0
\$\begingroup\$

I'm trying to follow the schematic on this page to create a preamp/impedance matcher for a balanced piezo contact mic. I've reproduced the schematic below.

enter image description here

I've built it on a breadboard and I'm finding that the output is at least 20dB below that of the contact mic on its own, and has considerable noise. I'm finding it difficult diagnosing this, not having very in depth electrical engineering knowledge myself. I wonder if someone could suggest the likeliest causes?

My JFETs are 2N3819, matched by hand using the technique suggested on that page. My R3 is 1K, and my R1,R2,R5,R6 are 3M3 (easier to source). My capacitors are polypropylene, rated for 100V, and all resistors are carbon.

Possible problems:

  • My JFET matching went wrong. I gather that unmatched JFETs would result in noise, but would they also result in a signal drop where there should be gain?
  • The assumptions the circuit design makes about the internal architecture of my audio interface are wrong, ie. "It takes advantage of the fact that phantom power is fed to the balanced XLR inputs via pair of closely matched 6.81k resistors. These resistors are used directly as the drain resitors in the differential amplifier rather than the more usual method of connecting an additional pair of matched resistors to each line to create a +48V supply rail and then feeding the signal back through DC blocking capacitors. Since the mixing desk already blocks DC at it’s inputs it’s not a problem that the signal is floating at +45V or so."
  • Some of my component selections were wrong: an exact 3M value is critical, or the resistors should be metal film.

Many thanks for all help and suggestions.

EDIT: Here's a photo of my current breadboard setup in case there are any glaring mistakes. preamp breadboard

\$\endgroup\$
  • 1
    \$\begingroup\$ A quick check with a multimeter will tell you if the +48V is present... There is usually a switch to enable it on the audio interface / mixing desk. \$\endgroup\$ – peufeu Dec 5 '17 at 19:35
  • \$\begingroup\$ I have indeed turned +48V on. Without it I don't hear anything. \$\endgroup\$ – Igid Dec 5 '17 at 19:55
  • \$\begingroup\$ OK, did you check the voltage on tail current setting resistor R3? Also the points XLR2, XLR3 on your schematic should be 48V - half of current through R3 times soundcard output resistance which is probably 6k8. So a little bit below 48V. \$\endgroup\$ – peufeu Dec 5 '17 at 20:03
  • 1
    \$\begingroup\$ Note carbon resistors are noisy and very inaccurate. Where did you find carbon resistors? This warrants checking the voltage between Q1/Q2 gates. It should be quite small. Imbalance in the 3M resistors could cause one of the FETs to turn off and then the whole thing no longer works... \$\endgroup\$ – peufeu Dec 5 '17 at 20:06
  • \$\begingroup\$ Ok, I'm measuring ~16.3V for the XLR terminals, and 0 across R3 (Though I don't quite get why I would expect to find a voltage there. Am I understanding where you're suggesting to measure?) I did subsequently read that carbon resistors are not ideal for audio applications, and so I was intending to replace them before I soldered it together, but I didn't think they'd make it break down altogether. So R1, R2, R5, R6 should be matched to within more than 5%? \$\endgroup\$ – Igid Dec 5 '17 at 20:14
0
\$\begingroup\$

The inverting feedback on each FET, from drain-to-gate, allows a gain of TWO.

Time constant on each gate is 1.5MegOhm * 220pF = 330uS, or 500 Hertz corner in that HIGH PASS FILTER. Thus your bass tones and midrange are strongly attenuated.

\$\endgroup\$
  • 1
    \$\begingroup\$ How are these comments relevant? The circuit is designed expressly to remedy the HPF caused by impedence mismatch, and I can only conclude that there must be something wrong in your reasoning in regard to that. \$\endgroup\$ – Igid Dec 6 '17 at 13:14

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.