I had an idea for a wearable electronic project. There would be a set of LEDs embedded in a long-sleeved shirt, controlled by an Arduino. I would wear gloves containing some sort of contact sensors. Then when I juggle, as I catch each ball, the sensor would trigger some kind of response in the LEDs.
The biggest unknown for me is the contact sensors. They would need to be very durable since they’ll be hit by a falling juggling ball thousands of times. They should also be as small and thin as possible, so that they don’t interfere with my ability to catch and throw the balls.
Anyone have ideas for the sensors, or thoughts in general about this project?
That’s a fairly difficult problem to solve.
I don’t know of anything off-the-shelf that will do what you want. Something like a carbon-black gel between two conductive layers might work.
But, I think that sensing the catch pneumatically might be easiest - put air-filled tubes or bladders in the fingers, and monitor the pressure with a silicon sensor. Catching the ball will register as a pressure pulse.
Can the balls be made of or wrapped in something conductive? A thin, flexible copper mesh would be perfect. Then you could pretty easily hobble together gloves that have a number of metallic pads on them. Half go to one wire, half to another wire and the ball landing in your hand would close close the circuit.
I can’t think of anything else easy enough to make that would trip no matter where or how the ball landed while also not tripping just by moving your hands a certain way. That’s my concern for any type of physical sensor or button. A good enough coverage is likely to end up with false positives.
If you could put a (strong enough) magnet in the ball, a reed switch sewn into the glove would probably work pretty well. But like the copper mesh idea, it would require you to only juggle items you’ve previously prepared.
I wonder if an accelerometer on the back of a glove would work? It could probably pick up the acceleration as you throw a ball, though picking up a catch would be harder.
If you have the budget for putting batteries, accelerometers, and radios (e.g. bluetooth) inside juggling balls, I think they would definitely work (as well as being able to tell when you caught a ball on the back of your neck, in your lap or wherever else).
I like Joey P’s idea the most- as it would be the easiest from a wiring and analysis perspective.
But there are a couple of other ideas that I had: Distance sensor but as it uses sound waves, I’m not sure teh “resolution” with moving hands.
But along those lines, Ambient light sensor should see whether you have something in your hand or it is closed. It is going to be very environmentally sensitive (bright room vs. outside vs. stage). But I think you could do it with dark balls and putting them in your palm. I also like this version as it looks like this and other light sensors are relatively small!
I’d mostly be using this at night, since LED displays in the daytime are not very impressive. That means I’d be juggling lighted balls, since I can’t juggle what I can’t see. I have a couple of sets of lighted balls and I don’t mind modifying them if necessary.
That suggests an idea which is basically Disheavel’s but inverted. I could use light sensors in my gloves which would mostly be seeing the ambient darkness, but would see a big burst of light when one of my lighted balls lands in my hand. Hm… I wonder how durable those ambient light sensors are.
JoeyP’s idea of wrapping them in mesh is interesting, but I wonder how I’d adhere the mesh to the ball. Something like epoxy seems like it would ooze out and cover/insulate the mesh. Maybe a very fine mesh would prevent that.
One advantage I see of an optical system is the sensor doesn’t need to be in the palm of my hand. It could be, say, at my wrist and oriented appropriately to “see” the ball when it lands. That could solve the durability issue.
The inversion of turning the gloves on when a lighted ball lands would definitely be the easiest to code and control. The only risk I see is a positive feedback from the gloves turning on and that setting off the sensor. But a cool effect would be something like green balls and red gloves?!? I bet you could find a multi colored sensor (or heck just put green cellophane over the sensor) to make it more sensitive to the green balls.
With regard to a mesh for conductivity, I, myself, was thinking a metal ball itself- copper? but if the balls are lit, that hampers that plan. BUT, I googled “conductive clear paint” and it looks like there are some latex and spray options that have graphene embedded and seem to work for projects. So no mesh, just a paint option! A bit spendy to test it looks like though and not sure about durability but you could always recoat. So I am back to liking the electrodes on fingers option for fewest moving parts.
What is the ultimate goal? If it is to make a light show synchronized to your juggling, there is no need to detect the ball. Just your arm motion. An accelerometer on your wrist will do that easily.
Sparkfun offers some flex sensors that were supposedly used in the Nintendo Power Glove. You’d need a control system but it might be pretty slick. You could, for example, vary the intensity or even the color based on how your fingers curl around the ball.
I’m mobile and can’t link but they’re Sparkfun item 8606. Sparkfun is a pretty good vendor for hobbyist sensors, microcontrollers, kits and other makers’ things. They have an actively moderated message board for project help.
That was my first though too. I moved over to suggesting the mesh based on the assumption that of all the things to juggle, conductive metal balls wouldn’t be real high up on the list. A mesh could be wrapped around bean bags, bowling pins, running chain saws…the normal stuff. And, if you can get it to work, it would be compatible with items that are conductive to begin with, like metal balls, with no further preparations required.
Yes, that’s basically the goal. And I’m starting to suspect you’re right. I thought there might be some obviously appropriate off-the-shelf sensor for this but it appears that’s not the case. There have been some interesting ideas in this thread that I will continue to explore, but it may be that a couple of accelerometers would be a simpler and more robust solution.
Don’t accelerometers count as off-the-shelf sensors? I am pretty sure there would be a detectable jolt when you catch a ball (I looked up a random chip, the ADXL345, and the datasheet claims you can set the sensitivity down to about 0.004 g), but either way it should be possible to pull the desired rhythm out of the data.
Sorry, I was unclear. I meant that I had thought there would be an off-the-shelf sensor that would detect the ball hitting my hand, like a contact switch. That’s what I had asked for in the OP. Not seeing such a thing now, I’m thinking a non-contact sensor like an accelerometer would be better. And yes, I don’t think it would be hard to extract what I want from accelerometer data. Probably the easiest would be to look for the large upward acceleration as I throw each ball.
I’d probably go with a little I2C accelerometer board in each ball. They’re cheap. It’ll hook right up to an Arduino with a few wires. Most have a gyro too, so if you want to get clever with spinning balls, you can detect that too.
When the ball is flying, it’ll record zero gees. In your hand, it should peak at >2. You could do clever things here as well, like changing hue/brightness based on acceleration.