Realistic high goal scoring rate
 

I keep seeing discussion about the high goal scoring rate required to maximize score. This is all well and good, but it's all theoretical. What do you think is a realistic rate of fire for a high goal shooter for an individual robot. An explanation of how you reach that number would be appreciated :)

Re: Realistic high goal scoring rate
 

I think realistically a well refined Ri3D 1.0 shooter could reach 3 balls/sec. And I believe there will be at least one powerhouse team with more than one shooter on their robot.

Re: Realistic high goal scoring rate
 

Well, that depends on how many balls you can shoot at the same time...

Re: Realistic high goal scoring rate
 

This is also the sentiment from my team. We are going into this season under the idea that great teams will have 2 or 3 shooters at around 3 fuel/second.

Re: Realistic high goal scoring rate
 

how about 12 fuel per second?

Re: Realistic high goal scoring rate
 

Are you guys French?

My team was testing out our shooter yesterday and we were able to do at least 3 per second pretty accurately

Re: Realistic high goal scoring rate
 

yes, sorry about that :rolleyes:

Re: Realistic high goal scoring rate
 

No problem I was just surprised lol

Re: Realistic high goal scoring rate
 

Yes, but accuracy. At that distance you're going to need much tighter control of the ball velocity. And the closer you get, the lower your rate of fire has to be.

Re: Realistic high goal scoring rate
 

we were shooting from that distance to see how the wheel slowed down. By adding mass to the flywheel and shooting from closer, accuracy will go up a lot

Re: Realistic high goal scoring rate
 

Part of why I'm asking. You tell me ;)

EDIT:
Very cool! How to you plan on implementing that hopper in a 36" (or 24") tall robot :p

Re: Realistic high goal scoring rate
 

I think top teams will get to a dozen per second, absolutely. Multiple ways to accomplish this shooting task this year - if the way you have in mind can't do it, think outside the box a little.

This isn't 2016 - volume and accuracy are tradeoffs where the right answer very well may be on the side of volume. Turreted shooter in 2009 could be more accurate than dumpers, but dumpers certainly were the more dominant design in general that year. This year's game has five times as many game pieces as 2009!

Re: Realistic high goal scoring rate
 

it seems possible for now to stay under 24", that would allow for a larger drive base, and a larger intake

Re: Realistic high goal scoring rate
 

I need to revisit my flywheel spinup spreadsheet and add a few new features to it. Think I'll do that tonight.

At any rate, if you're using an encoder to do closed loop control of your shooter speed, you want a specific gear ratio to maximize your rate of fire. You want a gear ratio that has the motor spinning at 1/2 its free speed when your wheel is at target speed, or something very close to this (err on the side of a bit more reduction). This puts the motor at its peak power point when its at firing speed, which will minimize the recovery time to target speed, or atleast minimize the speed drop when dumping an entire load of balls.

Interestingly, this is not the gear ratio that spins the motor up from zero to target speed the fastest. That depends on the system dynamics, but the example I'm working with has that ratio at 40% more reduction, and increases recovery time by 16%. Which would give you a rate of fire of 86% the theoretical max.

Re: Realistic high goal scoring rate
 

I don't think that's how it works. The motor is only at max power spinning half its free speed at full voltage, IE if the load the motor is under is slowing it down that much. If you're just spinning the motor at half its free speed with speed control, you're just applying a bit more than 6 volts to the motor (a bit more to account for friction losses etc). You definitely do want to be spinning a flywheel at less than 100% speed so it has some headroom to recover with, but I don't think it's exactly half voltage or half free speed either.

Re: Realistic high goal scoring rate
 

I am in complete agreement with this, actually. The important shooter metric is balls scored per second, not accuracy or balls fired per second. Erring on the side of volume is almost certainly preferred because it's probably easier to increase accuracy for a high volume system than to increase volume for a high accuracy system. I think it IS interesting that if you're pursuing a single stream shooter*, at some point you have to start backing away from the goal to increase your rate of fire without running balls into each other.

*How long do you think until the first joke about crossing the streams?
Like I said, I need to revisit my spreadsheet to better account for this. But two things are true about my statement:
1. If your wheel slows down by x% per ball, the fastest way to spin it back up to target speed is at at the peak mechanical power point. (Physics, that) And the peak power is at 12V and 50% free speed.
2. Peak power is the MOST power you'll ever get out of the motor. That's 337W for a single CIM. If it takes 33J to fire a ball, the most balls you can fire without slowing down is 10 balls per second. And that's going to be with the motor running at 1/2 free speed. If the motor's running any faster, it won't put enough energy back into the system and the wheel will slow until energy out = energy in.

Re: Realistic high goal scoring rate
 

Re: Realistic high goal scoring rate
 

14:06-14:13, so ~7 minutes

Longer than I expected actually.

Re: Realistic high goal scoring rate
 

He has to have some kind of intelligent filter that notifies him of joke opportunities, I swear.

Re: Realistic high goal scoring rate
 

That's the first time I've ever heard him referred to as "intelligent"

Re: Realistic high goal scoring rate
 

We are looking at about 12 balls per second at the moment with our protoype with consistency.

Re: Realistic high goal scoring rate
 

97.1 balls per second. ;)

Re: Realistic high goal scoring rate
 

next thing you know 254 does better than that

Re: Realistic high goal scoring rate
 

So, check my math, but assuming a single Fuel leaving the shooter wheel which is located at the top of a tall robot parked in the key against the boiler wall. My trajectory model calculates about 86 degree launch angle at about 5.7 m/s muzzle velocity. At a 5" pitch, that is 40 to 45 Fuels per second.

So that would be the max; of course, there would be a gap as you wouldn't be feeding the shooter as fast, so something less than that.

Am I missing something?

Re: Realistic high goal scoring rate
 

Does that include drag on the wiffleball?

Re: Realistic high goal scoring rate
 

Without checking your math, that seems like a solid upper limit "in flight". The feed process into the shooter will definitely be the limiting factor for top teams.

Re: Realistic high goal scoring rate
 

Peak rate or sustained rate? I could see some team with peak rates well above 30 balls/second.
Flywheel shooters are the only way to launch a ball

Re: Realistic high goal scoring rate
 

How fast is the ball traveling at the apex of your trajectory? The slowest speed on your trajectory is what you should use with pitch to calculate balls per second.

Re: Realistic high goal scoring rate
 

I can't hear you over my ACTUAL TRANSFORMER ROBOT*



*Whose design I didn't contribute to in the slightest

Re: Realistic high goal scoring rate
 

At 25 ft/s, that's about 12 lbs of thrust. I don't think you'll be going to space today with that engine.

Re: Realistic high goal scoring rate
 

Nah. We'll just fix it in code :D

Re: Realistic high goal scoring rate
 

Anybody tested the amount of bounce the balls have against the entrance?

Re: Realistic high goal scoring rate
 

The model is simple and does not calculate the influence of any drag or lift based on ball spin. In terms of rate, I use the muzzle velocity and pitch as representative of the rate of the balls leaving the shooter. Assuming all balls will have the same velocity profile and land in the Boiler, then what goes out of the shooter at a specific rate gets scored at the same rate. Right?

And yes, the feed rate to the shooter is the rate limiting step. Recognize, too, that feeding multiple balls to a multiple-ball wide shooter will increase the discharge rate. It may also create in-flight interactions which could reduce efficiency of scoring.

Re: Realistic high goal scoring rate
 

I'm thinking the best teams will be around 10 balls/second. During our Ri3D Testing, we were around 1.25 balls/second. This was only with 1 shooter on our robot, and a fairly unaggressive agitator. I imagine some teams will have three or more shooters on their robot and very aggressive agitators.

We're going to be seeing some robot machine guns out there... this game is going to be a fun one.

Re: Realistic high goal scoring rate
 

Those balls aren't traveling at a constant speed, even without drag. If you shoot a ball straight up, it doesn't keep going at the same speed, gravity slows it down. Vertical speed drops to 0 at the top of the arc. At the top of the arc, the balls are closer together because they only have the horizontal component of velocity left. At 86 degrees, you don't have much horizontal component. Specifically, 15 in/s. So you can only shoot 3 per second in a straight line without a traffic jam in mid air. There, I checked your math for you.

Re: Realistic high goal scoring rate
 

and this is why I'm hoping most shooters are shooting NOT from the front of the boiler, but instead further back

Re: Realistic high goal scoring rate
 

I would take a bet and say 3 balls a second will be the average scoring rate. Just a thought though.

Re: Realistic high goal scoring rate
 

Point blank shooters need to be 2 or 3 wide to make any sense, honestly.

Re: Realistic high goal scoring rate
 

I'll take 3 accurate bps over a spray and pray any day.

Re: Realistic high goal scoring rate
 

In this game, I don't know. If you're shooting at 3 bps, it takes you 16 seconds to shoot a Hopper's worth of balls for 16 points. If you shoot at 10 bps at 50% accuracy, you take 5 seconds to shoot a Hopper's worth of balls for 25 balls or 8 points in 5 seconds. It obviously depends on the accuracy and the shooting rate, but I'd take the second robot in this case.

Re: Realistic high goal scoring rate
 

especially with a fast ground pickup. Shoot two hoppers of balls, pickup your misses for cycle 3.

I think really effective shooters will either have figured out the distance, speed, accuracy problem, or will shoot layups massively quickly with open top hoppers to recycle the misses. If it takes you 5 seconds to empty your hotter, your first misses are already back in it by the end.

Re: Realistic high goal scoring rate
 

I consulted a very, very intelligent aero friend of mine and called in a favor. His model indicates an optimistic 7 bps directly in front of the goal from a single stream shooter. Coincidentally the tube below the funnel will accommodate 3 fuel with 1/4" between game pieces and between game pieces and the wall. I'm seeing something on the order of 18-21 fuel per second in our future potentially. I will go on record as saying that you need some kind of metering ahead of your shooter and that interface will be EXTREMELY critical if you're aiming for this kind of Lunacy™.

Re: Realistic high goal scoring rate
 

Once you work that out, you should wind it up to 8 or 9 per stream, just for the video of the midair traffic jam falling back on your robot.

Re: Realistic high goal scoring rate
 

50% accuracy doesn't qualify as spray and pray for me. If you can find a way in any scenario to more than triple your bps at an exchange of only half the accuracy, I'd take that deal too.

Re: Realistic high goal scoring rate
 

There's flywheel shooter spray and pray where you pump out as many balls as you can past a flywheel shooter at a somewhat sustained rate for a few seconds in somewhat the same area.

Then there's "I shoot my entire hopper at once" spray and pray where you fling 50 balls into the air at once and hope that a large portion of them land in the goal funnel.

Re: Realistic high goal scoring rate
 

*Waits to see the 50 ball catapult*

Re: Realistic high goal scoring rate
 

I suspect what whatever we build will be powered to accommodate this, and I promise a video when it happens. Generally we build this type of mechanism with our target to hit peak power on the motor curves when possible. This means generally we have twice the throughput that the system is designed for. Should be a hoot.

Re: Realistic high goal scoring rate
 

The biggest limitation here is not going to be the accuracy or power of your shooter, but the indexing of the fuel leading up to it. Getting the balls from a bulk storage in the robot to a single file line ready to go into the shooter is not trivial and can easily limit your otherwise perfect robot.

Re: Realistic high goal scoring rate
 

See Simbotics 2009 robot

Re: Realistic high goal scoring rate
 

Why must they be in a single file line?

Re: Realistic high goal scoring rate
 

I imagine he was thinking of a single wheel shooter similar to almost all the RI3D teams have been using

Re: Realistic high goal scoring rate
 

Wide wheel shooting is possible but harder to do well. Lots more things to control. It may be easier than indexing single file, but that is what build season is for.

Re: Realistic high goal scoring rate
 

I wonder if we will see multi stage shooters like the ones used in 2013 return. I could foresee it being more consistent at a high ROF.

Re: Realistic high goal scoring rate
 

The assumption you are making is that all balls are still lined up at the apex. Break that assumption, and things start to get way more interesting :)

Re: Realistic high goal scoring rate
 

A limited version of this might be a good way to control rof. Small sticky wheel just in front of your shooter spinning at a controlled rate. Conveyor system Infront of the rate control wheel can shove balls together at higher speed and provide a consistent stream for the rate control wheel to pay out.

Re: Realistic high goal scoring rate
 

Re: Realistic high goal scoring rate
 

Do you know what this model is exactly? I'd like to show it to my team.

Re: Realistic high goal scoring rate
 

Why is it harder to do well? Other than spinup time / speed loss (which is a throughput problem not a width problem), I don't necessarily think this is true at all, and it is certainly easier to get a large ball throughput this way.

Re: Realistic high goal scoring rate
 

My big thing is fitting everything together in a way that emphasizes the priorities in use of space. My old team, FRC 2970, came up with a good design in 2009 for ball throughput from a static hopper. Check us out on TBA for details. My biggest concern isn't so much dump/shooter design, but space and positioning everything else, which is a necessary detail that I feel most teams will need to acknowledge. Considering we can minimize the climb mechanism to around 4 or 5 inches tall at the top of the robot, and we've got around 9 inches on the bottom for drivetrain and intake, and one side has a gear delivery mechanism, how do we optimize the space so that:

- Intake of balls into the on-robot hopper/container is easy
- Output isn't limited by necessary frame space
- We have the power and space for actuators
- We don't lose our balls
- Dumping is consistent and fast

Our priority is not to emphasize use of the top target, so we can use a dumper rather than a powered shooter or turret, but we want to be able to deliver fuel effectively as a backup plan.

Re: Realistic high goal scoring rate
 

Absolutely. This is where the discussion of spin and the possible benefit/cost of trying to exploit the magnus effect began. Looks like we're going to provide an adjustable mechanical bias to test that hypothesis as well, primarily comparing performance in cases where either no spin or forward spin are imparted on the game piece.

Re: Realistic high goal scoring rate
 

We have found that forward spin helps tremendously in clearing out air space so that the balls are less likely to bunch up near apex. Back spin made the balls linger near the apex. These are just our findings; take them as you wish.