Pressure Sensors

[maria9876]

I am trying to build a portable marimba and therefore have a few inquiries about "pressure sensors". Does anyone know of any such thing that can sense/measure physical pressure (when a mallet strikes a bar, it needs to be able to measure how hard the hit was in order to output a certain volume).

I was looking at "Force Sensitive Resistors" sold on adafruit, but they are expensive (considering how many I would need for an entire mallet instrument). Does anyone know of a cheaper way of doing this? Thanks!

Force Sensitive Resistors: http://www.adafruit.com/products/166

[adafruit_support_bill]

FSR's are a lot cheaper than strain-gauges. If you search the web, there are a few sites that explain how to make your own FSR's from the anti-static foam sheet used to package chips.

[pburgess]

You might have some luck using piezo elements as strike sensors. This is what's inside the drum kits of the "Rock band" game controllers and that sort of thing. Or here's another example making a little percussion synthesizer.

The output is really "spiky" though. You have to process and filter the input a LOT to get clean(er) pressure measurements, and it's still not entirely perfect, so I can't say for certain how well this might work for your application. But it's inexpensive enough that it might be worth a trip to Radio Shack to get a couple test parts to experiment with.

FSRs are a whole lot "smoother" in that regard, but yeah, the cost will add up quickly on a big instrument.

[maria9876]

Thanks for the feedback. I was just thinking, if it is possible to somehow measure the pressure imposed on the mallets (only 2) instead of inserting FSRs into each and every bar on the instrument (about 50 bars), then it would be a world of a lot cheaper.

However, if I were to do this, would I still be using FSRs since the mallets are round? I'm not sure how that would work out exactly.

[maria9876]

Hmm, I read online that piezo elements are good for measuring vibrations (and with a mallet instrument that would be ideal). They are also about 7x cheaper than FSRs which is nice. Thanks for all the input! The gears are turning.

[adafruit_support_bill]

The other thing to look at is QTC (Quantum Tunneling Composite). It comes in several forms and is pretty easy to work with. But they are a UK company and it can be tough to find a supplier this side of the pond.

http://www.peratech.com/qtcmaterial.php

[maria9876]

Does anyone know of a switch that does not require a button (one that is flat)? To address a new issue, I need to figure out a way to have switches in each of the bars, that way when a bar is hit at any speed (like when you play the drums) it outputs a sound. So, it would have to be able to detect very fast hits (which essentially are rolls). I'm looking for something simple and accurate. Thanks.

[adafruit_support_bill]

http://www.rfmicrolink.com/ZOFLEXZL601.pdf

[maria9876]

That is the perfect solution to what I described, but it's a bit pricey. I was thinking of possibly coming up with a way to make a switch (in order to identify each bar as a different pitch) using the head of the mallet and each individual bar of the instrument.

Basically, when the mallet touches the bar, the switch would be completed, and the pitch of the bar would be identified. In a normal switch, there are two parts. However, (assuming that the mallet part of the switch always stays the same) that means that each bar would be a different half of the switch (so there would be variations to the second part of the switch).

In turn, these variations would render the different pitches. This is all just a theory. Is anything like this possible using conductive materials (preferably not brittle (at least not for the mallet materials, because they could be easily broken due to excessive force and use))?

[adafruit_support_bill]

That is the perfect solution to what I described, but it's a bit pricey.

Zoflex is not that expensive when you consider that a 6"x6" sheet yields 144 1/2" x 1/2" pads for $25.

[maria9876]

Oh, okay. So it comes it sheets that unfold to yield a greater amount of material? I was looking on their website for more information on that and couldn't find it, but that definitely is more reasonable than just a piece of material that is 6"X6". Thank you.

[adafruit_support_bill]

It comes as a 6x6" sheet of rubber-like material which is easily cut. To make a 'switch', just make a sandwich with it between two conductive elements.

For your marimba, you could probably put these switches under the bars, so a strike anywhere on the bar would activate the switch. You will probably need to experiment to find a configuration that gives you the right sensitivity. I'd start with a piece of foil tape on the bottom, a small patch of zoeflex, then a bare copper wire between the zoeflex and the bar.

[maria9876]

I've been looking into using some piezo elements for this project, and I've been told that they can be used as pressure sensors. However, I have not been able to find any information on pressure vs. resistance for piezo elements. Does anyone know where I can find this information for simple piezo elements? Thank you.

[adafruit_support_bill]

I have not been able to find any information on pressure vs. resistance for piezo elements.

Piezo elements generate a voltage when stressed. As PBurgess said in an earlier post:

The output is really "spiky" though. You have to process and filter the input a LOT to get clean(er) pressure measurements, and it's still not entirely perfect, so I can't say for certain how well this might work for your application. But it's inexpensive enough that it might be worth a trip to Radio Shack to get a couple test parts to experiment with.

[mtbf0]

seems as though in a marimba all the vibrations would cause a whole lot of noise in a piezo sensor.

perhaps you could measure force with an accelerometer in the mallet. that sudden stop when it strikes the bar ought to be detectable.