Freesound Forums

Contact microphones or measurement systems. This topic is foreign for me, so here is what I need.

I'd like to pick up a physical vibration from a solid object, through which sound goes. Like a drum (or table on which you can place vibration speaker). No sound from the air, only from vibration.

I need to pick up a frequency range starting at c.a. 70Hz, and ending below c.a. 4-6kHz. Plus - some cheap solution that would work in that range. Not DIY - rather something ready-to-use.

This is going to be rather a measurement system, that measures "acoustic impedance" so to speak (how the sound that goes through a solid object - is damped or amplified - in various, selected frequency bands).

Anyone has an experience with that or can navigate me?

p.s.:
I'm going to write my own software for that, because this is more specific topic.

You can get contact mics from shops selling guitars and guitar accessories.
These are clip-on and will work well to attach to 'thin' objects: metal sheets, tables, pipes, holloe objects (such as glasses, cans, jars).
You will not be able to attach it to something 'too thick' or 'too big' to clip on (bowlling balls, tennis balls, watermellons, walls, doors, etc)

Building a DYI contact mic from a piezoelectric disc (used as speakers in many toys, alarm clocks, etc) is very easy and much cheaper than the contact mic solution mentioned above. - and essentially uses the same sensing element, meaning it will work just as good (if not better).
Yu can use one from a broken electronic toy. Or you can buy them from electronics suppliers by the dozen for a few cents/pence each.
I have constructed some such devices. You can easily find instructions on the web.
Also, some internet shops will assemble these for you. Might still be cheaper than a professional contact mic.
VERY IMPORTANT: they need a coat of enamel paint or varnish to protect them from handling damage. The contacts where the solder attaches to the piezo disk are very sensitive.
BUT THE BEST IS...

... that these light devices can be attached using a variety of much more flexible solutions.
- you can glue them to a clothes peg (or other similar devide) and build a clip-on contact mic (equivalent to the professional contact mic I describe on the first paragraph).
- You can use a non-permanent glue to attach them to objects - gluetac, sticky tape,...
- In some cases, you can simply hold them in place with a weight on top. For example if measuring sound conduction on a tabletop or floorboard.
- You can permanently glue them to small ight objects to create 'modified' contact mics (styrofoam cups of various sizes, for example, - probably not very interesting for what you are trying to do.

ATTENTION:
Piezo discs will have a ressonance frequency. These frequencies are usually high, sometimes in the ultra-sounds. The piezo will naturally amplify frequencies around this ressonance frequency. Check the specs, if you are buying them. Measurements close to this frequency will be unreliable. Outside of this freq they often have a linear enough response.

Thanks for clarification.

So, standard piezo elements will work well with low frequency detection? Because this is what I was not sure about (that why remark on DIY's).

My second concern is about pairing (repeatability of parameters), to get general reference for output amplitudes.

Larger piezos should have a better response to lower freqs.
BEst thing to do is to check the specs and freq response. The responses are not always linear,sometimes there are notches on the freq response.

Also, you have to remember that these frequency responses relate to a free piezo disc. Once you attach it to something, these can be somewhat altered.
Some things can only be determined by experimentation.

If you really want to test it, you can attach one piezo to your hifi or monitor speakers and play a frequency sweep. You should have a fairly flat response from the HIFI speaker in the range of audio frequencies you were mentioning, and you can see what the response from the piezo looks like.

You might be able to calibrate your sensor this way and adjust if there is some loss of response in the lower freqs.
Then you can use this adjusted response curve to determine what is really going on when you attach the sensor to, say a table or a sheet of metal.

You may need to try out different discs from the same model to confirm if the response is the same for all of them - don't just assume it is: these are cheap components and absolue fidelity is not a manufacture requirement. If the response is significantly different you would need to number each piezo and calibrate individually, but I am hoping their accuracy is enough for your purpose.

There are other types of sensors (webserach for vibration sensors). These are used for example in the auto industry, but these would cost you probably many tens or even hundreds of £ per piece rather than pence!

Okay, thanks for tips.

Here's a good page that talks about how to preserve the low frequency response of the cheap brass piezo discs: http://www.megalithia.com/sounds/tech/piezo/index.html

Thanks!