Acoustic Rig For Detecting Mechanical Errors

Zach,

I don't think that you are going to have any scratching or colliding components in a (mobile phone) digital camera...

With the possible exception of an optical zoom mechanism, such a device is "solid state" and has no moving parts in the first place. Therefore, regardless of how sensitive the microphone or vibration sensor, there would be no "scratching or colliding" sounds to "listen" for.

Not to mention the complicating factors posed by the built-in ringer and "vibrate mode" annunciators in the phone...

---------------------------------------------------------------------------------

Just my $0.02

Thank you for your reply. I'm sorry, I should have been more clear: the part we make is the optical zoom mechanism. I'll give you more detail (I can't give out too much for legal reasons...):

The lens is encased in a lens carriage which is in turn encased in a square plastic chassis. There are metal flextures on both the top and bottom which is connected to the chassis and the lens carriage. Another part of the camera drives the lens up, which is forces back down by the flexure (it has arms on each side attached to the lens).

Also, the chassis we use often need to be modified by hand, which creates messy plasitic "burrs".

The sound I am trying to monitor are:

• Friction between the lens carriage and the chassis as it moves up and down,
• Burrs obstructing the mechanism or the flexures.

I hope this makes sense.

Zach

Glue a mic to it with epoxy then sit the whole thing on a bed of foam in a box full of foam, on a bed of foam, in another box of foam, in a cupboard away from any noise...

Wire the mic to a scope or a sepctrum analyser...via a preamp if necessary.

Or just use the damn mic that is already in the phone!!! ...after all that's presumably the purpose of the exercise..!

Del

Best to use a feedback loop to monitor the current of the motor that is driving the lens. This will more accurately let you know when the grease is dry or dirt has infiltrated. An increase in current would warn of problems. If you wait until it is making noise, the damage is already done. A good warning would be that it requires maintenance or cleaning.

An accelerometer might work. There are microphones that would work as well. Try PCB piezotronics at

http://www.pcb.com/

Thank you very much guys, all your tips have helped a great deal. Much appreciated.

Zach

A fiber optic sensor operating in interferometer mode makes an excellent acoustic sensor, and small ennough for your small structure. The interrogation system may be somewhat expensive.

A microphone with a focussing refloecot may be more practical.

I once participated to a research project where an array of microphones were placed on the radiuses of a circular pattern. I think there was a Fresnel function in their placement. The timing of the sound events was used to triamgulate the spource position.