3
\$\begingroup\$

My preamp in my acoustic guitar died after I put the battery in backwards. I need to replace it, and fast. Piezos have incredibly high output impedance (5+ mohm) so a buffer circuit is used to knock the impedance down to a sane level.

I tried building this circuit and while it works, it lacks any sort of bass amplification:

Preamp from Klon

I didnt have a 100k resistor, so r3 is 170k. I also didnt have a .1uf, so c1 is 4.7uf. I tried raising the r2 to 5mohm and it sounded the same. v+ is 9v and vb+ is 4.5v from a voltage divider. OpAmp is tl072

Any help or alteratives much appreciated!

EDIT

All caps are thin film, not electrolytic

The DI Box next in the chain has an input impedance of 1mohm

\$\endgroup\$
  • 1
    \$\begingroup\$ Is C1 now electrolytic? If so, you need to show the piezo and biasing. \$\endgroup\$ – Peter Smith Apr 22 '16 at 7:37
  • \$\begingroup\$ Googling for 100nF caps finds lots of sellers, with offers starting at $2 for 100 pcs with free shipping. Any sort of old electronics is also likely to have several such caps. Are you sure you can't get one? \$\endgroup\$ – Dmitry Grigoryev Apr 22 '16 at 7:42
  • \$\begingroup\$ Its what I had on hand. But doesnt raising the value of C1 lower the high pass filter on the input \$\endgroup\$ – Jonathan S. Fisher Apr 22 '16 at 7:44
  • \$\begingroup\$ @exabrial It does. I had an electrolytic cap in mind when writing my comment. \$\endgroup\$ – Dmitry Grigoryev Apr 22 '16 at 7:47
  • \$\begingroup\$ Which op-amp are you using? \$\endgroup\$ – Steve G Apr 22 '16 at 8:30
3
\$\begingroup\$

The 1 Mohm bias resistor is far too low - it will load the bass signals and attenuate them much more than mid range or the treble. This happens because the piezo has capacitance and this capacitance is in series with the signal being produced. It might only be a few hundred pF and, for instance, 1nF and an external 1 Mohm bias resistor will form a high pass filter with 3dB cut-off at 159 Hz. Too high for a guitar by about an octave (low E is 82.4 Hz for example).

Try replacing the 1 Mohm with 10 Mohm and see if you get an improvement. Here's a circuit I found that uses a 10 Mohm bias resistor and so it's clear to me that that is your problem: -

enter image description here

You can carry on using the op-amp but note the use of the 10 Mohm resistor for biasing. Also note that C1 is not required. Here's another guitar piezo JFET amp that uses a 10 Mohm (R2) for biasing: -

enter image description here

\$\endgroup\$
  • \$\begingroup\$ Thanks... I'll go pick up some 10mohm resistors right now. How high can I go and what problems will I encounter as I go higher? \$\endgroup\$ – Jonathan S. Fisher Apr 22 '16 at 19:23
  • \$\begingroup\$ 10M should be fine but if you need to go higher you might nhave to change the op-amp to one that has much higher input resistance although I believe the TL072 is reasonable. I've some some designs as high as 100 Mohm! \$\endgroup\$ – Andy aka Apr 22 '16 at 19:29
  • \$\begingroup\$ it definitely helped, however, now im driving way to much signal into the chain. Is it possible to reduce the gain without affecting the input impedance \$\endgroup\$ – Jonathan S. Fisher Apr 22 '16 at 21:25
  • \$\begingroup\$ Use a pot divider after the first amp or load up the 10 Meg resistor with maybe 1 to 10 nF in parallel. \$\endgroup\$ – Andy aka Apr 23 '16 at 9:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.