Preamp suggestions for piezo pickups?

[canman]

Hey gang,

Been a LONG time since I posted or built anything pedal related.  I've been building more instruments lately...lots of solid body ukuleles as of late.  I've been using rod piezo pickups under the saddle and they sound great through my Crate acoustic amp.

I've got a couple ukes heading up to Utah and I'd like to include a preamp to allow the user to plug the uke into any kind of computer type speaker...line level, right?  So anything from standard computer speakers to soundbars (like Bose, etc).

I did a bit of reading today and found that piezo pickups are super high impedance pickups.  I'm not good at figuring out impedance and I have always struggled to understand it.  So, I'm not sure what impact impedance will have on the device I want to build.

I've seen schematics for the MXR headphone amp and that seems like it would be a decent way to go.  I read that cabinet simulations are helpful with the headphone amp.  Would this be a decent solution, or would something else make for a better solution?

[FiveseveN]

I've had good results with the simple Tillman preamp. If you don't have FETs on hand, use an (FET input) op amp.
As for cab simulation, keep in mind that acoustic amps are wide-bandwidth. If you feel you need to, just incorporate filtering in the preamp.

[samhay]

The usual advice for a piezo preamp is (very) high input impedance.
However, lower input impedance will roll off bass from a piezo - the opposite effect to a typical guitar pickup. This may not be as much of a problem for a ukulele, particularly if the preamp is built into the instrument.

Personally, I like charge amplifiers for piezo preamps, but the Tillman preamp is not a bad choice and is easy enough to build for testing purposes.

[canman]

Nice, I'll check out the Tillman preamp.  Haven't heard of it before, thanks for the suggestions.

Regarding the bass rolloff...yeah, not a huge issue for a ukulele.  Having a little extra bass would be nice for a fuller sound, but not critical, especially since I just want it to work with computer speakers for convenience.

[Digital Larry]

I read that cabinet simulations are helpful with the headphone amp.

Cabinet simulators are most useful for guitar amps which generate distortion, as a regular guitar speaker naturally rolls off a lot of the high frequencies.  It's probably overkill for what you're doing.  If you find yourself needing to adjust the high frequencies down I'd suggest either a fixed RC low pass, adjust to taste, or a two-band tone control if you think you might want to change it from time to time.

[canman]

Thanks for that information...I may implement those filters just to have a little more control, but first, some debugging.  The circuit works, I built it off of IvIarks' layout.  The problem is that I'm getting a lot of hum.  I first tried it through computer speakers and it worked, I can hear the uke.  BUT, the hum is very loud.  So then I tried it through an acoustic amp and still got hum, so I know it isn't the speakers.

I don't have any 3M resisters, I used a 2.2M resistor instead.  I suspect that may have changed the input impedance a bit, but would that have caused a lot of hum?  The next size up I have is 4.7M, should I drop that in the 3M (2.2M, in this case) spot?

Here's a link to the layout, I don't remember what is and isn't kosher here so I don't want to post pics just in case: http://4.bp.blogspot.com/-W0Xq9uuEy4A/U5Jch8L7Y3I/AAAAAAAAHsM/FHdS0KRV6Ko/s1600/A+Pair+o%27+Preamps.png

[anotherjim]

You have a very exposed circuit due to it's high input impedance. The pickup and the board will need screening. If the pickup has a screened lead already, then that should be ok. Assume output cable to whatever amplifier will be screened too?
Are you powering from 9v battery?

2M2 resistor is probably ok - for now. Do you know of a spec for the pickup capacitance? If you know that, we can calculate the lowest input resistor. Problem if resistor is too low will be lack of bass. Could be a good thing to lose a bit of bass that way - reduce handling noise rumbles on an instrument that doesn't produce strong bass anyway.

As you like how the pickup works with your acoustic amp, see what it's manual says for the input impedance in the specifications list.

On stripboard, make sure you have no accidental shorts between the coppers. The track cut between fet gate and output cap is vital. The Tagboard layout looks good to me.



FET amplifiers don't always bias perfectly due to variations in the individual FET's. It's helpful to measure the the DC volts between the Drain and 0v to check if it's biased ok with those resistor values. If drain volts is close to +supply or 0v, then you will have trouble getting a good clean signal through. The gate input resistors value has no effect on the bias, so you needn't worry about changing that if you have to.

[canman]

I'll have to go over everything in a few hours when it cools down a bit more and check for any shorts in the board...you all know how hot it can get with a soldering iron!  But, I do have a couple of things to note in reply to anotherjim:

First, the pickup lead is shielded.  The ground braid completely wraps around the hot lead, so I'd assume no issues there.  Regarding the capacitance of this pickup, I do not know.  If it helps, it was a $5 piezo specifically for uke, off Amazon. 

Second, I don't know if the output cable to the amp is shielded.  It's one of those dumb cables you can't twist the end down to make any repairs, etc, so I don't know how shielded it is.  It's a pretty hefty cable, so I'm inclined to think it's fairly shielded.

Third, I've been powering from a OneSpot power supply.  I'll try a battery later and see if that helps.  If it does, I assume adding some power filtering to the board will help with power supplies.

Fourth, the amp is a Crate CA15 acoustic amp.  The manual states the input impedance is 1.5M ohms. 

Do any of these things stand out to anyone as problematic?  As I said, I'll go over the board this evening and check for shorts, copper burrs, etc., and I'll check the bias voltage.  Maybe it would be helpful to box it up as well, just to see if it eliminates the hum.

EDIT: I just realized you said screening, not shielding.  Hopefully they are synonymous.

[anotherjim]

If it helps, it was a $5 piezo specifically for uke, off Amazon.

Then we don't know, unless you can ask the seller for the info.

Fourth, the amp is a Crate CA15 acoustic amp.  The manual states the input impedance is 1.5M ohms. 

Then if it's ok with 1M5 it can only be better with 2M2 and we don't need to know the pickup capacitance.

Third, I've been powering from a OneSpot power supply.

Ordinarily I'd expect that to be just fine. The preamp already as a supply capacitor +9v to 0v, but it's performance can be improved with a resistor in series with the +9v going in - say 100R to 470R. The circuit take very little current so the extra resistor won't hurt.

I just realized you said screening, not shielding.  Hopefully they are synonymous.

Yep, same thing.

[roseblood11]

Shield the whole cavity inside the ukulele!

[deadastronaut]

can man,

ive added 'piezo rod' pickups into 2 classical guitars lately..and sound fantastic.

one for me and a friend, you can get the whole /pup/tuner/EQ graphic/and xlr/jack/ out kit
for around 5-6 Pounds gbp on ebay....nice little units. slot a 9v-pp3 in and off you go.

they are great.  build them into your uke's...

or butcher it to become a pedal..for the money its worth giving it a go.

http://www.ebay.co.uk/itm/Neewer-Acoustic-5-Band-EQ-Equalizer-Guitar-Preamp-Piezo-Pickup-Tuner-/192168073393?epid=1060601276&hash=item2cbe1c18b1:g:3dIAAOSwSypY~wLJ

[canman]

Right, so I did some testing.  All the traces look clean, no burrs or shorts.

I get these voltages on the transistor:

D: 2.8v
S: 2.8v
G: 0v

I suspect I need to change a couple resistors to get the bias up to the proper voltage.  2.8v seems too low, correct?  We should be around 4.5v right?  I'm not sure how the bias affects the amount of hum, but 2.8v definitely seems wrong.

I had considered buying one of those fabricated preamps but thought I'd try to build one first.  Will those allow you to plug into line level speakers?

EDIT: OK, so with a 1k resistor in place of the 2.2k resistor, the bias is at 4.2v and hum is gone through my acoustic amp.  Sounds fine...a little high pitched whine but I'm not worried about that right now.  The problem is still that when I play through computer speakers, the hum is there and now the uke hardly even comes through.  SO...I'm thinking the Tillman preamp isn't going to work.  Which is a bummer cuz it's so small and easy.

But this one looks like it should do the trick: http://tagboardeffects.blogspot.com/2017/05/barcus-berry-3000a-buffer-preamp-eq.html#comment-form

[blackieNYC]

The Tillman is about 3dB of gain.

[canman]

Does that mean I'll need an amp after the preamp?  I thought the powered computer speakers would work as an amp.  I'm not sure how to interpret "3db of gain."

[Eddododo]

Not a direct solution, but input buffer could start with this.
http://personalpages.tds.net/~fdeck/bass/quickand.pdf

[anotherjim]

Welcome to JFET hell.

Actually I built that amp a while ago for use with an acoustic bass guitar that had a bridge piezo but no onboard pre-amp. It worked fine driving a mixer line input which are a bit harder to drive than "consumer" line inputs such as in powered computer speakers.

Although this kind of amplifier is ideally biased halfway for maximum signal swing, it isn't always best or necessary to go for that. Note the Tillman article suggest +6v drain voltage. If the piezo produces a 1v max swing and gain is x3, then if bias gets the drain within about 3v of either +9v or 0v it will never clip.  A popular tactic is to use a variable resistance for the drain (R3) and adjust for the best sound. Take the variable out and measure with the DMM and fit a fixed resistor closest to the measured value - or fit the variable resistor (a pcb mount trimmer type) permanently.

Some things to raise the basic gain.

Increasing the drain resistance. You can find designs for 9v FET amps with that resistor around 15k or higher. Down side of making it higher is the output signal level can be loaded down by the input it is driving, negating the improvement. There is  interaction with bias since you will change the current in the source resistor.

The source resistor can be bypassed by a capacitor. Probably around 22uF. This resistor is needed to raise the source voltage relative to the gate voltage and is the primary DC bias element but it also reduces gain. A source bypass capacitor negates the gain reducing effect on the signal (which is AC) without changing DC bias.
Note that the smaller you make the source resistor, then the gain boost to be had is reduced, so not always worth doing. The bypass also increases distortion by removing the negative feedback effect. That distortion can sound nice on electric guitar, maybe not with a Uke.

The output has a 51k resistor to 0v. This "pull-down" resistor reduces connection noise when you plug it into something by keeping the negative plate of the output capacitor at 0v instead of floating around. It is adding some loading effect on the output in addition to the load of whatever it is driving. You can raise the pull down resistor value by a factor of 10 if you want. The increase in output may only be slight, depending on the input impedance of what it drives.

Note the tag layout for that Barcus Berry pre-amp has the green input wire in the wrong place. It should feed IC1 pin 3.

[canman]

Hmm...JFET hell sucks.  I had no idea I wasn't getting into something so complicated.  The circuit looks so easy!  All this information is really interesting though.  Seems like finding the perfect balance can be a HUGE pain, especially on some of the more complicated circuits.

I'm still debating whether or not to track down the hum with this thing through computer speakers or to just bag the idea and move on to something else.  Everything I read says that a preamp will bring an instrument level up to line level, so maybe I just need to dig a little deeper. 

[anotherjim]

Did you try boxing it up?

[Astronaurt]

Not sure if it's exactly what you're after, but I'm just about done with a project doing a complete overhaul of the Piezo Pre-amp for a friend's electro-acoustic guitar. My initial drawing was something like this:



Though I wound up changing the Active Baxandall Tone Stack to be closer to the one posted here:
http://www.learnabout-electronics.org/Amplifiers/amplifiers42.php

Filled in and changed some component values after testing it all out on a breadboard, used a J201 for the input Buffer JFET and a LM4562 Dual Op-Amp,
With plenty of adjustments to the Tone Caps and Tone Pots to get the Bass and Treble controls responding how I wanted.

It's designed more to drive a cable to a guitar amp than to drive a speaker, but if you have any kind of line input on your computer speakers then this should be more than sufficient to plug it in. Though there's some white noise from a Leaky FET and general resistor noise going on, it's well below what you'd normally expect from a single coil pick-up and definitely not what you'd call "hum."

[canman]

Sorry for my absence, I've been busy and it has been a humid summer thus far...tends to suck the motivation right out of you!

I have not tried boxing it up.  I probably should, just to see, but I've more or less given up on the tiny preamp.  It works through a regular amp, with no hum, so I'm wondering if maybe my powered speakers have something else going on that I'm not aware of.  Kinda doubt it, but you never know.  I did plug in a Noisy Cricket into the speakers and that worked fairly well...a bit of hum, but not bad. 

Thanks for posting that schematic Astronaurt, I'll take a look at it and give it a try!  It's been a while since I studied a schematic so it might take me a while.  Gotta dust off the cobwebs!