I wanted to explore what it takes to run a shooter wheel in the range of 1000-1500 rpm. We touched on this slightly in another thread a few days ago when discussing how fast the old red quad encoder would run, but my original setup for that test was less than idea so put together a new test rig shown below.
393 motor on high speed
first stage gear 60:36
second stage 30:12 using chain
third stage 30:12 using chain
total gearing is 1.66 * 2.5 * 2.5 = 10.41
Theoretical speed of the omniwheel is 160rpm * 10.41 = 1666rpm.
Actual speed of the omniwheel is about 1350rpm
The motor is running at about 80% free speed, 130rpm. Current at this speed is about 1.1A so almost max efficiency and we are well below the PTC trip point.
As others have already noted, safety glasses will need to be strictly enforced this year, even at this speed it’s quite a beast (compared to what we are used to) and shaft collars or other small objects coming loose will/may be a problem.
A couple of notes.
Don’t start the motor at maximum power, ramp up gently over 2 or 3 seconds, current is high as the system comes up to speed and you want to minimize PTC heating.
There’s a lot of momentum when the power is removed, I hope that’s not going to be an issue, I plan to ramp down the motor control to reduce stress on the system but we have no control of robots being disabled at the end of the match.
There’s almost no torque on the output omniwheel (not surprising). The 9.2 in-lb of torque the motor has been reduced to 0.88 in-lb, as the omniwheel has a 2 inch radius there is very little force needed to stop it. Anything that jams the shooter will trip the PTC almost instantly as it’s so easy to stall the motor.
Very interesting. Just curious: why did you use chains instead of gears? I would think the added side force of the chain might put more friction on the axles???
No particular reason, I wanted to avoid the 12 tooth pinion gear, I wanted some slop in the system, decoupling the final drive using a chain helps with tolerances (ie. things don’t have to be aligned quite so well) and I just happened to have sprockets and chain lying around. Started thinking about bicycles etc. and it turned out this way. At some point I will try some alternatives.
As I don’t have any parts to play with at the moment, it’ll be interesting to see how this works! How much power does it have at that speed, is it very easy to stall?
Another safety thing to watch out for would be to make sure you don’t lean over the robot and get loose shirts or clothing stuck in the wheel. It probably wouldn’t do too much harm as there isn’t much force, but still something to be careful about.
On a 26" wheel spinning that fast it’s around 115mph.
Thank you for making this thread and posting the videos. I notice that you are only using 1 motor per side for 2 motors total. I’m assuming that to fire a ball another identical shooter wheel would go on the other side? Do you think those 2 motors will have enough power to launch a ball across the field or will you have to switch to 4 motors? Is the internal motor gearing more efficient than the external compound gearing? If so what would be the affect of using turbo speed internal gearing?
I’m not going to draw too many conclusions from this initial test, it’s way too early in the design process and you all should be doing this really not me. However, this is what I think so far
I did this in response to the idea of measuring output shaft speed directly using a quad encoder. Based on this, I think that’s a bad idea.
One motor per side of a two wheel shooter may work, that could be the next test. The two wheels should probably be mechanically coupled.
Internal vs external gears, no real opinion on that, never tried the turbo gears.
Unfortunately, like everyone else, I don’t have game objects. Wish I had grabbed one from Grant at worlds, Oh well.
It was just there so I had something to put that small white arrow on
Seriously, don’t read too much into this design, I was curious about a few things and put this together in about an hour just to see what would happen. Would the motor be able to spin the wheel, how much current would it use, how easy to stall, how much vibration, things like that. To measure the speed of the wheel I shot the video at 60fps and then measured (approximately) the movement of that small white arrow between frames.
Great video. I agree this will be a noisy game since we will likely have 2 or 3 wheels driven seperately.
Now x that by 4 robots and the drivers and audience will be wearing ear muffs like NASCAR races:D
At 1500 rpm, I estimate about 150 mph! 