Can I replace a condenser microphone with a piezo in this circuit?

Thread Starter

zirconx

Joined Mar 10, 2010
171
In general, can you swap in a piezo element in place of a condenser microphone?

For example this pizeo: https://www.digikey.com/product-detail/en/0/490-7713-ND
into this circuit board: https://www.banggood.com/10Pcs-Micr...ection-Module-p-1254927.html?cur_warehouse=CN

I'm trying to monitor water flow (simple on/off, not rate) in a copper water pipe. I thought this audio detection circuit board would work. I'm thinking a piezo element, mounted to the pipe, might work better.

When I say "doesn't work", I mean that even when zip tied tight to the copper water pipe, it does not get triggered by the sound of the water. The sensor itself does work, I can trigger it by tapping on it, speaking loudly, or clapping.

 

AlbertHall

Joined Jun 4, 2014
12,345
Generally speaking, no. The piezo device requires to feed a high impedance. Electret microphones include a FET to produce a low impedance output.
 

Thread Starter

zirconx

Joined Mar 10, 2010
171
I'm open to other ideas on how to solve my problem (detecting water flowing through a copper pipe). The main question I'm trying to answer is - how long are my kids showers.:)

I'm guessing the reason this circuit board isn't working is because it's not very resonant at the frequency the water sound is making. To my own ear it is very loud, even several feet away from the pipe. If I put my ear to the pipe the sound level is unpleasant. I'm surprised it doesn't trigger the microphone. I was think if of mounting it more rigidly with a drop of glue.
 

MisterBill2

Joined Jan 23, 2018
18,179
To have that microphone pick up enough sound to trigger the device you can have it facing a noisy part of the plumbing, sich as a valve. Fluid noise is mostly produced by turbulence, and a valve is typically the most turbulent part of a standard plumbing system. And enclosing the sensing assembly may allow more sound to be captured. Having the micro-phone a shrt distance away from the noisy portion may improve the pickup as well.
 

Audioguru again

Joined Oct 21, 2019
6,674
The Chinese module that uses an electret microphone probably has a fairly low input impedance since an electret mic has a Jfet as an impedance converter. The output of the circuit is not audio, instead it is a digital high if the sound level is below the threshold adjustment and is a digital low if the sound level is above the threshold adjustment.

The Instructable using a piezo sensor has a 1M resistor at the input of an LM613 dual opamp (it is obsolete).

But a piezo used as a microphone picks up only strong vibrations and high audio frequencies when it is feeding a very high impedance.
A piezo used as a microphone resonates and produces a squeaky output signal at its resonance frequency (4.6kHz for the one you showed).
 

MisterBill2

Joined Jan 23, 2018
18,179
All of those reasons are why I instead suggest methods of increasing the sound into the present microphone unit. A simple funnel will increase the sensitivity. If you are able to provide a straight-on photo of that circuit board we can figure out the circuit. If you have another microphone cartridge like that you can experiment with pointing it at different parts of the piping .
And in what direction was it ty-wrapped to the water pipe? Those cartridges are only straight-in sensitive without some enclosure to make the angle wider. Thus aiming matters a whole lot. AND, I see an adjustment pot. Is it set for maximum sensitivity? That adjustment matters also.
 

Thread Starter

zirconx

Joined Mar 10, 2010
171
All of those reasons are why I instead suggest methods of increasing the sound into the present microphone unit. A simple funnel will increase the sensitivity. If you are able to provide a straight-on photo of that circuit board we can figure out the circuit. If you have another microphone cartridge like that you can experiment with pointing it at different parts of the piping .
And in what direction was it ty-wrapped to the water pipe? Those cartridges are only straight-in sensitive without some enclosure to make the angle wider. Thus aiming matters a whole lot. AND, I see an adjustment pot. Is it set for maximum sensitivity? That adjustment matters also.
It was parallel to the pipe, so I could zip tie the canister part of the mic to the pipe. I thought sound was very omnidirectional, and I thought the physical coupling would make the pointed direction irrelevant.

I can try again (in a few weeks, I'm in the middle of another project right now) with it pointed directly at the pipe.
 

MisterBill2

Joined Jan 23, 2018
18,179
It was parallel to the pipe, so I could zip tie the canister part of the mic to the pipe. I thought sound was very omnidirectional, and I thought the physical coupling would make the pointed direction irrelevant.

I can try again (in a few weeks, I'm in the middle of another project right now) with it pointed directly at the pipe.
The diaphram in those microphones is very lightweight, and so mechanical coupling to the case, off axis, will not generate much signal.
 

Audioguru again

Joined Oct 21, 2019
6,674
A problem with those Chinese modules is that they have no spec's and no schematic for us to see their sensitivity. This one probably can hear a window or door breaking loudly, not water running in a metal pipe. The electret mic it comes with is omni-directional (picks up sounds all around it).
 

MisterBill2

Joined Jan 23, 2018
18,179
With a bit of cutting and two more resistors and one capacitor you should be able to utilize that unused comparator section, pins 5,6. &7 as another gain stage. a gain of 10 or 20 would probably solve the problem. And the reference bias for the + input can come from the junction of the 1K and the power on LED. OOPS! the LM393 is open collector, so you would also need a resistor between pins 7 and 8. Depending on your soldering skills that may be achievable with surface mount parts or 1/8 watt sized resistors. But that seems like the simple electronic solution .
 

MisterBill2

Joined Jan 23, 2018
18,179
I see a few audio problems with the simple mic preamp:
Certainly right. But the thing may oscillate without those caps, since there is no positive feedback to provide any hysteresis. It should be possible to unsolder them to verify that. Since flow noise is mostly higher frequency , removint the cap tied to the base of the transistor should improve the sensitivity without affecting the comparator operation.
 
Top