Impedance Switching Possible?

[thehallofshields]

Hey guys. I'm going down the road of learning audio electrics by playing with piezo-electric transducers on a banjo and bass. I'm an still a novice.

I'd like to modify the pzt preamp schematics floating around to allow me to switch between different resistor values. -This would (theoretically) allow me to use instruments rigged with single pzt (10megaohm resistor) , parallel pzt (3megohm resistor) and magnetic-coil pickups (1megohm resistor) with the same preamp circuit. As a side-note I plan to both make a mint-box pre-amp and figure out how to mod a Boss GE-7 for multiple input impedance values.

But since the main 1megohm 3megohm or 10megohm resistor of most preamp circuits is connected directly to the ground; I sense danger in switching this value while the signal is active or the pedal plugged in.

I haven't figured out how to wire the switches yet, but I was hoping you guys could alert me to problems that I'll likely encounter.

Is switching input impedance possible and/or safe?

Thanks.

[peskywinnets]

I'd be inclined just to pitch for the middle value @4.7Meg ...you'll get a tad less output on the piezos - but they've a pretty strong signal anyway. I'd imagine you main hindrance isn't the input resistor...but if sharing the same preamp, then balancing the output levels - there's going to be a chunky variation between the mag & the piezo.

[David Schwab]

When you increase the loading on piezos you lose low end and the top end gets scratchy.

The best way to do this is with separate buffers for each pickup that then get mixed together, instead of sharing a buffer stage.

[thehallofshields]

Thanks for the input.

So, would it be safe to build a preamp pedal with a Variable-Resistor pot on the input side of the circuit so I could turn an external knob to go from 10megohm (Piezo) to 1megohm (Magnetic)?

Most JFET circuits that I have found have the impedance-bridging input resistor between the positive input and the input ground. Something tells me that turning this knob while a pedal is engaged might not work so well. - I sense danger!

BTW. I'm not planning on mixing pickup types inside of instruments, I simply want to design a preamp that can bridge the impedance of different types of pickups.

[dpm]

I've done mosfet preamps with variable input impedance for magnetic pickups, it's a very cool way to tune things IMO. For your purposes I think you should go with with a switch as there's no real need for 'tuning'. You might also think about putting switchable capacitors to ground on the input to emulate cable capacitance, maybe 100pF, 220pF, 470pF for example. You'll need an output level control or pad switch if you want consistent level from this thing, piezos can be a lot louder than magnetics. Why the jfet pre? Opamps work well and are very forgiving. I only used mosfets because my application called for multiple onboard preamp stages so I wanted to keep the current draw low.

[Rick Turner]

I've done this with piezos, and it's no big deal. The way to do it is to have a 10 meg load resistor permanently in the circuit, and then switch in additional resistors in parallel to drop the load impedance to whatever levels you want. Works fine to roll off the low end and control rumble stuff. It's all quite "pickup design" specific...you'll have to just tweak by ear.

[thehallofshields]

Ok, you guys rule. I've been dying for this information. You guys have all answered my question well, but I might try to squeeze out some additional information while I'm at it. If you'd rather that I start new threads for these questions; I'll understand.

This project is specifically for my PZT rigged banjo, which actually sounds quite good for not yet having the impedance bridged and being a 'first-timers' electrical project. I've got a 3-way pickup selector and a volume knob running straight to a 1/4" jack, but I have a serious moral objection to housing a 9V battery inside the body for an impedance bridging preamp.

Which brings me to my next question: Does cable length matter as much for Piezo's?
I've read a great deal of conflicting information on other forums on this topic. Many suggest that the calculations for frequency loss (5ft max for a 1megohm source) for magnetic pickups don't apply to PZT (being like a capacitor rather than an inductor.) Of course, I imagine that after running my signal through volume and tone pots into a 1/4" out traditional electric guitar style, it might be the same.

[thehallofshields]

You'll need an output level control or pad switch if you want consistent level from this thing, piezos can be a lot louder than magnetics. Why the jfet pre? Opamps work well and are very forgiving.

I was hoping that I could use a compressor pedal as a limiter to prevent the banjo's volume levels from becoming explosive. Again, this is a first-time project, so don't be afraid to shoot down my ideas.

I've read a bit of debate about different transistors , and the jfet seemed to have the most consensus for impedance bridging. I am willing to be talked out of it, or perhaps it would be better if you could direct me to some quality information regarding preamps.

[Rick Turner]

Best to buffer as close to the source as you can. Piezo polymers are much more sensitive to cable capacitance loss than piezo ceramics, but still the losses and the potential for noise pickup makes it best to buffer in the instrument. Passive volume and tone controls really suck with piezos, too. You can always mount one or more batteries off-board and run single ended power up on the "ring" channel of a stereo cord while bringing buffered signal down. That's what I do for ukes using my D-TAR Timberline piezo system.

Also, too much shielding is never enough with piezos. Working at those high impedances, the pickups and lead out wires are incredibly sensitive to hum pickup, but if you get it all right, it's amazing how quiet piezo systems can be...much quieter than even well shielded and matched coils with humbuckers. Plus, with piezo ceramics, you can get a lot of signal voltage...too much for a lot of systems.

Once you get the basics down, piezo pickup design becomes all about geometry...the geometry of the pickup sensitivities and the vibration input be it directly from strings or from a diaphragm.

[thehallofshields]

Works fine to roll off the low end and control rumble stuff.

I've seen a pot titled 'rumble' on the Tech21 Reverb Pedal. Is it just the opposite of a tone-knob (low-cut vs high-cut)? I thought it was a badass name, but I haven't been able to specifically find the meaning of the term.

[thehallofshields]

You can always mount one or more batteries off-board and run single ended power up on the "ring" channel of a stereo cord while bringing buffered signal down. That's what I do for ukes using my D-TAR Timberline piezo system.

That is so badass. If I were to mount a small preamp circuit inside the instrument, do you think it would be possible to rig an Equalizer Pedal to
feed the preamp? Is there somewhere you can send me to reaseach this so I don't have to bother you guys with a zillion more questions about it?

[dpm]

Mr. Turner is really the expert here, my own experience is limited and the way I come to understand this stuff is through trying and listening, I don't seem to grasp the theory until I hear the effect for some reason.

a compressor pedal as a limiter to prevent the banjo's volume levels from becoming explosive

I imagine limiting is a good idea with piezo equipped banjo. What I meant re. the level control is purely an output volume to allow you to adjust to the same overall volume whatever you're plugging in, ceramic piezos putting out a lot of voltage like Rick says.

Moving on to the low cut thing, this is giving away some ideas I have for an onboard pre, but if you want to be able to mess with input impedance and basic eq predictably you'll need a dual stage design. This is because altering the impedance will also alter the shelving frequency of a given low cut capacitor if it's on the input too. Let's say you have the impedance control at one setting and your low cut is at 100Hz, if you change the impedance a little your low cut might move to 200Hz. Putting a low cut cap on the output side will mean that the frequency will vary depending on what you're plugging into next - it's input impedance will effect the shelving frequency.
So you need to run one buffer stage with input impedance control, then throw a low cut on the front of a second buffer with a known fixed input impedance. This works for me in the context of magnetic pickups. With piezos playing with impedance to reduce rumble might be fine, but with mags I want to tune the character of the pickup via the preamp input impedance, then cut the bass at a known frequency.

What really helped me was running spice models of simple preamps with frequency response plots. 5spice is free and easy to use, and there are plenty of others.

BTW, electronics gurus please feel free to correct any errors in what I've said.

[Steve Conner]

Does cable length matter as much for Piezo's?[/B]

Apologies if anyone else already addressed this above, but:

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.

[Rick Turner]

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

Please get terminology right and draw up your exact signal path in a block by block diagram and figure out just what is going to be in the instrument and what is going to be off-board.

For instance, piezo>buffer>cable>EQ? Why the EQ? Where are your various stages of preamplification? What are the source and load impedances from one stage to the next?

Don't put in more stuff than you absolutely need. I've seen too many piezo-based systems where the only reason why EQ is needed is because the basic impedance matching is screwy from the get-go. I've seen a lot of instrument to amp signal chains that have so much crap that isn't needed that you can't even hear how good the pickup may be...and that includes a lot of active magnetic pickup-equipped instruments. There's nothing quite like inappropriate electronics to screw up a decent pickup and instrument. Just look at Yamaha and Takamine. The Tak acoustic pickups are frighteningly good...if you junk the electronics and substitute a high head room buffer. But they do a great job of f...ing up a good pickup design by adding too many bells and whistles.

EQ in general is way overused. EQ in the instrument, EQ in an off-board preamp or amplifier, EQ on a PA channel, EQ on the PA master....that's four stages of EQ in a typical system. What is the purpose of each stage? Can you get rid of any of it? What are you EQing? The instrument? The PA or amplifier speakers? The room? (which is a losing game anyway...)... Are you boosting bass in one place and cutting it in another?

What you really want is pickups and instruments that sound good flat. If you need to EQ a pickup to make it sound good, then the pickup is at fault.

[dpm]

^ I couldn't agree more. Also, with active onboard EQ and pedal based stuff you've got a practical current draw limit set by the battery. Really high headroom and low noise EQ draws a lot of current so battery life would be very short. IMO you're best off with a buffer that immediately sounds good, then the least EQ possible via a nice channel strip or outboard pre, something not designed around getting 6+ months out of a 9v battery.