1
\$\begingroup\$

I need to amplify the microphone output to some level about 3-4 Vpp without a DC offset. The output voltage will only be sensed (it will pass through a peak detector). So I assume I don't need any current supplying ability.

I first thought of using a common emitter with around 400 gain, but there is also an option of using a non-inverting or inverting (I don't care phase shift) opAmp circuit with the required gain.

I wonder are there benefits of using one over the another, in which aspects do they differ?

Or, are both these two options a bad idea?

\$\endgroup\$
8
  • 1
    \$\begingroup\$ do you want it compressed? \$\endgroup\$ Commented Dec 19, 2019 at 1:51
  • \$\begingroup\$ archive.org/details/… \$\endgroup\$ Commented Dec 19, 2019 at 2:56
  • 3
    \$\begingroup\$ I haven't looked over the links here, but the general advice I recall from analysis done a decade ago is to use a JFET with an electret at the first stage. The load is well matched for noise -- better than BJT (or MOSFET, which is not even in the running.) Once the pre-amplification is done, the rest is lots easier. \$\endgroup\$
    – jonk
    Commented Dec 19, 2019 at 6:08
  • \$\begingroup\$ @TonyStewartSunnyskyguyEE75 your link direct me to a list of electronics books \$\endgroup\$
    – muyustan
    Commented Dec 19, 2019 at 7:24
  • 1
    \$\begingroup\$ Precision rectifiers with gain are common. But where are your specs for decay time? \$\endgroup\$ Commented Dec 19, 2019 at 7:32

1 Answer 1

3
\$\begingroup\$

To get a gain of 400 with a Common Emitter (bipolar transistor) circuit you will probably need 2 stages and a dozen or so passive components. This is what I used in the last device I built that amplified and peak-detected a microphone input. But getting the component values right was a pain, and wiring it was no fun either.

An op amp can do the same job with fewer components, and is easier to 'tune'. But you need to choose one that works on the voltage you have available. If the power supply is 5V then to get 4Vpp you will probably need an op amp with 'rail-to-rail' output.

Other op amp specs depend on the circuit used and the required performance. If it is operated with a DC gain of 400 then you want a low input offset voltage (<1mV for 0.4V output offset), and if 'without a DC offset' means only amplifying the positive half of the waveform then the inputs will need to work at or slightly below ground. Gain Bandwidth Product should be high enough to amplify the highest frequency you are interested in (eg. at least 4MHz for 52dB at 10kHz). A dual op amp would give you an extra one for the peak detector. The MCP6022 is an example that may be suitable.

\$\endgroup\$
3
  • 1
    \$\begingroup\$ hmm, thanks. Any other specs I should have be looking while choosing the opAmp other than output voltage swing ability? \$\endgroup\$
    – muyustan
    Commented Dec 19, 2019 at 7:18
  • 1
    \$\begingroup\$ Bruce's opamp recommendation also tends to have low output impedance that might be required to drive a simple diode-capacitor peak detector. You do need current to accommodate fast dV/dt charging of peak-detector capacitance. \$\endgroup\$
    – glen_geek
    Commented Dec 19, 2019 at 15:38
  • 1
    \$\begingroup\$ @muyustan I have added some more op amp specs and an example device that should be suitable for supply voltage <=5.5V. \$\endgroup\$ Commented Dec 19, 2019 at 18:20

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