Interfacing a piezo to an arduino

  • Thread starter Thread starter Rx7man
  • Start date Start date
  • Tags Tags
    Arduino Piezo
Click For Summary

Discussion Overview

The discussion revolves around interfacing a piezo crystal with an Arduino microcontroller for a diesel engine timing monitor application. Participants explore various methods for signal conditioning, including the use of specific integrated circuits and the challenges associated with high voltage outputs from piezo devices.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested
  • Mathematical reasoning

Main Points Raised

  • One participant suggests using the MAX9924 chip for noise filtration and zero crossing detection, but expresses uncertainty about its performance with peaky inputs from the piezo.
  • Another participant raises the need for a charge amplifier to properly interface the piezo with an A/D converter, particularly for monitoring pressure pulses in a diesel injection line.
  • Concerns are mentioned regarding the dynamic range of the piezo signal, with one participant indicating that they need to build a piezo clamp first to determine this.
  • A participant questions the mechanical contact method for generating force impulses in the piezo transducer and requests a datasheet for the specific piezo device being used.
  • It is noted that piezo transducers can produce high voltage outputs only under significant force, and simple voltage clamp diodes may suffice for clipping voltage impulses.
  • One participant shares their experience with an Envirodiy Mayfly Arduino board, highlighting its features for data acquisition and the importance of input impedance for accurate measurements.
  • Another participant suggests that an A/D converter may not be necessary if the magnitude of the pulse is not required, but acknowledges that a charge amplifier might still be needed.

Areas of Agreement / Disagreement

Participants express differing views on the necessity of specific components like charge amplifiers and A/D converters, indicating that multiple competing approaches are being considered without a clear consensus.

Contextual Notes

Participants have not yet established the dynamic range of the piezo signal, and there are unresolved questions regarding the specific piezo transducer to be used and its datasheet. The discussion also reflects varying levels of understanding about the interfacing requirements and signal processing needs.

Rx7man
Messages
425
Reaction score
189
I'm looking to build a device that takes a pulse from a piezo crystal and interface it to a microcontroller.. I know the piezo is a very high voltage device, so I need to find some way to make it compatible.. I looked it up on the Arduino site and it seems like a terribly primitive method to use a 1meg resistor in parallel with it.
I was thinking perhaps a VR sensor IC chip like the MAX9924 which is capable of some noise filtration, zero crossing detection, and I need them in my project anyhow.
https://datasheets.maximintegrated.com/en/ds/MAX9924-MAX9927.pdf

The application is a diesel engine timing monitor, I'll have a VR sensor on the cam gear (10* per tooth), with a hall effect sensor to sense #1 cylinder position, the Piezo crystal.. Something like what's shown here https://www.amazon.com/dp/B00063Y1LU/?tag=pfamazon01-20I may need to play with the resistor values a bit, start with more resistance and lower it until it works
1602359332530.png


Open to other ideas, maybe there's a specific chip for this? I didn't see anything when I searched Mouser
 
Engineering news on Phys.org
What's the dynamic range? The first thing that comes to mind is a log amp coupled with an automatic gain control, etc. Sorry I don't have anything specific.
 
  • Like
Likes   Reactions: Rx7man
I'm really not sure what the dynamic range is going to be! I'm going to have to build the piezo clamp first.

Only thing that makes me question the MAX9924 chip is it usually gets a sinusoidal input and I don't know how it'll like a very peaky input..might be some trial and error there
 
I sent a tech support question to Maxim.. maybe they have an idea.. worth a shot
 
Sorry, I'm not understanding the piezo transducer part. AFAIK, a piezo transducer converts force impulses to capacitive voltage spikes. Where are you making mechanical contact with something to provide those force impulses to the piezo transducer? I looked in your links, and did not see a datasheet for the piezo transducer that you want to use -- can you post a link to the datasheet? (sorry if I missed it)
Rx7man said:
I know the piezo is a very high voltage device
Only if the force impulses are really big, and even then the source impedance is very high. Simple voltage clamp diodes to the rails are generally enough to clip the voltage impulses from a piezo transducer.
 
  • Like
Likes   Reactions: Rx7man
I'm not sure which transducer to use at this point
Here's the reply from Maxim.. that last line is a little hard to english
1602695365492.png
 
Some comments:
My experience with piezoelectric transducers is that they need a charge amplifier (search the term) to properly interface with an A/D converter.

If you want to monitor pressure pulses in a diesel injection line, you will need a sample rate fast enough to capture the pulses. Figure on at least 5, and preferably 10, samples per pulse duration. And a processor fast enough to process the signal in between samples. You may also need an anti alias filter.

I did some tinkering with an Envirodiy Mayfly Arduino board: https://www.amazon.com/dp/B01F9B4WCG/?tag=pfamazon01-20. This board is designed for data acquisition. It comes with a real time clock, 10 bit and 16 bit A/D converters, multiple power supply options, and has a built in micro SD card slot. The A/D input impedance needs to be less 30K ohms for good accuracy and repeatability. I learned this after trying to measure an external battery voltage using a 1 Meg / 300K voltage divider. A 100K / 30K voltage divider worked well, while a 1 Meg / 300K divider had an unacceptable level of both random noise and bias error.
 
  • Like
Likes   Reactions: Rx7man and berkeman
That's an interesting board, though I don't need to know the magnitude of the pulse, so there's got to be a way of doing this without an A/D converter and repeated sampling..

A charge amplifier may be necessary though, and will be kept in mind for sure
 

Similar threads

  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 11 ·
Replies
11
Views
5K
  • · Replies 1 ·
Replies
1
Views
4K
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
966
Replies
2
Views
2K
Replies
4
Views
1K
  • · Replies 13 ·
Replies
13
Views
2K