Detecting hit location of table tennis balls

Hi,

I would like to detect the location of a table tennis ball hitting the table.

My idea is to

1) mount a number of Arduino-connected piezoelectrical elements under each side of the table. 2) interpolate the location from the amplitudes of the signal red from the piezo elements.

Do that plan sound feasible or do any of you have other suggestions?

Regards, Tuk

PS: I am new to this forum, I hope that I post the right place.

That should work, but the coding could be a bit tricky. I suppose it would involve comparing them to find the one reading the highest, and then seeing which ones near it is the highest, etc. etc. to get as accurate a location as possible.

If you are mounting "a number of" piezo sensors, you need to keep in mind that there are a limited number of input pins available on the Arduino Duemilanove.

Also, the top of a table tennis table is pretty sturdy. I'm not sure that you are going to be able to differentiate which sensor is nearest the ball. If I remember correctly, a table tennis ball is pretty light. I'm not sure how much movement it will cause in a 3/4 plywood table top.

@TchnclFl:

Hopefully one in each corner of each side is enough. With a bit of luck I can get the location by interpolating between the values read there.

@PaulS:

If you are mounting “a number of” piezo sensors, you need to keep in mind that there are a limited number of input pins available on the Arduino Duemilanove.

Yes, I will probably need the Arduino Mega with its 16 analog inputs…

Also, the top of a table tennis table is pretty sturdy. I’m not sure that you are going to be able to differentiate which sensor is nearest the ball. If I remember correctly, a table tennis ball is pretty light. I’m not sure how much movement it will cause in a 3/4 plywood table top.

Actually I am thinking about those that work like contact microphones (pictured here: http://en.wikipedia.org/wiki/Contact_microphone). They are very cheap and very sensitive.

This also makes me a bit worried: I fear the sound will transfer too well through the table so the sensor readings gets too similar. Testing will tell, I guess.

Even 16 inputs will only give you 8 per side. Whether or not that is enough depends, I guess, on why you care where the ball hits the table.

I'd start with 4 on one side, to see if there is any detectable difference in sensor readings. I doubt it, but I could be wrong. I was once before.

I could be wrong. I was once before.

Sorry, but that made me laugh out loud :P. I love how you’ve been wrong once :D! (Don’t take that offensively, I just love the way you worded it)

I'd start with 4 on one side, to see if there is any detectable difference in sensor readings. I doubt it, but I could be wrong. I was once before.

Yeah, I'll try it and report my results. Thanks so far.

I think you will be lucky to get any signal out of a piezo sensors from this setup but best of luck anyway.