Software control or gearing

[clandry94]

My team is going to be using a linear shooter with two wheels and is still deciding on whether we want to run the front wheel faster than the back wheel. Anyway, if we want to run one slower is it possible to put a limit on a Victor motor controller to have a max amount of output as to make the rear wheel run slower than the front instead of running max power, but having the rear wheel geared down. Are there any pros or cons to this?

[Johnbot]

Yes, you can cause the rear wheel to run slower if it is driven off of a second Victor. You don't necessarily have to put a limit on it, but you could have it scale. (I.e. if you run the front wheel at speed X, you can run the rear wheel at speed 0.80*X)

Anyways, I don't see anything wrong with doing either, but that's just on the software side.

[connor.worley]

If both wheels have the same gearing, making one slower in software will have no benefit.

[Ether]

Originally Posted by connor.worley

If both wheels have the same gearing, making one slower in software will have no benefit.

Why not ?

[theprgramerdude]

Originally Posted by Ether

Why not ?

^This. Software control should be used anyways along with some feedback mechanism for determining RPM, as invariably the power level required to get a wheel to spin at a certain speed will change over time due to various factors.

[Redo91]

Which option would be better for your team?

[Hebrew]

Has there been a tested advantage to having a wheel move slower than the other? Our team will start our shooter prototype on Monday (Bit of an ordering delay for needed parts) and I am curious about this now...

[nathan_hui]

I believe RobotIn3Days did something like that, but basically what the issue was that without a controller, the first wheel slowed down quite a bit first, so a second wheel was used to boost the muzzle velocity. Having the first wheel move slower would theoretically allow more overhead for a controller to work to improve the performance, but then again using the proper controller for a shooter wheel would improve the performance without the need for a second wheel.

Alternatively, without a controller, you could simply use the slower wheel to gradually speed up the wheel, so you might be able to shove more shots through.

[Ether]

Originally Posted by nathan_hui

using the proper controller for a [linear] shooter wheel would improve the performance without the need for a second wheel

I haven't seen any videos of a successful linear shooter with only one wheel.

With a linear shooter, the frisbee has only a fraction of the contact time that a curved one does. Because of this, the acceleration in the linear shooter must be much greater to achieve the same exit velocity.

Beyond a certain acceleration, excessive slippage will occur, resulting in poor performance.

Adding a second wheel increases the contact time, reducing the required acceleration, and reducing the slippage.

[F22Rapture]

Any thoughts on the viability of a hybrid shooter? IE, one that starts out with a small curve (say a 45 degree arc) and then straightens out and gets boosted by a second CIM?

[clandry94]

Originally Posted by F22Rapture

Any thoughts on the viability of a hybrid shooter? IE, one that starts out with a small curve (say a 45 degree arc) and then straightens out and gets boosted by a second CIM?

We considered that, but couldn't determine a design that would fit within our robot perimeter unfortunately.

[Mr V]

Originally Posted by connor.worley

If both wheels have the same gearing, making one slower in software will have no benefit.

Yes it will, the slower wheel can accelerate the disc a certain amount w/o slippage pass it on to the second wheel which can accelerate in more w/o slippage.

[Mr V]

Originally Posted by nathan_hui

Having the first wheel move slower would theoretically allow more overhead for a controller to work to improve the performance, but then again using the proper controller for a shooter wheel would improve the performance without the need for a second wheel.

The system can not respond fast enough for a "proper controller" to maintain its speed when accelerating the disc. By the time the control system can sense that it needs to add more power the contact event is most of the way over, at least on a linear shooter particularly if you are using a PID loop.

[RRLedford]

Originally Posted by Ether

I haven't seen any videos of a successful linear shooter with only one wheel.

With a linear shooter, the frisbee has only a fraction of the contact time that a curved one does. Because of this, the acceleration in the linear shooter must be much greater to achieve the same exit velocity.

Beyond a certain acceleration, excessive slippage will occur, resulting in poor performance.

Adding a second wheel increases the contact time, reducing the required acceleration, and reducing the slippage.

I haven't seen any such videos either, but I also haven't seen any one wheel axis, linear shooters where the pinch edge opposing the one wheel is capable of dynamically flexing and pivoting enough, as the disk passes through, to significantly expand length of the arc segment of contact with the single wheel.

I also haven't seen much reporting on the seriously wide variations of wheel-to-disk friction grip that different formula urethane wheel materials can produce.

We have already found that premium skate wheels at durometer a measured firmness of 75A have dramatically better grip than other wheels only slightly firmer, and with a seemingly less "grippy" urethane formula.

In the wold of skate wheels (boards, inlines, and quads) the range or tweaking of urethane formulas to achieve grip, roll and rebound performance benefits is quite incredible. I highly advise all teams who are doing small wheel (one or two) shooters consider evaluating their options in the skate wheel category. Lots of size selections also available.

These efforts could prove to be as valuable as RPM tweaking, or at least supplement them.

-Dick Ledford