I'm not sure where this should go So I'm starting it here. If y'all think it belongs under another thread please point the way.
I was going to keep this to myself and try to market a board to solve some of the problems but decided I'd just as soon make it public and if there is enough demand maybe proceed with a design. I've already built one board to test with and initial testing looks like I'm on the right track.
From what I've read, people are unhappy with two characteristics of Piezo pickups.
(1) Feedback and runaway squeal?
(2) really nasty sound when overdriven.
I'm going to address issue 2 first as I feel like i have a better hand on this. Although, I think issue 1 is related to issue 2.
I picked up a Radio Shack 273-0073 to experiment with. First test I hooked it up to an oscilloscope (instrument to visually display the relationship between Voltage (and possibly current) with respect to time. I removed the piezo element from the plastic housing and laid it flat on a table top. I connected the wires to the oscilloscope and pressed the piezo element down on the table top, then lightly tapped the table with a spoon.
Ding! 3V peak output. 3V is equivalent to a Humbucker output. Most inductive pickups are much lower, typically around 0.5V.
So the Piezo has a high output voltage drive (with little current as it has a high impedance).
Next I struck the table a smart rap with the spoon.
Output went right off the limits! I re-adjusted the scope and tried again. 20Volts peak! Damn! These things can overdirve the hell out of amp inputs.
In fact, some Piezo elements can produce several thousand volts when properly configured and are used to ignite propane grills.
So, what to do?
Most people are dampening the response of the element by mounting it with softer adhesives, and carefully placing them near the bridge and away from the more resonant back panel (I believe this to be the current best case recommendation, yes? no?)
Limit the output when heavily driven to (1) protect the amp inputs, (2) limit the overdirve to produce more pleasant sound.
How do you limit the output? Clamps or compressors.
The simplest clamp is a pair of diodes in parallel with the piezo output.
A better clamp would be to add a resistor in series with the diodes to limit the current.
Even better would be to buffer the output, them add the clamps, follow it with a buffer.
Finally, a compressor that detects the output level and gradually decreases the gain of the amplifier to control amount of compression.
Right now I have a TL082 with the first op amp configured as a gain of 0.5 (yes,negative gain since the output of the piezo element is so strong). The output of the first stage drives a 10K resistor, then a 10K resistor to two switching diodes (1N914) to the ref level. The connection of the two 10K resistors is the output and goes to a 1K resistor to two 2.6V zener diodes in parallel but cathode to anode and anode to cathode. These also tie to the ref level for the op amps. FInally the second amp (TL082 has two op amps in one 8 pin package) is used to buffer the output.
My initial test showed that this does indeed provide some soft clipping.
Next up is to build a LM380 power amp and mount the piezo element under the bridge of my two string slide I built last night.Small First Two String CBG_2.JPG
Replies
I use the Tillman preamp design with some extra stuff on the input stage to even out the variance in FETs and a trim pot before the output to bring it down a little if necessary. I use 25k pots for tone and volume, 0.047 cap for tone. So far as mounting piezos goes, I've tried a few things, but now I just cut out a circle of ply slightly bigger than the disc, supa-glue the piezo to the underside of the lid in the bridge position (maybe just slightly forward and slightly toward the bass strings), run hot blue around the edge of it and sit the ply circle on top, then run hot glue around the ply edge so the disc is completely sealed off. Quick, easy, sounds good, no feedback.
Using two piezos in parallel will halve your impedance, which can't be a bad thing. Just splice red to red, and black to black.
I prefer the larger size discs because you get a more even response across all strings, but I guess if you use two discs parallel the small ones are fine. Mount them side by side and they cover the whole bridge width.
I wondered if anyone has tried this as I thought that sticking a strip on a bridge might be quite effective?
Any one tried? Anyone like to offer an opinion?
Thanks
Yeah, the films need LOTS of pressure. I can see how sitting a bridge directly on them would achieve that. So on a lid they need to be sandwiched on really tight. I have a couple I haven't been able to use because I can't get enough out of them but will try putting them between bridge and lid. Thanks Randy!
I suspect the issues have been addressed by mechanical mounting techniques.
I posted my start of this thread after reading a lot if intro stuff (2002-2004) , but before finding a lot of late stuff (2007-2009). As an engineer I would like to understand the mechanics behind the resonance issues. As an EE I tend to look for and electrical/electronic solution rather than a mechanical one.
Right now I"m off on a tube design tangent. I've built two CBGs and have the second tuned Drop D (DADG). So I'm trying to learn to play it.
Sven
Michael Puckett said:
Since it is being used as a microphone, this is not nearly an optimal configuration. Experience has shown that removing the piezo element from the plastic case and mounting to wood in the guitar is a better solution. I have mine mounted under the bridge, and glued to the neck that runs through the guitar. It logically makes sense to mount it in this location since the bridge will conduct the sound/mechanical vibrations directly to the piezo element at this location.
William Bannier said:
Also, where's the "best" place for putting the transducer?
I finally finished my 4 string with piezo pickup. (The 'kids' had to check it out).
I built a pre-amplifier to buffer the signal from the Piezo element, and finally got it working (forgot a ground and spent half a day chasing a 60hz hum problem).
It appears that a good strum on the top E string produces a nice 1V peak signal that rapidly decays. 1V is nominal for the input to most amps and should not overdrive the input. So with the piezo element mounted under the bridge the piezo element produces a good nominal amplitude signal. This seems to be the recommended location now days for the piezo element. (I mounted it with hot glue per the nom suggestion.
I'll mount another piezo element on the back of the cigar box to see if it produces a greater signal strength, as some have reported what I interpreted to be excessive drive and induced clipping at this location. I'll check the amplitude at this location and report what I find later in the week.
I hope to get the power amp going this week as well, but I need to pick green beans and can them so I may be a bit off on my expectations.
First_4-String.JPG
First_4-String_1.JPG
-
1
-
2
of 2 Next