Juggling Ball Arduino

I started another topic on 1 aspect of this and people seemed interested so I thought I would make myself a master thread to discuss my project in.

I want to build juggling balls that can change color when (or a number of MS after) they are thrown, caught, and reach the top of their toss. I am still a novice juggler so I am starting this project now so that it is finished when I have the skills to use them.

Here are the issues I am going to need tackle:

  1. Sensing the toss, catch, and peak.

I think this can all be done with a 3 axis accelerometer.

  1. Making sure the ball will hold up to 1000s of throws, catches, and drops.

  2. Power and recharging the ball. I would love to power it via the kinetic energy I am giving it somehow, but I don't think this will be feasible.

  3. Keeping the ball well balanced.

  4. Making a good light diffuser so the ball's color looks uniform.

If anyone else can think of any other issues that might come up feel free to chime in.

Also mods, if this topic is in the wrong place please move it. I didn't see any forum for projects in the planning phase.

i would use a piezo speaker thing to sense the throws and catches

much cheaper and easy work with

as for making it durable i guess that comes down to how you plan on protecting them

awesome idea though - i like it :slight_smile:

i would use a piezo speaker thing to sense the throws and catches

I have a big piezo speaker I can play with. Do I just wire it up backwards to make a simple sensor out of it?

as for making it durable i guess that comes down to how you plan on protecting them

I was thinking of casting it in something that is non conductive. Like silicon or acrylic. Then it would be totally solid and should be a good diffuser naturally. The only thing exposed to the outside of the ball will be a mini usb connection. I think I will try to make it soderless. Then there would be less to break.

I wired up my pizobuzzer backwards and it seems to be able to read when I flick it with my finger but not as much if I move it around. So I think it is the pizo working, but could be my loose wires moving around.

I guess the pizo is working like omni-directional accelerometer where compression on any axis will give a reading.

I may try to make something with a tennis ball soon that I can see what a toss looks like to my speaker.

NICE project. Will be adding to my favourites to follow progress.

OK apologies if I ramble off or try to tell you what to do, but these are things I'd be thinking about - in absolutely no order whatsoever.

  • You could use a foam ball from a craft shop for the prototype - the foam is an EXCELLENT diffuser, but will be dimmer. They obviously weight (almost) nothing, so may need to be weighted, but at least you know the added electronics won't tip the balls into the "too heavy" range.

  • Using an accelerometer to sense the throw will be the most powerful way to do it, but it will be a lot more energy intensive that other simpler techniques such as the peizo element or a spring inside a loop.

  • I think the peizo would only detect the impact of the ball when caught in your hand - it wouldn't detect the throw - this maybe OK. Detecting impact using a peizo is how the Nike+iPod sensor works.

  • You could have some sort of "score" or feeback system at the end of each juggling round - perhaps flashing one time for each successful 10 throws or something.

  • Not sure if you could detect the different phases of the throw using an accelerometer - the ball is always spinning. Although because it's over such a small time period, MAYBE you could use dead-reckoning to keep track of which direction is down, then measure the acceleration in that direction.

  • Even if you are just detecting the impact, you could have it so that as soon as you detect an impact, it changes to a know set colour, then you could set a timer so that it fades through a colour range which you assume happens during the flight phase.

  • It would be relatively trivial to add a buzzer and a small vibration motor to provide additional forms on interaction/feedback if you can think of something cool to do with them.

Any ways cool idea - keep the design coming!

I cut a tennis ball in half and my pizeo and arduino fit in perfectly upside down. I think I can glue things in place long enough to test it out.

Counting catchs was one of my first ideas when I thought of an arduino ball. Also changing lighting patterns based on some signal, like a squeeze. So each routine can have its own lighting.

What I would really like to do is use a lightweight wireless protocol so they can talk to each other and other devices while in use.

I have thought of using a ball within a ball approach too. If the inner ball could rotate freely inside the outer then it should remain relativity stable. You could even figure out the orientation of the outer ball with respect to the inner.

One step at a time though.

Very interesting! I look forward to seeing how your experiments progress :).

As a side note: It is known to be deviously tricky to detect the peak of a spinning object, so maybe, as others have suggested, you just don't even try to do that, and focus on catches and throws.

As a juggler myself, I look forward to your progress with this project.

This idea is especially close to home since this year I received a set of LED juggling balls from my Secret Santa. It didn't take long before I started thinking about how an atmega could get crammed in there.

My silly video shows what a cool effect led juggling balls can create. To celebrate 15 years of juggling, I plan on tackling the 5 ball cascade in 2010.

good luck

Cool video, the pullouts in the dark look really good.

What I really want is like a waterfall effect. Throw up the balls black or red or something and have them fall down blue.

how would an accelerometer be able to tell if its been thrown or caught in a spinning object?

how would an accelerometer be able to tell if its been thrown or caught in a spinning object?

When you catch a ball the spin suddenly stops.

When you throw a ball the spin should increase.

I am going to do a test tonight with my pizo speaker as a cheap accelerometer. I don't expect to have any trouble picking out the catch singnal, but we will see. I am not expecting to see the throw with this test.

if yolu got a sensitive enough piezo i bet you see the throw as well as the G force will effect it?

there cheaper as well!

You people are motivating me to much. I am suppose to be working right now, not gluing things into tennis balls.

Anyway the glue is setting now. So in a few hours I should be able to toss a ball with a big long wire sticking out of it.

I will try to see if I can get some video with the sensor readings on screen as I toss & catch.

Weird Peizo question now. The buzzer I am using takes from 2v-24v I think. Do you think I would get better readings out of it, if I feed it higher voltage and put in resistor to get it in the < 5v range before the arduino reads it. Or do you think using the 5v line directly is just as good.

My silly video shows what a cool effect led juggling balls can create. To celebrate 15 years of juggling, I plan on tackling the 5 ball cascade in 2010.

Holy cow how many hands do you have? That video was impossible to tell which ones you were holding and which ones were seemingly floating. Well done :O.

Thanks guys.

Holy cow how many hands do you have? That video was impossible to tell which ones you were holding and which ones were seemingly floating. Well done .

Just the two. After reviewing the video, I realized I could have just had the missus move a ball around and no one would have been the wiser. :sunglasses:

Cool video, the pullouts in the dark look really good.

What I really want is like a waterfall effect. Throw up the balls black or red or something and have them fall down blue.

...so that's what those are called.

That effect will be awesome. Did you have any luck testing last night?

if yolu got a sensitive enough piezo i bet you see the throw as well as the G force will effect it?

You will see no effect from a piezo sensor from any G force. You only get an output when the crystal in the sensor is deformed. When you flick the sensor with you finger you are in effect deforming it. There might just be a chance that sudden deceleration might be enough to trigger it but I doubt it. If you look at how you juggle, the catching hand moves to track the ball before grasping it, there is no sudden stop. So, sorry, but it would very much surprise me if you could detect anything inside a ball with this sensor.

An accelerometer would work and the amount of spin on these balls would have a negligible effect so don't worry about it.

By the way this section is for showing off things you have made not for projects in developing.

Did you have any luck testing last night?

I didn't give my glue enough time so I had to redo it. Then I went to a killer party. I will post an update today though.

You will see no effect from a piezo sensor from any G force. You only get an output when the crystal in the sensor is deformed. When you flick the sensor with you finger you are in effect deforming it. There might just be a chance that sudden deceleration might be enough to trigger it but I doubt it. If you look at how you juggle, the catching hand moves to track the ball before grasping it, there is no sudden stop. So, sorry, but it would very much surprise me if you could detect anything inside a ball with this sensor.

Stop being so Grumpy Mike :wink: I know this isn't going to be perfect, but I have all the parts laying around, so I might as well try. Even if it doesn't work I will learn from it.

Luckly I am not a great juggler yet, the balls still hit my hands with a noticeable sound. Hopefully the tennis ball enclosure will help magnify the effect.

By the way this section is for showing off things you have made not for projects in developing.

I mentioned in the first post that I didn't think this was the correct place. The forum needs a section just for planning projects, the hardware and software ones are too specific.

Thinking about it more, I think it might be possible to sync colours to phases of the flight.

On startup, it just displays a single solid colour, and when juggling starts, say the first 5 tosses per ball are used to measure the average time in between the impacts of catches.

You could then "hard code" a portion of this time, say the first 20% of that time is when the ball is "in hand".

The remaining time is "flight time", and you could split that in half for the up and down halves.

Hmmm... if i didn't have so many other damn projects on the go!

If you use a larger ball you could have a rechargable battery and a means to charge it while its in use. But it may be hard to scale this down so you can effectively charge the3 battery.

Another option would be a small crank charger. But have the crank handle removable so its out of the way.

This all is probably too much to cram into the ball.

My tennis ball prototype didn't work well in multiple ways. I am going to have to put this on hold a bit until I get some funds and more concrete ideas.

I would gladly keep discussing it here in the mean time.