We’ve been testing our shooter, and we’ve been noticing that the height the ball hits our target at changes frequently, even though nothing about the robot’s positioning or the robot itself changes. We even tested with the same ball, entering the shooter in the same way, and the height the ball reaches still changes. The shooter is made up of two smooth wheels under the ball, both on the same axle and powered by two CIMs without a gearbox, that give it backspin and a hood that gives the balls the correct angle (sorry, but I don’t have any pictures). What are some reasons that these inconsistencies could be appearing?
How are you controlling the motor/wheel speed? (Ideally this should be with an encoder, but at a minumum you should do voltage compensation.)
How much play is in your hood?
Are the two wheels mechanically linked to the same speed? (You say same axle, but some might say this when they’re just rotating on the same axis next to each other; just to be sure.)
What type of wheels (in particular, to know if are they compliant or relatively solid - for shooting compliant balls, solid wheels are better and more consistent)?
1 Like
The wheels, which are the AndyMark SmoothGrip 6" wheels, are on the same axle, powered on each side by a CIM, with no gearboxes. An encoder on the axle gives us our RPM, which we are using to change the distance we shoot at. The RPM has not changed with different heights of shots, however. We’ve also watched our hood while it shoots, and it does not flex in any noticeable amount.
If you’re shooting at a relatively steep angle, then the problem is likely related to the velocity out of the shooter. If you’re shooting at a relatively shallow angle then it’s probably related to your release angle. Do you have any pictures of your setup that you’re willing to share?
Sorry, I don’t see the hood. By “slightly less than 45 degrees” do you mean closer to the horizontal or closer to the vertical?
Also, how much compression are you doing (That is, what is the ball diameter minus the distance between the wheel surface and the inside of the shooter hood.)?
Closer to horizontal.
There is 1" of compression.
I haven’t worked with the game piece myself, but that is rather light for a shooter this year. As I’ve read it, most teams have been at or a little over 2".
Do you, by chance, know how much of a difference that change in compression makes? My team didn’t really experiment too much with the compression. Would the compression affect the ball’s height?
It can make a difference. I would add something in there that would make some thickness that is flexible that can follow your shooter. Something like a few pieces of cardboard that you can tape on and see if it makes a significant difference for you.
1 Like
A low compression will result in lower speeds and greater variation in speed. That variation in speed will change the vertical postion at the target, with the variation increasing as you get farther from the target. How far away are you typically when taking shots?
Yes, that. I was thinking of closed cell foam or corrugated plastic as other possibilities.
3 Likes
We’ve been shooting from the area between the sector line and the color wheel. When we shoot from the sector line, our ball’s height can vary about a foot either direction. Usually, the variance is close to 6", however.
We’ll definitely try that tomorrow, thanks!
We’re relatively consistent up close, but the small errors are magnified the farther we get from the power port
That’s likely to be true of any issue, but it’s certainly consistent with a speed inconsistency. Just as a sanity check, you may want to have your programmers report the speed when the ball is pushed in to the flywheel, and the minimum speed of the wheel during the shot, to see which (if either) correlates to shot altitude from a fixed range.
Is your robot rocking? Many teams choose to use 4 wheeled drivetrains or small drop centers in order to keep their robot from rocking when shooting.
The robot does sway slightly, but the variance still appears when we shoot a single ball, then let the robot settle, and shoot the same ball again.
It looks like you’re using a KoP chassis. This normally experiences minimal rocking. If this is a concern to you, weight the robot to either the front or rear so that you are certain that you are on the same four wheels when shooting.
1 Like
We have found slow motion video (easy from many smart phones these days), to be very useful in finding causes of error in fast moving mechanisms.
Also log your flywheel speed, there may be more variation there than you expect, but you will not know unless you measure it.
1 Like