PVC Pipe Impact Sensing

Forum 2005-2010 (read only) Hardware Interfacing
1

Hi, I need help finding the best way to tell if someone has hit the end of a PVC pipe. I'm working on a project for my scout troop that will be used in our gateway at the 2010 national jamboree. Our gateway theme is the blue man group so we decided to put PVC pipe instruments for people to play. What I want to do is have lights inside each pipe so when you push one of the four buttons on the panel the pipes will start lighting up in a certain order to tell you wich ones to hit to play a song. When a button is pushed the pipes you need to hit light up and stay lit until you hit them. then the next pipes in the sequence will light up.

I first tried to use piezo elements to sense the hit but the holder we made for the pipes acts as a shock absorber and the piezo picks up almost no vibration. I am thinking of using some infrared emiters and detectors and mounting them on the inside of the pipe so that they point out the pipe. Then, when a paddle hits the pipe, it will reflect the IR back in to the detector. the problem with that is that you don't need to hit the pipe for the sensors to detect it. Are there any other ways to determine if the pipe has been hit? Any help is appreciated.

2

If the mounting and strike location are right, you could use a laser module mounted at one end, aimed at a photodiode mounted at the other. When the pipe flexes, the photodiode moves out of the "line of fire" of the laser. You'd probably need to make sure the diode is behind a very small aperture to minimize the amount of flexing needed.

Depending on how the pipe flexes, you may find that the light is diminished, but not cut to zero, because it reflects off the inside of the pipe such that part of it still reaches the diode. You'll probably need to do some experimenting to set "on" and "off" thresholds.

Ran

3

I take it that one end of the pipe is sealed? wonder if there is a detectable pressure change when the end is struck?

4

daveg, the pipes do not have to be sealed in one of those pvc instruments. They can be, as covering the end lowers the pitch an octave.

To the OP, would an IR or ultrasonic distance finder work? Mounted on the inner rim of the pipe, it would sense the paddle thing.

5

Why not use a spring, like those vibration activated light ball thingies? When it gets hit, the spring vibrates and touches another contact ...

6

But the pipes themselves aren't going to be moving much.

7

I think sciguy's idea of a rangefinder would work. Another possibilty is turn it on its head and detect the impact from the paddle. Obviously this wouldn't be as good as hitting the wrong pipe would work.

8

Actually yeah, why not just use a microphone to pick up the noise of the tube being hit?

9

What if there is other noise? Like other pipes first of all?
It probably could be done with mics, just would need a bunch of calibration.

10

Thanks for all the suggestions

I take it that one end of the pipe is sealed?

No both ends are open. Like sciguy said, doing that would lower the pitch and we don't want that.

would an IR or ultrasonic distance finder work?

Yes, that would work but i'm doing this with all of my own parts and any new parts I have to buy. So doing that wold be too expensive.

why not just use a microphone to pick up the noise of the tube being hit?

Well that might work but I think it would be too unreliable because if the kids around it were making a lound noise, that might trigger it.

11

Well what kind of mic are you thinking of? I was thinking of just an electret mic.

12

That would be good if I was working with only one pipe but I'm working with 8.

13

Why not use a force sensor? You have to actually touch it and they are very thin. I found them here:

http://www.robotshop.com/interlink-05-circular-fsr-2.html

14

Thanks everyone for all of your suggestions. I ended up just going with the IR emiter/detecter method. It works pretty well.

Here is a video of it working:

system 15