Handbook for Sound Engineers

(Wang) #1
Attenuators 781

amplifier is –91 dB and out of the booster amplifier is
–58 dB [–110 + (+33) + (–14) + (+33)]. The mixer
noise going into the booster amplifier is –125 dB; there-
fore, the output noise is –92 dB [–125 + (+33)] or 34 dB
below the signal.
In Fig. 22-28B, the input signal of –110 dBm is atten-
uated 20 dB in the mixer so the signal to the booster is
–111 dB and the signal output is –78 dB. The mixer
input noise is still –125 dB into the booster and –92 dB
out of the booster, a difference between the signal and
the noise of only 14 dB, hardly enough to be useful.
An active mixer is one that uses operational ampli-
fiers (op-amps) or some other active device along with
resistors and/or potentiometers to control gain or
attenuation.
A unity-gain current-summing amplifier is used for a
standard active mixer. The mixer is usually designed for
an input impedance of about 5–10 k:, an output imped-
ance of less than 200:, and a gain of 0 to 50. A typical
active mixer is shown in Fig. 22-29.


In unity-gain current-summing amplifiers feedback
to the minus or inverting input presents an extremely


low apparent input impedance or virtual ground on the
inverting input.
The positive input is also essentially ground since
the current through Rn will only produce about 0.5 mV.
While the positive input can be grounded, it is better to
make the Rn a value about the same as the parallel
combination of R 1 +R 2 +Rf to reduce offset voltage.
Any small, positive-going input applied to the input
of R 1 is amplified by the high-gain op-amp driving the
output negative since the input signal is on the inverting
input. The output signal is fed back through Rf, the feed-
back resistor, and it continuously attempts to drive the
voltage on the input to ground.
Since the input is a virtual ground, the input imped-
ances are determined by R 1 and R 2. The gain of the
circuit is

(22-53)

(22-54)

If the gain of both inputs were to be the same, R 1 and
R 2 would remain constant and Rf would be varied.
Mixers, however, usually require separate gain control
for each input so R 1 and R 2 are varied to change the
gain of the system. Increasing R 1 or R 2 decreases the
gain. The main disadvantage of this system is that the
input impedance varies with gain.
The advantage of an active mixer is that gain is
included in the mixing circuit; therefore, it does not
need a gain makeup amplifier that amplifies both the
signal and the mixing noise after the mixer. With active
mixing, the mixing noise is also reduced along with the
signal, improving the SNR, particularly at low level.

22.2.16 Summing Amplifiers^1

A standard audio circuit function is the linear combina-
tion of a number of individual signals into a common
output without crosstalk or loss. This function is well
suited for the summing amplifier, which is often
referred to as active combining network. Summing
amplifiers operate much like the mixer in Section
22.2.15. Fig. 22-30 shows a 10-input summing ampli-
fier using one op-amp. Channel isolation is important in
summing amplifiers to eliminate crosstalk.
The primary determinant of interchannel isolation is
the nonzero summing-bus impedance presented by the
virtual ground of the inverter and, to a lesser extent, by

Figure 22-28. Signal-to-noise analyses of passive
attenuators.

Figure 22-29. An active mixer block diagram.

14 dB

14 dB

B. Passive attenuator with 20 dB attenuation.

A. Passive attenuator with 0 attenuation

110 dBm

0 dB

Mixing loss

33 dB

Signal 58 dB
Noise 92 dB

Mixer noise ( 125 dB)

110 dBm 33 dB
Mixing loss

Mixer noise 
125 dB)

33 dB

Signal 78 dB
Noise 92 dB

33dB

20 dB

Rf

Rn

R 1

R 2

Input 1

Input 2

Output
+

input 1gain

Rf
R 1

----- -=

input 2gain

Rf
R 2

----- -=
Free download pdf