3
\$\begingroup\$

I am designing an Amplifier circuit for the microphone. For this, I am using a buffer in between two stages for isolation and I am confused between 2 circuits. I know, in voltage buffer circuits, the input impedance is very high, and the output impedance is very low but I am confused with impedance matching if I use voltage buffer.

In the first circuit, I have used a voltage buffer, with this OPAMP U6 and U7 will be isolated. But, the output impedance of OPAMP is always low, the equivalent impedance will be low if seen from 100k resistance. Will there be any loading? Also, is there an impedance mismatch between OPAMP U6 and U7?

In the second circuit, I used an inverting amplifier with a low value of resistance. If I use this, the impedance matching is there[the equivalent of the output impedance of OPAMP and 100k (divided by gain, if miller approximation is used) is low, and the input impedance of U7( considering virtual R16 connected to virtual ground) is low)]. But it is not a voltage buffer actually.

Which circuit should I use? Also, I am working on low low-frequency circuit, so will impedance matching really be an issue? enter image description here

enter image description here

\$\endgroup\$
1
  • \$\begingroup\$ Thanks a lot guys, I'll change circuit accordingly. \$\endgroup\$ Dec 4, 2023 at 16:00

3 Answers 3

5
\$\begingroup\$

I am working on low low-frequency circuit, so will impedance matching really be an issue?

When we are dealing with signals whose maximum useful frequency has a wavelength significantly greater than the distances between components then we don't care about matching impedances. The general rule of thumb is that we wouldn't normally match impedances if the wavelength was more than ten times the maximum transmission distance. So, 1 MHz has a wavelength of 300 metres and, put into perspective we can immediately say that even at 20 kHz we would never match impedances.

Of course there are exceptions, maximum power is transferred when we match impedances and, long AC power transmission networks that form a loop must do so very carefully but, with audio design we never match impedances.

Which circuit should I use?

So, your first circuit is the best route but, if you wanted to use an inverting op-amp (as per the 2nd circuit) then use higher values for resistors R16 and R17. Typically we wouldn't want to load an op-amp output with an impedance less than 1 kΩ hence, 100 Ω, is too low for most op-amps.

is there an impedance mismatch between OPAMP U6 and U7?

Yes there is but, it's irrelevant for your audio application.

\$\endgroup\$
2
\$\begingroup\$

Yes, there is impedance mismatch.

What you don't realize that it is what you want.

Ideal op-amps have zero output impedance and infinite input impedance. Those are good things.

Imagine if you did match the impedances. It would mean the zero output impedance would drive a zero input impedance. Or that infinite output impedance would drive infinite input impedance. Both are completely unusable for any purpose.

Audio signals are also usually voltage signals, not current or power signals. Usually you want to transfer 100% or as much voltage as you possibly can between devices. Unless for some specific reason you don't.

\$\endgroup\$
2
\$\begingroup\$

A couple of things

Using TL072 as a voltage buffer doesn't offer any benefits if the preceding stage also is TL072. Load driving capacity stays the same.

You do not need impedance matching to avoid reflected waves. In audio frequency circuit boards the propagation delays on the board are so short that no remarkable pulse form distortions are caused.

Impedance matching to maximize signal power vs noise is remarkable in audio, too. But that cannot be achieved in this way. Tricky opamp circuit noise analysis hopefully shows it. In addition you should check if 100 Ohm load could be driven with your signal voltages (unknown here). If the signal voltage causes too much output current the inverting buffer turns from useless to harmful.

Impedance matching becomes critical in mic inputs - that's between the electromechanical part of the mic and the first amp stage. If you use a 200 Ohm dynamic mic you could get a substantial signal to noise ratio boost by inserting a voltage lifting transformer between the opamp input and the mic. So called High Z dynamic mics have such transformers inbuilt.

Electret mics have the first stage inbuilt, but there's still numerous possibilities to spoil the remaining noise performance with a poor next stage design. Get some proper audio design handbook. See also application notes of opamps which are designed for audio applications.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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