AM/FM NOISEBLAN KER WI TH SEVERAL
TRIGGER CONTROLS
■
PROGRAMMABLE MULTI PATH D ETECTOR
■
QUALITY DETECTOR OUTPUT
TDA7405
TQFP44
ORDERING NUMBER: TDA7405
Digital control:
■
I2C-BUS INTERFAC E
DESCRIPTION
The devic e includes a high performanc e audiopr ocessor
and a st ereodecoder-noiseblanker combination with the
whole low frequency signal processing necessary for
state-of-the-art as well as future carradios. The digital
control allows a programming in a wide range of all the
filter characteristics. Also the stereodecoder part offers
several possibilities of programming especially for the
adaptation to different IF -devices.
separate second source-selector
Beepinternal Beep with 3 frequencies + diagnostic setting (19kHz tone)
Mixing stageBeep, Phone,Navigation and FM mixable to all speaker-outputs (see Figure 20)
programmabe Voice-Band Filter
Loudnessprogrammable center frequency and frequency response
soft-step control with programmable times
Bass2nd order frequency response
center frequency programmable in 8 steps
DC gain programmable
± 15 x 1dB steps
Treble2nd order frequency response
center frequency programmable in 4 steps
± 15 x 1dB steps
Equalizertwo stereo equalizing-filters for separate front/rear adaption
1st filter center-frequency programmable i n 16 steps ( 4 steps/octav e, min 63Hz, max
840Hz)
2nd filter center-frequency programmable in 16 steps (4 steps/octave, min 300Hz,
max 4kHz)
quality factor programmable in 4 steps
± 15 x 1dB steps
selectable flat-mode
Speaker4 independent speaker controls in 1dB steps
control range 95dB
separate Mute
Mute Functionsdirect mute
digitally controlled SoftMute with 4 programmable mute-times
Pause Detectorprogrammable threshold
Companderdynamic range compression for use with CD
2:1 compression rate
programmable max. gain
4/56
Page 5
TDA7405
Table 1. ELECTRICAL CHARACTERISTICS
(VS=9V; T
SymbolParameterTest ConditionMin. Typ.Max.Unit
INPUT SELECTOR
R
in
V
CL
S
IN
G
IN MIN
G
IN MAX
G
STEP
V
DC
V
offset
DIFFERENTIAL STEREO INPUTS
R
in
G
CD
CMRRCommon Mode Rejection RatioV
e
NO
DIFFERENTIAL MONO INPUTS
R
in
CMRRCommon Mode Rejection RatioV
BEEP CONTROL
V
RMS
f
Beep
MIXING CONTROL
M
LEVEL
G
MAX
A
MAX
A
STEP
=25°C; RL=10kΩ; all gains=0dB; f=1kHz; unless otherwise specified)
amb
Input Resistanceall single ended Inputs70100130kΩ
Clipping Level2.22.6V
Input Separation80100dB
Min. Input Gain -10+1dB
Max. Input Gain131517dB
Step Resolution0.511.5dB
DC StepsAdjacent Gain Steps-515mV
G
MIN
to G
MAX
-10110mV
Remaining offset with AutoZero0.5mV
Input Resistance
Differential70100130kΩ
(see Fig. 1)
Gainonly at true differential input-101dB
-5-6-7dB
-11-12-13dB
= 1V
CM
= 1V
V
CM
@ 1kHz4670dB
RMS
@ 10kHz4660dB
RMS
Output-Noise @ Speaker-Outputs 20Hz - 20kHz, flat; all stages 0dB915µV
=25°C; RL=10kΩ; all gains=0dB; f=1kHz; unless otherwise specified)
amb
(continued)
SymbolParameterTest ConditionMin. Typ.Max.Unit
S/NSignal to Noise Ratioall gains = 0dB
= 2V
flat; V
O
RMS
bass, treble at +12dB;
a-weighted; V
ddistortionV
S
Channel Separation left/right80100dB
C
Total Tracking ErrorAV = 0 to -20dB-101dB
E
T
= 1V
IN
V
= 1V
OUT
= -20 to -60dB-202dB
A
V
; all stages 0dB0.0050.1%
RMS
; Bass & Treble = 12dB
RMS
= 2.6V
O
RMS
106dB
100dB
0.050.1%
9/56
Page 10
TDA7405
2DESCRIPTION OF THE AUDIOPROCESSOR PART
2.1 Input stages
In the basic configuration two full-differential, tw o mono-differential, one single ended stereo and one singleended mono input are available. In addition a dedicated input for the stereodecoder MPX-signal is present.
Figure 1. Input stages
2.1.1 Full-differential stereo Input 1 (FD1)
The FD1-input is implemented as a buffered full-differential stereo stage with 100kΩ input-impedance at each
input. The attenuation is programmable in 3 steps from 0 to -12dB in order to adapt the incoming signal level.
A 6dB attenuation is included in the differ ential stage, the additi onal 6dB are done by a following r esi stive di vider. This input is also configurable as two single-ended stereo inputs (see pin-out).
2.1.2 Full-differential stereo Input 2 (FD2)
The FD2-input has the same general structure as FD1, but with a programmable attenuation of 0 or 6dB embedded in the differential stage.
10/56
Page 11
TDA7405
2.1.3 Mono-differential Input 1 (MD1)
The MD1-input is designed as a basic differential stage w ith 56kΩ input-impedance. This input is configurable
as a single-ended stereo input (see pin-out).
2.1.4 Mono-differential Input 2 (MD2)
The MD2-input has the same topology as MD1, but without the possibility to configure it to single ended.
All single ended inputs offer an input impedance of 100kW. The AM-pin can be connected by software to the
input of the stereodecoder in order to use the AM-Noiseblanker and AM-High-Cut feature.
2.2 AutoZero
The AutoZero allows a reduction of the number of pins as wel l as external c omponents by c anceling any offset
generated by or before the In-Gain-stage (Please notice that externally generated offsets, e.g. generated
through the leakage current of the coupling capacitors, are not canceled).
The auto-zeroing is started every time the DATA-BYTE 0 is selected and needs max. 0.3ms for the alignment.
To avoid audible clicks the Audioprocessor is muted befor e the loudness stage during this time. The AutoZerofeature is only present in the main signal-path.
2.2.1 AutoZero for Stereodecoder-Selection
A special procedure is recommended for selecting the stereodecoder at the main input-selector to guarantee
an optimum offset-cancellation:
(Step 0: SoftMute or Mute the signal-path)
Step 1: Temporary deselect the stereodecoder at all input-selectors
Step 2: Configure the stereodecoder via IIC-Bus
Step 3: Wait 1ms
Step 4: Select the stereodecoder at the main input-selector first
The root cause of this procedure is, that after muting the stereodecoder (Step 1), the internal stereodecoder
filters have to settle in order to perform a proper offset-cancellation.
2.2.2 AutoZero-Remain
In some cases, for example if the µP is executing a refresh cycle of the IIC-Bus-programming, it is not useful
to start a new AutoZero-action because no new source is selected and an undesired mute would appear at the
outputs. For such applications the A631 could be switched in the AutoZero-Remain-Mode (Bit 6 of the subaddress-byte). If this bit is set to high, the DATABYTE 0 could be loaded without invoking the AutoZero and the
old adjustment-value remains.
2.3 Pause Detector / MUX-Output
The pin number 40(Pause/MUX) is configurable for two different functions:
1. During Pause-Detector OFF this pin is used as a mono-output of the main input-selector. This signal is
often used to drive a level-/equalizer-display on the carradio front-panel.
2. During Pause-Detector ON the pin is used to define the time-constant of the detector by an external capacitor.
The pause-detector is driven by the internal stereodecoder-outputs in order to use pauses in the FM-signal for
alternate-frequency-jumps. If the signal-level of both stereodecoder channels is outside the programmed voltage-window, the external capacitor is abruptly dis charged. Inside the pause-condition the capacitor is sl owly recharged by a constant current of 25µA. The pause information is also available via IIC-Bus (see IIC-Bus
programming).
11/56
Page 12
TDA7405
2.4 Loudness
There are four parameters programmable in the loudness stage:
2.4.1 Attenuation
Figure 2 shows the attenuation as a function of frequency at fP = 400Hz.
Figure 2. Loudness Attenuation @ fP = 400Hz
0.0
-5.0
-10.0
dB
-15.0
-20.0
-25.0
10.0100.01.0K10.0K
Hz
2.4.2 Peak Frequency
Figure 3 shows the four possible peak-frequencies at 200, 400, 600 and 800Hz
Figure 3. Loudness Center frequencies @ Attn. = 15dB
0.0
-5.0
dB
-10.0
12/56
-15.0
-20.0
10.0100.01.0K10.0K
Hz
Page 13
2.4.3 Loudness Order
Different shapes of 1st and 2nd-Order Loudness
Figure 4. 1st and 2nd Order Loudness @ Attn. = 15dB, fP=400Hz
TDA7405
0.0
-5.0
dB
-10.0
-15.0
-20.0
10.0100.01.0K10.0K
Hz
2.4.4 Flat Mode
In flat mode the loudness stage works as a 0dB to -19dB attenuator.
2.5 SoftMute
The digitally controlled SoftMute stage allows muting/demuting the signal with a I2C-bus programmable slope.
The mute process can either be activated by the SoftMute pin or by the I2C-bus. This slope is reali zed in a special S-shaped curve to mute slow in the critical regions (see Figure 5). For timing purposes the Bit of the I2Cbus output register is set to 1 from the start of muting until the end of de-muting.
Figure 5. Softmute-Timing
Note: Pleas e notice that a s tarted Mute -action is alway s terminated and could not be interrupt ed by a change of t he m ute -signal.
13/56
Page 14
TDA7405
2.6 SoftStep-Volume
When the volume-level is changed audible clicks could appear at the output. The root cause of those clicks
could either be a DC-Offset before the volume-stage or the sudden change of the envelope of the audiosignal.
With the SoftStep-feature both kinds of clicks could be reduced to a minimum and are no more audible. The
blend-time from one step to the next is programmable in four steps.
Figure 6. SoftStep-Timing
1dB
0.5dB
SS Time
-0.5dB
-1dB
Note: For steps more than 0.5dB the SoftStep mode should be deactivated because it could generate a hard 1dB step during the blend-time
2.7 Bass
There are four parameters programmable in the bass stage:
2.7.1 Attenuation
Figure 7 shows the attenuation as a function of frequency at a center frequency of 80Hz.
Figure 7. Bass Control @ fC = 80Hz, Q = 1
15.0
10.0
5.0
dB
0.0
-5.0
14/56
-10.0
-15.0
10.0100.01.0K10.0K
Hz
Page 15
2.7.2 Center Frequency
Figure 8 shows the eight possible center frequencies 60, 70, 80, 90, 100, 130, 150 and 200Hz.
Figure 8. Bass center Frequencies @ Gai n = 14dB, Q = 1
15.0
12.5
10.0
7.5
dB
5.0
2.5
0.0
TDA7405
10.0100.01.0K10.0K
Hz
2.7.3 Quality Factors
Figure 9 shows the four possible quality factors 1, 1.25, 1.5 and 2.
Figure 9. Bass Quality factors @ Gain = 14dB, fC = 80Hz
15.0
12.5
10.0
7.5
5.0
2.5
0.0
10.0100.01.0K10.0K
15/56
Page 16
TDA7405
2.7.4 DC Mode
In this mode the DC-gain is increased by 4.4dB. In addition the programmed center frequency and quality factor
is decreased by 25% which can be used to reach alternative center frequencies or quality factors.
Figure 10. Bass normal and DC Mode @ Gain = 14dB, fC = 80Hz
15.0
12.5
10.0
7.5
5.0
2.5
0.0
10.0100.01.0K10.0K
Note: The center frequency, Q and DC- m ode can be set full y independently.
2.8 Treble
There are two parameters programmable in the treble stage:
2.8.1 Attenuation
Figure 11 shows the attenuation as a function of frequency at a center frequency of 17.5kHz.
Figure 11. Tr ebl e C ontrol @ fC = 17.5kHz
15.0
10.0
5.0
0.0
-5.0
16/56
-10.0
-15.0
10.0100.01.0K10.0K
Page 17
2.8.2 Center Frequency
Figure 12 shows the four possible center frequencies 10k, 12.5k, 15k and 17.5kHz.
Figure 12. Treble Center Frequencies @ Gain = 14dB
15.0
12.5
10.0
7.5
5.0
2.5
0.0
TDA7405
10.0100.01.0K10.0K
2.9 EQ-Filter
There are two EQ-Filters present in the A631: one for the High-Frequency-Range and one for the Low-Fr equency-Range with a certain overlap. They ar e programmable in c enter -frequeny ( 4 frequencies /octave) , in Q(4 settings) and i n Attenuation (1dB-steps). In addition several configurations are possible to use the filters i n the fr ontor rear-path.
Table 2. Gain, Center Frequency and Quality Factor of Equalizer Filters
ParameterMinMaxUnit
Gain-1515dB
Center Frequency Filter 163840Hz
Center Frequency Filter 23004000Hz
Quality Factor14
17/56
Page 18
TDA7405
2.9.1 Equalizer-Setup
The two Filters can be configured in mul tiple ways in order to cov er as most as possible appl ications. Both filter s
can be programmed to be either in the front- or in the rear-path, respectively. This featur e enables to have e.g.
the High-Filter in the front- and both filters in the rear-path.
Figure 13. Equalizer Configuration
2.9.2 Attenuation
Figure 14 shows the attenuation as a function of frequency at a center frequency of 625 Hz.
Figure 14. Gain/Attenuation of EQ-Filter
15
dB
10
5
0
-5
-10
-15
2e+01
1e+02
1e+03
1e+04
Hz
2e+04
18/56
Page 19
2.9.3 Frequencies
Figure 15 shows the different center frequencies of the EQ-Filter at 12 dB gain
Figure 15. Cen te r- Frequencies of EQ - Fi lter
14
dB
12
10
8
6
4
2
0
TDA7405
-2
2e+01
1e+02
1e+03
2.9.4 Q-Factor
Figure 16 shows the four possible quality factors 1, 2, 3 and 4.
Figure 16. Different Q-factors of Equalizer-Filter
14
dB
12
10
8
6
4
2
0
1e+04
2e+04
Hz
-2
2e+01
1e+02
Hz
2e+04
1e+041e+03
19/56
Page 20
TDA7405
2.10Compander
2.10.1Signal-Compression
A fully integrated signal-compressor with programmable Attack- and Decay-times is present in the A631 (see
Figure 17). The compander consists of a signal-level detection, an A/D-Converter plus adder and the normal
SoftStep-Volume-stage. First of all t he left and the right InGai n-signal is recti fied, respec tively, and the logar ithm
is build from the summed signal. The following low-pass smooth the output-signal of the logarithm-amplifier and
improves the low-frequency suppression. The low-pass output-voltage then is A/D- converted an added to the
current volume-word defined by the IIC-Bus. Assuming reference-level or higher at the compander input, the
output of the ADC is 0. At lower levels the voltage is increasing with 1 Bit/dB. It is obvious that with this configuration and a 0.5dB-step volume-stage the compression rate is fixed to 2:1 (1dB less at the input leads to 0.5dB
less at the outpu t). The inter nal refer ence-level of the compander is p rogrammable in three steps from 0.5V
to 2V
the programmed reference-level after the InGain-stage. E.g. at a configured refer ence-level of 0.5V
put of the InGain-stage has to have also 0.5V
. For a proper behavior of the compr ess ion-circui t it is mandatory to hav e at a 0dB input-sign al exactl y
RMS
at 0dB source-signal (Usually the 0dB for CD is defined as
RMS
RMS
the maximum possible signal-lev el). To adapt the external level to the internal reference-level the programmable
attenuation in the differential-stages and the InGain can be used.
Figure 17. Compander Block Diagram
RMS
the out-
2.10.2Anti-Clipping
In a second ap pli cation the compander-circuit can be used for a Anti-Clipping or Limiting function. In this case one of
the dedicated inputs (AM or MPin) is connected directly to the Clip-Detector of the Power-Amplifier. If no clipping is
detected, the open-collector output of the Power-Amplifier is highohmic and the input-voltage of the rectifier is V
REF
The level-detector interprets this as a very small signal and react s with the maximum programmed compander-gain.
In the application this gain has to be compensated by decreasing the volume wit h the same value in order to get the
desired output-level. In clipping situation the open-collect or-current generates a voltage-drop at the rectifier-input,
which forces the compander to decrease the gain until the clipping disappears .
It is even possible to run the compression-mode and the Anti-Clipping mode in par allel. In this case the maximum compander-Gain should be set to 29 dB.
20/56
.
Page 21
TDA7405
2.10.3Characteristic
To achieve the desired compression characteristic like shown below the volume has to be decreased by 4dB.
Figure 18. Compander Characteristic
0
-10
Output Level
-20
dB
-30
-40
-50
-60
0
-8dB
2:1
-38dB
15dB
-10-20-30-40-50-60
Input Level
dB
2.10 .4I² C -BUS-Timing
When the Compander is active a volume- word coming from this stage is added to the I2C-Bus volume-w ord
and the volume is changed with a soft slope between adjacent steps (SoftStep-stage). As mentioned in the description of this stage it is not recommended to change the volume during this slope. To avoid this behaviourwhile the Compander is working, and the volume has to be changed, the compander-hold-bit is implemented
(Bit 7 in the subaddress-byte). The recommended timing for changing the volume during compander-ON is the
following:
1. Set the compander-hold-bit
2. Wait the actual SoftStep-time
3. Change the volume
4. Reset the compander-hold-bit
The SoftStep-times are in compander- ON condition automatically adapted to the attac k-time of the Compander.
In the following table the related SoftStep-times are shown:
Attack-TimeSoftStep-Time
6ms0.16ms
12ms0.32ms
24ms0.64ms
48ms1.28ms
2.10.5AC-Coupling
In some applications additional signal manipulations are desired, for example surround-sound or more-bandequalizing. For this purpose an AC-Coupling is placed before the speaker-attenuators, w hich can be activated
or internally shorted by I²C-Bus. In short condition the input-signal of the speaker-attenuator is available at the
AC-Outputs. The input-impedance of this AC-Inputs is 50k
Ω
.
21/56
Page 22
TDA7405
2.10 .6O utput Selec tor
The output-selector allows to connect the main- or the second-source to the Front-, Rear-speaker-attenuator,
respectively. As an example of this programming the device is able to connect via software the main-source to
the back (rear) and the second-source to the front (see Figure 19).
Figure 19. Output Selector
2.10.7S peake r-Attenu ator and Mixing
A Mixing-stage is placed after each speaker-attenuator and can be set independently to mixing-mode. Having
a full volume for the Mix-signal the stage offers a wide flexibility to adapt the mixing levels.
Figure 20. Mixing Configuration
2.10.8Aud iopro cesso r Testi ng
During the Testmode, which can be activated by setting bit D0 of the stereodecoder testing-byte and the audioprocessor testing byte, several internal signals are available at the FD2R- pin. During this mode the input resistance of 100kOhm is disconnected from the pin. The internal signals available are shown in the Data-byte
specification.
22/56
Page 23
TDA7405
3STEREODECODER-PART
3.1 Features:
■
no external components necessary
■
PLL with adjustment free, fully integrated VCO
■
automatic pilot dependent MONO/STEREO switching
■
very high suppression of intermodulation and interference
■
programmable Roll-Off compensation
■
dedicated RDS-Softmute
■
Highcut- and Stereoblend-characterisctics programmable in a wide range
■
FM/AMNoiseblanker with several threshold controls
■
Multipath-detector with programmable internal/external influence
■
I2C-bus control of all necessary functions
Table 3. ELECTRICAL CHARACTERISTICS
V
= 9V, deemphasis time constant = 50µs, MPX input voltage V
S
modulation frequency = 1kHz, input gain = 6dB, T
= 27°C, unless otherwise specified.
amb
SymbolParameterTest ConditionsMin. Typ. Max.Unit
= 500mV (75kHz deviation),
MPX
V
MPX Input LevelInput Gain = 3.5dB0.51.25V
in
R
G
G
G
SVRRSupply Voltage Ripple RejectionV
Input Resistance70100130kΩ
in
Min. Input Gain1.53.54.5dB
min
Max. Input Gain8.5111 2.5dB
max
Step Resolution1.752.53.25d B
step
= 100mV, f = 1kHz55dB
ripple
αMax. Chan nel Separatio n3050d B
THDTotal Harmonic Distortionf
SN+
--------------
Signal plus Noise to Noise RatioA-weighted, S = 2V
N
=1kHz, mono0.020.3%
in
rms
8091dB
MONO/STEREO-SWITCH
V
PTHST1
V
PTHST0
V
PTHMO1
V
PTHMO0
Pilot Threshold Voltagefor Stereo, PTH = 1101525mV
Pilot Threshold Voltagefor Stereo, PTH = 0152535mV
Pilot Threshold Voltagefor Mono, PTH = 171217mV
Pilot Threshold Voltagefor Mono, PTH = 0101925mV
PLL
rms
∆f/fCapture Range0.5%
23/56
Page 24
TDA7405
Table 3. ELECTRICAL CHARACTERISTICS
V
= 9V, deemphasis time constant = 50µs, MPX input voltage V
S
modulation frequency = 1kHz, input gain = 6dB, T
(continued)
= 27°C, unless otherwise specified.
amb
= 500mV (75kHz deviation),
MPX
SymbolParameterTest ConditionsMin. Typ. Max.Unit
DEEMPHASIS and HIGHCUT
τ
DeempF
M
M
FM
τ
DeempA
M
M
AM
Deemphasis Timeconstants FMV
Highcut Timeconst ant Multiplier FM
Deemphasis Timeconstants AMV
Highcut Timeconstant Multiplier AMV
LEVEL
V
LEVEL
V
LEVEL
V
LEVEL
V
LEVEL
LEVEL
V
LEVEL
V
LEVEL
V
LEVEL
LEVEL
>> V
>> V
>> V
>> V
<< V
>> V
>> V
>> V
>> V
<< V
HCH
HCH
HCH
HCH
HCL
HCH
HCH
HCH
HCH
HCL
255075µs
4462.580µs
5075100µs
70100130µs
3
37.5µs
47µs
56µs
75µs
3.7
REF5VInternal Reference Voltage4.755.3V
L
L
maxs
L
Gstep
VSBL
VSBL
VSBL
VHCH
VHCH
VHCH
VHCL
VHCL
VHCL
min. LEVEL Gain-101dB
min
max. LEVEL Gain567dB
LEVEL Gain Step Resolutionsee section 2.70.20.40.6dB
Min. Voltage for Monosee section 2.8172023%REF5V
min
Max. Voltage for Monosee section 2.8627078%REF5V
max
Step Resolutionsee section 2.81.63.35.0%REF5V
step
Min. Voltage for NO Highcutsee section 2.9374247%REF5V
min
Max. Voltage for NO Highcutsee section 2.9586674%REF5V
max
Step Resolutionsee section 2.94.28.412.6%REF5V
step
Min. Voltage for FULL High cutsee section 2.9151719%VHCH
min
Max. Voltage for FULL High cutsee section 2.9293337%VHCH
max
Step Resolutionsee section 2.92.14.26.3%REF5V
step
Carrier and harmonic suppression at the output
α19Pilot Signalf=19 kHz4050d B
α38Subcarrier f=38kHz75dB
α57Subcarrier f=57kHz62dB
α76Subcarrier f=76kHz90dB
24/56
Page 25
TDA7405
Table 3. ELECTRICAL CHARACTERISTICS
V
= 9V, deemphasis time constant = 50µs, MPX input voltage V
All parameters mesured in FM mode if not otherwise specified.
SymbolParameterTest conditionsMin.Typ.Max.Unit
V
V
V
RECT
RECT
T
T
Rectifier VoltageV
RECT
Deviation dependent
DEV
Rectifier Voltage
Fieldstrength
FS
controlled Rectifier
Voltage
Suppression Pulse Duration FMSignal HOLDN
SFM
Suppression Pulse Duration AMSignal HOLDN
SAM
=0mV0.50.91.3V
MPX
V
=50mV, f=150kHz1.51.72.1V
MPX
V
=200mV, f=150kHz22.52.9V
MPX
meas.with
=500mV
V
MPX
(75kHz dev.)
meas.with
=0mV,
V
MPX
V
<< V
LEVEL
(fully mono)
in Testmode
in Testmode
SBL
11
10
01
00
11
10
01
00
00
01
10
11
00
01
10
11
0.5
0.9
1.7
2.5
0.5
0.9
1.7
2.1
0.9 (off)
1.2
2.0
2.8
0.9 (off)
1.4
1.9
2.4
38
25.5
32
22
1.2
800
1.0
640
1.3
1.5
2.3
3.1
1.3
1.5
2.3
3.1
V
V
V
V
OP
OP
OP
OP
V
V
V
V
µs
µs
µs
µs
ms
µs
ms
µs
V
RECTADJ
SR
V
ADJMP
R
G
AMIF,min
Noise rectifier discharge
adjustment
Noise rectifier
PEAK
charge
Noise rectifier adjustment
through
Multipath
AM IF Input resistance355065kOhm
AMIF
min. gain AM IFSignal AM-RECTIFIER in
Signal PEAK in
Testmode
Signal PEAK in
Testmode
Signal PEAK in
Testmode
00
01
10
11
0
1
00
01
10
11
0.3
0.8
1.3
2.0
10
20
0.3
0.5
0.7
0.9
6dB
Testmode
G
G
f
AMIF,max
AMIF,step
AMIF,min
max. gain AM IF20dB
step gain AM IF2dB
min. fc AM IFSignal AM-RECTIFIER in
14kHz
Testmode
f
AMIF,max
max. fc AM IF56kHz
5.All thresholds are measured using a pulse with TR = 2 µs, THIGH= 2 µs and TF = 10 µ s. The repe ti ti on rate mu st not increa se the PE AK voltage.
V/ms
mV/µs
V/ms
27/56
Page 28
TDA7405
Figure 21. Noiseblanker Test-Pulse
V
in
DC
V
op
T
R
Figure 2 2. Trigger Threshold vs. VPEAK
VTH
MIN. TRIG. THRESHOLDNOISE CONTROLLED TRIG . THRESHOLD
T
HIGH
T
F
Time
260mV (00)
220mV (01)
180mV (10)
140mV (11)
28/56
8 STEPS
65mV
30mV
1.5V0.9V
V
PEAK [V]
Page 29
Figure 23.
V
PEAK
[V ]
OP
2. 8
2. 0
1. 2
0.9
TDA7405
00
01
10
D ete cto r off ( 1 1)
20
32.54575
Figure 24. Fiel dst renth Control l ed Tri gger Adjustment
V
PEAK
MONOSTEREO
≈
3V
NOISE
ATC_SB OF F (11)
noisy signal
2.4V (00)
1.9V (01)
1.4V (10)
good signal
DEVIATION [KHz ]
0.9V
E'
29/56
Page 30
TDA7405
MULTIPATH DETECTOR
Features:
■
internal 19kHz band-pass filter
■
programmable band-pass- and rectifier-gain
■
selectable internal influence on Stereoblend and/or Highcut
Table 5. ELECTRICAL CHARACTERISTICS
SymbolParameterTest ConditionsMin.Typ.Max.Unit
f
CMP
Center frequency of MultipathBandpass
G
BPMP
G
RECTMP
I
CHMP
I
DISMP
Bandpass GainG16dB
Rectifier GainG17.6dB
Rectifier Charge Current0.25
Rectifier Discharge Current4mA
Quality Detector
A
Multipath
Influence Factor
stereodecoder locked on Pilottone19kHz
G212dB
G316dB
G418dB
G24.6dB
G30dB
0.5
00
01
10
11
0.70
0.85
1.00
1.15
µA
4FUNCTIONAL DESCRIPTION OF STEREODECODER
Figure 25. Blockdiagram of the Stereodecoder
30/56
Page 31
TDA7405
The stereodecoder-part of the A631 (see Figure 25) contains all functions necessary to demodulate the MPXsignal like pilottone-dependent MONO/STEREO-switching as well as "stereoblend" and "highcut". Adaptations
like programmable input gain, rol l-off compensation, selectable deemphasis time constant and a programmable
fieldstrength input allow to use different IF-devices.
4.1 Stereodecoder-Mute
The A631 has a fas t and easy to contr ol RDS-Mute function which is a combination of the audi oprocessor's SoftMute and the high-ohmic mute of the s t ereodecoder. If the stereodecoder is selec ted and a SoftMute command
is sent (or activ ated through the SM-pin) the stereod ecoder will be set automatical ly to the high-ohmi c mute condition after the audio-si gnal has been softmuted. Hence a checking of alternate frequenc ies could be performed.
Additionally the PLL can be set to "Hold"-mode, which disables the PLL input during the mute time. To release
the system from the mute condition simply the unmute-command must be sent: the stereodecoder is unmuted
immediately and the audioprocessor is softly unmuted. Figure 26 shows the output-signal VO as well as the internal stereodecoder mute sign al. This influence of S oftMute on th e stereodecoder mute can b e switched off by
setting bit 3 of the SoftMute byte to "0". A stereodecoder mute command (bit 0, stereodecoder byte set to "1")
will set the stereodecoder in any case independently to the high-ohmic mute state.
If any other source than the stereodecoder is selected the decoder r emains muted and th e MPX-pin is c onnected to V
should be applied.
Figure 26. Signals during stereodecoder's SoftMute
to avoid any discharge of the coupl ing capac itor through leakage c ur rents. No further mute c ommand
ref
Figure 27. Signal-Control via SoftMute-Pin
31/56
Page 32
TDA7405
4.2 InGain + Infilter
The InGain stage allows to adjust the MPX-signal to a magnitude of about 1Vrms internally which is the recommended value. The 4.th order input filter has a corner frequency of 80kHz and is used to attenuate spikes and
noise and acts as an anti-aliasing filter for the following switch capacitor filters.
4.3 Demodulator
In the demodulator block the left and the right channel are separated from the MPX-signal. In this stage also the
19-kHz pilottone is c ancelled. For reaching a high channel separ ation the A631 offer s an I
roll-off adjustment which is able to compensate the lowpass behavior of the tuner section. If the tuner's attenuation at 38kHz is in a range from 7.2% to 31.0% the A631 needs no external network in front of the MPX-pin.
Within this range an adjustment to obtain at least 40 dB channel separation is possible. The bits for this adjustment are located together wi th the fieldstrength adjustment in one byte. This gives the possibility to perform an
optimization step during the production of the carradio where the channel separation and the fieldstrength control are trimmed. The setup of the Stereoblend characteristics which is programmable in a wide range is described in 2.8.
4.4 Deemphasis and Highcut
.The deemphasis-lowpass all ows to cho ose a time co nstant between 37.5 and 100µs. The high cut control range
will be 2 x t
Deemp
or 2.7 x t
dependent on the selected time constant (see programming section). The bit
Deemp
D7 of the hightcut-byte will shift timeconstant and range.
Inside the highcut control range (between VHCH and VHCL) the LEVEL signal is converted into a 5 bit word
which controls the lowpass time constant between t
Deem p.. .3
(3.7) x t
. Thereby the resolution will remain
Deemp
always 5 bits independently of the absolute voltage range between the VHCH- and VHCL-values. In addition
the maximum attenuation can be fixed between 2 and 10dB.
2
The highcut function can be switched off by I
C-bus (bit D7, Highcut byte set to "0").
The setup of the highcut characteristics is described in 4.9.
2
C-bus programmable
4.5 PLL and Pilottone-Detector
The PLL has the task to lock on the 19kHz pilottone during a stereo-transmission to allow a correct demodulation. The included pilottone-detector enables the demodulation if the pilottone reaches the selected pilottone
threshold VPTHST. Two different thresholds are available. The detector output (signal STEREO, see Blockdiagram) can be checked by reading the status byte of the A631 via I
2
C-bus. During a Softmute the PLL can be
set into "Hold"-mode which freezes the PLL's state (bit D4, Softmute byte). After releasing the Softmute the PLL
will again follow the input signal only by correcting the phase error.
4.6 Fieldstrength Control
The fieldstrength input is used to control the highcut- and the stereoblend-function. In addition the signal can be
also used to control the noiseblanker thresholds and as input for the multipath detector. These additional functions are described in sections 3.3 and 4.
4.7 LEVEL-Input and -Gain
To suppress undesired high frequency modulation on the highcut- and ster eoblend-control signal the LEVEL
signal is lowpass filtered firs tly. The filter is a combination of a 1.st-order RC-lowpass at 53kHz (working as antialiasing filter) and a 1.st-order switched -capacitor-low pass at 2.2k Hz. The second s tage is a programmable gain
stage to adapt the LEVEL signal internally to different IF-devices (see Testmode section 5: LEVELHCC). The
gain is widely programmable in 16 steps from 0dB to 6dB (step=0.4dB). These 4 bits are located together wi th
the Roll-Off bits i n the "Stereodec oder-Adjus tment"-byte to si mplify a possible adjustment during the produc tion
of the carradio. This signal controls directly the Highcut stage whereas the signal is filtered again (fc=100Hz)
before the stereoblend stage (see Figure 25).
32/56
Page 33
TDA7405
4.8 Stereoblend Control
The stereoblend control bloc k con ver ts the internal LEV EL-volt age (LEVELSB) i nto an demodu lator compati ble
analog signal which is used to control the channel separation betwe en 0dB and the maxi mum separation. Internally this control range has a fixed upper limit which is the inter nal reference voltage REF5V. The lower limit can
be programmed between 20 and 70% of REF5V in 3.3% steps (see Figure 28, Figure 29).
To adjust the external LEVEL-voltage to the internal range two values must be defined: the LEVEL gain LG and
VSBL (see Figure 29). At the point of full channel separation the external level signal has to be amplified that
internally it be comes equal to REF5V. The s econd point (e.g. 10dB channel sep.) is then adjusted w ith the VSBL
voltage.
Figure 28. Internal stereoblend characteristics
The gain can be programmed through 4 bits in the "Stereodecoder-Adjustment"-byte. All necessary internal reference voltages like REF5V are derived from a bandgap circuit. Therefore they have a temperature coefficient
near zero.
Figure 29. Relation between internal and external LEVEL-voltages for setup of Stereoblend
70
20
4.9 Highcut Control
The highcut control setup is similar to the stereoblend control setup : the starting point VHCH can be set with 2
bits to be 42, 50, 58 or 66% of R EF5V whereas the range can be set to be 17, 22, 28 or 33% of VHCH (see
Figure 30).
33/56
Page 34
TDA7405
Figure 30. Highcut characteristics
5FUNCTIONAL DESCRIPTION OF THE NOISEBLANKER
In the automotive environment the MPX-signal as well as the AM-signal is disturbed by spikes produced by the
ignition and other radiating sources li ke the wiper-motor. The aim of the noiseblanker part is to cancel the audible influence of the spikes. Therefore the output of the stereodecoder is held at the actual voltage for a time
between 22 and 38µs in FM (370 and 645µs in AM-mode). The blockdiagram of the noiseblanker is given in
Figure 31.
Figure 31. Block diagram of the noiseblanker
In a first stage the spikes must be detected but to avoid a wrong triggering on high frequency (white) noise a
complex trigger control is implemen ted. Behind the triggerstage a pulse former generates the "blanking "-pulse.
5.1 Trigger Path FM
The incoming MPX signal is highpass-filtered, amplified and rectified. This second order highpass-filter has a
corner-frequency of 140kHz. The rectified signal, RECT, is integrated (lowpass filtered) to generate a signal
called PEAK. The DC-charge/discharge behaviour can be adjusted as well as the transient behaviour(MP-discharge control). Also noise with a frequency 140kHz increases the PEAK voltage. The PEAK voltage is fed to
a threshold generator, which adds to the PEAK-voltage a DC-dependent threshold VTH. Both signals, RECT
and PEAK+VTH are fed to a comparator which triggers a re-triggerable monoflop. The monoflop's output activates the sample-and-hold circuits in the signalpath for the selected duration
5.2 Noise Co ntrolled Threshold Adjus tment (NCT)
There are mainly two independent possibilities for programming the trigger threshold:
1. the low threshold in 8 steps (bits D
2. and the noise adjusted thr eshold in 4 steps (bits D
34/56
to D3 of the noiseblanker-byte I)
1
and D5 of the noiseblanker-byte I, see Figure 22).
4
Page 35
TDA7405
The low threshold is active in combinati on with a good MPX signal w ithout nois e; the PEAK voltage is les s than
1V. The sensitivity in this operation is high. If the MPX signal is noisy (low fieldstrength) the PEAK voltage increases due to the higher noise, which is als o recti fied. With inc reasing of the PEAK volta ge the trigger thr eshold
increases, too. This gain is programmable in 4 steps (see Figure 22).
5.3 Additional Threshold Control Mechanism
5.3.1 Automatic Threshold Control by the Stereob lend vo ltage
Besides the noise controlled threshold adjustment there is an additional possibility for influencing the trigger
threshold which depends on the stereoblend control.
The point where the MPX signal star ts to become nois y is fixed by the RF part. Ther efore als o the starting point
of the normal noise-controlled trigger adjustment is fixed (Figure 24). In some cases the behavior of the noiseblanker can be improved by increasing the threshold even in a region of higher fieldstrength. Sometimes a
wrong triggering occurs for the MPX signal often shows distortion in this range which can be avoided even if
using a low threshold. Because of the overlap of this range and the range of the stereo/mono transition it can
be controlled by stereoblend. This increase of the threshold is programmable in 3 steps or switched off.
5.3.2 Over Deviation Detector
If the system is tuned to stations with a high deviation the noiseblanker can trigger on the higher frequencies
of the modulation or distortion. To avoid this behavior, which causes audible noise in the output signal, the
noiseblanker offers a deviation-dependent threshold adjustment. By rectifying the MPX signal a further signal
representing the actual deviation is obtained. It is used to increase the PEAK voltage. Offset and gain of this
circuit are programmab le in 3 s teps with the bi ts D
off the detector, see Figure 23).
and D7 of the noiseblanker-byte I ( bit c ombination '00' turns
6
5.3.3 Multipath-Level
To react on high repetitive spikes caused by a Multipath-situation, the discharge-time of the PE AK voltage can
be decreased depending on the vol t age-level at pin M Pout. The A63 1 offers a linear as wel l as a threshold driven control. The linear influence of the Multipath-Level on the PEAK-signal (D
a discharge slewrate of 1V/ms
switches on the 18kOhm discharge if the Multipath-Level is below 2.5V (D
1
The slewrate is measured with R
1
. The second possibility is to activate the threshold driven discharge which
Discharge
= nfinite and V
MPout
= 2.5V
of Multipath-Control-Byte) gives
7
of noiseblanker-byte II-byte).
7
5.3.4 AM mode of the Noiseblanker
The A631 noiseblanker is also suitable for AM noise canceling. The detector uses in AM mode the 450kHz unfiltered IF-output of the tuner for spike detection. A combination of programmable gain-stage and lowpass-filter
forms an envelope detector which drives the noiseblanker's input via a 120 kHz 1.st order highpass. In order to
blank the whole spike in AM mode the hold-times of the sample and hold circuit are much longer then in FM
(640µs to 1.2ms). All threshold controls can be used like in FM mode.
5.4 Functional Description of the Multipath-Detector
Using the Multipath-Detector the audible effects of a multipath condition can be minimized. A multipath-condition is detected by rectifyi ng the spectr um around 19kH z in the fieldstrength si gnal. An exter nal capacitor is used
to define the attack- and decay-times for the Stereoblend (see blockdiagram, Figure 32). Due to the very small
charge currents this capacitor should be a low leakage current type (e.g. ceramic). U sing this configuration an
adaptation to the user's requirement is possible w ithout effecting the "normal" fieldstrength input (LEVEL) for
the stereodecoder. This application is given in Figure 32. Another (internal) time constant is used to c ontro l the
Highcut through the multipath detector
Selecting the "internal influence" in the configuration byte the Stereo-Blend and/or the Highcut is automatically
invoked during a multipath condition according to the voltage appearing at the MP_OUT-pin.
35/56
Page 36
TDA7405
Figure 32. Bl ock d iagram of the M ul tipath-Detect or
5.5 Quality Detector
The TDA7405 offers a quality detector output which gives a voltage representing the FM-reception conditions.
To calculate this voltage the MPX-noise and the multipath- detec tor output are summed according to the following formula :
V
Qual
= 1.6 (V
-0.8 V)+ a (REF5V-V
Noise
Mpout
).
The noise-signal is the PEAK-signal without additional influences (see noiseblanker description). The factor 'a'
can by programmed to 0.7 .... 1.15. The output is a low impedance output able to drive ex ternal circui try as well
as simply fed to an AD-converter for RDS applications.
36/56
Page 37
TDA7405
5.6 Dual-MPX Mode
The TDA7405 is able to support a twin tuner con cept v ia the Dual-MPX-Mode. In thi s configuration the MPX-pin
and the MD2G-pin ar e acti ng as MPX1 and MPX2 inputs . The D C-Voltage at the MD2-pin contr ols whether one
or both MPX-signals are used to decode the stereo FM-signal. It is designed as a window-comparator with the
characteristic shown in Figure 33 (Please note that the thresholds have a hysteresis of 500mV). In this mode
the stereodecoder highohmic-mute mutes both inputs in parallel.
For use with front-ends which do not support a weak-signal-mute function the TDA7405 offers this feature as
well. If this mode is enabled the pins 26 and 27 are used as an AC-coupling behind the Mute-Stage (see blockdiagram). In parallel pin 12 (AMIF) is switched internally to realize a mute time-constant with fast attack - and
slow decay-time.
During the Testmode, which can be activated by setting bit D0 and bit D1 of the stereodecoder testing-byte,
several internal signals are available at the FD2R+ pin. During this mode the input resistance of 100kOhm is
disconnected from the pin. The internal signals available are shown in the Data-byte specification.
38/56
Page 39
6I²C BUS INTERFACE
6.1 Interface Protocol
The interface protocol comprises:
-a start condition (S)
-a chip address byte (the LSB bit determines read / write transmission)
-a subaddress byte
-a sequence of data (N-bytes + acknowledge)
-a stop condition (P)
CHIP ADDRESS SUBADDRESS DATA 1 ... DATA n
TDA7405
MSB
S1000110R/WACKCAZIAAAAAACKDATAACK P
LSBMSBLSBMSBLSB
S = Start
R/W ="0" -> Receive-Mode (Chip could be programmed by µP)
"1" -> Transmission-Mode (Data could be received by µP)
ACK = Acknowledge
P = Stop
MAX CLOCK SPEED 500kbits/s
6.1.1 Auto increment
If bit I in the subaddress byte is set to "1", the autoincrement of the subaddress is enabled.
6.1.2 TRANSMITTED DATA (send mode)
MSBLSB
ST
XXXXXP
SM
SM = Soft mute activated
ST = Stereo
P = Pause
X = Not Used
The transmitted data is automatically updated after each ACK. Transmission can be repeated without new chi-
paddress.
6.1.3 Reset Condition
A Power-On-Reset is invoked if the Supply-Voltage is below than 3.5V. After that the following data is written
automatically into the registers of all subaddresses :
MSBLSB
1
111111
0
The programming after POR is marked bold-face / under lined in the programming tables.
With this programming all the outputs are muted to V
Note: 1. Using the Mu l t i p ath Time-C onstants for St ereo-Bland and High-Cut
Table 34. Testing Stereodecoder(31)
MSBLSB
D
Note :
D
7
0
1
This byte is used for testin g or evaluation purposes onl y and must not set to other valu es than "
D
D
D
D
D
6
5
4
3
2
D
1
0
Multipath Influence on High-Cut
0
On
1
Off
Multipath Influence on Stereo-Blend
0
1
0
1
1
x
On
Off
Level-Input over Multipath-Detector
1
On
x
Off
Dual MPX Mode
0
1
On
Off
Weak-Signal-Mute
0
1
on
off
WSMute Threshold
0
1
0.3 V
0.5V
WS-Mute-Depth
0
1
0
1
D
6
D
5
D
D
4
3
D
2
1
-20 dB
-24 dB
-28 dB
-32 dB
D
0
Main Testmode
0
off
on
1
Stereodecoder Testmode
0
1
off
on
Testsignals
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
1
0
1
1
0
1
0
1
0
1
0
1
1
1
1
1
1
1
1
1
1
0
1
1
0
0
0
1
1
0
1
1
0
0
0
1
1
0
1
1
0
0
0
1
1
0
1
1
F228
NB threshold
Level for Stereo-Blend
Pilot magnitude
VHCCL
Pilot threshold
VHCCH
REF5V
HOLDN
NB Peak
AM-Rectifier
VCOCON; VCO Control Voltage
VSBL
Pilot threshold
Level for High-Cut
REF5V
Audioprocessor Oscillator
Information furnished is believed to be accurate and reliable. However, STMicroelectronics assumes no responsibility for the consequences
of use of such information nor for any infringement of patents or other rights of third parties which may result from its use. No license is granted
by implic ation or oth erwise under any patent or patent rights of STMicroelectronic s. Specific ations mentioned in this public at ion are subj ect
to change without notice. This publication supersedes and replaces all information previously supplied. STMicroelectronics products are not
authorized for use as critical components in life support devi ces or systems wi thout express written ap proval of STMi croelectro nics.
The ST logo is a registered trademark of STMicroelectronics
2001 STMi croelectr oni cs - All Righ ts Reserved
Australia - Brazil - Canada - Chi na - F i nl and - France - Germany - Hong Kong - Indi a - Israel - Ital y - J apan -Malaysia - Malta - Morocco -
Singap ore - Spain - Sw eden - Switze rl and - United K i ngdom - Unit ed S tates.
STMicroelectronics GROUP OF COMPANIES
http://www.s t. com
56/56
Loading...
+ hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.