Impedance Switching Possible?

[peskywinnets]

A piezo looks like a capacitor, so it reacts to cable length differently than a magnetic pickup, which is a great big inductor.

Specifically, a long cable just reduces the piezo's output voltage without altering its tone, because the cable is just another load of capacitance shunting the piezo's own. Too low input resistance on the preamp is what kills bass.

+1.

about 18 months ago, I went into this a fair bit towards designing myself a preamplifier for marrying up with a Variax hex piezo bridge with a mag (a blender if you like) ...the piezo signal output was a bit feisty for the associated circuit ...but could be tamed by simply slapping in more capacitance across the piezo.

[Rick Turner]

Piezos, particularly ceramic based ones, can have huge voltage outputs and incredibly wide dynamic range, and if you shield everything well, the noise floor is way, way down there. At an LA music party one evening, one of the engineers at Ocean Way Recorders in LA told me a story about Ry Cooder coming in to do the sound track for "Tresspass". The guy knew that I did guitar repair for him, and we were swapping Cooder stories. Ry was using his beam-like slide instrument for that session; it's basically a lap steel with a five foot string length and eleven strings on one side, and thirteen on the other if you flip it over. The engineer set initial levels on the console the usual way...by listening for the background noise and hum, and guessing what gain level to set. The Ry strummed the thing, and the engineer said the woofer cones leaped forward and were heading to the back wall, the output and low end response was so huge. I then told the engineer that I had made the pickups...

Properly done, piezos are like perfect Victorian children...they do not speak until spoken to...

[David Schwab]

Rick, you should write a book. You always have some good musician stories.

[Rick Turner]

Yeah, my publisher would like that book, too! I'm only 10 years past deadline... And the company has been sold twice. Now it's BackBeat. Then it was Miller-Freeman. At least they haven't asked me to return the advance!

[Guitarist]

I love Ry's sound and playing. I think he's still managed through Feldman here in Vancouver (with who I had a brief discussion while interning there.)

With those ceramic disks, where is a good starter area to set the pickup on an strung instrument. Things can get boomy inside a guitar or ukelele even! Is axis important?
Shielding one of those things must be fun. Oh yeah, I did finally buy some copper tape.

Another question while (sort of) on the topic. Do you find that string ground is still necessary?

[Rick Turner]

I didn't use ceramic discs for the pickup in Ry's "Big Thunder"; I made individual string pickups...yes, 24 of them closely coupled to the strings, and wired the arrays, one for 11 strings, and one for 13 in parallel with separate outputs for each array. I get my piezo ceramic elements custom made to my specs.

I'm not a huge fan of soundboard "transducers"; the location issues are just too twitchy for my taste. 1/16" one way or another and it's heaven or hell. I've heard some sound good and many sound like a cell phone at the bottom of a well. It's not the pickup technology per se; it's like real estate...location, location, location. In my opinion, the best of the accelerometer-type pickups are the Trance Audio system which derived from the old FRAP 3-D pickup and our D-TAR Perfect Timbre, and yet they, too, are incredibly sensitive to placement as well as the degree of damping or transfer of vibration through the adhesive medium. Even Bob Taylor, who probably makes the most consistent acoustic guitars in the world, could not plan the ideal position for his soundboard transducers for the Expression System. One would be on the money, and the next one, with pickups put in exactly the same positions according to blueprints, would sound like crap. Why? Because each piece of wood is different. Same reason why Stradivari graduated each top just a bit different and so all the subsequent violin makers who tried to exactly copy Strad's measurements fell far short of the master. Now that modern violin makers understand that it's about something other than graduation measurements, they are able to make truly remarkable violins which rival "old master" instruments.

[thehallofshields]

I do not quite understand "equalizer pedal feeding preamp".

I meant using the 9V current from an EQ Pedal and routing sending it to the input via Right Channel of a Stereo Cable to power an onboard Preamp. This was just an idea for portability and simplicity, but I'm here to listen. All this sweet information you guys are giving me is really making me change what I want to try out.

[thehallofshields]

Why the EQ? Where are your various stages of preamplification? What are the source and load impedances from one stage to the next?

I thought the eq would help me cut any shrill frequencies and was planning on using it as the base for a preamp/buffer, but I'll accept the wisdom that the buffers should be as close to the pickups as possible.

I've read that your average 22-27 Brass PZT has an impedance of around 1 megohm, so the plan was:
1.) 3 Brass PZT (1megohm each) ->
2.) 5-Way Switch (is this the same Z as putting them in parallel?) ->
3.) simple JFET Preamp/Buffer (5-10megohm in, 1kohm out) ->
4.) equipment designed for electric guitar.

Suggestions I've recieved:
Give each piezo it's own buffer.
Don't use a JFET.

[dpm]

Don't use a JFET

An opamp such as the TL071 or 72 is a JFET input device

[Rick Turner]

If you want to experiment in this field, you first need a really good "lab" buffer...one on which you can easily change input impedance and which has a lot of dynamic range. Running it on 18 or more volts for your "lab" work is good. Then experiment with the pickup designs as a totally separate issue from electronics. You need to change one thing at a time and one thing only. For this kind of work, you need that "lab" quality buffer and the only thing you'll change on it is the input impedance. Forget EQ for the time being; it helps you lie to yourself. Your goal should be to get the absolute best wide range sound from the pickup before you start mucking around with low end cutoff (via deliberate impedance mismatching) or high end roll off (super simple with passive treble cut done in the active circuit). If that doesn't get you in the ball park, you need to redesign your pickup, not fix a bad pickup with more and more stages of electronics. You need a really good wide range system to which to listen. Literally studio quality. You can always use guitar amps later. You need the audio microscope that will reveal everything, and it should be capable of going fairly loud.

This is how I've designed everything I've done in the "acoustic" realm, and the pay-off has been that when I go to gigs where one of my instruments or pickup systems is being used, I'll often go to whoever is mixing the sound, and I'll introduce myself. I then ask what they're doing at the console to deal with the signal from my stuff. Almost invariably, I'm told, "Oh, I'm just running it flat; it was the easiest thing I had to deal with tonight." And that is how I know I've done my job right way up stream in the signal path.

BTW, there are few real secrets about this stuff. It's very basic to first buffer the signal, and then to put in an EQ stage, a gain stage, a mix stage, or whatever, and then in some systems, to have a line driver/buffer AFTER a volume control but before the output jack.

You're giving away no ideas that haven't been done for decades, and tweaking frequency response by playing with impedance matching and mismatching is pretty old news, too, for both piezo and magnetic sources...and microphones of all types.

About twenty years ago, my pal Bob Wolstein designed a custom direct box for a studio that had variable guitar/bass load impedance on the input as well as variable source impedance on the output. It enabled long, long cable runs while presenting both the guitar AND the amp with "normal" 20 foot guitar cable response...or whatever other load/source tweakage one could want. It was killer for doing guitar tracks from a control room with a screamingly loud amp in an isolation room with both guitar and amp "thinking" the guitar was close. He could also split the output signal into multiple amps without loading down the guitar from it all.

[dpm]

BTW, there are few real secrets about this stuff. It's very basic to first buffer the signal, and then to put in an EQ stage, a gain stage, a mix stage, or whatever, and then in some systems, to have a line driver/buffer AFTER a volume control but before the output jack.

You're giving away no ideas that haven't been done for decades, and tweaking frequency response by playing with impedance matching and mismatching is pretty old news, too, for both piezo and magnetic sources...and microphones of all types.

About twenty years ago, my pal Bob Wolstein designed a custom direct box for a studio that had variable guitar/bass load impedance on the input as well as variable source impedance on the output. It enabled long, long cable runs while presenting both the guitar AND the amp with "normal" 20 foot guitar cable response...or whatever other load/source tweakage one could want. It was killer for doing guitar tracks from a control room with a screamingly loud amp in an isolation room with both guitar and amp "thinking" the guitar was close. He could also split the output signal into multiple amps without loading down the guitar from it all.

Oh yeah, I'm absolutely aware this is all old news, the only 'innovative' (but not really) thing I'm thinking of incorporating is making a few things controllable and selectable onboard in a very simple manner, if needed at all. This is a very interesting discussion, I haven't really got into piezo circuits in any depth because they're not that relevant on a personal level. I'm one of those either purely electric or purely acoustic kind of guys so I get a little frustrated with amplified acoustic sound, but then I'm no performer so haven't needed to find a solution.

[thehallofshields]

Forget EQ for the time being; it helps you lie to yourself.

You're giving away no ideas that haven't been done for decades...

Hey thanks; I appreciate the brutal honesty. So obviously I've got a lot to learn about audio, and I like your methodology. Are there any resources (threads, books, schematics) you could direct me to that could help me get a basic design figured out?

I don't know what a lab buffer could mean. Are you suggesting something like an external preamp/direct box type of thing? My original intention was to make external buffer/preamp that would have a somewhat universal use, so that I could play with different pickup placements and circuitry designs easily.

Where do you recommend I go from here?

[dpm]

Rick, if you have time, how much does a full range piezo guitar signal benefit from high headroom 'hifi' opamps such as those offered by Burr Brown and Analog Devices? And how much is compromised by going to onboard 9v power with long battery life? I've only read comparisons of these things in a hifi context which I take with an enormous grain of salt. I see traditional magnetic electric sounds as fairly forgiving, lack of dynamic range and limited frequency response not being too much of an issue. I'm more concerned about noise in that context.

[Rick Turner]

Practically everything I've done with piezos in the past 20 years has used 18 volt supplies, either single ended ( + 18 V ) or bi-polar ( + 9, -9). That extra headroom is very useful.

[peskywinnets]

Make of this waht you will, Line6 use six single supply 5V preamps with their Variax hex peizo bridges (piezos made by LR Baggs) ...the actual opamps they use are TI OPA4340s.

I guess it comes down (in part) to the size piezo used...the larger the piezo the feistier the signal - the hex piezos used by Line6 are rather small.