Loading Piezo Tranducers

[thehallofshields]

Okay I know it's not exactly stompbox, but I didn't know where else to turn.

I wanted to experiment with an onboard piezo pickup on a project guitar, and I was thinking to do a JFET Buffer for the Piezo and the Magnetic Pickups. The simple Tillman design should do the trick.



But... to make the Volume Controls Non-Interactive, I'd need to place the Volume Pots on the Input side of the Circuit. However that will ruin the High Input Z of the JFET Preamp, which won't be a problem for the Magnetic Pickups, but might be for the Piezo Disc.

PZT Transducer (Pickup Cavity) → Volume Pot → Tillman → Output Jack

Single Coil (Neck) → Pickup Switch → Volume Pot → Tillman → Output Jack

Single Coil (Bridge) →     ↑

Anyone have any ideas? Criticisms? I've got 1 small 1MOhm Audio Pot lying around, but I was hoping to be able to do everything with Guitar-Sized 500kA Pots.

[JerS]

What if you used a single pot at the input of the circuit as a voltage divider to set the balance between the two pick-ups - you could add a second pot at the end of the circuit to control overall volume.

Alternatively, building a second Tillman would not be a major effort, and then each pick-up could be controlled completely independently.

[thehallofshields]

You know a mixer pot is not a bad idea. I'm glad I posted.

[thehallofshields]

So how is a 500k Volume control going to interact with a Piezo Disk?

[petemoore]

 A 500k pot turned atw up will sound the same as a 470k or 560k to ground.
A 500k pot turned down 'some' will sound like a resistor in series with signal path, some' being what R value the ''top portion'' [the two lugs that the signal wires connect through] between 0k and 500k, is set to.

[thehallofshields]

As I understand it: the Transducer is Capacitor-like, so we get High-Pass with the Load Resistance, which is the Input-Z of the following stage.

Usually Piezo buffers have 2.2M or greater Input Impedance.

So with 500k, I can't tell if I'm going to run into any significant Bass Attenuation.

[tubegeek]

You'd want to have the hi-Z-input buffer immediately after the piezo, then do any mixing or attenuation after it. You could always buffer after those features again. The standard pot values aren't high enough to go into direct from a piezo, or at least, that's the recommendation I've always seen. I haven't built anything like this, just going on "book learnin' ."

I'd guess you don't need any pull down resistor at all (no R1) - use the very-high input Z of the FET by itself to get good response from the piezo.

So like this maybe?

Single Coil (Neck) → Pickup Switch → Volume Pot → Tillman2 → Output Jack

Single Coil (Bridge) →     ↑

PZT Transducer→ Tillman1  ↑

[thehallofshields]

I think that 2 Independent Volume Controls is going to require 3 Buffers.
1 for the Piezo, 1 for the Piezo Volume Control, and 1 for Magnetic Pickups loaded by a Volume Pot.

Maybe this would be easier with a Mixer Pot, but I can't imagine them mixing well un-buffered.

[thehallofshields]

Here is a comment I pulled from a sucky youtube video:

Grant Anderson
6 months ago

Hello again Christian.  I'm an electrical engineer here in Edinburgh, Scotland.  I've been into modifying guitars and making electronic effect pedals since, well before you were born actually.. so i took a quick look at the radioshack part Piezo Transducer 273-073.  All it looks to me is that it's an ordinary piezo with two wires coming out of it.

I can tell you something right off the bat that there is one flaw in your design.  See, the piezo is a funny thing indeed.  When you tap on it, it excites the crystal latice structure and this produces electrical impulses that, as you know, relate to those waveforms that your guitar is producing (acoustically).  Piezo only responds to longitudal waveforms.  Thats why it only works when you glue it to the body of the guitar and as you learned, how you glue it!  Wheras your guitar pickup responds to Transverse waveforms.  However, the problem is that your piezo (being a funny thing it is) will after a short time being stimulated and losing some of it's energy, will suddenly reverse it's polarity and will switch back and forth (almost like alternating current).  It gets faster and faster at this until it's lost all of it's energy.

You've mixed this funny piezo device into your standard guitar signal without any sort of buffer.  Which for the most part is ok and sounds ok, but I bet it has an odd "thin-ness" to the sound that you kinda notice once in a while.  It's hard to get it right where you want it?  If you've ever wired up a guitar pickup out of phase to the one next to it and then had them both on and Wow.. were did the sound go?  phase cancelation is to blame.  So your piezo is contributing a small amount of phase cancelation to your overall circuit.  Not good. Plus the piezo is subject something known as the Wolf Tone.  You'll have to Wiki it but it has to do with the string of the guitar resonating in step with the body of the guitar which does odd things to a piezo (which only picks up longitudal waves), giving a "shrill" or a "wolf cry" as they call it..

Anway, go to radioshack and buy a small pack of high speed signal diodes.  Tiny little red glass diodes, they are.  They have a little black band on them.  the positive wire from the piezo will have the band furthest away and the negative wire has the band closest to the piezo negative wire.  That will block negative current traveling down the positive wire and vise versa.  It should clean up your sound a little. Or at least, take away any slight annoying artifacts that you're picking up.  You can also try using Germanium Diodes such as the 1N34A which will effect the tone of the piezo transducer and may give interesting results.  However, you will never have 100% satisfaction with a peizo as a pickup.  Another experiment is to disconnect the negative piezo wire from the circuit entirely because frankly, it's not entirely required but it will lower your output volume but again might lend an interesting sound that may just do the trick

Ultimately, you would want to totally control the mix of both your guitar signal and the piezo signal by using a low noise quad operational amplifier which can help to both normalize and compress the irratic behavior of the piezo (it has a harsh voltage spike and then loses it's energy very quickly - it's a crystal) which will get it's signal levels porportional to your pickup levels.  Sometimes though this takes the fun out of guitar.  Sometimes it's good to just have it all wild and crazy and unpredictable. Be a little like Jimi Hendrix in that sense!  Take it easy.

[thehallofshields]

I wonder if anyone has tried putting Diodes in Series with the Piezo as described above. I've got not idea what this would do.

[Quackzed]

i'd recommend a tillman directly after the piezo , another buffer after the pickup switch for the pickups(tillman or whatever you like for mag pickups). then  a voltage divider for one or the other or both to get each to approximately normal passive level at their respective outputs. then you could have 2 separate vols -1 for each with a passive mixer at the end to keep em from interacting... just 2 pots and the 2 buffers... the buffers should handle the passive mixing without losing any treble.  the piezo is happy with a buffer before it hits anything else. and anything you plug into gets hit with a 'normal' signal level and behaves as expected...

[thehallofshields]

So 2 Volume Pots each with a Ground connection leading into a Passive Mixer Pot?

[thehallofshields]

Or I suppose I could MPF102 Buffer each, place a Trimpot Divider at each output, run them into a 100k Pot for Blending, then a J201 Tillman for 3db Gain, then Master Volume/Tone to the output.

[Quackzed]

i'd still say tillman after the peizo, any benefits you get from the tillman circuit will be lost after a  buffer, unless its also very high z buffer, like the tillman...
then make the last buffer a mpf102 booster for whatever makeup gain you need... though you might not need it, the piezo should have volume to spare,and any buffer after the mag pickups could just as easily be a booster for a little added gain... adjust the piezo output trimmer to around 3db boost via  its voltage divider/trimmer so its got enough boost, make the mag pickup buffer a booster and adjust its trimmer for a  3db boost then both outs into a blend pot then into vol/tone pots... no extra boost/buffer needed...

[thehallofshields]

i'd still say tillman after the peizo, any benefits you get from the tillman circuit will be lost after a  buffer, unless its also very high z buffer, like the tillman...

I figured the Series Z of the Mix Pot would cause problems or interact poorly with a Volume Pot.

You think I'd be fine with just 2 Tillmans?

[Quackzed]

I figured the Series Z of the Mix Pot would cause problems or interact poorly with a Volume Pot.

the tillmans should be able to drive the signal through the mix pot fine, there will be slightly less volume from each when at 50% mix pot, but thats the nature of mix pots...
the volume pot may have slightly less volume change up at the top of the taper 10 to 7 because its a higher value than is needed with the buffers, treble bleed mods with a parallel resistor cap also do this to some extent, and most (me included) seem to think of it as a good feature, rather than a negative, gives you a little more fine control between full and slight rolloff. if you find the volume rolls off too slowly at the top of the pot you can always sub in a smaller value volume pot later on, down to about 10k even should be ok with the buffers driving it, but you may need to re-adjust the trimmers to get a bit more boost back as a lower pot value will rob more level... i dont think you'll need it though... because of how a guitar volume works with a series resistance getting bigger and a  resistance to ground getting smaller as you turn it...

[thehallofshields]

Okay last question. What is the difference in current draw of Jfet Buffers and Opamp Buffers?

[Quackzed]

they're both really low... low enough to ignore the difference iirc... like in the pico amperes range if the opamp is a jfet opamp. 9v battery will last a long time!

[thehallofshields]

Awesome. A google search turned up a forum where one member claimed JFET Opamps used like 2.5mA even when idle. That seemed like far too much power dissipation. I was hoping they'd both be similar and in the Pico range.

I've got a design to run both pickups into a Dual Opamp Buffer, Mix Pot, JFET then regular Volume/Tone. I'm looking for something useful to do with a 4th Potentiometer.

The layout is very compact and I'll post it when I get it done.

[thehallofshields]

I can't get past reading that engineers suggestion to put Diodes in series with the Piezo.

I've been trying to research this and I'll I'm coming up with are Diodes used in Parallel with the Opamp Rails to clamp down the Piezo's Voltage spikes.

Did that long youtube comment I posted make sense to anyone?