Piezo wire

[David Schwab]

Yeah I saw those.. haven't looked through them yet. And some other interesting ones too. But you can see that Highlander sells them, and they were the first to market with an under-saddle coax pickup... been out since the mid 90's. I believe Rick Tuner designed the Highlander.

[David Schwab]

OK.. I found the answer as to where and how the first piezo coax cable pickups were made... it's an interesting story!


Interview with Rick Tuner

Tell me about the pickups you've designed for these Renaissance guitars, and how that design was developed.
Well, one day David Crosby comes into Westwood Music and says, "Turner, if you design a better way to amplify an acoustic guitar, and I don't get it first, I'll have your balls cut off." (laughter) So what are you going to do? It was something that needed to be done anyway, and I had already done work in that area for Gibson. One really great thing I had going for me at that time was that Jackson Browne basically put his studio at my disposal, which was incredible. If there wasn't a session happening I could go in and listen to anything, and record it to twenty-four track tape, and listen back. Some of the finest analog gear in the world is in his studio. So that was my sound laboratory when I was in LA. I'd go down to Jackson's and we'd try stuff all day, and record it, and listen back.

During this period I came across this passive-sonar-array technology, this coaxial wire. I'm good at finding things, and I'm good at talking to engineers at other companies and getting them excited about new applications for their products. You've got to be willing to jump out there. Now, with the internet it's amazing. You can find stuff pretty easily. Anyway, I find this stuff, and it looks pretty interesting.

Is it the wire that they trail behind submarines?
Right, and listen for Red October. It's a coaxial wire but instead of an insulator, what would be the insulator is rubber impregnated with piezo-ceramic dust. I stumbled on the material, and then worked with Bob Wolstein on its application. I had started working with Bob on hex pickups when I was at Westwood Music and doing the Gibson stuff. He is as fine an electronic-circuit designer as anybody I've ever met.

If the piezo material is impregnated in rubber, how does the signal develop.
Well, there's a center conductor and an outer shield with this stuff in-between. It develops a voltage potential between the center conductor and the outside shield, and it does so radially. One of the things that distinguishes the piezo-cable pickups from flat piezo pickups is that they give you vibrational input from the walls of the saddle slot and the saddle instead of just from the saddle, so they develop a signal more three dimensionally, where a flat piezo responds more two dimensionally. The end result is a more complex signal. You couple that with Bob's electronics, which I haven't heard anyone beat, and you've got a very, very nice system.

We developed the Highlander pickup at Westwood Music and in Jackson's studio. My two contributions to it were finding the technology, finding the coaxial pickup material and working the deal with the company that makes it to get what we needed, and then, since most luthiers shouldn't ever be allowed to handle a soldering iron, nor to go anywhere near a fine guitar with a soldering iron, I came up with the idea to develop a packaging system that was soldering-iron-free, and that also took care of shielding and so on. I don't know whether it was Bob or me who thought, what if we could design it to install completely from the outside, through the end pin hole? Anyway, Bob said, "Oh yeah, I can do that." Of course, now the other pickup companies, Fishman, and so on, have knocked off this preamp-in-an-end-pin concept from Highlander.
Bob had started working on circuit design with computers at that point. He switched over from designing circuits in the traditional way to using computers in no time flat. The guy's a friggin' genius. He had circuit-simulator software, and he was able to write a program which set up the circuit for the preamp with all the parts values, and then it changed the parts values slightly and tested each virtual circuit just as though he was working with real-world parts. The amazing thing was that he was able to set his computer up to test 10,000 virtual circuits overnight! By testing all these variations he could determine the parts values we wanted. That makes a difference not only in performance, but also in how much the circuit is going to cost. You can get parts that are rated to various tolerances, 20% parts, 10% parts, 5%, 2%, and 1%, but the tighter the values are, the higher the cost. For instance, he discovered that the input-impedance-setting resistor was the most important resistor in the circuit in terms of noise, and the most important in precision in terms of reproducible results. So bam, that's a 2% part. Everything else can be 5%. Then he designed this long skinny circuit card that fit in a metal tube that soldered directly onto an end-pin jack, and bam, we were there! I wanted to call it a tube preamp, but that was taken (laughter).

So that was the start of Highlander. Bob and I were partners in that until my guitar making thing, which was starting to take off simultaneously, kicked in to the point where I just said, "Bob, we can be partners or we can be friends, but I can't start two companies at the same time. You take Highlander and do it." I've continued to use Bob as my primary supplier for electronic circuits, but he and Holly have taken the company and made it roll.

From http://www.renaissanceguitars.com/interview.html

[Just Bob]

It is an interesting story. Some of the factoids are a little out of kilter regarding the piezo coax, but close. The stuff they use for intrusion detection (which is also used for saddle pickups) is not the same stuff they use on subs. You can't buy that on the open market. You also can't simply take piezoelectric particles and mix them with rubber. Particles would all have to radially align their sensitive axis with the core and the rubber would need to be somewhat conductive and other minor issues. I spent a good 20 or 25 years designing electronics and sensors for government use, including these kind of sensors.

The piezo coax you can get is made by wrapping a plastic tape around a wire core and applying a braided shield. The plastic (PVDF) is stretched to make it piezoelectric. Its made to bury in the ground or in concrete slabs and detect footsteps. Much less challenging than sonar. Anyway, the rest of the story is interesting.

The piezo coax is compressed into the saddle slot and you detect the volumetric compression of the coax. Whether that's better or worse than a flat piezo pickup, I couldn't say. The signal may be larger, but so is the noise because that piezo plastic is noisier than ceramic piezoelectrics. What matters more is the signal-to-noise ratio rather than just the absolute signal level. Anyway, can get too easily bogged down in all the technology when what counts is the sound. If it makes the sound you like, its good.

[ake]

Piezo wire

another piezopickup made off piezowire http://www.headwayelectronics.com/

[David Schwab]

The piezo coax is compressed into the saddle slot and you detect the volumetric compression of the coax. Whether that's better or worse than a flat piezo pickup, I couldn't say.

Thanks for the info... very interesting.

Tuner claims that because the coax also touches the sides of the slot, it also picks up side to side motion, as well as the up and down stuff between the saddle and bridge.

[Just Bob]

Thanks for the info... very interesting.

Tuner claims that because the coax also touches the sides of the slot, it also picks up side to side motion, as well as the up and down stuff between the saddle and bridge.

That's an interesting concept, but I'm not sure of the real necessity. Depends on whether you're trying for "accurate" sound or "good" sound. Any force on the bridge in any direction which causes compression of the coax will give signal. So, you're measuring stuff that does not necessarily contribute to the acoustic sound. Good? Bad? I'm not sure. The bridge tries to move every which way but the guitar top constrains it to move only in certain directions. The top radiates your sound. I would think that ideally the bridge is tight in the saddle and they move as a unit. A solid piezo sensor has a high modulus. It does not squish much under the force of the strings. Millionths of an inch. The piezo coax is much softer and does deform. Probably in the thousandths of an inch range. That will have some effect on the sustain and tone.

Anyway, the whole business of sensors is complicated. The sensor changes the instrument. It's really a different instrument when you put any sort of sensor on it. The sensor does not measure what you hear acoustically. A contact mic or soundhole or endpin mic does not hear what the player or the audience hears. The amp and speakers color the sound. You just configure the whole package and see if you like the sound.

[David Schwab]

Very true. One interesting thing he says is that his Renaissance guitars sound more like an acoustic than most acoustics with under saddle transducers because his top doesn't vibrate under the bridge (he has a solid block under the bridge like an ES-335). His reasoning is the vibrating top will cancel out some bridge vibrations.

it is a complex situation.

I'm planning on making a coax under saddle pickup for a solid body fretless bass. I'm going to replace the metal Schaller bridge with a wooden or aluminum acoustic guitar style bridge.

[Just Bob]

That sounds pretty cool. I never tried a piezo on a solid-body. A piezo transducer is basically a capacitor. It generates charge in response to strain. Any other capacitance in parallel with the sensor (i.e. cable) will form a capacitive voltage divider and reduce your available output voltage. Also, the input impedance of the preamp forms an RC filter with the pickup. The capacitance of that little piece of coax is probably only 2 or 3 pF. Ceramic piezo materials have a much higher dielectric constant. If you want good output on the low end (RC constant around 20 Hz), you need an input impedance for the preamp that's huge. Around 1e11 ohms. I'd do a MOSFET preamp right in the saddle. Don't know if you can do surface mount, but that's what I use for most stuff these days.

Piezo materials respond to strain. How much motion you need to get a certain strain depends on the modulus of elasticity of the piezo material. Ceramics are stiff but the plastic piezo stuff is soft. Probably also a good amount of hysteresis. You need a lot more actual motion. Piezo materials are generally used so they provide the restoring force (think springs) in a mechanical circuit. So, the trick is to constrain the bridge so it can't flop around too much but still put pressure on the piezo stuff. Maybe a tall Delrin bridge in a deep saddle slot. Or, something more like a hinge using stainless shim stock.

[David Schwab]

I haven't used any surface mount stuff yet, but I am planning on using a FET or MOSFET based circuit.

I once saw an interesting project that used two pieces of copper clad circuit board with a thin piece of foam rubber in between that was used for an electronic drum pad. When you hit it, it changed the capacitance and the circuit that went with it produced a trigger pulse.

I'm going to make a bridge with some type of synthetic saddle, like delrin or tusq. The idea of using aluminum as the base (and maybe even the saddle) is it has a similar modulus as wood.

Turner does an interesting thing... he angles his saddle back about 13°, which is supposed to improve the pressure.

[Joe Gwinn]

I haven't used any surface mount stuff yet, but I am planning on using a FET or MOSFET based circuit.

Be aware that MOSFETs are very noisy as linear amplifiers, but make good switches. JFETs (and bipolar transistors) can be very quiet.

[Just Bob]

Be aware that MOSFETs are very noisy as linear amplifiers, but make good switches. JFETs (and bipolar transistors) can be very quiet.

Yes, but there's a lot of art to this. MOSFET current noise can be very low and a piezo preamp needs very high input impedance. Designing a low-distortion MOSFET front end isn't the simplest thing. A fairly good off-the-shelf solution is an OPA129 opamp (available in surface mount for about $6). It has input bias current in the femptoamp range. As a non-inverting amplifier, the input impedance can be in the 1e15 area with noise voltage around 25 nV/root Hz. Rather high, but the larger signal available by using the high impedance makes the overall SNR at the output pretty good. With inverting amplifiers, the necessary feedback resistor is huge (over 10 gigohms) and is a trick in itself.

Layout for these sort of amplifiers is critical. Using .050 pitch SO parts, it is possible to hand solder if you're good. Smaller pitch is much harder, I did it for years before getting the hot-air equipment. I just tinned one pad for a component, then soldered one end of the part while holding it with tweezers. Once it's in place, the other pads can be soldered. Takes a little practice and best to try it before you've had 6 cups of coffee. A piezo preamp with the OPA129 only has about 5 parts. Could easily make it on a piece of PC board with a dremel. Around 0.2 by 0.3 inches. Solder the piezo coax right to the board.

[David Schwab]

Be aware that MOSFETs are very noisy as linear amplifiers, but make good switches. JFETs (and bipolar transistors) can be very quiet.

I generally use JFETs, both transistors and op amps, and some bipolar transistors... I haven't used any MOSFETs as of yet. I have a few circuits that use them that I have been wanting to try out, but I might just use an op amp anyway... also I've been wanting to try some of the nice audio op amps from Analog and Burr Brown.

[Joe Gwinn]

I generally use JFETs, both transistors and op amps, and some bipolar transistors... I haven't used any MOSFETs as of yet. I have a few circuits that use them that I have been wanting to try out, but I might just use an op amp anyway... also I've been wanting to try some of the nice audio op amps from Analog and Burr Brown.

OK. The consequence of device noise dependes on where in the signal chain the device is used, and for what. MOSFETs used as variable resistors (versus amplifiers) or switches don't cause excess noise.

[jimi]

I'm surprised somebody said this stuff sounds like crap. Maybe it's the type of wire used. I've played around with it a bit and to me it sounds really nice, very natural. I compared it directly to an EMG undersaddle pickup and it was miles better. I left the outer braid on. I use a simple preamp with a jfet and a transistor and had no problems with noise. In fact it's very quiet. It has a slightly lower output than usual so needs to be amplified more.
It's usually recomended to rout a half round groove in the bottom of the saddle slot for it to sit in although I didnt do this and had no problems.
I know of two companys that make pickups with it, Highlander and Headway in the UK so I don't know if there's any patent on it.

[Just Bob]

Piezo pickups aren't really bad, they just tend to be a bit more noisy than I like. A lot of companies make them and they wouldn't sell if they were really bad. A lot depends on on your amp and speakers, of course. If you play through equipment meant for electric rather than acoustic, you may not hear any noise at all. I use flat studio monitors and no effects or compression or anything and the hiss can be somewhat audible. They can also make some strange noises if they're not firmly bedded in the saddle, but that's easily fixable. They have their advantages.