Guitar Piezo / Mic Blend tips wanted

[Ufdah]

A about two years ago I put a piezo element in my acoustic guitar and built a preamp for it using a LT1012 Op-Amp. It has sounded decent but I really think blending in an electret condenser mic would really add to the sound. As I see it I have three options: (or four if you have a beter one...)

1) Keep the piezo circuit the same and add a passive blend circuit w/pot and mic.

2) The same as 1) but with an active blend / buffer amp.

3) I could design a new high impedance blend circuit (greater than 1M ohm, because of the piezo, although ideally there'd be around 10K resistance on the mic...); then run the combined signal into my existing preamp.

If ya'll audio gurus could give me some opinions and and direction on this project I'd appreciate it. The Mic is a panasonic wm-61a electret, and the piezo disc is just a little 20mm disc I had laying around. I'd love to keep the power cunsumption as low as possible as well as running off of only one 9V. I have two holes is the guitar for rotary controls, I'd like one to be blend, and the other master volume.

I've been lurking around here for a while and have gained A LOT of good information from everybody here, but I couldn't find anything in the archives that was able to help me make these decisions. So, here's my first post! (a few additional keywords for future searchers: preamp summing combine schematic potentiometer pot volume )

 

[Scodiddly]

I've found that with most piezos, including those disk elements, a higher impedance helps a lot. My own cheapass little circuit can be found here.

You might check the microphone Meta for stuff on the little Panasonic capsules. I've used them a fair amount and they're pretty easy to work with. Somewhat noisy, but inside a guitar should be loud enough to make that a non-issue at least live.

 

[mike_relay]

I played a gig the other day with my upright and my microphone broke, so I had to use my pickup, which I can't stand. The sound guy used the Behringer phantom powered DI.. and I have to say, it sounded much better than I've heard it before. I keep on meaning to try Scott's circuit myself before I buy a more expensive direct box.

-mike

 

[Ufdah]

Thanks for the replies guys. I looked and I have a handful of the N5457 JFET's so I think I'll have to give your circuit a try Scott. So if you were me would you buffer the piezo then blend them together and run it through another one of your circuits with a little gain to offset the loss in the sum. Also how would you blend them together?

 

[Scodiddly]

I'd build each one as a separate circuit and use a regular audio mixer, at least while developing the idea.

 

[paddy]

hi ufdah,

*lots* of info here re mics:

http://www.harmony-central.com/Guitar/Faqs/pickups/acoustic.condenser.mic.txt

I always found both internal mics and piezos to be a poor compromise on sound. Try getting one of the more modern pickups that are sensitive to vibration instead of pressure. The lr baggs element active is excellent, keep an eye out on ebay, I got my last one $80 shipped. others to sonsider are the highlander and bband varieties. Of course if you like the james taylor twang stick to the piezo

 

[paddy]

have a look at the linkwitz mod too:

http://www.linkwitzlab.com/sys_test.htm#Mic

 

[Ufdah]

Thanks paddy, I've actually already read both those pages multiple times and have performed the linkwitz mod on my electret. The thing that I'm having a hard time figuring out is the best way to blend the piezo / mic...

 

[Scodiddly]

[quote author="Ufdah"]Thanks paddy, I've actually already read both those pages multiple times and have performed the linkwitz mod on my electret. The thing that I'm having a hard time figuring out is the best way to blend the piezo / mic...[/quote]

Well, you could look at some of the "blender" products made by Fishman, Rane, etc. for features. A polarity switch on one/both of the sources is definitely a good thing. I myself would have a low-cut switch on the microphone side to stop low-end rumble through the mic - maybe as high as 300-400Hz. A sweepable low-cut isn't very hard to implement, either.

 

[Ufdah]

I guess I should have mentioned this at the beginning: The guitar in question is a $300 Samik and the goal is to see how good I can get it to sound without really spending any money... I know that there are pre built units that do what I want but that just goes against everything in me, I'd much rather make something comperable...

 

[Ufdah]

So what do you think about this?
The FET's are 2N5457's

I'm using an Orban Pan Circuit to blend the Mic and Piezo then going into another 2N5457 to give the signal the last little kick before it heads out the door to the volume control (not pictured, probably a 500K Audio pot). I'm also using the Orban circuit to set the output impedance of the Mic and Piezo circuits. Each varies from 25K to 31K at either end of the pan.

I think I'll also mess around with different value caps for the 2.2uF on the mic to block the low end.

If you have any input as to how I could improve this I'm all ears!

 

[bcarso]

Way smaller resistors in the sources of the FETs would be a good start, and then bias up the gates above ground until you get some voltage at the sources to work with. With 5457's you will have the source positive w.r.t. the gates by a volt or so at a decent but still ~low-current operating point.

You might want to a.c. couple to the second FET so as to bias the gate up a bit.

The piezo doesn't need a coupling cap otoh.

The symbol for a N channel JFET is an arrow pointing at the flat section---as drawn those look like MOSFETs.

You could also consider getting gain out of the FETs by tying the sources to ground, or to ground through a bypassed source resistor, and taking the signals off of a drain resistor pullup to +9. Maybe your voltages are large enough already that this wouldn't make sense.