LME49740 Gain Calculation

[chris319]

Disclaimer: I'm a total dunce at this kind of thing.

Looking at the mic preamp circuit on page 15 of the data sheet for the National LME49740:

http://www.national.com/ds/LM/LME49740.pdf

If I used 3 megohm resistors for R2 - R7, I would get 96 dB of gain according to the formula given. Can that be right? Each amp has an open loop voltage gain of 140 dB.

I'm contemplating a functional high-gain (> 80 dB) mic preamp with this chip (or a reasonable alternative) so any help is greatly appreciated.

Thank you.

 

[JohnRoberts]

That 140 dB of open loop gain is at DC, if you look at the bode plot there is a -6dB/oct gain rolloff starting pretty low. At 20 kHz it has 60 dB of open loop gain which is respectable, but not going to cut it with a closed loop gain of 96.

There are ICs designed for use as mic preamps, and Sam has a thread with a number of examples of classic mic preamp topologies.

A single (even a very good) opamp generally won't be optimal, unless you put a transformer in front of it, and if you don't like transformers, that's not optimal either.

There are numerous mic preamp threads here and about.

JR

 

[dale116dot7]

For audio, I don't like to see any one gain stage going over about 40dB. Exactly what John said... if the open-loop gain at 20kHz is 60dB, I like to have 20dB worth of extra gain just so the circuit behaves as designed - no unintended phase shifts. You can cascade a pair of them, though, and be close to what you want, I think. I'm not too sure of the LME49740 will be quite quiet enough for a preamp. The SSM2019 quieter (1nV vs 2.7nV per rtHz). Another point is that if your feedback resistors are that high, you might be getting in too much Johnson noise, and parasitics (capacitance) will start to diddle with your frequency response - and possibly stability. I don't like to see over 100k in an audio circuit - the lower the better. Well, maybe in an LF EQ stage where the noise is swamped by a capacitor, then those impedances are ok.