AD SSM2122

Dynamic Range
Processors/Dual VCA
SSM2120/SSM2122
a
FEATURES
0.01% THD at +10 dBV In/Out
100 dB VCA Dynamic Range
Low VCA Control Feedthrough
100 dB Level Detection Range
Log/Antilog Control Paths
Low External Component Count
APPLICATIONS
Compressors
Expanders
Limiters
AGC Circuits
Voltage-Controlled Filters
Noise Reduction Systems
Stereo Noise Gates
FUNCTIONAL BLOCK DIAGRAM
V+
SIGNAL
OUT
–VC
+VC
V+
V+
SIGNAL
INPUT
36kΩ
V+
IREF
GENERAL DESCRIPTION
The SSM2120 is a monolithic integrated circuit designed for the
purpose of processing dynamic signals in various analog systems
including audio. This “dynamic range processor” consists of two
VCAs and two level detectors (the SSM2122 consists of two
VCAs only). These circuit blocks allow the user to logarithmically
control the gain or attenuation of the signals presented to the
level detectors depending on their magnitudes. This allows the
compression, expansion or limiting of ac signals, some of the
primary applications for the SSM2120.
36kΩ
CURRENT
MIRRORS
V–
PIN CONNECTIONS
22-Pin Plastic DIP
(P Suffix)
THRESH 1 1
22 GND
LOG AV 1 2
21 V+
CONOUT 1 3
20 SIGOUT 2
SIGOUT 1 4
+VC1 5
19 +VC2
SSM2120
18 CFT 2
CFT 1 6
TOP VIEW 17 –VC2
–VC1 7 (Not to Scale) 16 SIGIN 2
SIGIN 1 8
15 RECIN 2
RECIN 1 9
14 CONOUT 2
IREF 10
13 LOG AV 2
V– 11
12 THRESH 2
16-Pin Plastic DIP
(P Suffix)
GND 1
16 GND
SIGOUT 1 2
15 V+
+VC1 3
CFT 1 4
14 SIGOUT 2
SSM2122
13 +VC2
TOP VIEW
–VC1 5 (Not to Scale) 12 CFT 2
SIGIN 1 6
IREF 7
V– 8
11 –VC2
10 SIGIN 2
9 GND
REV. C
Information furnished by Analog Devices is believed to be accurate and
reliable. However, no responsibility is assumed by Analog Devices for its
use, nor for any infringements of patents or other rights of third parties
which may result from its use. No license is granted by implication or
otherwise under any patent or patent rights of Analog Devices.
© Analog Devices, Inc., 1995
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 617/329-4700
Fax: 617/326-8703
SSM2120/SSM2122–SPECIFICATIONS
(@V = 615 V, T = +258C, I
S
A
ELECTRICAL CHARACTERISTICS unless otherwise noted)
Parameter
REF
= 200 mA, +VC = –VC = GND (AV = 0 dB). 0 dB = 1 V rms
Conditions
Min
POWER SUPPLY
Supply Voltage Range
Positive Supply Current
Negative Supply Current
SSM2120/SSM2122
Typ
Max
±5
8
–6
VCAs
Max ISIGNAL (In/Out)
Output Offset
Control Feedthrough (Trimmed)
Gain Control Range
Control Sensitivity
Gain Scale Factor Drift
Frequency Response
Off Isolation
Current Gain
THD (Unity-Gain)
Noise (20 kHz Bandwidth)
± 300
RIN = ROUT = 36 kΩ, –30 dB ≤ AV ≤ 0 dB
Unity-Gain
± 325
±1
± 750
–85
± 18
10
–8
V
mA
mA
± 350
±8
µA
µA
µV
dB
mV/dB
ppm/°C
kHz
dB
dB
%
dB
+40
6
–3300
250
100
Unity Gain or Less
At 1 kHz
+VC = –VC = 0 V
+10 dBV IN/OUT
RE: 0 dBV
–0.5
0.005
–80
LEVEL DETECTORS (SSM2120 ONLY)
Detection Range
Input Current Range
Rectifier Input Bias Current
Output Sensitivity (At LOG AV Pin)
Output Offset Voltage
Frequency Response
IIN = 1 mA p-p
IIN = 10 µA p-p
IIN = 1 µA p-p
90
0.085
+0.5
0.04
95
2800
4
3
± 0.5
± 3.4
1000
50
7.5
CONTROL AMPLIFIERS (SSM2120 ONLY)
Input Bias Current
Output Drive (Max Sink Current)
Input Offset Voltage
5.0
± 85
7.5
± 0.5
Units
dB
µA p-p
nA
mV/dB
mV
kHz
± 175
± 4.2
nA
mA
mV
Specifications are subject to change without notice.
ORDERING GUIDE
ABSOLUTE MAXIMUM RATINGS
Supply Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ± 18 V
Operating Temperature Range . . . . . . . . . . . . –10°C to +55°C
Junction Temperature . . . . . . . . . . . . . . . . . . . . . . . . . +150°C
Storage Temperature . . . . . . . . . . . . . . . . . . –65°C to +150°C
Maximum Current into Any Pin . . . . . . . . . . . . . . . . . . 10 mA
Lead Temperature Range (Soldering, 60 sec) . . . . . . . +300°C
Package Type
θJA1
θJC
Units
16-Pin Plastic DIP (P)
22-Pin Plastic DIP (P)
86
70
10
7
°C/W
°C/W
Model
Temperature
Range
Package
Description
Package
Option
SSM2120
SSM2122
–10°C to +50°C
–10°C to +50°C
22-Pin Plastic DIP
16-Pin Plastic DIP
(N-22)
(N-16)
NOTE
1
θJA is specified for worst case mounting conditions, i.e., θJA is specified for
device in socket for P-DIP.
CAUTION
ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily
accumulate on the human body and test equipment and can discharge without detection.
Although the SSM2120/SSM2122 features proprietary ESD protection circuitry, permanent damage
may occur on devices subjected to high energy electrostatic discharges. Therefore, proper ESD
precautions are recommended to avoid performance degradation or loss of functionality.
–2–
WARNING!
ESD SENSITIVE DEVICE
REV. C
SSM2120/SSM2122
+VC1
SSM2122
V+
THRESH 1
CONOUT 1
|I IN |
INPUT 1
OUTPUT 1
FULL
WAVE
RECTIFIER
RECIN 1
2V
–VC1
CFT 1
LOG AV 1
V–
V+
CONOUT 2
+VC2
THRESH 2
|I IN |
INPUT 2
OUTPUT 2
FULL
WAVE
RECTIFIER
RECIN 2
2V
–VC2
CFT 2
V–
LOG AV 2
Figure 1. SSM2120 Block Diagram
VOLTAGE-CONTROLLED AMPLIFIERS
VCA PERFORMANCE
The two voltage-controlled amplifiers are full Class A current
in/current out devices with complementary dB/V gain control
ports. The control sensitivities are +6 mV/dB and –6 mV/dB. A
resistor divider (attenuator) is used to adapt the sensitivity of an
external control voltage to the range of the control port. It is
best to use 200 Ω or less for the attenuator resistor to ground.
Figures 2a and 2b show the typical THD and noise performance
of the VCAs over ±20 dB gain/attenuation. Full Class A operation
provides very low THD.
0.03
VCA INPUTS
The signal inputs behave as virtual grounds. The input current
compliance range is determined by the current into the reference
current pin.
THD – %
0.01
REFERENCE PIN
0.003
The reference current determines the input and output current
compliance range of the VCAs. The current into the reference
pin is set by connecting a resistor to V+. The voltage at the
reference pin is about two volts above V– and the current will be
–20
I REF
[(V +) – ((V – ) + 2 V )]
=
RREF
–10
0
GAIN – dB
10
20
a. VCA THD Performance vs. Gain
(+10 dBV In/Out @ 1 kHz)
The current consumption of the VCAs will be directly proportional to IREF which is nominally 200 µA. The device will
operate at lower current levels which will reduce the effective
dynamic range of the VCAs. With a 200 µA reference current,
the input and output clip points will be ± 400 µA. In general:
–70
ICLIP = ± 2 IREF
NOISE – dBV
–80
VCA OUTPUTS
The VCA outputs are designed to interface directly with the virtual
ground inputs of external operational amplifiers configured as
current-to-voltage converters. The outputs must operate at virtual
ground because of the output stage’s finite output impedance.
The power supplies and selected compliance range determines
the values of input and output resistors needed. As an example,
with ± 15 V supplies and ±400 µA maximum input and output
current, choose RIN = ROUT = 36 kΩ for an output compliance
range of ± 14.4 V. Note that the signal path through the VCA
including the output current-to-voltage converter is noninverting.
–90
–20
–10
0
GAIN – dB
10
20
b. VCA Noise vs. Gain (20 kHz Bandwidth)
Figure 2. Typical THD and Noise Performance
REV. C
–3–
SSM2120/SSM2122
TRIMMING THE VCAs
Note: It is natural to assume that with the addition of the
averaging capacitor, the LOG AV output would become the
average of the log of the absolute value of IIN. However, since the
capacitor forces an ac ground at the emitter of the output
transistor, the capacitor charging currents are proportional to
the antilog of the voltage at the base of the output transistor.
Since the base voltage of the output transistor is the log of the
absolute value of IIN, the log and antilog terms cancel, so the
capacitor becomes a linear integrator with a charging current
directly proportional to the absolute value of the input current.
This effectively inverts the order of the averaging and logging
functions. The signal at the output therefore is the log of the
average of the absolute value of IIN.
The control feedthrough (CFT) pins are optional control feedthrough null points. CFT nulling is usually required in applications
such as noise gating and downward expansion. If trimming is
not used, leave the CFT pins open.
Trim Procedure
1. Apply a 100 Hz sine wave to the control point attenuator.
The signal peaks should correspond to the control voltages
which induce the VCAs maximum intended gain and at least
30 dB of attenuation.
2. Adjust the 50 kΩ potentiometer for the minimum
feedthrough.
(Trimmed control feedthrough is typically well under 1 mV rms
when the maximum gain is unity using 36 kΩ input and output
resistors.)
USING DETECTOR PINS REC IN, LOGAV, THRESH AND
CONOUT
Applications such as compressor/limiters typically do not require
control feedthrough trimming because the VCA operates at
unity-gain unless the signal is large enough to initiate gain
reduction. In this case the signal masks control feedthrough.
When applying signals to RECIN (rectifier input) an input series
resistor should be followed by a low leakage blocking capacitor
since RECIN has a dc voltage of approximately 2.1 V above
ground. Choose RIN for a ± 1.5 mA peak signal. For ± 15 V
operation this corresponds to a value of 10 kΩ.
This trim is ineffective for voltage-controlled filter applications.
A 1.5 MΩ value of RREF from log average to –15 V will establish
a 10 µA reference current in the logging transistor (Q1). This
will bias the transistor in the middle of the detector’s dynamic
current range in dB to optimize dynamic range and accuracy.
The LOG AV outputs are buffered and amplified by unipolar
drive op amps. The 39 kΩ, 1 kΩ resistor network at the
THRESH pin provides a gain of 40.
LEVEL DETECTION CIRCUITS
The SSM2120 contains two independent level detection
circuits. Each circuit contains a wide dynamic range full-wave
rectifier, logging circuit and a unipolar drive amplifier. These
circuits will accurately detect the input signal level over a
100 dB range from 30 nA to 3 mA peak-to-peak.
An attenuator from the CONOUT (control output) to the
appropriate VCA control port establishes the control sensitivity.
Use 200 Ω for the attenuator resistor to ground and choose
RCON for the desired sensitivity. Care should be taken to minimize
capacitive loads on the control outputs CONOUT. If long lines
or capacitive loads are present, it is best to connect the series
resistor RCON as closely to the CONOUT pin as possible.
LEVEL DETECTOR THEORY OF OPERATION
Referring to the level detector block diagram of Figure 3, the
RECIN input is an AC virtual ground. The next block implements the full-wave rectification of the input current. This
current is then fed into a logging transistor (Q1) whose pair
transistor (Q2) has a fixed collector current of IREF. The LOG
AV output is then:
DYNAMIC LEVEL DETECTOR CHARACTERISTICS
V LOG
kT  |I IN|
AV =
ln 

q
 I REF 
Figures 4 and 5 show the dynamic performance of the level
detector to a change in signal level. The input to the detector (not
shown) is a series of 500 ms tone bursts at 1 kHz in successive
10 dBV steps. The tone bursts start at a level of –60 dBV (with
RIN = 10 k) and return to –60 dBV after each successive 10 dB
step. Tone bursts range from –60 dBV to +10 dBV. Figure 4
shows the logarithmic level detector output. The output of the
detector is 3 mV/dB at LOG AV and the amplifier gain is 40
which yields 120 mV/dB. Thus, the output at CONOUT is seen
to increase by 1.2 V for each 10 dBV increase in input level.
With the use of the LOG AV capacitor the output is then the log
of the average of the absolute value of IIN.
(The unfiltered LOG AV output has broad flat plateaus with
sharp negative spikes at the zero crossing. This reduces the
“work” that the averaging capacitor must do, particularly at low
frequencies.)
39kΩ
1kΩ
THRESH
V+
RIN
INPUT
|I I N |
RECIN
FULL
WAVE
RECTIFIER
IREF
CONOUT
Q2
RCON
TO VC
200Ω
Q1
2V
LOG AV
CAV
V–
RREF
V–
Figure 3. Level Detector
–4–
REV. C
SSM2120/SSM2122
2V
The decay rates are linear ramps that are dependent on the
current out of the LOG AV pin (set by RREF) and the value of
CAV. The integration or decay time of the circuit is derived from
the formula:
1s
100
90
Decrementation Rate (in dB/s) =
I REF × 333
C AV
10
0%
Table I. Settling Time (tS) for CAV = 10 mF. tS = tS (CAV = 10 mF)
Figure 4. Detector Output
10 dB Step
20 dB Step
30 dB Step
40 dB Step
50 dB Step
60 dB Step
2V
100
90
5 dB
3 dB
2 dB
1 dB
11.28 ms
16.65
18.15
18.61
21.46
26.83
28.33
27.79
30.19
35.56
37.06
37.52
(+144 µs)
(+46 µs)
46.09
51.46
52.96
53.42
APPLICATIONS
10
The following applications for the SSM2120 use both the VCAs
and level detectors in conjunction to assimilate a variety of
functions.
0%
50ms
The first section describes the arrangement of the threshold
control in each control circuit configuration. These control
circuits form the foundation for the applications to follow which
include the downward expander, compressor/limiter and
compandor.
Figure 5. Overlayed Detector Output
DYNAMIC ATTACK AND DECAY RATES
Figure 5 shows the output levels overlayed using a storage
scope. The attack rate is determined by the step size and the
value of CAV. The attack time to final value is a function of the
step size increase. Table I shows the values of total settling
times to within 5 dB, 3 dB, 2 dB and 1 dB of final value with
CAV = 10 µF. When step sizes exceed 40 dB, the increase in
settling time for larger steps is negligible. To calculate the attack
time to final value for any value of CAV, simply multiply the
value in the chart by CAV/10 µF.
THRESHOLD CONTROL
Figure 6a shows the control circuit for a typical downward
expander while Figure 6b shows a typical control curve. Here,
the threshold potentiometer adjusts VT to provide a negative
unipolar control output. This is typically used in noise gate,
downward expander, and dynamic filter applications. This
potentiometer is used in all applications to control the signal
level versus control voltage characteristics.
THRESHOLD
THRESHOLD
CONTROL
RIN
L
VT
V+
RECIN
+
V–
RT
|I IN |
39kΩ
CONOUT RCON
RLL
TO +VC
1kΩ
200Ω
2V
MONO – RIN = 10kΩ
STEREO – RIN = 20kΩ
VCON
MONO
OR R
RIN
LOG AV
–
CAV
1.5MΩ
V–
V–
*
*LOWER LIMIT CAN BE FIXED
BY CONNECTING A RESISTOR
RLL FROM RECIN TO GROUND
VIN – dB
a. Control Circuit
b. Typical Downward Expander
Control Curve
Figure 6. Noise Gate/Downward Expander Control Circuit and Typical Response
REV. C
–5–
SSM2120/SSM2122
In the noise gate, downward expander and compressor/limiter
applications, this potentiometer will establish the onset of the
control action. The sensitivity of the control action depends on
the value of RT.
as shown in Figure 8a, with its response in Figure 8b. The value
of the resistor RPV will determine the maximum output from the
control amplifier.
For a positive unipolar control output add two diodes as shown
in Figure 7a. This is useful in compressor/limiter applications.
Figure 7b shows a typical response.
STEREO COMPRESSOR/LIMITER
The two control circuits of Figures 6 and 7 can be used in
conjunction to produce composite control voltages. Figures 9a
and 9b show this type of circuit and transfer function for a
stereo compressor/limiter which also acts as a downward
expander for noise gating. The output noise in the absence of a
Bipolar control outputs can be realized by adding a resistor from
the op amp output to V+. This is useful in compandor circuits
THRESHOLD
V+
THRESHOLD
CONTROL
VT
V+
MONO
OR R
RIN
RPV
TO –VC
RT
REC IN
39kΩ
|I IN |
*
+
RCON
200Ω
CON
OUT
VCON
RIN
L
1kΩ
2V
MONO – RIN = 10kΩ
STEREO – RIN = 20kΩ
CAV
–
THRESH
LOG AV
1.5MΩ
V–
*UPPER LIMIT CAN BE FIXED BY
VALUE OF PULL UP RESISTOR (RPV)
CONNECTED TO POSITIVE SUPPLY
V–
VIN – dB
a. Control Circuit
b. Typical Compressor/Limiter Control
Curve
Figure 7. Compressor/Limiter Control Circuit and Typical Response
V+
+
V+
MONO
OR R
V–
VT
RECIN
RPV
RT
39kΩ
|I I N |
RLL
CONOUT
1kΩ
TO +VC
OR –VC
MONO – RIN = 10kΩ
STEREO – RIN = 20kΩ
CAV
VT = 0
–
THRESH
1.5MΩ
VT > 0
200Ω
2V
LOG AV
VT < 0
VCON
L
RIN
*
GAIN
RIN
*
V–
V–
*UPPER AND
LOWER LIMITS CAN
BE ESTABLISHED BY
VALUES OF RPV AND
RLL, RESPECTIVELY
VIN – dB
a. Control Circuit
b. Typical Compandor Control Curves
Figure 8. Compandor Control Circuit and Typical Curves
EXPANSION
THRESHOLD
COMPRESSION
THRESHOLD
THRESHOLD EXP.
FIGURE 6
L
+VC
VOUT – dB
200Ω
MONO
OR R
THRESHOLD COM.
FIGURE 7
–VC
200Ω
*
VIN – dB
a. Control Circuit
b. Input/Output Curve
Figure 9. Control Circuit for Stereo Compressor/Limiter with Noise Gating and Input/Output Curve
–6–
REV. C
SSM2120/SSM2122
10pF
+VC
SIGNAL
INPUT
10pF
–VC
36Ω
200Ω
36kΩ
200Ω
36kΩ
TRANSMISSION
OR
STORAGE
MEDIUM
2200pF
47Ω
36kΩ
SIGNAL
OUTPUT
2200pF
47Ω
200Ω
1µF
–VC
V+
200Ω
V+
RECIN
39kΩ
10kΩ
RE
10kΩ
1kΩ
1µF
V+
V+
10kΩ
|I I N |
+VC
1µF
10kΩ
RC
RECIN
39kΩ
|I I N |
1kΩ
LOG AV
4.7MΩ
LOG AV
1µF
V–
4.7MΩ
V–
V–
V–
Figure 10. Companding Noise Rejection System
being used. As an extreme example, a household tape player
would require a higher compression/expansion ratio than a
professional stereo system.
signal will be dependent on the noise of the current-to-voltage
converter amplifier if the expansion ratio is high enough.
As discussed in the Threshold Control section, the use of the
control circuit of Figure 6, including the RPV to V+ and two
diodes, yields positive unipolar control outputs.
20
OUTPUT SIGNAL LEVEL – dB
COMPANDING NOISE REDUCTION SYSTEM
A complete companding noise reduction system is shown in
Figure 10. Normally, to obtain an overall gain of unity, the
value of RC is equal to RE. The values of RC/E will determine the
compression/expansion ratio.
Table II shows compression/expansion ratios ranging from 1.5:1
to full limiting with the corresponding values of RC/E.
An example of a 2:1 compression/expansion ratio is plotted in
Figure 11. Note that signal compression increases gain for low
level signals and reduces gain for high levels while expansion
does the reverse. The net result for the system is the same as the
original input signal except that it has been compressed before
being sent to a given medium and expanded after recovery. The
compression/expansion ratio needed depends on the medium
0
IREF ≈ 3µA
RREF = 4.7MΩ
OVERALL
RESPONSE
25dB
2:1
EXPANSION
–20
2:1
COMPRESSION
–40
–60
–80
–80
–60
–40
–20
INPUT SIGNAL LEVEL – dB
0
20
Figure 11. Companding Noise Reduction with 2:1
Compression/Expansion Ratio
Table II.
Input Signal
Increase (dB)
Gain
(Reduction
or Increase)
(dB)
Compressor
Only
Output Signal
Increase (dB)
Expander
Only
Output Signal
Increase (dB)
Compression/
Expansion Ratio
RC/E V
DVCONTROL –
(mV/dB)
20
20
20
20
20
20
20
20
6.67
10.00
13.33
15.00
16.00
17.33
18.00
20.00
13.33
10.00
6.67
5.00
4.00
2.67
2.00
0
22.67
30.00
33.33
35.00
36.00
37.33
38.00
40.00
1.5:1
2:1
3:1
4:1
5:1
7.5:1
10:1
AGC*/Limiter
11,800
7,800
5,800
5,133
4,800
4,415
4,244
3,800
2.0
3.0
4.0
4.5
4.8
5.2
5.4
6.0
*AGC for Compression Only.
REV. C
–7–
SSM2120/SSM2122
DYNAMIC FILTER
Figure 12 shows a control circuit for a dynamic filter capable of
single ended (nonencode/decode) noise reduction. Such circuits
usually suffer from a loss of high frequency content at low signal
levels because their control circuits detect the absolute amount
of highs present in the signal. This circuit, however, measures
wideband level as well as high frequency band level to produce
a composite control signal combined in a 1:2 ratio respectively.
The upper detector senses wideband signals with a cutoff of
20 Hz while the lower detector has a 5 kHz cutoff to sense only
high frequency band signals. This approach allows very good
noise masking with a minimum loss of “highs” when the signal
level goes below the threshold.
V+
AUDIO
INPUT
10kΩ
2.2µF
RECIN
9
THRESHOLD
CONTROL
V–
160kΩ THRESH
|I IN |
1kΩ
FC ≤ 20Hz
(WIDEBAND)
39kΩ CONOUT 12kΩ
1
3
LOG AV
3.3µF
2
V–
1.5MΩ
V–
160kΩ
V+
10kΩ
3300pF REC
IN
15
THRESH
|I IN |
1kΩ
FC = 5kHz
(HIGH FREQUENCY)
39kΩ CONOUT 5.6kΩ
14
12
200Ω
LOG AV
3.3µF
5
13
V–
+VC
36kΩ
V–
1.5MΩ
36kΩ
36kΩ
SIGIN
8
5
36kΩ
AUDIO
OUTPUT
47Ω
2200pF
100pF
SIGOUT
7
–VC
200Ω
Figure 12. Dynamic Noise Filter Circuit
–8–
REV. C
SSM2120/SSM2122
HIGH-FREQUENCY SIGNAL LEVEL – dB
Figures 13a–c show the filter’s 3 dB frequency response with the
threshold potentiometer at V+, centered, and V–. Data was
taken by applying a 300 Hz signal to the wideband detector and
a 20 kHz signal to the high-frequency band detector simultaneously. These figures correspond to filter characteristics for
50 dB, 70 dB and 90 dB dynamic range program source
material, respectively. The system could thus treat signals from
anything ranging from 1/4" magnetic tape to high performance
compact disc players.
Note that in Figure 13a the control circuit is designed so that
the minimum cutoff frequency is about 1 kHz. This occurs as
the control circuit detects the noise floor of the source material.
Dynamic filtering limits the signal bandwidth to less than 1 kHz
unless enough highs are detected in the signal to cover the noise
floor in the mid- and high frequency range. In this case the filter
opens to pass more of the audio band as more highs are detected.
The filter’s bandwidth can extend to 50 kHz with a nominal
signal level at the input. At other signal levels with varying high
frequency content, the filter will close to the required bandwidth. Here, noise outside the band is removed while the
perceived noise is masked by other signals within the band.
Even in this system, however, a certain amount of mid- and high
frequency components will be lost, especially during transients
at very low signal levels. This circuit does not address low
frequency noise such as “hum” and “rumble.”
20
50.6
50.6
50.6
50.6
50.6
50.6
26
26
26
8.3
11.7
11.7
11.7
10
0
–10
17.8
6
–20
1.9
2.75
3.9
5.5
5.5
5.5
1.0
1.0
1.2
1.7
2.4
2.4
2.4
1.0
1.0
1.0
1.0
1.0
1.1
1.1
1.1
–50
–40
–30
–20
–10
0
10
WIDEBAND SIGNAL LEVEL – dB
20
–30
–40
–50
HIGH-FREQUENCY SIGNAL LEVEL – dB
a. VTHRESH at V+
20
50.6
50.6
50.6
50.6
50.6
50.6
50.6
50.6
50.6
50.6
49.2
49.2
49.2
49.2
10
0
–10
–20
48
15.1
22
22
22
22
22
4.9
7.1
10
10
10
10
10
1.5
2.2
3.1
4.2
4.2
4.2
4.2
4.2
–50
–40
–30
–20
–10
0
10
WIDEBAND SIGNAL LEVEL – dB
20
–30
–40
–50
HIGH-FREQUENCY SIGNAL LEVEL – dB
b. VTHRESH Centered
20
50.6
50.6
50.6
50.6
50.6
50.6
50.6
50.6
50.6
50.6
50.6
50.6
50.6
50.6
50.6
10
0
–10
–20
50.6
50.6
50.6
50.6
50.6
50.6
40
41
41
41
41
41
41
12.3
17.3
17.8
17.8
17.8
17.8
17.8
17.8
–50
–40
–30
–20
–10
0
10
WIDEBAND SIGNAL LEVEL – dB
20
–30
–40
–50
c. VTHRESH at V–
Figure 13. 3 dB Filter Response
REV. C
–9–
SSM2120/SSM2122
V+
THRESHOLD
V–
V+
SIGNAL
INPUT
RIN1
200Ω
12kΩ
160kΩ
39kΩ
RECIN
CONOUT
IN |I I N |
+VC
1kΩ
FC ≤ 20Hz
36kΩ
12kΩ
LOG AV
CAV1
1.5MΩ
V–
2200pF
V–
V+
RIN2
SIGNAL
OUTPUT
47Ω
–VC
160kΩ
RECIN
200Ω
39kΩ CONOUT 5.6kΩ
IN |I I N |
200Ω
1kΩ
FC = 5Hz
DOWNWARD EXPANDER
LOG AV
+VC
CAV2
1.5MΩ
36kΩ
V–
36kΩ
V–
100pF
36kΩ
36kΩ
36kΩ
47Ω
2200pF
–VC
200Ω
Figure 14. Dynamic Filter with Downward Expander
The dynamic filter and downward expander techniques used
together can be employed more subtly to achieve a given level of
noise reduction than would be required if used individually. Up
to 30 dB of noise reduction can be realized while preserving the
crisp highs with a minimum of transient side effects.
+20
–30
–30
–40
–45
–50
Downward expansion uses a VCA controlled by the level
detector. This section maintains dynamic range integrity for all
levels above the user adjustable threshold level. As the input
level decreases below the threshold, gain reduction occurs at an
increasing rate (see Figure 15). This technique reduces audible
noise in fade outs or low level signal passages by keeping the
standing noise floor well below the program material.
This technique by itself is less effective for signals with
predominantly low frequency content such as a bass solo where
wideband frequency noise would be heard at full level. Also,
since the level detector has a time constant for signal averaging,
percussive material can modulate the noise floor causing a
“pumping” or “breathing” effect.
+20
–60
OUTPUT – dB
A composite single-ended noise reduction system can be
realized by a combination of dynamic filtering and a downward
expander. As shown in Figure 14, the output from the wideband
detector can also be connected to the +VC control port of the
second VCA which is connected in series with the sliding filter.
This will act as a downward expander with a threshold that
tracks that of the filter. Although both of these techniques are
used for noise reduction, each alone will pass appreciable
amounts of noise under some conditions. When used together,
both contribute distinct advantages while compensating for each
other’s deficiencies.
INPUT – dB
DYNAMIC FILTER WITH DOWNWARD EXPANDER
–60
–75
Figure 15. Typical Downward Expander I/O Characteristics
at –30 dB Threshold Level (1:1.5 Ratio)
–10–
REV. C
SSM2120/SSM2122
FADER AUTOMATION
The SSM2120 can be used in fader automation systems to serve
two channels. The inverting control port is connected through
an attenuator to the VCA control voltage source. The noninverting
control port is connected to a control circuit (such as Figure 6)
which senses the input signal level to the VCA. Above the
threshold voltage, which can be set quite low (for example
–60 dBV), the VCA operates at its programmed gain. Below
this threshold the VCA will downward expand at a rate determined by the +VC control port attenuator. By keeping the release
time constant in the 10 ms to 25 ms range, the modulation of
the VCA standing noise floor (–80 dB at unity-gain), can be
kept inaudibly low.
The SSM2300 8-channel multiplexed sample-and-hold IC
makes an excellent controller for VCAs in automation systems.
Figure 16 shows the basic connection for the SSM2122 operating
as a unity-gain VCA with its noninverting control ports grounded
and access to the inverting control ports. This is typical for fader
automation applications. Since this device is a pinout option of
the SSM2120, the VCAs will behave exactly as described
earlier in the VCA section.
The SSM2122 can also be used with two or more op amps to
implement complex voltage-controlled filter functions. Biquad
and state-variable two-pole filters offering low pass, bandpass
and high pass outputs can be realized. Higher order filters can
also be formed by connecting two or more such stages in series.
+15V
0.1µF
10pF
36kΩ
1/2
TL082
SIGOUT 1
1
16
2
15
3
14
4
13
36kΩ
200Ω
V+
220kΩ
50kΩ*
200Ω
SSM2122
200Ω
V–
–VC1
10pF
1/2
TL082
220kΩ
50kΩ*
12
5
200Ω
36kΩ
SIGIN 1
2000pF
47Ω
6
11
7
10
V–
150kΩ
V+
36kΩ
–VC2
SIGIN 2
2000pF
0.1µF
9
8
–15V
SIGOUT 2
V+
47Ω
*OPTIONAL CONTROL FEEDTHROUGH TRIM
Figure 16. SSM2122 Basic Connection (Control Ports at 0 V)
REV. C
–11–
SSM2120/SSM2122
OUTLINE DIMENSIONS
Dimensions shown in inches and (mm).
C2088–2–11/95
16-Pin Plastic DIP
(N-16)
0.840 (21.33)
0.745 (18.93)
16
9
1
8
0.280 (7.11)
0.240 (6.10)
0.060 (1.52)
0.015 (0.38)
PIN 1
0.210 (5.33)
MAX
0.325 (8.25)
0.300 (7.62) 0.195 (4.95)
0.115 (2.93)
0.130
(3.30)
MIN
0.160 (4.06)
0.115 (2.93)
0.022 (0.558)
0.014 (0.356)
0.100
(2.54)
BSC
0.015 (0.381)
0.008 (0.204)
0.070 (1.77) SEATING
0.045 (1.15) PLANE
22-Pin Plastic DIP
(N-22)
1.080 (27.43)
1.020 (25.91)
22
12
1
11
0.280 (7.11)
0.240 (6.10)
0.060 (1.52)
0.015 (0.38)
PIN 1
0.325 (8.25)
0.300 (7.62) 0.195 (4.95)
0.115 (2.93)
0.210
(5.33)
MAX
0.022 (0.558)
0.014 (0.356)
0.100
(2.54)
BSC
0.070 (1.77) SEATING
PLANE
0.045 (1.15)
0.015 (0.381)
0.008 (0.204)
PRINTED IN U.S.A.
0.160 (4.06)
0.115 (2.93)
–12–
REV. C