Connecting piezo to a 3.5mm 4-pin input

Discussion in 'The Projects Forum' started by Goldylox, Feb 25, 2015.

  1. Goldylox

    Thread Starter New Member

    Feb 25, 2015
    3
    1
    Hi all! I want to write a desktop app that receives vibration data from a piezo, so I need to transmit that data from the piezo via the microphone port. I'm naively connecting to the mic + ground pins of a 3.5mm 4-pin cable but that's not working because my Macbook Pro doesn't detect that a microphone is plugged in at all.

    I've been told that I need to get the impedance to ~1 kOhms for the MBP's microphone detection to work properly. I've also been told that piezos have much more impedance than that. My current thinking is that if I fix this impedance problem, everything will be dandy.

    First: am I thinking about this correctly?

    Second: I found this page describing a circuit to convert the piezo's impedance to something more reasonable. Would this work for me? And if so, can someone give me a quick explanation of what's happening here (or just yell at me to read the semiconductor book)? I read the entire DC and AC textbook on AAC but I haven't gotten into the semiconductor textbook yet so this circuit doesn't make a lot of sense to me.

    In general, I'm trying to learn holistically as I go, but impedance and impedance matching is a new subject for me.

    Thanks!
     
    Rolland B. Heiss likes this.
  2. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,542
    1,251
    For the circuit you have, the concept is correct but the DC output impedance still is too high to awaken an Apple input stage. You'll need to follow your buffer circuit with another one to drive the mic input. You probably won't need much, or any voltage gain. Do you have any idea what kind of signal bandwidth you need? If it is relatively low, like almost DC to 100 Hz or 1 KHz, you might be able to do the whole thing with a CMOS opamp. they don't have a lot of gain-bandwidth, but they have a very high input impedance.

    ak
     
  3. Goldylox

    Thread Starter New Member

    Feb 25, 2015
    3
    1
    Thanks for the quick reply! I'm going to be using this to detect hits on a drum practice pad, so it's not even really a microphone as much as it is a vibration sensor. Given that, I can be extremely flexible about what the final output signal looks like, since the user will either be smacking the drum or not (and we'll see spikes accordingly). If possible, I'd like to stay as accurate as possible (so I can sense volume, for example), but priority 1 is getting it working.

    What would a CMOS opamp circuit for this look like?

    Thanks again, this is very new territory for me... I have some reading to do.
     
  4. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,173
    397
    My thought is to use an emitter follower with something like a MPSA14, darlington, bias input with a pair of 1M resistors, emitter load 4k7, cap couple with 1 uF into a 1 k load, the 1 k is the input to mic. input. Power, 5V or battery 5 to 9 V. Drawing tomorrow if desired
     
  5. Goldylox

    Thread Starter New Member

    Feb 25, 2015
    3
    1
    A drawing would be amazing. This is a good incentive to learn about semiconductors, but I don't yet have the skillset right now to construct a circuit from your description =(
     
  6. Ramussons

    Active Member

    May 3, 2013
    557
    92
    Hello Goldylox, I hate to go tangentially .:mad:

    You have posted a lovely superimposed photo of your construction. I have spent a long time trying to do this superimposing and have failed - maybe I'm dumb to the obvious :eek: Would you be kind enough to explain the steps taken to superimpose the 2 scans?

    Thanks.
     
  7. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,173
    397
    Simple Z match ckt: Gain about 90 %. Piezeo clipped to edga of desk, slapping desk gave about 1 V PP signal, output about the same. R 1 any value from 100 k to 1 M, R 2 = 3 X R 1. R 4, what ever it takes to enable mic input. If piezeo is located at more than a ft. shielded twisted pair or coax should be used Z match 00000.jpg
     
Loading...