3
\$\begingroup\$

I'm currently working on an experimental string musical instrument. It isn't my invention though. The source of inspiration called Dvina has an unusual pickup system. It senses the current induced in a string itself vibrating in a magnetic field, almost identical to a ribbon microphone. To couple the string to an amplifier an impedance matching transformer is used (as well as in ribbon mics). Can I avoid designing a transformer and use an active preamplifier instead? Here is my naive take on its schematic

Schematic

The resistance of a string is a fraction of Ohm. It's decoupled and weakly tied to the ground via high value resistors to suppress capacitively coupled external interference. What else I have missed? What drawbacks does it have in comparison to a transformer?

\$\endgroup\$
4
  • 1
    \$\begingroup\$ It might be worth looking into a transimpedance amplifier circuit. \$\endgroup\$
    – Theodore
    May 17 at 21:37
  • 1
    \$\begingroup\$ In §8.5.9 of The Art of Electronics (3rd Edition), they tried to design an amplifier for a ribbon microphone. The problem was that most transistors and opamps have too much noise voltage. The amplifier needed a voltage noise below 100 pV / √Hz. They used 16 power transistors like the ZTX851 in parallel. \$\endgroup\$
    – jy3u4ocy
    May 21 at 7:07
  • \$\begingroup\$ Can you estimate the maximum generated voltage here? This would narrow down the parameters for a search. \$\endgroup\$
    – Jens
    May 21 at 17:30
  • \$\begingroup\$ @jy3u4ocy I now understand why even active mics have a transformer \$\endgroup\$
    – e_asphyx
    May 22 at 13:38

3 Answers 3

3
\$\begingroup\$

Can I avoid designing a transformer and use an active preamplifier instead?

If you use a transformer, even though they may be undesirable because of their size, you will get better signal to noise ratio on the output compared to using an op-amp voltage amplifier. Of course that doesn't mean you won't get acceptable results for your instrument but, if signal levels are inherently low (or you require a large dynamic range) you might find the op-amp circuit will not work as good as a transformer.

\$\endgroup\$
2
  • \$\begingroup\$ Yes, you're right, I haven't thought about a noise. Transformers just freak me out :) \$\endgroup\$
    – e_asphyx
    May 17 at 11:15
  • 1
    \$\begingroup\$ @e_asphyx maybe you need to raise a question on transformers so that you can become less-freaked-out by them. \$\endgroup\$
    – Andy aka
    May 17 at 11:34
3
\$\begingroup\$

All amplifiers have a level of noise, and an ideal 'noise resistance'.

With a very low voltage, low resistance source, you are best paired to an opamp with a low noise resistance, for best SNR.

A hack used by a colleague of mine to get super-low noise voltage and resistance was to use an LM394 'super-match' transistor pair. This IC is actually designed for exquisite temperature tracking between the two devices. It does this by having thousands of transistor die, connected in a chequerboard fashion so that any thermal gradients are averaged out. However, this also happens to result in transistors that are actually several thousand transistors in parallel, giving them very low voltage noise, at the expense of high current noise - a low noise resistance. Get some initial gain with one of these in front of a more conventional opamp, if you are troubled for SNR.

\$\endgroup\$
1
\$\begingroup\$

You might pick up a lot of HF energy from mobiles with this nice antenna. Add 2 capacitors (1nF) to GND before the energy can reach the opamp. There are ugly HF rectification effects in some an opamp input stages. Ferrit beads in series with the input will also help. Yes, noise will be a problem here, as well as finding a matching transformer.

\$\endgroup\$

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.