acoustic preamp needs BOOST...help!

[Jimbo78]

Hi all, new here...

I recently built a preamp/buffer for my piezo pickup. Here's the schematic I followed:

http://analogguru.an.ohost.de/193/sc...rcus_3000A.gif

It sounds great tone-wise, the eq and volume controls work fine.
Just one problem--it attenuates the signal!

I have to have both the preamp and my PA channel maxed out to get a decent sound. My bandmates with onboard preamps (acoustic guitars as well) are popping the speakers at 5/10, I've tried different channels on the PA, my guitar amp too--same story. Great sound, no guts.

Could one of you electronic gurus take a lok at the schematic and suggest something I could change/add to boost the output a bit? I've put a seperate preamp in-line to the PA, but that's kinda clumsy and muddies up the tone....

thanks for your expertise!

[Steve Conner]

There's no value given for R4. Did you install R4? What value did you use?

Do you have anything between the piezo and the preamp? A long cable, a volume control?

Also, this preamp only has a gain of 1 at full volume. It's more of a buffer with tone controls, than an amp. You'll probably find that it makes the signal no louder than the piezo by itself, it just adds bass.

Depending on your pickup, a gain of 1 may not be enough.

[J M Fahey]

Agree and add: yes, it has unity gain, yet that's normal in Piezo World, because they have very high output, into a high impedance, so the problem must be something else.
1) Check that the preamp is properly biased, you should have around 4.5V DC on pins 1 and 7.
2) recheck all component values, specially capacitors. An easy to make error, specially when breadboarding, would be to switch C6 and C8 positions.
Everything would work fine, you'd have excellent sound , lows, highs, everything, but the piezo output would be brutally padded.
3) Although *I know* it can work as-is, I'd put a 10uF capacitor between Pin1 and the union of R1/R7.
In theory it should be a bipolar one, but in practice a regular electrolytic will do, pointing either way.
4) Your piezo itself may be shorted or whatever.
Connect it straight, no preamp, into the input of any guitar amp, which will probably be 1M impedance.
The typical mixer line in will probably be 50K (even 22K) and have 1/20 the sensitivity of a Mic input, so it's not advisable.
5) If you have a multimeter with a sensitive AC scale (200 mV), your piezo straight into it should provide over 100mV while strumming open chords.
If you get that, connect the preamp, check that it does not kill that output, then check that you still have it on Pin 1 and Pin 7; then max. volume and find it on the output jack.
In a nutshell, chech that you have signal to begin with, and that you do not lose it along the way.
Good luck.

[Jimbo78]

Thanks guys,

I'll check out the things you've noted and post results. The piezo is slightly louder directly into amp, but sounds, well, like an unbuffered piezo...ugly!

r4 appears to have be absent in the original device that the schematic was taken from, maybe a variant for 12 string or bass or something. I had a look at some other buffer/preamps and tried 10M resistor. Made no difference. Cable is 2 metres from piezo to buffer, and just a 20cm link to amp. When I gig I use a 6 foot cable into PA (6.5mm, unbalanced input). I do get useable volume out of it, but I don't like being maxed out with nowhere to go when it gets noisy in the restaurant.

Will decoupling the two amp stages do anything for the gain/tone? Or is it simply a precaution/best practice?

Assuming it all checks out, and simply the piezo is just too weak with unity gain, is there anything simple I could do to increase the gain...like wiring in a little FET or transistor stage at some point in the circuit? I'm guessing messing with a circuit that's already optimized isn't wise.

Otherwise I might try adding a second piezo transducer (in series?) to boost the input.

will keep you posted, thanks a lot.

[Jimbo78]

I may have identified at least one problem...c8, is supposed to be 100n, but the little mtk box has 124k100 on the top...should be 102k100, shouldn't it?

[J M Fahey]

1)

decoupling the two amp stages

DC coupling , as is now, will let any offset from IC1 be amplified by IC2, which in itself isn't terrible, *but* tone controls now act as DC gain ones.
Any DC across controls ends up causing scratchiness. Adding the cap I suggested minimizes that big way.
Yes, it's "good practice" doing that.
2) You tested your piezo straight into what kind of input? You don't say.
Guitar amp or PA/mixer?
3) To have 10x gain (much more than what you need) pull the link from pin 1 to pin 2, replace it with a 100K resistor which also has a 100pF in parallel, and wire pin 2 to ground, through a 10K resistor and a 1uF capacitor in series.
Negative leg to ground, positive to the 10K resistor.
If *too* much gain, (I guess so), raise those 10K to 22K, 47K or 100K.
Post your results.
EDIT:

I may have identified at least one problem...c8, is supposed to be 100n, but the little mtk box has 124k100 on the top...should be 102k100, shouldn't it?

No, it's fine. *Should* be 104 something, yours is slightly higher, no problem.
In fact I would use a 1uF or 10uF electrolytic there.

[Jimbo78]

Hi JM and Steve,

perfect 10! Added the cap and resistor across pins 1 and 2 and grounded through 10k and 1uF. It's on strip board already, so there was no room for decoupling cap before tone controls, but Ill make another, smaller version on perfboard including that mod as well.

Results: amp now powers up nicely arounf the 3/10 mark, so plently of headroom there. The powered mixer (old TOA 601) is good at halfway on both channel and main. Still can't make the input trim light flash, so it's not too powerful...but 100% better.

Upshot: when I hit the footswitch from rhythm to lead, I'll be able to cut through.

Thanks heaps guys...this forum is excellent! Hope one day I'll be able to help out another member.

By the way, JM, what's the basic theory behind the gain mods you suggested? Is there a rule of thumb you're going by?

Regards, Jim F

[J M Fahey]

Well, not "rule of thumb" exactly.
If you are interested, google "operational amplifier", you'll find different levels of explanations, needn't get into NASA stuff here, in any general purpose explanation you'll find that op amps are *very* useful "building blocks", to build zillions of things.
In this case, it's a "non inverting amplifier" and its gain is "A= (R1/R2)+1" in this case being:
A= amplification factor or gain
R1 is the one from pin 1 to pin 2 , going from output to "negative/inverting" input
R2 is the one from the inverting input to ground.
So, here we have: A=(100K/10K)+1=11 so you'll have on pin 1 11x the signal received on pin 3.
I *think* you piezo is weak somehow, because now not from theory but from experience 11x gain is seldom or never needed, 1x is the de facto standard, but, oh well, if it works don't fix it.
And what are the capacitors for? They do not appear in the formula (in this basic case).
Well, we live in the *real* world, which by definition is not perfect.
1) your wiring can catch some of the heavy interference that surrounds us everywhere, unless you live in an isolated house in de middle of nowhere, so those 100pF reduce the stage gain above 16000 Hz, where there's not much useful audio.
2) That op amp will have a small "offset voltage" ; you don't want it amplified 11x, that 10uF capacitor in series with the 10k prevents that.
For the same reason, the tone control will benefit too from its own capacitor in series, as suggested.