Using an Accelerometer to Change LED modes

First let me say what Im trying to do, and that I'm pretty new to building electronics. My intent is to make an LED Contact staff that is interactive with the user. Im assuming most of you guys wont know what that is so here's a video of a person using a normal contact staff.

For my project im intending to use the Teensyduino board from PJRC converted to run at 3.3v. Im looking at using a battery like the UltraFire 3.7v 2400mah lithium ion battery for the power source.

I would like to use this accelerometer Triple Axis Accelerometer Breakout - MMA8452Q - SEN-10955 - SparkFun Electronics from spark fun to control different flash modes for the Led. Things like direction changes and the like would create different patterns.

My question is, would this accelerometer be a good choice for this project? or would something like a simple tilt switch be a better choice? The option needs to be fairly robust, as the staff spends almost as much time hitting the ground as it does on the users arms haha.

I would suggest that your best bet is to contact a supplier.

I remember that when I first put a gyro in a radio controlled car, the manufacturer mentioned they are not very robust. Having said that my experience is that they are robust enough to survive the occasional 30Mph crash in my cars and they are used extensively in radio controlled helicopters which are also a harsh environment.

So your best bet might be to forget the advice about contacting a supplier and just get yourself an accelerometer and I would also suggest a gyro and see which(or a combination of the two) gives the most meaningful and reliable input in your application.

One other idea which I know nothing about but which would give your users control over the animation would be to investigate 'capacitive touch', using this you could add 'hot zones' to the device which would change the animation or trigger a particular sequence of events.

See here for more Arduino Playground - HomePage

Duane B.

I think most accelerometers would be tough enough for the task unless that take the force of dropping directly. Have you selected an I2C version on purpose? They're more finicky to get running but you'll often get data faster and with higher resolution. Not sure with that particular one. At any rate, I think you're on the right track. It would be interesting to try a gyro as the previous poster mentioned but they ain't cheap.

A gyro might be cool but I'm not sure I need the angular data they provide. I really just need to know when the staff changes direction. It changes angle and orientation so often that that data would have to be ignored.

I chose that accelerometer for 2 reasons 1- its slightly cheaper then the others and 2- it has the words don't panic.. wait I mean its a lot smaller then others :D. 2 is the most important reason as all this has to reside in a tube with a .777 ID. I could get tube that is slightly larger if need be but that's the size I prefer.

As far as capacitive touch, that really sounds awesome. I have some extra tube laying around, I may pick up some resistors and capacitors tonight and play around with it some. Im pretty new to all this, so the possibilities are overwhelming and exciting to say the least.

I started playing with the CapSense a couple nights ago and I think the possibilities are pretty awesome. But I would still like to go down the accelerometer route, if not now in the very near future.

Are there any accelerometers that are easier to implement then the i2C's i specified earlier? Would This one be easier to use? The sample sketch certainly looks like it.

Remember the data im looking for is changes in direction right now and possibly changes in orientation later.

thanks for your help!.