular motors for the other side?

@Liam Matthews
All of the motors are geared the same. Regular Motors. At the beginning we had all of the motors at 127, but then we tried just a P loop where it changed the motor speeds dependent on how many clicks the encoders are off (at this point we only had encoders on the output gear, so it may work using all six). But, using just the top encoders, a P loop by itself didn’t work

Makes sense, but I’m sorry pal. I’m out of ideas.

Clarification:
Is the lift crooked when you lift the arm up (aka, does it seem the left and right side are lifting at different rates?)

Knowing all your gears and programming is correct, the first thing I’d check is if all your motors are even running. Test them individually. It could be something as simple as a motor controller going out.

As we have a record of destroying motor controller, check them, also check if all of your motors are working and if everything is plugged in.

@idenzer
The lift is crooked when you lift the arm up, they are lifting at different rates. And we did check, all of the motors are running, we ran all of them separately. Also, we tried running pairs of motors, say the bottom ones on each side, and it still lifted crooked. (We did this for the three pairs) The gearboxes are exactly the same. So I am just guessing that it is the motors running at different speeds, we haven’t gotten new motors in awhile

Check to see if the friction is different on both gear ratios. We had this problem and it was because one side had the wrong size spacer. Once it was fixed we had no problems

@Gameon2546
The friction should be the same. The sides are identical, down to how tight the shaft collars are

have you checked your bearing blocks, sometimes they maybe the problem…

@vex2.0
I have, we have checked the tightness of everything, including the motors. They are rigid when the lift moves up and down

Yeah if I look at the picture closely one of the gears is shifted further down than the others. Could be missing a bearing entirely.

Try using spacers. You’re only using collars (I think some people call them shaft locks or something).

@AcidEpicice @vex2.0
This was an old picture, we noticed that too and fixed it. I probably should have mentioned that in the first place, but that is true, spacers could help. But does anyone have experience using PID? Any pointers?

Well… Here are some of my thoughts:

  • Have you checked that the side that is crooked has ALL of the motors going the same direction? Maybe you should tamper the polars or reverse in programming and do trial/error.
  • Sometimes one can be faster because it has more friction. On the crooked side, try checking if the axles aren’t on an angle from one hole to the other. How about trying to loosen friction on the shaft collars next to the bearing flats? For the gears on axles without motors how about changing the gears to have the round-hole plastic inserts instead of the square metal inserts? Or maybe try lubrication?
  • Just tamper the speed of the sides (Like motor=joystick*0.9) rather than having to figure out what a PID is then trying to make a good enough program with PID which will take a long time to understand…

Try rebuilding the entire thing with UNUSED vex parts, I believe that either your c- channel is ever so slightly bent, causing more friction on one side, or that some if the gears are more worn than others

That wouldn’t cause so much asynchrony. Metal likely isn’t the issue.

You shouldn’t need to use PID for the lift. Something mechanical or electrical HAS to be wrong.

Hmm… I see collar nuts next to 12 tooth pinions. This has given me tonnes of problems before; often times the gear meshing with the pinion gets caught on the collar nut. Have you tried using black spacers in between those and the gear?

I apologize if this was mentioned before, but did you check the motor controllers? My 6th grade team had their lift leaning to one side and we later found out that one of the motor encoders was bad, so basically it was 2 lifting on one side and 3 lifting on the other.