How to give my fast and dirty Piezo pick up more b@lls and bass?

[Skruffyhound]

I'm making a little amp/effects box to take with me on holiday. It's going great, cabinet polished, speaker attached, headphone socket working, Ruby sounding great and a Mindbender (Negative ground Tonebender) that sparked up sounding fantastic first time. Well it sounded great with my SG, when I switched back to the Martin "backpacker" with a piezo taped to the front it sounded a bit tinny and lost all those great full tones. The Ruby sounds good enough, at least through headphones, the 7 centimeter speaker is never going to be a wow, but it'll do for a bit of fun.

How do I change the characteristics of the piezo to make it less trebly. I'm thinking to try and buffer it very close to the piezo and run it into a tonestack to start with.I may also experiment with Piezo placement and using more of them. I haven't really thought this project through, I'm just giving it a go because I don't have time to get all OCD on it, and I'm having fun.

Can anyone suggest something to try? (This seems like a job for Mark Hammer if you ask me, he'd definitely have a hypothesis.)

[Digital Larry]

I think for one thing, piezos like to feed into a super high input impedance, like 10 Meg.  So, lots of time you'd see them going to an op-amp set up as non-inverting so it can go directly to the (+) input.  When you use an op-amp in an inverting configuration, your input impedance is set by the input resistor. 

And you might want to use some low pass filtering to dump some of the high end to ground.

And now we are at the edge of my ability to contribute meaningfully.

[Skruffyhound]

Thanks for the suggestions Larry, I'll try them out today

[italianguy63]

Hey, let us know what you figure out.  I am getting ready to build an electric Viola for my son.  I have the parts, and trying to figure out if I want to put the piezo on the bridge, or under the bridge inside the instrument.  (I'm leaning to inside).  I would like a way to control the tone as well-- maybe with just a cap/resistor combination.  I don't want it "buzzy."

MC

[GibsonGM]

Piezo's typically output thru a series capacitance of about 15nF.  With a low(ish) input impedance that we'd get away with using an inductive pickup, the piezo presents a HPF response, with a cutoff of just a few hundred Hz.  So yes - try a much higher input impedance.   

I'd use a buffer dedicated to this task. 

[Skruffyhound]

Good info Mike, I'll get on it this evening.

I can't promise an exhaustive experimental approach Mark since if I'm going use it at all it will be in Chiang Mai next Saturday, but I report back if anything useful turns up 

[italianguy63]

Thanks for any insight.

Chang Mai?  Nice.  I visited China once back in 2000.  What a cool trip....

[bluebunny]

I was asking a similar question a while back, Aston.  R.G.'s onboard preamp came up as a solution, along with the Tillman preamp and Jack's JFET buffer at AMZ.  Unfortunately, I've not got around to implementation yet, so I can't give an opinion, but I'll probably build all three and road-test them.

[Skruffyhound]

Hi Marc, yes, I was thinking about the Tillman and Jacks Buffer. I've looked at both of those before but didn't know about R.G.'s onboard preamp. Do you have a link? Geofex can take some navigating.

[Skruffyhound]

This thread yielded more corroboration and this interesting hypothesis:

It is a fact that the capacitance of the piezo driver forms a high pass filter with the input impedance of the following amplifier stage. If the capacitance of the piezo disk is large enough, then the input impedance of a normal recording equipment (but not a PC sound card) might be enough to keep the bass response clean. Bigger piezo disks typically have larger capacitance (because the larger area is directly related to capacitance), so if going towards this path, try to find a "big" piezo disk somewhere.

So - 1 onboard buffer (when I was building my reverb plate I noticed that the prof's buffered the piezo's very close to where they were attached.)
       2 High input impedance.
       3 Low pass filter
       4 Big Piezo.

With the big piezo I'm wondering if placing two or even three in parallel would increase the capacitance and allow more bass through. Guess I'll find out tonight if I get a move on.

[Skruffyhound]

Anderton weighs in :

http://www2.gibson.com/News-Lifestyle/Features/en-us/Make-Acoustic-Guitar-Piezo-Pickups-Sound-Great.aspx

[Skruffyhound]

Found R.G.'s onboard preamp thanks

[Skruffyhound]

I think I'm going with the Tillman with a 10M resistor as R1
Now to check out low pass filters ...

[GibsonGM]

Nice...step-by-step, Aston...you may find the high input Z is enough, or might then need to tweak more with the filter.

Anyway, it shouldn't be hard to get that sound back to where you want it

[jatalahd]

With the big piezo I'm wondering if placing two or even three in parallel would increase the capacitance and allow more bass through.

This was good reasoning from you! Actually I have done a bit more reading lately related to piezos and this thesis work contains some nice information:

http://tdx.cat/bitstream/handle/10803/6285/THESIS.pdf?sequence=1

At appendix page A1.8 (page 246 of 277) it clearly says that:

"If the required input impedance is very high, there are two ways to prevent the cut off frequency
from rising:
1. use a parallel capacitor (C p ) externally coupled to the piezoelectric sensor.
2. use a stack of discs in parallel (stack), and thus the total capacitance will be higher and
lower the required input impedance.The capacitance of a parallel stack increase
proportionally to the number of stacked discs..."

So yeah, there are tweaks to get larger capacitance for a piezo sensor. However, I have not done this in practise myself, so this is just what the theory predicts.

[Skruffyhound]

This was good reasoning from you!

Thank you jatalahd, I thought it might be worth a try. I can say that two was not enough of a "stack" even though they were much bigger than the first one I was using. The Tillman wasn't enough either.
So ...I'm left with the low pass filter and now your suggestion of the cap in parallel (thanks for that). I have selected a very simple Bandaxall tone control built point to point on a 500k dual pot as my first attempt. Other suggestions for lowpass filters gratefully received. I'm not really sure where to set the frequency for a guitar application.

Time is running out for me though as flight time approaches
It's been really enjoyable to do a bit of soldering again after about 2 years out learning to play guitar properly. I'm pretty pleased to have three circuits that I "perfed" on the fly from schematics that all fired up immediately.
Albeit simple circuits.

[DrAlx]

With the big piezo I'm wondering if placing two or even three in parallel would increase the capacitance and allow more bass through.

...  " use a stack of discs in parallel (stack), and thus the total capacitance will be higher and
lower the required input impedance.The capacitance of a parallel stack increase
proportionally to the number of stacked discs..."

A few years ago I did some experiments with wiring 2 piezos in parallel.  My thinking went along these lines...
If I wire the two piezos in parallel, then I sort of have one large piezo with double the area, so the capacitance will double, but the signal should get larger too because more vibrational energy is being picked up.

What I found though was things didn't sound louder when I put things in parallel and if anything they got quieter.
I came to the conclusion that because the piezos were put at different locations on the guitar body they were subjected to different vibrations.
In an extreme case, you could imagine having just one piezo stuck to the guitar body and the other one that is in parallel to it being left dangling in the air so it picks up nothing at all.
The piezo that is just dangling in the air will take whatever voltage is produced by the other piezo and convert it into its own vibration (i.e. it will act like a little speaker that sucks the signal).
I haven't looked through the link you describe but I suspect that gluing the piezos firmly together in a stack stops that situation from happening 'cos all piezos sense the same vibration.

In the end I decided to keep my two piezos separate under the treble and bass side of the bridge which meant I could record things in stereo.

[Skruffyhound]

I like the info that is coming up in this thread.
Can anyone tell me what cutoff frequency would be good for guitar so I can calculate R and C values.
I breadboarded the Bandaxall last night but was not impressed, I'll check for mistakes and have another go today.

[jatalahd]

EDIT: removed my answer because was not sure what I was answering to 

[ashcat_lt]

I like the info that is coming up in this thread.
Can anyone tell me what cutoff frequency would be good for guitar so I can calculate R and C values.
I breadboarded the Bandaxall last night but was not impressed, I'll check for mistakes and have another go today.

You're looking for appropriate cutoff for the LPF?  I'd start from the idea that most magnetic pickups cutoff around 5KHz give or take a K or two.