3-axis accelerometer

This is also going to be part of my Iron Man suit.

For anyone who has seen the first movie, near the end, Tony is not wearing his helmet. In order to block bullets, he rotates his arm at the elbow, whipping his forearm towards the outside of his body, and a shield folds open from his armor.

In order to replicate this, I think an accelerometer is my best option, but I've never used one before, and have no idea which one to order for my purposes. As far as I can tell, a 3g one would be sufficient to detect the motion described above, but I'm not sure. I also only want the shield to activate when I provide enough force, and not just when I'm waving my arms around, or make the slightest movement. Any suggestions would be greatly appreciated, and any information on how to do this would also really help.

Thanks in advance,


Are you trying to deploy some kind of mechanical shield? You mention "he rotates his arm..."--perhaps what you are looking for is a gyro-sensor. Or maybe I just don't understand what you are trying to accomplish here. Please add more info(drawing/picture).

Yeah, sorry, it’s a little difficult to describe. I’ve found a picture from the movie, and have drawn a terrible image to describe the motion I need to to do activate the shield.

And, here’s the motion, if it will work:

As I rotate my arm quickly, the shield should open. The rotation stems from the elbow. Hopefully that helps.

It seems like the movement is a change of pitch. In which case you're good to go with a 3 axis accelerometer since those things can detect roll and pitch angles. However, if the motion is horizontal from side to side (yaw). Then you will be better off with a compass module. :slight_smile:

In any case an accelerometer could work… Even when detecting rotation. As long as it is placed away from the axis of rotation, an accelerometer can detect the resultant of rotation(centrifugal ‘force’(or in this case, the acceleration that causes the cf)).

Physics Time:
a = w2r
a = the acceleration “felt” by the accelerometer
w(closest I could get to the omega symbol) = the rotational velocity(rad/sec)
r = the radius(distance from axis of rotation to accelerometer)
So… w = SQRT(a/r); (watch out for negatives!)

The issue with using this method of measurement is that you can get extraneous readings. Any acceleration measured along the axis you use for the accelerometer would be interpreted as rotation by the MCU.