Tennis Ball Machine

As you can expect, I'm relatively new to the Arduino scene. I have a BSEE with a focus in small circuit design and programming in c,c++,c# was a hobby of mine growing up.

My goal is to learn about microcontrollers, motors, and parts assembly while being productive with my efforts at the same time. This wasn't difficult as my wife is starting to take Tennis lessons from me and a Tennis Ball Machine would combine all of the mentioned subjects. I'm drawing everything up in Autodesk Inventor and the wiring schematics will be done in CADSoft or Visio. I am not afraid to 3D print any enclosures or structure pieces if necessary.

Requirements: - 2 spinning vertical wheel design - Tennis ball must travel up to 80mph from machine- I may use it for myself as well - Enclosure must oscillate 15 degrees left/right and up/down - Must be portable and run on rechargeable batteries - Light weight (max 40lbs) - She may want to bring it out herself. - Remote control of at least 80 ft - length of tennis court - Controller is its own part. I do not wish to have a phone or computer as the controller for these parts could change while a dedicated controller would not. - Any wireless communications method is permitted but nearby signals should not interfere with range/quality of signal - Motor must be of varying speeds to allow for top wheel RPM to differ from bottom and using the remote control, increase/decrease rpm's of top & bottom motors independently

I am not here for guidance on how to physically build the structure. Instead I'm here for the electronics controls portion and motor selection. Which parts to buy and how to implement.

Research: I've done an extensive bit of research already. I know that I can build a gearbox to increase torque coming out of a motor but I wish to maintain high RPM due to the ball speed requirement. I understand the wheel RPM will decrease as the ball is compressed and passed through the opening between the wheels and that electric current draw is the main reason for battery drain. This is due to the torque required to "spin-up" the wheel from a lower RPM to the set point value. I may need to drive the motors with a separate single power supply (some kids toys have ideal rechargeable batteries in them such as the cars/bikes that are motorized).

I assume that a DC motor is best for the rotation of the main wheels, while a stepper motor would be best for creating the oscillation requirement of turning the contraption 15 degrees left/right/up/down.

A good starting point for me is which parts should I be interested in from Arduino and other companies interfacing with the Arduino as well as which motor would best suite this project. I'm a huge fan of math and physics so if you wish to provide, please do not be shy.

Thanks!

I suspect your main issues are going to me the power requirements.

Have you worked out how many watts of power the motor requires.

E.g 1000w

You probably need to use LIPO batteries to make the unit portable, but I think it could end up being quite heavy.

BTW

I presume you have seen this design

http://www.rs-online.com/designspark/designshare/eng/projects/121/view/stage/brief/

Using the tool from http://www.robotshop.com/blog/en/drive-motor-sizing-tool-9698, I came up with a few parameters:

User input:
Total mass of wheel: 1 lbs
Number of Drive motors: 2
Radius of wheel: 4 inch.
Velocity: 176 ft/s = 120 mph
Supply voltage: 12v

Calculated results:
Angular Velocity: 5,044 RPM
Torque: 0.78 Nm
Total Power: 199.68 W
Maximum Current: 16.64 Amps
Battery Pack: 49.9 Ah

I’m likely confusing those input parameters for something I’m not trying to do but it was a guideline for selecting a motor. If I’m incorrect, please let me know. This isn’t my field of study.

That design I have seen before and they are using a phone as their remote control. I am considering now to have another mini arduino board with LED and buttons in a plastic container as my remote controller since button finding will be difficult. Not sure. What I did notice was that they are powering the two devices from the L298 IC on the one board. Intriguing.

Battery Pack: 49.9 Ah

With lipos, I was trying to do a weight calculation, and it would be around 7 pounds in weight, and would cost you in the order of $700 just for the batteries.

http://www.hobbyking.com/hobbyking/store/__53878__Walkera_5200mAh_3S_Lipoly_Battery_Pack_for_QR_X350_PRO.html

Using lead acid would blow your weight budget completely. I suspect the NiMh would also be too heavy albeit slightly cheaper.

What was your $ budget for this project ??

How about using a pneumatic piston to fire the balls and a tank of compressed air to drive it? Just concentrate on short, high torque strokes and it won't empty as fast as an air-cannon would.

You could fill it up at a gas station or from a car electric tire pump. That's cheap, portable power.

From what I recall reading various articles on energy storage, that the energy density of compressed air is not very good. Thats why they don't run cars on compressed air, they use Lipo batteries

Also, I recall a MythBusters episode when they needed to make a machine to throw balls of some sort, and after trying the piston one, they found its results were not very repeatable, hence they had to build a 2 wheel version like most of these machines are (and the OP intends to build)

I presume you have seen this design

http://www.rs-online.com/designspark/designshare/eng/projects/121/view/stage/brief/

It looks as though that design allows for topspin and backspin; it would be nice if it were possible to rotate the firing wheels assembly so that you could do left & right too.

wildbill: It looks as though that design allows for topspin and backspin; it would be nice if it were possible to rotate the firing wheels assembly so that you could do left & right too.

I intend to place a stepper motor mounted on the base plate to the wheel mount and have it rotate on a central pivot point. Keeping a repeatable balance between each shot means I'll need to come up with either a large pivot point or a ball bearing system in a circular track. Obviously, the stepper motor would be connected to a pully near the front of the enclosure and mounted to the enclosures base structure such that each incremental rotation of the stepper motor will pull a chain/rope/string attached to an anchor point on the swivel structure holding the wheels. Thus pulling the structure forward in a rotational manner. The trick is, getting it to rotate back; I may use springs or two steppers, not sure.

Using this same method, I plan to create a vertical oscillation as well (varying speeds in the program to accomodate for distance).

GoForSmoke: How about using a pneumatic piston to fire the balls and a tank of compressed air to drive it? Just concentrate on short, high torque strokes and it won't empty as fast as an air-cannon would.

You could fill it up at a gas station or from a car electric tire pump. That's cheap, portable power.

This approach would not allow for top spin or slicing simulations because the tube needed to guide the ball from the hydraulic shoot would always create a topspin coming out of the shoot. Also, it won't last me 1.5 hours (typical tennis lesson length) due to compression being non-repeatable without accumulators to replenish. Just adds too much complexity that 2 wheels don't...

rogerClark: What was your $ budget for this project ??

I am hoping to keep the budget to something less than a brand new tennis ball machine to justify my quest. New ball machines are typically $800-1500 range. I'd like to stay below $500.

nbourg8:

rogerClark: What was your $ budget for this project ??

I am hoping to keep the budget to something less than a brand new tennis ball machine to justify my quest. New ball machines are typically $800-1500 range. I'd like to stay below $500.

I'd start with the assembly for getting the ball to eject at 125MPH first. That is very fast so you are trying to keep a lot of energy at bay. Just getting that working will be challenge, tuning the wheels (two motors) to get your max speed, controlling down to reasonable pass speeds and top/bottom spin.

Controlling the ball release, its horizontal angle, motor speed control and even an RF remote will be relatively easy once you get the "physics package" done.

I guess you guys are training to play against the Williams sisters.

BulldogLowell: I guess you guys are training to play against the Williams sisters.

Haha! After reviewing this requirement, I'm dropping it to 80mph. The ball machine is ground level; serves come from 7-8 feet vertical which means they can reach higher speeds and stay in the court.

This means my motor calculations are adjusted.

Thanks for helping me realize this.

BulldogLowell: Controlling the ball release, its horizontal angle, motor speed control and even an RF remote will be relatively easy once you get the "physics package" done.

As expected. I attempted to draw it up in Autodesk Inventor but due to PC complications, when I make the wheels, I get graphics card errors.

To Google Sketchup I go. Don't worry, I'm fluint in both and both are capable of 3D printing.

Perhaps a portable machine like what you spec'ed is more like several thousand dollars, so even if you spent $600+ building it you would save a lot of money.

I think that the power is the issue after you get those motors working together.

You'll need access to a machine shop for the aluminum structure in the high energy parts.

Your wife will look like Serena Williams lifting that car battery into the trunk of her car!

nbourg8: I intend to place a stepper motor mounted on the base plate to the wheel mount and have it rotate on a central pivot point. Keeping a repeatable balance between each shot means I'll need to come up with either a large pivot point or a ball bearing system in a circular track.

Hardware stores like Ace carry Lazy Susan bearings. The big ring is rated to 1000lbs.

Obviously, the stepper motor would be connected to a pully near the front of the enclosure and mounted to the enclosures base structure such that each incremental rotation of the stepper motor will pull a chain/rope/string attached to an anchor point on the swivel structure holding the wheels. Thus pulling the structure forward in a rotational manner. The trick is, getting it to rotate back; I may use springs or two steppers, not sure.

If the pivoting base is round then a motor with a small wheel can turn against it in either direction. That would give the stepper a long lever arm, reducing power needed but yes it would slow azimuth adjustment. I would put brakes on the base that engage before a shot. The stepper would not have to hold it steady but be sure to release the brakes before running the stepper and maybe use a sensor to make sure they do. A current sensor on the stepper might also be a good idea. In a perfect world, nothing gets stuck. I don't live there.

@GoForSmoke The lazy susan bearing item is excellent. Being capable of 4, 6, 12 inches diameter is perfect adaptability. This is going in my design. I wonder if we can somehow adapt a gear path on the circular base for which the stepper motor can ride?

With gears, I could simply measure feedback of gear location if a ball being shot causes the gear to move. Brakes is an interesting concept but added weight. Not sure about brakes atm.

@all I'm still unclear as to which Arduino controllers I should be interested in to operate with a remote controller containing dials, switches, buttons and an LED screen for value readouts.

start with an UNO, lots of capability and @GoForSmoke can help you squeeze every byte out of it!!!

(also, you could try a worm gear to rotate the base)

Really a small wheel turning a much big one only needs friction. Cheap bicycle motors work that way and accelerate people, even go uphill.

Painting stripes or putting index holes on the pivoting part and use a led and sensor will let you be sure about position. You wouldn't even need to use a stepper, just a regular speed-controllable motor.

A brake can be as simple as a cam that jams against the turntable or a pin that goes in a hole though that would limit the angles that the machine could fire.

A small stepper that doesn't have a reduction gear drive will not hold against a hard push, gear teeth or not.

Hi, I just wonder if there is some product out of this discussion or at least what was the blocker not finishing the idea. Thanks a lot for info (I'm thinking to build something really similar now.)