Piezo sensors... "Pressure, pushing down on me"

[tedmich]

when I play my guitar with both hands hammering on I get three notes per string: one from the nut to my left hand finger, one from my right hand finger to the bridge and the note in between my fingers. I can hear them all until I plug in then the bridge note takes over. If I could have a PU up on the 12th fret and one near the nut I could amplify all these notes, but the wound coils have to be very small and the wires run under the fretboard or (better) down the side of the neck in an ultra thin flat wire... I don't think piezos could work as there is no way to get the tension except at the fret which has two notes, one at each side, and the fret moves...

Regarding headstock vibration, the Sustainiac Model C simply clamps to any stringed instruments head stock,
SUSTAINIAC MODEL C
and vibrates it at the same pitch as the note, as headstocks are pretty much tuning forks. Cable is a pain though...

[Rick Turner]

David,

I beg to differ. It's parallel resistance that rolls off low end by lowering the impedance load.

You can play with parallel load resistance to tame too much sub-30 Hz response. With the stuff I do, we generally use a 1 meg Ohm load for magnetic pickups and a 10 meg load for piezos, but we've gone as high as 100 meg (the D-TAR Wavelength system), and I've put three way switches in some instruments to purposely knock some low end off when needed.

Series resistance will just make piezo systems noisier.

One nice thing is that within reason, you can passively parallel individual piezo pickups. I've done this with up to 11 elements on a wild sort of experimental "beam" lap steel for Ry Cooder. That one had 9 strings on one side and 11 on the other. Yes, a 9 string pickup and an 11 string pickup, and low end that frightened the hell out of a recording engineer in LA. He was used to approximately setting levels on electric guitars based on hearing the noise floor. This instrument with an 18 volt buffer is incredibly quiet just sitting there, so the guy cranked the gain. He said he thought the woofer cones were going to hit the back wall when Ry strummed the thing. That'll teach him...

[Rick Turner]

One of the advantages had by Baldwin, the early '60's Gibson designers, Ovation, and Takamine is that they did not have to design pickups that could be retrofitted into Martins, etc. That's a real design restraint; I've worked a lot within that realm, but it's really great when I can be free of the requirement of the after-market. It's hard enough selling coaxial cable pickups that work best in a cove-bottomed slot tilted back about 7 to 9 degrees.

Thanks to the miracle of 3D printing and CNC capability, I can now go back to some of my earlier individual string piezo designs and re-engineer for cost-effective limited production.

[Leo_Gnardo]

Another good classic piezo pickup...the Takamine system with the six piezo cylinders...especially if you bypass the Tak electronics and use a really high quality buffer. Frighteningly good, and one of the instruments/pickups that taught me just how badly you can screw up a good source with bells and whistles electronics. I tested four generations of Tak on-board preamps using a custom high voltage DC servod 10 Meg input buffer as a control, and the best of the Tak preamps were the earliest and simplest. Through the reference preamp, the Tak pickup sounded amazing. This was a Jackson Browne's studio at about the same time I did Willie's guitar bridge.

I'm glad your ears heard what mine did Rick - first year or two out those Takamine "Palathetic" pickups sounded so good it was uncanny. And never since. But the bells & whistles sure did abound. Guitar-side preamps first with semi-parametric mid (a good thing as long as it's carefully used) then with tuner built in, then another with reverb effects. Nothing matches those early ones though. Remember my slogan "simplicity yields clarity." That's all very well and good but 1) you gotta have new products to show off at NAMM & Musikmesse, 2) modern guitarists demand more knobs and sliders, so they can personally wreck their sound while the crowd watches (and the star berates their techs and monitor mixers). How many times I've been through that last one... feh!

[Rick Turner]

I've put piezo pickups under string nuts; one for Fred Frith on one of my guitars and one for classical guitarist Mesut Ozgen for a "prepared guitar" piece written for him by Benjamin Verdery. Interesting for special effects, but it's mostly (but not completely) enharmonic. It would be very interesting for slide guitar effects. I'll have to do one for Martin Simpson who does a lot of "behind the slide" effects...spooky stuff. Might be interesting to hear what Sonny Landreth could do with this.

I've put them in the neck joint of a Fender...don't bother.

I inlaid a magnetic pickup between the 1st and 2nd fret for a guy in 1970 or '71...don't bother.

Piezos in the peghead?...don't bother...

And read up on "accelerometer". In piezo land, it's a piezo element with a weight load. That's what the FRAPs were...in 3D...now reborn as the Trance Audio Amulet system.

Also, bear in mind that a lot of the piezos of the past were poorly interfaced, but that does bring us to the 1960s Baldwin and then Gibson C1E piezo systems which were actually outstandingly good. And the Baldwin had an onboard preamp...yes, 50 years ago. It's what Willie Nelson has in his Martin, "Trigger", for which I made a new bridge in 1990. Another good classic piezo pickup...the Takamine system with the six piezo cylinders...especially if you bypass the Tak electronics and use a really high quality buffer. Frighteningly good, and one of the instruments/pickups that taught me just how badly you can screw up a good source with bells and whistles electronics. I tested four generations of Tak on-board preamps using a custom high voltage DC servod 10 Meg input buffer as a control, and the best of the Tak preamps were the earliest and simplest. Through the reference preamp, the Tak pickup sounded amazing. This was a Jackson Browne's studio at about the same time I did Willie's guitar bridge.

[David Schwab]

given that most people think of them as weak and tinny - at least, they are when not used properly.

Anything not used properly has a tendency to sound bad, because it's not being used properly!

If you put resistance in series with a piezo you lose low frequency response. They need a very high impedance load.

[David Schwab]

I read of Frank Zappa having piezo's in his headstock but who knows if that ever made it to record - might have worked if squeezed under the nut.)

The burnt Hendrix Strat had Barcus Berry Hot Dots in the neck. You can hear that guitar on the solo in Apostrophe'. It's very clanky sounding on top of the regular pickup sound.

[John_H]

...What magnetic pickups seem to do, and what acoustic instrument definitely do is to "time dis-align" stringed instrument string signals. That is not necessarily bad...this phase shifting adds interesting non-linearities. I just think we should understand that that is what's going on and then choose how to deal/work with it artistically.

Great explanation, and I'd like to add an example where it applies to me. I had a 12 string Ovation guitar that recorded really well. I would mic the bridge, and neck, and record the piezo too. Mic placement is very critical. You not only need to have your mic's in phase, but it's necessary to balance them with the piezo as well. When you get it right, the difference is like night, and day.

[Rick Turner]

Tell me what would happen if you ran a magnetic pickup into the input of a PA or recording console without using a direct box. That's going to put typically a 100 K or considerably less load on the guitar. "Weak and tinny" might be a appropriate response. If you do not interface a piezo...any piezo...into the proper buffer preamp stage, it will be weak and tinny. That's not the pickup; it's the impedance match that is bad. Piezos must be buffered to get out what they can deliver.

I've made piezos that I could see were tracking well down into subsonic ranges...using a fully DC coupled amplification with no capacitors in the signal path. You're not going to get that with magnetic pickups. Of course, below about 30 Hz is fairly unusable and just messes up power amps and loudspeakers, but it's nice to know you can get it if you need it.

Now for phase response...

Magnetic pickups divide the string by a ratio that is constantly changing as you play up and down the neck. You thus have nodes and antinodes swishing around with cancellations and reinforcements that have nothing much to do with what the string is actually doing. It's OK, we've gotten used to it; we even like it, but it's not an accurate representation of what the string is doing. Then there are group delay issues that I can hear but not adequately "scientifically" monitor...yet...but I know they are there.

A bridge saddle coupled piezo receives and transduces what the string is doing at the bridge quite accurately. In fact, it gets the same string signal that is then sent into the top of an acoustic instrument. And it gets that "information" before the rest of the guitar top and body gets it.

What magnetic pickups seem to do, and what acoustic instrument definitely do is to "time dis-align" stringed instrument string signals. That is not necessarily bad...this phase shifting adds interesting non-linearities. I just think we should understand that that is what's going on and then choose how to deal/work with it artistically.

[rhgwynn]

It was surprising to me when Rick said that piezos held auditory superiority over magnetic pickups; given that most people think of them as weak and tinny - at least, they are when not used properly. But after I took a moment to think about it, it kinda makes sense. There is no electronic interference picked up, which cleans up the signal; noise tends to get worse the more pedals you use, and can be a big problem when you are dependent on them for your sound. The downside might be if you are too noisy against the soundboard or wherever the sensor is mounted, that sound would come through. I'm sure there are more but I'd rather not speculate.

[big_teee]

Thanks Bruce.
T

[Bruce Johnson]

Here's an analogy that may help:

Think of a piezo element as being a hollow rubber bulb with a tube coming out of it. When you squeeze the bulb repeatedly, you get pulses of air coming out the tube. The harder you squeeze, the bigger the amplitude of the pulse. But if you just glue the bulb firmly onto a vibrating surface, you won't get much signal. The bulb needs to be squeezed, squashed, or significantly bent between parts that move in relation to each other, to get pulses of signal.

The mechanical side of designing a piezo pickup is all about applying physical movement to the element. You can't just hang the element somewhere and expect it to pick up sound. It has to be mounted down solidly AND have a separate vibrating part pressing on it, causing it to squash or bend. The bulb has to be squeezed. Mounting element in the bridge, with a saddle block on top of it, is one of the obvious ways to get that squashing action. If you mount it somewhere else, it has to be between moving parts.

Now, some pre-made piezo pickups already have been made to create that squashing/bending internally. For example, most large disk pickups are made to be attached to a surface. But, they depend upon the disk flexing and bending to make signal. The amplitude of the movement applied to the disk will determine the amplitude of the signal. Mount it squishy and the signal will be squishy. Mount it to something solid and immoveable, and there's no signal.

I learned all this from working with Rick on some systems for my basses, back in the 1990's.

[Rick Turner]

A few pithy comments:

If you're working a piezo pickup in pressure mode, then they work best with a decent pre-load of string pressure to mechanically bias them into a good linear output mode.

If you're working in shear mode, you just need enough down pressure to keep extraneous rattles out of the picture.

The best results happen when the foundation/mounting of the pickup is very, very solid.

The best results happen when the vibration input to the top of a piezo block is as isolated as possible from the foundation/mounting of the pickup.

Too much shielding is almost enough.

There can be "magic ratios" of closeness to the strings vs. closeness to the top on acoustic instruments.

Area counts.

When closely coupled to the strings, piezo pickups have significantly better phase response to string vibration than the instrument itself...for better or worse.

The frequency response of piezo pickups makes magnetic pickups sound like early telephones. We're talking whales to bats here.

If you do everything right and still don't like the sound, then it's location, location, location, OR you just don't like to hear phase coherent string signals. And then you'll have to go into the DSP realm to deal with it. That's why I proposed the Mama Bear DSP acoustic guitar preamp, first to Martin Guitars (Chris didn't like it when I said it was possible to make an Ovation sound like a D-28...but it is...), and then to my pals at Seymour Duncan. Hence D-TAR and the Mama Bear project.

Don't bother with all the math. Make, properly buffer, and listen.

Play with load impedance to control low end. It can be too much. Shunt resistance across the pickup makes for a high pass filter. See previous comment re. the math. Just substitute different resistors. I've even done three position load switches for extreme low end tone control.

[rhgwynn]

I used to have a Washburn bass that was all piezo. It had the strip under the bridge assembly, and each individual saddle had a piezo button epoxied into a recess on the bottom. The tiny wires broke off one or two, so I had a chance to take a look. It seems to me that for a piezo disc or button to work well, it needs to be firmly secured and have solid contact with the vibrating surface. I managed to clean them all out, and they were VERY firmly installed. Maybe a recess and epoxy is the answer here? I'm not sure about different epoxies abilities to help or hinder the buttons ability to pick up the vibrations but, there you have it. The bass did have a preamp - their so called Advanced Bass Technology.

[Rick Turner]

Catching up here...

First, the Highlander was and is NOT an electret element. The original which I sourced and had custom made by NTK in Japan was a coaxial wire with piezo ceramic dust impregnated rubber as what you'd think of as being the "insulator" between the center conductor and the shield. The newer ones as well as the D-TAR (Duncan Turner Acoustic Research) is a piezo polymer (Kynar) in the same coax configuration.

Second, it's all about location, location, location and interface geometry. And then it's about proper buffering and possibly pre-amplification (for lower output sensors). There are a couple of approaches to the buffering...high-Z input FET preamps and charge amps, but if you don't get it right at that very first stage of electronics, you'll never get it sounding good. Piezos have incredible dynamic range, hence my insistence of 18 Volt buffer/preamps in the instrument.

Also, piezos make utterly lousy line drivers. You can kind of get away with passive (into a high Z external preamp or "acoustic" amplifier with a very high Z input) with piezo ceramic pickups. You can't at all with piezo polymer pickups. Kevin Beller measured a 9 dB loss in only 10 feet of decent guitar cable. It's the voltage loss to capacitive loading that kills you.

The other issue is that too much shielding is barely enough with piezos. As a source impedance, they are almost like an open circuit, and it's just so easy for hum fields to get in.

The advantage of undersaddle pickups is that they get the guitar string signal from the same location (bridge saddle) as the guitar gets the signal.

I could write a book on this stuff, but I think that once the basics are understood...then you should develop your own style.

I do not much favor SBTs...Sound Board Transducers. They are so location sensitive that they all too often sound like the sound is coming from the bottom of a resonant well. They also turn the top of a guitar into a mic diaphragm, and thus they are feedback monsters. The best of them...like the old FRAP and the Trance Audio systems...wind up mounted in pairs directly under the bridge saddle on the bridge plate.