Re: What a Parrot Brained Fuckwit

Bruce Burke wrote: 
For R=10 ohms

Take a voltage increase across R from 2 volts to 4 volts.

P1 = 2^2/10 = 0.4watts
P2 = 4^2/10 = 1.6watts

dB = 20*log(4/2) = 6.02 (voltage case)
dB = 10*log(1.6/0.4) = 6.02 (power case)

So for electronics, a dB, is a dB......

Yes, but I think what upsets the purists is that a typical audio amplifier doesn't involve the same value of R at the input and the output.

A power amplifier with an input impedance of 10K and the capability of driving a 4 ohms load has far more power gain than implied by just squaring the voltage gain. Similarly the amplifier in a condenser mic may have unity voltage gain but a lot of power gain because of current gain, and in that context it's important. You have to be very careful about what you mean when you talk about the gain in dB of a voltage follower. I guess that you could say it has a voltage gain of 0dB and a current or power gain of <some positive number> dB.

In practice I don't think it matters much, as long as audio engineers understand the derivation of the 20*log formula and the difference between voltage and power.

Of course, you're probably thinking acoustics,  where the dB is
representing the change in sound pressure intensity. Because it is a
square law relationship, we use the 20*log.... formula.

Actually sound power (as opposed to pressure) is quite a useful concept here. If you have a stereo pair of speakers with the same power coming out of each and you disconnect one, the drop in sound level is 3dB, because the *power* simply adds - no square law required unless you choose to work in units of pressure.

Anahata
.