TDA5212

Wireless Control Components
ASK / FSK Single Conversion Receivers
TDA 521X Version 1.1
Application Note February 2004
Revision History
Current Version: 1.1 as of 11.02.04
Previous Version: Page
(in previous
Version)
Page
(in current
Version)
Subjects (major changes since last revision)
4-36
4-36 - 4-37
calculation changes of the slicing level time constant
This Application Note describes the Single Conversion Receiver TDA 5210 in design step A3, TDA5211 in design step B4 and TDA5212 in design step C3.
ABM®, AOP®, ARCOFI®, ARCOFI®-BA, ARCOFI®-SP, DigiTape®, EPIC®-1, EPIC®-S, ELIC®, FALC®54, FALC®56, FALC®-E1, FALC®-LH, IDEC®, IOM®,
IOM®-1, IOM®-2, IPAT®-2, ISAC®-P, ISAC®-S, ISAC®-S TE, ISAC®-P TE, ITAC®, IWE®, MUSAC®-A, OCTAT®-P, QUAT®-S, SICAT®, SICOFI®, SICOFI®2, SICOFI®-4, SICOFI®-4µC, SLICOFI® are registered trademarks of Infineon Technologies AG.
ACE™, ASM™, ASP™, POTSWIRE™, QuadFALC™, SCOUT™ are trademarks of Infineon Technologies AG.
Edition 02.04
Published by Infineon Technologies AG,
Balanstraße 73,
81541 München
© Infineon Technologies AG February 2004.
All Rights Reserved.
Attention please!
As far as patents or other rights of third parties are concerned, liability is only assumed for components, not for applications, processes and circuits implemented within components or assemblies.
The information describes the type of component and shall not be considered as assured characteristics.
Terms of delivery and rights to change design reserved.
Due to technical requirements components may contain dangerous substances. For information on the types in question please contact your nearest
Infineon Technologies Office.
Infineon Technologies AG is an approved CECC manufacturer.
Packing
Please use the recycling operators known to you. We can also help you – get in touch with your nearest sales office. By agreement we will take packing
material back, if it is sorted. You must bear the costs of transport.
For packing material that is returned to us unsorted or which we are not obliged to accept, we shall have to invoice you for any costs incurred.
Components used in life-support devices or systems must be expressly authorized for such purpose!
Critical components1 of the Infineon Technologies AG, may only be used in life-support devices or systems2 with the express written approval of the
Infineon Technologies AG.
1 A critical component is a component used in a life-support device or system whose failure can reasonably be expected to cause the failure of that lifesupport device or system, or to affect its safety or effectiveness of that device or system.
2 Life support devices or systems are intended (a) to be implanted in the human body, or (b) to support and/or maintain and sustain human life. If they
fail, it is reasonable to assume that the health of the user may be endangered.
1
Table of Contents
1 Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
i
2 Product Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2-1
2.1
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2-2
2.2
Product Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2-2
2.3
Package Outlines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2-4
3 Pin Configuration and Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-1
3.1
Pin Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-2
3.2
Pin Definition and Function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-3
4 Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-1
4.1
Low-Noise Amplifier (LNA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-2
4.2
AGC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-8
4.3
Mixer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-12
4.4
Overall Performance of the Front End . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-12
4.5
PLL Synthesizer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-14
4.6
Reference Oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-16
4.7
IF Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-24
4.7.1 IF Amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-24
4.7.2 IF Filtering. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-24
4.8
ASK/FSK Switch Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-27
4.8.1 FSK Mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-27
4.8.2 ASK Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-29
4.9
Demodulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-30
4.9.1 ASK Demodulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-30
4.9.2 FSK Demodulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-32
4.10
Data Filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-34
4.11
Data Slicer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-35
4.12
Principle of the Precharge Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-44
4.13
Quiet Data Output during no Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-48
4.14
Decoder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-51
4.15
Settling Time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-51
4.16
Spurious Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-54
4.17
Sensitivity Measurements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-55
4.17.1 Test Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-55
4.17.2 Sensitivity of RKE Receivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-56
4.17.3 Dependance of the ambient Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-57
4.17.4 Sensitivity depending on the IF Filter Bandwidth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-58
4.17.5 Dependance of Data Filter Bandwidth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-59
5 Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5-1
5.1
Test Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5-2
5.2
Test Board Layouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5-3
5.3
Bill of Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5-5
6 List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
i
7 List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
i
2
Product Overview
Contents of this Chapter
2.1
2.2
2.4
Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
Product Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
Package Outlines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-5
TDA 521X
Product Overview
2.1 Abstract
This application note describes the operation of the TDA 5210 and TDA5211
evaluation boards. It demonstrates the design of a low-cost receiver for applications in wireless I.S.M. data communication systems. Various application
considerations are presented to assist system designers in implementing the
device.
The application board of the TDA5210 can be operated in one of the assigned
frequency bands for short-range devices (SRD) at either 434MHz or 868MHz,
the application board of the TDA5211 at 315MHz. The receivers have been
optimized for single-channel operation in systems using both amplitude shift
key (ASK) modulation and frequency shift key (FSK) modulation. The board
complies with the I-ETS 300 220 regulations.
2.2 Product Description
The TDA 5210/TDA5211 has been implemented in a 25GHz silicon bipolar process. It supports all low-power device (LPD) wireless applications with ASK and
FSK modulated signals with data rates of up to 120kbit/s. The maximum achievable data rate also depends on the IF filter bandwidth and the local oscillator
tolerance values.
As can be seen from the block diagram in Figure 2-1, the basic concept of the
TDA5210 is a single conversion receiver with an on-chip fully integrated PLL
frequency synthesizer and an IF of nominally 10.7MHz. The 10.7MHz IF was
selected because of the availability of low-cost ceramic filters in a variety of
bandwidths between 60kHz and 280kHz. The user is free to select other IFs
and/or filters that are compatible with the 3MHz - 25MHz bandwidth provided by
the 90dB limiting IF-strip. The IF-limiter provides a received signal strength indication (RSSI) with over 80dB dynamic range. The RSSI output is used as the
demodulator for the ASK signals. In case of FSK a PLL is demodulating the signals. The output of the ASK demodulator is DC-coupled internally to the data
slicer. An on-chip 2nd order low-pass filter is provided at the demodulator output
for both ASK and FSK modulation. Its upper frequency limit should be set to
meet the baseband system requirements. The data slicer is an one-bit analogto-digital converter that makes the bit decision and provides a digital data signal. In accordance with the code being used for modulation, there is a choice
between two different internal analog-to-digital converters. The conventional
Adaptive Data Slicer utilizes a large capacitor to provide a DC reference for the
bit decision. It should be used for the digital conversion of coded signals with no
or only a small DC component. The alternative Clamping Data Slicer references
its bit decision on the positive peak level of the data signal. It can be used with
all unsymmetrical codes containing DC content. Since the clamping data slicer
does not have to charge a large capacitor, it requires a shorter preamble and
Wireless Control Components
2-2
Application Note, February 2004
TDA 521X
Product Overview
exhibits a nearly instantaneous response time with some slight loss of sensitivity. The local oscillator (LO) is a single-channel PLL frequency synthesizer. It is
fully integrated on the chip.
Table 2-1 Summary of the Key Parameters of the Device
Selectable frequency range
TDA5210: 400-440 and 810-870 MHz
TDA5211: 310-350 MHz
Fully integrated VCO and PLL frequency synthesizer
ASK and FSK demodulation at data rates up to 120kbit/s NRZ
Supply voltage range 5.0 V ± 10%
Low supply current (6 mA typ. FSK mode, 5.3mA typ. ASK mode)
Reference frequency
TDA510: 6.7 MHz or 13.4 MHz
TDA511: 5.1 MHz or 10.2 MHz
IF frequency range 3 - 25MHz
Power down mode
Input sensitivity < -107 dBm in ASK mode
Input sensitivity < -100 dBm in FSK mode
Adaptive (RC) and peak data slicer
Low-pass filter with selectable cut-off frequency
Wireless Control Components
2-3
Application Note, February 2004
TDA 521X
Product Overview
2.3 Block Diagram
VCC
IF
Filter
MSEL
LNO
MI
6
LNI
RF
3
MIX
8
9
LIM
IFO
12
LIMX
17
FFB
18
15
OPP
22
SLP
21
SLN
19
20
LNA
+ FSK
- ASK
+
-
TAGC
FSK
PLL Demod
SLICER
+
+
LIMITER
OP
25
DATA
4
PEAK
26 PDO
DETECTOR
TDA 521X
OTA
:1/2
VCC
VCO
: 128 / 64
Φ
DET
UREF
CRYSTAL
OSC
AGC
Reference
23
THRES
24
3VOUT
14
Bandgap
Reference
Loop
Filter
DGND
13
2,7
5,10
VCC AGND
11
FSEL
16
1
28
27
PDWN
CSEL
Crystal
Functional_diagram.wmf
Figure 2-1
Wireless Control Components
Functional Block Diagram
2-4
Application Note, February 2004
TDA 521X
Product Overview
2.4 Package Outlines
P_TSSOP_28.EPS
Figure 2-2
Wireless Control Components
P-TSSOP-28-1 package outlines
2-5
Application Note, February 2004
3
Pin Configuration and Function
Contents of this Chapter
3.1
3.2
Pin Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
Pin Definition and Function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
TDA 521X
Pin Configuration and Function
3.1 Pin Configuration
CRST1
1
28
CRST2
VCC
2
27
PDWN
LNI
3
26
PDO
TAGC
4
25
DATA
AGND
5
24
3VOUT
LNO
6
23
THRES
VCC
7
22
FFB
MI
8
21
OPP
MIX
9
20
SLN
AGND
10
19
SLP
FSEL
11
18
LIMX
IFO
12
17
LIM
DGND
13
16
CSEL
VDD
14
15
MSEL
TDA 521X
Pin_Configuration_521X.wmf
Figure 3-1
Wireless Control Components
IC Pin Configuration
3-2
Application Note, February 2004
TDA 521X
Pin Configuration and Function
3.2 Pin Definition and Function
Table 3-1 Pin Definition and Function
Pin No.
Symbol
1
CRST1
Equivalent I/O-Schematic
Function
4.15V
1
Connection 1 to the symmetrical reference oscillator
circuit. The reference oscillator is of the negative
impedance converter type.
It represents a negative
resistor connected in series
to an inductor between the
CRSTL1 and CRSTL2 pins.
50uA
2
VCC
5V DC bias supply.
3
LNI
RF input to the LNA. This
input is DC-coupled to the
base of the common emitter
input stage of the LNA cascade configuration.
57uA
3
500uA
4k
1k
Wireless Control Components
3-3
Application Note, February 2004
TDA 521X
Pin Configuration and Function
4
TAGC
This pin is used for the gain
control of the LNA. The gain
of the LNA can be reduced
by approx.18dB. The
threshold voltage for the
gain control function is
1.3V. The control sensitivity
is -1dB/8mV
See Section 4.2
4.3V
3uA
4
1k
1.4uA
1.7V
5
AGND
6
LNO
Ground connection for the
analog section
1.7V
5V
2k
2k
1k
8
6
Output of the receiver RF
low-noise amplifier (LNA).
Collector of the common
base output stage of a cascade configuration. A DC
path to VCC can be supplied
by the output matching
9
network.
400uA
7
VCC
8
MI
5V DC bias supply.
Symmetrical input to the
mixer. The inputs are DCcoupled to the base of the
input stage of a Gilbert Cell
mixer configuration.
1.7V
2k
9
2k
MIX
8
9
400uA
10
AGND
Wireless Control Components
Ground connection for the
analog section
3-4
Application Note, February 2004
TDA 521X
Pin Configuration and Function
11
FSEL
TDA5210:This pin is used
to select the desired operating frequency range of the
receiver.
FSEL ≤ 0.2V will give
access to the 868MHz frequency range. FSEL ≥ 1.4V
or an open will set the
receiver to the 434MHz
mode.
TDA5211: Not applicable,
has to be left open!
750
1.2V
2k
11
12
IFO
300uA
2.2V
RF mixer output. This pin is
the single-ended IF output
of the mixer. The output
impedance is set internally
to 330Ω. It interfaces
directly with 10.7MHz standard ceramic IF filters.
60
12
4.5k
13
DGND
Ground connection for digital electronics.
14
VDD
DC bias supply to the digital
section
15
MSEL
ASK/FSK Modulation Format Selector
1.2V
3.6k
15
Wireless Control Components
3-5
Application Note, February 2004
TDA 521X
Pin Configuration and Function
16
CSEL
1.2V
80k
16
17
Input to the IF amplifier/limiter. The input is DC-coupled to the base of the first
differential stage of the IFF
amplifier strip. The differential input impedance is set
internally to 330Ω to meet
standard 10.7MHz ceramic
filter requirements.
LIM
2.4V
15k
17
18
TDA5210/TDA5211:
A logic low (CSEL< 0.2V)
applied at this pin sets the
internal frequency divider
for a reference frequency of
13.xx MHz/10.xx MHz.
A logic high (CSEL > 1.4V)
or an open will be applied
for a reference frequency of
6.xx MHz/5.xx MHz.
LIMX
75uA
330
18
15k
19
SLP
15uA
100
3k
19
Output of the low-pass filter,
directly coupled to the noninverting input of the data
slicer. An external RC lowpass filter connected to the
inverting input SLN, pin 20
of the data slicer sets the
reference level for the data
slicer/comparator to the
average DC level of the
data bit stream.
80µA
Wireless Control Components
3-6
Application Note, February 2004
TDA 521X
Pin Configuration and Function
20
SLN
5uA
10k
20
21
OPP
5uA
This pin is used for setting
the data slicer reference
level. A RC low-pass filter
from the filter output SLP,
pin19 provides the comparator with an average DC
level of the data bit stream.
When applying the peak
detector interface, the peak
voltage of the data signal at
PDO, pin 26 is tapped to
determine the bit decision
threshold voltage.
Non inverting input of the
low-pass filter operational
amplifier. A capacitor to
ground will be applied as
part of a second order
Sallen-Key low-pass filter.
200
21
22
FFB
5uA
100k
Access point to place a
capacitor to the output of
the low-pass filter. This
capacitor is part of the network building the SallenKey low-pass filter.
22
23
THRES
5uA
10k
23
Wireless Control Components
3-7
The voltage at this pin sets
the receiver input level to a
value where the AGC circuit
comes into operation.
THRES is the inverting
input of a differential operational transimpedance
amplifier that is used to
compare the internal RSSI
voltage of the IF amplifier
with the voltage applied to
THRES. The voltage at
THRES can be set by a
voltage divider attached to
the reference voltage at
3VOUT, Pin 24.
Application Note, February 2004
TDA 521X
Pin Configuration and Function
24
3VOUT
24
20kΩ
3.1V
25
Highly stable 3.0V voltage
source. This voltage reference output is derived internally from a band gap
voltage reference. It is held
constant over variations in
supply voltage and operating temperature. A resistive
voltage divider will be used
to precisely set the trigger
level at THRES, Pin 23 for
the AGC access point
Data output from the
demodulator. The output
level is TTL/CMOS-compatible. The fall and the rise
time of the output pulse will
be approx. 3µs when
loaded with 10pF.
DATA
500
25
40k
26
PDO
200
26
Wireless Control Components
3-8
An external capacitor at this
pin will be charged to the
peak level of the data filter
output voltage. The decision
threshold of the data slicer
will be set below this voltage by an amount given by
the division ratio of the voltage divider coupled to this
output. An external time
constant at the PDO output determines the decay
time of the reference voltage. It should be set in
accordance with the lowest
signal component within the
data string.
Application Note, February 2004
TDA 521X
Pin Configuration and Function
27
PDWN
Enable pin for the receiver
circuit. PDWN ≤ 0.8V or an
open turns off all receiver
functions. PDWN ≥ 2.8V
powers up all receiver functions
27
220k
220k
28
CRST 2
4.15V
28
50uA
Wireless Control Components
3-9
Connection 2 to the symmetrical reference oscillator
circuit.The reference oscillator is of the negative
impedance converter type.
It represents a negative
resistor in series to an
inductor between the
CRSTL 1 and CRSTL2
pins.
Application Note, February 2004
4
Functional Description
Contents of this Chapter
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
4.10
4.11
4.12
4.13
4.14
4.15
4.16
4.17
Low-Noise Amplifier (LNA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
AGC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8
Mixer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-11
Overall Performance of the Front End . . . . . . . . . . . . . . . . . . . . . . . 4-12
PLL Synthesizer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-13
Reference Oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-15
IF Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-22
ASK/FSK Switch Functional Description . . . . . . . . . . . . . . . . . . . . . 4-25
Demodulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-27
Data Filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-31
Data Slicer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-32
Principle of the Precharge Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-43
Quiet Data Output during no Transmission . . . . . . . . . . . . . . . . . . . 4-46
Decoder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-49
Settling Time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-49
Spurious Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-51
Sensitivity Measurements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-52
TDA 521X
Functional Description
4.1 Low-Noise Amplifier (LNA)
The low-noise amplifier is an on-chip high-gain amplifier operating at a current
of 0.5mA. The gain can be reduced by approx. 18dB by applying a high state to
the TAGC input, pin 4. The S-parameters of the LNA and the input to the mixer
are shown in the following table.
Table 4-1 S-Parameters of the LNA
315 MHz
Parameter
434 MHz
868 MHz
S11 LNA, high gain
0.895
-25.5deg
0.873
-34.7deg
0.738
-73.5deg
S11 LNA, low gain
0.918
-25.2deg
0.899
-35.4deg
0.772
-80.2deg
S21 LNA, high gain
1.577
150.3deg
1.509
138.2deg
1.419
101.7deg
S21 LNA, low gain
0.193
153.7deg
0.183
140.6deg
0.179
109.1deg
S12 LNA, high gain
0.003
128.2deg
0.023
172.3deg
S12 LNA, low gain
0.001
-153.5deg
0.022
173.4deg
S22 LNA, high gain
0.897
-10.3deg
0.886
-12.9deg
0.866
-24.2deg
S22 LNA, low gain
0.907
-10.5deg
0.897
-13.6deg
0.868
-26.3deg
S11 MIX
0.954
-10.9deg
0.942
-14.4deg
0.918
-28.1deg
Matching the LNA input to the source impedance of the generator and matching
the LNA output to the mixer input has to be done by a LC network. Both networks used on the evaluation boards have been designed to achieve best
selectivity at a designed loss of 2 to 3dB for each filter. The low-loss and hence
low Q design is mandatory in order to keep the circuit tuning-free. This low-loss
design achieves a high voltage gain of the LNA and hence a good sensitivity of
the receiver. Higher losses result in better selectivity at the expense of gain and
sensitivity.
It is good practice to design such a network utilizing an appropriate linear CAE
design tool. Even a very simple version can be used very efficiently. A very
practical way to design the network will be shown below.
As an example, the design of the input-matching network for an 868MHz application will be demonstrated in detail.
Wireless Control Components
4-2
Application Note, February 2004
TDA 521X
Functional Description
50Ω
C1
UG
C3
L1
C2
TDA5200
S11LNA
IG
RP1
CI'
C2
L1
RP2
C3'
RP3
LNA_matching..wmf
Figure 4-1
LNA input matching network
As shown in the circuit diagram in Figure 4-1, the input filter represents a single
tuned parallel resonance circuit loaded by three resistors: RP1, the generator
resistor transformed to the circuit, RP2, which collects the filter component
losses, and RP3, the LNA input resistor transformed to the circuit.
By dimensioning
RP1 = RP3
the overall performance of the filter will be optimized: the best selectivity will be
achieved at lowest losses. The design example will be done for assumed filter
losses of 2-3 dB.
The power efficiency of the network in Figure 4-1 is:
 1 − QL
η = 
 QU




2
with the QL and QU, the loaded and unloaded quality factors of the circuit.
The unloaded QU of the circuit is given by:
QU =
Wireless Control Components
4-3
RP2
Z
Application Note, February 2004
TDA 521X
Functional Description
The characteristic impedance of the circuit is
1
1
 L 2
Z =   = ω0 L =
C
ω
C
 
0 tot
The effective total capacitance adds up to:
C tot = C1'+C 2 + C 3'
The resonance frequency fo is
ω 0 = 2πf 0
The loaded QL of the circuit will be:
QL =
( RP1 // RP 2 // RP3)
Z
The 3dB bandwidth B for the single tuned LC network can be calculated from:
QL =
f0
B
A tuning-free design is required on grounds of cost. In order to minimize the
complexity of the circuit, the application example uses a single tuned LC parallel
circuit for both the antenna input matching network and the matching network
between the LNA output and the mixer input.
The requirement for a tuning-free implementation provides the basis for dimensioning the networks.
Due to component and manufacturing tolerances, the frequency of the resonance circuit may vary by ∆f from the designed frequency. This will result in loss
of gain of the LNA, and therefore in reduced sensitivity of the receiver. In order
to limit this sensitivity loss, the two circuits in the application example are dimensioned in such a way that each of them exhibits a maximum additional loss of
3 dB under worst-case tolerance conditions.
Typical relevant tolerance values are:
Table 4-2 Tolerance Values of the LNA Input Matching Circuit
∆f/f
Tolerance of C
± 2%
± 1%
Tolerance of L
± 2%
± 1%
Manuf. tolerance
Board
Placing the components on the pads
± 2%
± 2%
Total frequency tolerance
Wireless Control Components
4-4
± 6%
Application Note, February 2004
TDA 521X
Functional Description
Detuning losses will be low for a higher 3dB bandwidth and hence for a low
loaded QL of the circuit. With an assumed drop in gain by max. 3 dB per circuit,
the required QL is
QL ≤ f r / 2∆f = 8.3
QL : loaded Q of resonance circuit
fr : operating frequency
∆f : frequency offset of resonance circuit
Using chip inductors of size 0805, an unloaded QU of
QU ≈ 30 (434MHz )
40 (869MHz )
can be achieved for the resonance circuits. The Q of the capacitors are greater
by at least a factor of 5, so their losses are not taken into account.
The filter losses for each circuit at the LNA input and output will be:
2
 1 −Q L 
 = .523 ≅ − 2.8dB (434MHz)
a = 
 QU 
.628 ≅ − 2.0dB (868MHz)
This value applies to optimum dimensioning, where the generator resistance
and the load resistance are both transformed into an equal conductance of
RP1 = RP3 in parallel to the resonance circuit.
The selectivity of these circuits is not impressive due to their low loaded QL. In
the example, the image frequency rejection for ∆f = 2* fIF = 21.4MHz is therefore only
aif = 2 × 2.2dB = 4.4dB (434MHz )
2 × 0.6dB = 1.2dB (868MHz )
This result can be improved (still with a tuning-free design) by using high Q, pretuned resonators, e.g. based on SAW structures or ceramic filters.
It is rather unusual to design an LC filter for a specified loaded QL. The following
description outlines a very practical method for the design of such a filter. This
procedure is not as attractive as a CAE design but it is very effective. The example applies for the input matching circuit of the LNA with a frequency of 868MHz.
As a first step, a convenient combination of C1 and C3 has to be found to
transform the generator resistance and the LNA input resistance to the same
conductance RP1=RP3 across the parallel resonance circuit. This can be very
elegantly done on the PCB by use of a VNA. This method offers the advantage
that all parasitic elements of the board are captured.
Experimental values of C1 and C3 have been found on the evaluation board
with a frequency of 868MHz as
C1 = 1 pF
C 3 = 5.6 pF
Wireless Control Components
4-5
Application Note, February 2004
TDA 521X
Functional Description
In the next step the value of L1 will be calculated.
Considering the formula of QL and QU and the above decided values of QL and
QU yields the following equation for the ratio QU/QL
QU
40
RP 2
=
= 4 .8 =
Q L 8 .3
RP1 // RP 2 // RP3
and therefore
RP 2 = 3.8 × (RP1 || RP3)
The resultant value of the conductance has been measured as
RP1 = RP3 = 300Ω
which yields the value of RP2
RP 2 = 570Ω
The inductance L1 is calculated, using the formular for QU and Z given above
Z ≥ ω0 L =
RP 2 570Ω
=
= 14.25Ω
QU
40
and
L1 ≥
Z
ω0
= 2.6nH
The (standard) value selected is
L1 = 3.3nH
The size of the capacitor C2 required to set the circuit to resonance is once
again most easily found empirically. The resonant frequency in general will not
be hit accurately enough with standard component values. There is still some
room for adjustment by changing the positions of L1 and C3 on the board and
by changing the orientation of L1 (!). If these variation options are not satisfactory, then C1 or C3 can be changed slightly, violating rule (1). The resultant
slight loss of receiver sensitivity can be tolerated.
The values of the components imparted in the circuit should fall within a range
where they can be handled electrically. The circuit itself gives some freedom in
the determination of the characteristic impedance Z and hence in the component values. The component values may be selected to conform to certain criteria.
One selection criterion for the filter components is that their value remains
within technically manageable limits. For example, it may be specified that the
C values should not exceed 10pF. Up to this value they are available with absolute tolerances of ± 0.1pF. Frequency-determining C’s should not fall below
2.2pF because of the ± 0.1pF tolerance specification. Inductances below 3.3nH
also pose problems due to the increasing influence of lead inductances. Once
the inductance value exceeds approximately 33nH with a frequency of
868MHz, their effective impedance is capacitive due to their self-resonance.
Wireless Control Components
4-6
Application Note, February 2004
TDA 521X
Functional Description
The overall frequency response of the LNA is shown in the following figures.
The midband voltage gain from the receiver input to the mixer input is +21dB/
+18dB with a frequency of 434/868 MHz, for instance.
LNAvoltage_vsfreq.wmf
Figure 4-2
LNA voltage gain vs. frequency of 434MHz receiver
Inputref_vsfreq1.wmf
Figure 4-3 Input reflection and relative gain of the LNA vs. frequency
at 869MHz
Wireless Control Components
4-7
Application Note, February 2004
TDA 521X
Functional Description
4.2 AGC
The receiver incorporates an “automatic gain control” (AGC) function to change
its gain in accordance with the RF input level. Applying this function will improve
the power handling capability of the receiver by almost 20dB. Increasing the
voltage at the TAGC control input (Pin 4) beyond a level of approx. 1.3V will
reduce the gain of the LNA by 19dB/18dB with 434/868 MHz by shifting the bias
point to a low-level current. The low gain mode of the LNA will result in a power
handling capability of the receiver for input levels up to 0dBm without degradation. A low signal or an open at TAGC sets the LNA to the high gain mode.
C18
R4
R5
Uthreshold
3VOUT
THRES
24
23
RSSI (0.8 - 2.8V)
20kΩ
VCC
OTA
+3.1 V
Iload
RSSI < Uthreshold: Iload= -1.5µA
4
TAGC
UC
C5
LNA
Gain control
voltage
RSSI > Uthreshold: Iload=4.2µA
Uc:< 2.6V : Gain high
Uc:> 2.6V : Gain low
Ucmax= VCC - 0.7V
Ucmin = 1.67V
LNA_autom.WMF
Figure 4-4
AGC circuit
The AGC function is controlled by a comparator connected to the RSSI voltage.
The decision threshold for the start of the AGC function is set by the comparator
reference voltage. This voltage is tapped down from the precision 3.0V voltage
source at the 3VOUT port (Pin 24) via the R4-R5 voltage divider. If the RSSI
voltage reaches the comparator threshold voltage, then the TAGC voltage is
pulled up by the transconductance comparator output current. The LNA passes
over to the low gain mode.
The threshold of the AGC should be set to the lowest possible receiver input
level, in order to optimize the large-signal response of the receiver for the widest
possible input level range. However, it must be set higher than the receiver sen-
Wireless Control Components
4-8
Application Note, February 2004
TDA 521X
Functional Description
sitivity limit by at least the value of the gain reduction, so that the mixer does not
operate at its sensitivity threshold. With the specified LNA gain reduction of
approx. 18 dB, it is recommended that the threshold be dimensioned to a
receiver input level of at least 25...30 dB higher than the sensitivity limit value.
Figure 4-5 shows the AGC function for a transition level set to -85dBm by a
combination of R4=330kΩ and R5=330kΩ.
3
2,5
RSSI / V
AGC disabled
2
AGC
implementation
1,5
1
0,5
0
-130
-110
-90
-70
-50
-30
-10
Pe / dBm
RSSI.wmf
Figure 4-5
RSSI voltage as a function of input level for AGC implementation
The AGC function is filtered via C5. C5 is charged by the current source at the
comparator output by a 4.2µA source and a 1.5µA sink current for a fast attack
and slow decay time. The AGC operation should not be triggered by the data
signal. Since the AGC operates in a linear mode without any hysteresis, there
always is some range of the receiver input level, where the AGC will be affected
by the data signal.
The AGC gain at TAGC is 10dB/80mV. Due to the very high total gain within the
AGC, the loop will only be stable in case of ASK for
C 5 ≥ 10nF
Because of the RF-level is independent from the logik level in case of FSK, the
value of C5 is not so critical.
For ASK the following has to be considered:
A recommended value for C5 is 47nF
Wireless Control Components
4-9
Application Note, February 2004
TDA 521X
Functional Description
For a receiver input level above the static AGC threshold level, the RSSI data
signal at SLP (Pin19) will show a characteristic overshoot at the positive pulse
edge. It is caused by the AGC loop trying to level down the RSSI signal to the
preset threshold value. The amplitude and duration of the overshoot is directly
related to C5. In order to keep the overshoot below 30% of the signal amplitude,
C5 should be designed for a value as shown in the subsequent table.
Table 4-3 Dependence of TL Value on C5
TL
C5
0.25ms
10nF
1ms
47nF
2ms
150nF
where TL is the longest period of no signal change within the data sequence.
The design of the data slicer has to consider the distorted pulse shape. This will
be quite unproblematic when applying the adaptive data slicer. For most coding
schemes, C5 can even be set to 1/5 of the above value then without a significant
degradation of performance.
Using a too small capacitor and receiving an ASK signal with a field strength
that is within the control range of the AGC may result in a distortion of the signal
at the data filter output.
After the positive edge of the signal the AGC reduces the LNA gain and therefore also the data filter output voltage. The transition from high to low (no RFsignal) induces the AGC to enhance the LNA gain and subsequently the data
filter output voltage again. Depending on the time constant of the slicing level a
data (slicer output) signal as shown in the Figure 4-6 can be the result.
Wireless Control Components
4 - 10
Application Note, February 2004
TDA 521X
Functional Description
Example for a too small AGC capacitor relating to the largest low time:
-70dBm_low=9ms_duty-cycle=50%.wmf
Figure 4-6
AGC time constant; RF-Level=-70dBm
When applying the peak data slicer, however, the threshold has to be reduced
to cover the overshoot voltage. A combination of R2=100kΩ and R3=390kΩ
should be imparted then. Setting the threshold to this lower level will reduce the
receiver sensitivity by approx. 6dB, however.
4.3 Mixer
The Double Balanced Mixer is based on a Gilbert Cell configuration with a symmetrical input and a single-ended output. The output presents a 330Ω termination which directly interfaces to 10.7MHz ceramic filters. A LC matching network
is used to connect the LNO to the MI or the MIX input. The conversion voltage
gain of the mixer with the internal load of 330Ω is 21dB. The receiver uses lowside LO injection. This avoids interference caused by signals at the image frequency range when operated with high-side injection.
Increasing the external load of 330Ω would generally lead to an improved
1dB compression point and also an improvement of the IIP3.
Wireless Control Components
4 - 11
Application Note, February 2004
TDA 521X
Functional Description
4.4 Overall Performance of the Front End
Table 4-4 shows the overall performance of the LNA and of the mixer at a frequency of 434MHz. Table 4-5 summarizes the same results for the evaluation
board operated at 868 MHz.
Table 4-4 Measured Mixer Performance at 434 MHz
Parameter
low gain
Gain
1 dB
LNA
high gain
21dB
Receiver sensitivity
LNA&Mixer
low gain high gain
22 dB
42 dB
- 95 dBm
- 112 dBm
IIP3
-10 dBm
- 10 dBm
-28 dBm
-48 dBm
1dB compression
- 18 dBm
- 15 dBm
- 31 dBm
- 53 dBm
Table 4-5 Measured Mixer Performance at 868 MHz
Parameter
low gain
Gain
0 dB
LNA
high gain
19 dB
Receiver sensitivity
LNA&Mixer
low gain high gain
19 dB
40 dB
- 95 dBm
- 112 dBm
IIP3
-5 dBm
- 14 dBm
-26 dBm
-35 dBm
1dB compression
- 6 dBm
- 15 dBm
- 34 dBm
- 55 dBm
Table 4-6 Measured Mixer Performance at 315 MHz
Parameter
low gain
Gain
2 dB
LNA
high gain
21 dB
Receiver sensitivity
LNA&Mixer
low gain high gain
23 dB
42 dB
- 95 dBm
- 113 dBm
IIP3
-13 dBm
- 10 dBm
-25 dBm
-43 dBm
1dB compression
- 7 dBm
- 14 dBm
- 35 dBm
- 54 dBm
The IIP3 of the mixer and subsequently of the whole system can be shifted to a
higher level by adding a resistor between the mixer output and GND. But of
course this additional resistor results in a higher current consumption.
Wireless Control Components
4 - 12
Application Note, February 2004
TDA 521X
Functional Description
Table 4-7 Measured IIP3 at 434 MHz vs addition resistor
Additional Supply
Current (FSK)
low gain
high gain
Resistor value
no resistor
IIP3
LNA&Mixer
low gain high gain
0
-20 dBm
-40 dBm
4k7
320µA
- 15.5 dBm
- 35.5 dBm
1k2
850µA
-11 dBm
-31 dBm
4.5 PLL Synthesizer
The basic circuit of the PLL frequency synthesizer consists of a VCO operating
at a nominal frequency of 840MHz in case of TDA5210 and 330MHz in case of
TDA5211 respectively, a frequency divider with a division ratio of either 64 or
128, a frequency/phase discriminator and a reference oscillator. The reference
oscillator of the TDA5210 operates at either 13.4MHz or 6.7MHz, the oscillator
of the TDA5211 at either 10.2MHz or 5.1MHz. In case of operation at 868.4
MHz the VCO operates at 857.7MHz, for example, This frequency is divided by
64 for operation at a reference frequency of 13.4015625MHz or by 128 at a reference frequency of 6.70078125MHz. For 315MHz, for which the TDA5211 has
do be used, the VCO frequency of 651.4MHz can also be divided either by 64
requires a reference frequency of 10.178125MHz or by 128 requires a referenc
frequency of 5.0890625MHz.
The VCO signal is directly applied to the mixer stage when operating the
receiver at 868MHz. For operation at 434MHz and 315MHz respectively, the
VCO signal is divided by two to build the injection signal to the mixer. The VCO
of the TDA5210 covers a typical frequency range of 765MHz to 910MHz at the
limits of the tuning voltage of 4.7V and 2.4V, the one of the TDA5211 covers a
frequency range of typically little less than 590MHz to 720MHz inbetween its
tuning voltage range. An example of a typical tuning curve is presented in Figure 4-7, the phase noise spectrum of the VCO signal measured with a resolution bandwidth of 10kHz in Figure 4-8. The phase noise spectrum shows the
characteristic noise suppression within the loop bandwidth of 150kHz. Sideband noise outside the loop bandwidth at a frequency offset of ±200kHz can
be specified at -87dBc/Hz. This noise sets the limit of the adjacent channel suppression of the receiver.The selectivity of the IF filter is bypassed this way.
Wireless Control Components
4 - 13
Application Note, February 2004
TDA 521X
Functional Description
TDA5210A3
915
910
905
900
895
890
885
880
875
870
865
860
855
850
845
840
835
830
825
820
815
810
805
800
795
790
785
780
775
770
765
760
755
750
0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9
1 1,1 1,2 1,3 1,4 1,5 1,6 1,7 1,8 1,9
2 2,1 2,2 2,3 2,4 2,5 2,6
VCO_Tuningbereich.wmf
Figure 4-7
Typical VCO tuning curve of TDA5210
Phasenoise.wmf
Figure 4-8
Wireless Control Components
Phase noise spectrum of local oscillator
4 - 14
Application Note, February 2004
TDA 521X
Functional Description
4.6 Reference Oscillator
The receiving frequency and its stability are determined by the reference crystal
and the associated oscillator circuit.
The oscillator is a symmetrical configuration of the negative impedance converter type. A resistor and a capacitor are “sign-inverted” to give a negative
resistance in series with an inductance between the oscillator ports CRST1 and
CRST2, pin1 and pin28. The equivalent impedance parameters of the oscillator
presented in Section 5.1.3 of the Specification have been taken between pin1
and pin28 of the TDA521X on the evaluation board.
The value of the capacitor necessary to achieve that the oscillator is operating
at the intended frequency is determined by the reactive (inductive) part of the
negative resistance of the oscillator circuit and by the crystal specifications
given by the crystal manufacturer.
CS
CRST2
Input
impedance
Z1-28
Crystal
28
TDA521X
1
CRST1
Crystal_loard1.wmf
Figure 4-9
Determination of Series Capacitance Value for the Crystal Oscillator
A crystal is specified with a load capacitance CL. The series capacitor CS
needed to achieve the wanted oscillation frequency in presence of the above
mentioned series reactance imposed by the oscillator circuit can be calculated
according to the following formula.
CS =
1
1
+ 2π f X L
CL
with CL the load capacitance (refer to the quartz crystal specification).
Examples:
6.7 MHz:
CL = 12 pF
13.401 MHz: CL = 12 pF
XL=695 Ω
CS = 9.56 pF
XL=1010 Ω
CS = 5.94 pF
These values may be obtained in high accuracy by putting two capacitors in
series to the quartz, such as 27pF and 15pF in the 6.7MHz case and 22pF and
8.2pF in the 13.401MHz case. The calculation of CS or the two serial resistors
respectively for TDA5211 and a oscillator frequency of either 5.1MHz or
10.2MHz can be done in the same way, of course.
Wireless Control Components
4 - 15
Application Note, February 2004
TDA 521X
Functional Description
The frequency stabilities of both the receiver and the transmitter and the modulation bandwidth set the limit for the bandwidth of the IF filter. To achieve a high
receiver sensitivity and efficient suppression of adjacent interference signals,
the narrowest possible IF bandwidth should be realized.
The frequency stability of the center frequency of the receiver is affected by a
number of factors:
■
Tuning tolerance of the crystal
■
Temperature stability of the crystal
■
Aging of the crystal
■
Tolerance of the load capacitor CL
■
Tuning tolerance of the oscillator circuit
■
Temperature stability of the oscillator circuit
The first three items are parameters of the crystal specified by the manufacturer.
The so called pulling sensitivity describes the magnitude of the frequency
change relating to the variation of the load capacitor.
δf s '
f
C1
δD
=−
= s
2
δCL
δ
C
2 ⋅ (C0 + CL )
L
It seems do be the best to chose CL as large as possible to get a small pulling
sensitivity and subsequently keep the influence of the serial capacitor and of its
tolerances as small as possible. But CL mustn’t mixed up with the serial capacitance Cs. CL is the effective value of the capacitance applied to the crystal and
is calculated by dividing the impedance in this particular case the capacitive
reactance by the angular frequency “ω”.
CL =
1
 1

 CS

 − ω 2 Losc

Where Losc is the inductivity of the oscillator occuring on the output between pin
1 and pin 28.
With the aid of this formular it becomes obviously that the higher the serial
capacitance Cs, the higher the influence of Losc. Subsequently the tolerances
of the oscillator isn’t only described by the serial capacitor but also by the tolerances of des oscillator inductivity and even in particular by the absolute value
of this inductivity.
Wireless Control Components
4 - 16
Application Note, February 2004
TDA 521X
Functional Description
Relative frequency change per changing of Cs
δf S '
(
fS
C1
=−
× 1 + ω 2 LOSC C L
δC S
2 × (CO + C L )
=
(
SD
× 1 + ω 2 LOSC C L
SC L
)
)
Relative frequency change per changing of Losc
δf S '
fS
C1
= −(2πf S ')2 ×
× CL 2
2
SLOSC
2 × (CO + C L )
=
δD
× (2πf S ')2 × C L 2
δC L
In the suitable range for the serial capacitor, either capacitors with a tolerance
of 0,1pF or with a tolerance of 1% are available.
The tolerance of the internal oscillator inductivity is much higher, so the inductivity is the dominating value for the tolerance over a large CL/C0 range.
Assuming the crystal parameters,
fs’ = 13,40155MHz, C1=4,75fF, C0=1,29pF
and an inductivity of the oscillator of 12µH,
Figure 4-10 shows the frequency error of the oscillator caused by a 20% deviation of the inductivity as a function of CL/C0. Below the frequency error caused
by a deviation of the serial capacitor Cs by 0.1pF from the correct value shown
in Figure 4-11 and by 1% from the correct value shown in Figure 4-12, both as
a function of CL/C0. The entire error caused by the inductivity and Cs is represented in Figure 4-13, one the one hand with the constant deviation ∆Cs of
0.1pF, one the other hand with the constant relative failure ∆Cs/Cs of 1%
Wireless Control Components
4 - 17
Application Note, February 2004
TDA 521X
Functional Description
.
∆ frel(dL) vs CL/C0
12
,5
11
,5
10
,5
9,
5
8,
5
7,
5
6,
5
5,
5
4,
5
3,
5
2,
5
-5
1,
5
0,
5
0
df/fs' [ppm]
-10
-15
-20
-25
-30
-35
-40
CL/C0
dfrel vs CL/C0
Tol_caused_by_Losc.wmf
Figure 4-10
Tolerances caused by ∆L of 20%
∆ frel(dCs=const) vs CL/C0
12
,5
11
,5
-10
10
,5
9,
5
8,
5
7,
5
6,
5
5,
5
4,
5
3,
5
2,
5
1,
5
0,
5
0
df/fs' [ppm]
-20
-30
-40
-50
-60
-70
-80
CL/C0
dfrel(dCs=const) vs CL/C0
Tol_caused_by_dCs=0.1pF.wmf
Figure 4-11
Wireless Control Components
Tolerance of the oscillator caused by ∆Cs of 0.1pF
4 - 18
Application Note, February 2004
TDA 521X
Functional Description
∆ frel(dCs/Cs=const) vs CL/C0
5
5
5
5
6,
7,
8,
9,
,5
5
5,
12
5
4,
,5
5
3,
11
5
2,
,5
5
1,
10
5
0,
0
-1
df/fs' [ppm]
-2
-3
-4
-5
-6
CL/C0
dfrel(dCs/Cs=const) vs CL/C0
Tol_caused_by_dCs=1%.wmf
Figure 4-12
Tolerance of the oscillator caused by ∆Cs of 1%
-∆
∆ fges/fs' vs CL/C0
80
70
-dfges/fs' [ppm]
60
50
40
30
20
10
5
5
5
5
6,
7,
8,
9,
,5
5
5,
12
5
4,
,5
5
3,
11
5
2,
,5
5
1,
10
5
0,
0
CL/C0
for fixed dCs/Cs and dL/L (see table)
for fixed dCs and dL/L
Tol_caused_by_Losc_and_Cs.wmf
Figure 4-13
Wireless Control Components
Tolerance of the oscillator caused by ∆L and ∆Cs
4 - 19
Application Note, February 2004
TDA 521X
Functional Description
Especially in the case of constant relative failure ∆Cs/Cs (1%) the frequency
error increases with an increasing CL relating to C0, although the pulling sensitivity decreases.
But decreasing CL is limited by the fact, that the smaller CL the higher the resistance Rs appearing on the pins of the crystal, which is presented in Figure 4-14.
R1 is the dynamic resistance of the equivalent circuit of the crystal.
Rs/R1 vs CL/C0
10,000
9,000
8,000
Rs/R1
7,000
6,000
5,000
4,000
3,000
2,000
1,000
,5
12
,5
11
,5
10
9,
5
8,
5
7,
5
6,
5
5,
5
4,
5
3,
5
2,
5
1,
5
0,
5
0,000
CL
Rs/R1 vs CL/C0
Rs_to_R1_vs_CL_to_C0.wmf
Figure 4-14
Rs/R1 vs CL/C0
In the following table an assessment of the worst case overall frequency spread
to be expected in case of operation at 868MHz with a 13.4MHz Jauch reference
crystal as denoted in the Bill of Materials in Table 5-1 is shown. The calculation
is taking into account the tolerance of the crystal and of the components in the
oscillator circuit which are determining the tuning tolerance and temperature
stability of the circuit. Note that the result is a sum of the squares of the indivual
terms.
A spreadsheet1 may be obtained from Infineon which can be used to predict the
total frequency error (3σ) by simply entering the crystal specification.
The Table 4-8 refering to the crystal specification described above.
1.available for download on the Infineon RKE Webpage www.infineon.com/rke, also
included on evalkit CD-ROM
Wireless Control Components
4 - 20
Application Note, February 2004
TDA 521X
Functional Description
Table 4-8 Assessment of the Frequency Error of the Crystal Oscillator
tuning ± 10ppm
± 10ppm
temperature stability ±20ppm
± 20ppm
tolerance of the series load capacitor ±1%
± 3ppm
tolerance of the oscillator circuit (3 σ spread of
internal component values assumed),
calculated with spread sheet
±33ppm
without temperature drift, according to spread sheet
± 34ppm
with temperature drift
± 54ppm
Tolerance of
the crystal
Tolerance of
the circuit
Total
frequency
error (3 σ
spread)
There is also the possibility to force the oscillator with an external signal, shown
in Figure 4-15 and Figure 4-16.
1
CRST1
1k
Signal
Generator
TDA521X
50Ω
28
CRST2
50Ω
Force_X-tal_without.wmf
Figure 4-15 Forcing the crystal oscillator without a transformer
Wireless Control Components
4 - 21
Application Note, February 2004
TDA 521X
Functional Description
Transformer
1
CRST1
1k
Signal
Generator
TDA521X
50Ω
28 CRST2
50Ω
Force_X-tal_with.wmf
Figure 4-16 Forcing the crystal oscillator with a transformer
4.7 IF Section
4.7.1
IF Amplifier
The IF section is an AC-coupled high-gain differential input, single-ended output
amplifier. It utilizes three identical gain stages each with a Received Signal
Strength Indicator detector. The RSSI signal of the IF amplifier is obtained by
summing the individual detector signals. The differential input resistance has
been set internally to 330Ω. Single-ended operation of the amplifier presents
the nominal load to the ceramic IF filter.
Figure 4-17 shows the frequency response of the IF amplifier. It can be used at
all IF frequencies within the range of 3MHz and 25MHz without significant degradation of the overall performance of the receiver.
Wireless Control Components
4 - 22
Application Note, February 2004
TDA 521X
Functional Description
10
a/dB
0
-10
-20
-30
-40
-50
-60
1
10
100
f/MHz
Frequency response IF.ampl.wmf
Figure 4-17
4.7.2
Frequency response of the IF amplifier.
IF Filtering
The TDA 521X has been designed to be used with a low-cost ceramic IF filter.
Those filters are supplied with a wide variety of bandwidth between 60kHz and
300kHz. The nominal input and output impedance is specified at 330Ω. The frequency response of such a filter with a nominal bandwidth of 230kHz imparted
to the evaluation board is shown in Figure 4-18. The filter characteristic may be
degraded by oscillator and signal feed-through to the input of the IF amplifier.
Both signals may convert to the IF frequency at the IF amplifier, bypassing the
IF filter. This effect can be clearly demonstrated at high input levels. A simple
low-pass filter in front of the IF amplifier may keep the RF signals from entering.
In most cases a careful layout of the board gives adequate decoupling.
Please note that the far off suppression may be as low as 30 dB. Filters like the
Murata SFE 10.7 MA5-A even show a peak at 4.5MHz in the frequency
response. Thus beside the excepted IF of 10.7MHz also unwanted signals
6.2MHz away from the IF may influence the following demodulation chain.
Wireless Control Components
4 - 23
Application Note, February 2004
TDA 521X
Functional Description
Frequencyresponse.wmf
Figure 4-18
IF frequency response
The bandwidth of the IF filter should be set to a value where the modulation signal is reliably transferred under the influence of the frequency tolerance of the
transmitter and the receiver.
Table 4-9 Design Example: IF Bandwidth Calculation (3σ value)
Transmit frequency
434.4MHz
Modulation ASK4kbit/s
4kbit/s
Frequency tolerance of the
transmitter
± 63ppm
Frequency tolerance of the
receiver
± 54ppm
Tolerance of the center
frequency of the IF filter
± 30kHz
Total tolerance
Spectrum of modulation
± 69ppm
±108ppm
±1.5*4kbit/s = ± 6kHz
±14ppm
The IF bandwidth therefore should be:
BIF ≥ ± (108ppm+14ppm) * 434MHz = ± 53kHz = 106kHz
Wireless Control Components
4 - 24
Application Note, February 2004
TDA 521X
Functional Description
4.8 ASK/FSK Switch Functional Description
The TDA5210 is containing an ASK/FSK switch which can be controlled via
Pin 15 (MSEL). This switch is actually consisting of 2 operational amplifiers that
are having a gain of 1 in case of the ASK amplifier and a gain of 11 in case of
the FSK amplifier in order to achieve an appropriate demodulation gain characteristic. In order to compensate for the DC-offset generated especially in case
of the FSK PLL demodulator there is a feedback connection between the
threshold voltage of the bit slicer comparator (Pin 20) to the negative input of
the FSK switch amplifier. This is shown in the following figure.
15
MSEL
RSSI (ASK signal)
ASK/FSK Switch
Data Filter
FSK PLL Demodulator
AC
0.18 mV/kHz
+ ASK
RF1 int
+ FSK
RF3 int
100k
DATA Out
RF2 int
+
v=1
Comp
100k
25
-
300k
RF4 int
DC
typ. 2 V
30k
1.5 V......2.5 V
FFB 22
21
OPP
SLP
19
20
SLN
ASK mode : v=1
FSK mode : v=11
C12
C14
R1
C13
ask_fsk_datapath.WMF
Figure 4-19
4.8.1
ASK/FSK mode datapath
FSK Mode
The FSK datapath has a bandpass characterisitc due to the feedback shown
above (highpass) and the data filter (lowpass). The lower cutoff frequency f2 is
determined by the external RC-combination. The upper cutoff frequency f3 is
determined by the data filter bandwidth.
The demodulation gain of the FSK PLL demodulator is 180µV/kHz. This gain is
increased by the gain v of the FSK switch, which is 11. Therefore the resulting
dynamic gain of this circuit is 2mV/kHz within the bandpass. The gain for the DC
content of FSK signal remains at 180µV/kHz. The cutoff frequencies of the
bandpass have to be chosen such that the spectrum of the data signal is influenced in an acceptable amount.
In case that the user data is containing long sequences of logical zeroes the
effect of the drift-off of the bit slicer threshold voltage can be lowered if the offset
voltage inherent at the negative input of the slicer comparator (Pin 20) is used.
The comparator has no hysteresis built in.
Wireless Control Components
4 - 25
Application Note, February 2004
TDA 521X
Functional Description
This offset voltage is generated by the bias current of the negative input of the
comparator (i.e. 20nA) running over the external resistor R. This voltage raises
the voltage appearing at pin 20 (e.g. 1mV with R = 100kΩ). In order to obtain
benefit of this asymmetrical offset for the demodulation of long zeros the lower
of the two FSK frequencies should be chosen in the transmitter as the zerosymbol frequency.
In the following figure the shape of the above mentioned bandpass is shown.
gain (pin19)
v
v-3dB
20dB/dec
-40dB/dec
3dB
0dB
f
DC
f1
f2
0.18mV/kHz
f3
2mV/kHz
frequenzy_char.WMF
Figure 4-20
Frequency characterstic in case of FSK mode
The cutoff frequencies are calculated with the following formulas:
f1 =
1
R1× 330kΩ
× C13
2π
R1 + 330kΩ
f 2 = v × f1 = 11× f1
f 3 = f 3dB
f3 is the 3dB cutoff frequency of the data filter - see Section 4.10.
Example:
R1 = 100kΩ, C13 = 47nF
This leads to f1 = 44Hz and f2 = 485Hz
Wireless Control Components
4 - 26
Application Note, February 2004
TDA 521X
Functional Description
4.8.2
ASK Mode
In case the receiver is operated in ASK mode the datapath frequency charactersitic is dominated by the data filter alone, thus it is lowpass shaped.The cutoff
frequency is determined by the external capacitors C12 and C14 and the internal 100k resistors as described in Section 4.10.
0dB
-3dB
-40dB/dec
f
f3dB
freq_ask.WMF
Figure 4-21
Frequency charcteristic in case of ASK mode
4.9 Demodulation
4.9.1
ASK Demodulation
After passing the IF filter, the IF signal is fed to the limiter. The limiter serves
two functions: amplification and demodulation of the filtered IF signal. The limiter rectifies the IF in order to demodulate the received signal. The demodulated
signal is referred to as the RSSI signal. Figure 4-22 shows the relation between
the RSSI voltage level and the limiter input IF level at LIM, pin17. As can be
seen, the RSSI function is linear to the log. of the limiter input level over a range
of 80dB. The receiver can detect a modulated carrier over an input signal
dynamic range of more than 80dB. Applying the integrated AGC function, this
range will be extended by another 18dB to a total dynamic range of 95dB. The
maximum input level that can be detected by the receiver is approx. 0dBm. This
value greatly depends on the depth of modulation of the transmitter signal.
Wireless Control Components
4 - 27
Application Note, February 2004
TDA 521X
Functional Description
3
RSSI / V
2,5
2
1,5
1
0,5
0
-120
-100
-80
-60
-40
-20
0
20
Pe / dBm
Limiter_char.wmf
Figure 4-22
Limiter demodulator characteristics
3
RSSI / V
2,5
2
RSSI LNA ON
RSSI LNA OFF
1,5
1
0,5
0
-120
-100
-80
-60
-40
Pe / dBm
-20
RSSI_vs_LNA.wmf
Figure 4-23
RSSI Voltage vs. Receiver Input Level
Figure 4-18, Figure 4-22 and Figure 4-23 give some interesting information
about the interaction of the different gain blocks of the receiver.
The voltage gain between the antenna input and the limiter input is 40dB.
The LNA block adds approx. 7dB of noise at the mixer input to the receiver.
Following the formula for the noise figure NFc of cascaded blocks with individual noise figures NF1 and NF2 and the gain G1
NFC = NF1 +
Wireless Control Components
4 - 28
NF 2 − 1
G1
Application Note, February 2004
TDA 521X
Functional Description
it is quite evident that the given factor of 7dB ≅ 5 is the factor between the two
terms of the above formula. Applying this formula, it can be concluded that the
LNA gain, GLNA can be reduced by 3dB without degrading the overall noise
characteristics of the receiver significantly. The factor between the two terms in
the formula will then be 4dB ≅ 2.5, then resulting in a total noise figure of
NFC = NFLNA +
NFMX - 1
1
= NFLNA ( 1 +
) = 1.4 NFLNA
G LNA
2.5
The noise of the LNA (having a noise figure of NFLNA) will then contribute just
1.4 ≅ 1.4dB to the total noise of the receiver.
The converter block (LNA in cascade with the mixer stage) adds approx. 10dB
of noise to the receiver at the input of the limiter. Applying the above formula for
the cascaded noise figure again, it can be concluded that the converter gain
could be reduced by 6dB without reducing the overall sensitivity of the receiver
significantly. The receiver sensitivity will drop by 1.4dB.
The circuit thus gives some margin to apply filters with higher losses at the IF
and at the LNA output.
4.9.2
FSK Demodulation
The FSK-Demodulator realized in the TDA521X is a PLL-Demodulator.
The phase of the IF-signal and VCO output signal are compared against one
another in the “phase detector circuit” (PD). The voltage on the phase detector
output is proportional to the phase difference of the two input signals. This voltage is feed via the loop filter to the VCO-control input, to track and match the
VCO-frequency to the IF-frequency. Applying a FM/FSK-signal the loop filter
output signal is an image of the original NF-signal modulated onto the carrier
coupled with a DC-voltage as a function of the center frequency. To reduce the
influence of the center frequency on the DC-voltage on the output of FSK-switch
and data filter subsequently, a negative DC-feedback is realized as explained
in Chapter 4.8.1 (see also Figure 4-19).
Because of the wide bandwith of the PLL of ±500kHz, not only fast catching, but
also demodulation of a RF-frequency causing an IF-frequency in that band is
enabled. Therefore deviation of the IF-frequency from the nominal value - inside
the band of ±500kHz - , caused by the tolerances of the reference oscillator for
instance, does not effect the FSK-demodulation.
Wireless Control Components
4 - 29
Application Note, February 2004
TDA 521X
Functional Description
IF-Signal
NF-Signal
PD
Loop Filter
VCO
FSK-PLL-Dem.wmf
Figure 4-24
Block diagram of the FSK-PLL-Demodulator
The amplitude on the PLL-Demodulator output is not depending on the IF-signal
amplitude, but only on the deviation of the FSK-signal. Using a data signal with
a frequency between f2 and f3 according Figure 4-20 in Chapter 4.8.1, not too
close by one of this, the amplitude on the output of the FSK-switch is still only a
function of the deviation. Outside of this bandwith the amplitude is also a function of the frequency of the data signal, as shown in Figure 4-20. The upper cutoff-frequency f3 is round about 52kHz as indicated in Table 4-10.
Table 4-10 Data-Frequency response of the FSK-switch
Frequency of the
data signal
Amplitude on the FSK-switch out
Deviation
[kHz]
[mV]
[dBr]a
[kHz]
5
714
0
±150
20
659
-0.7
±150
30
636
-1
±150
43
567
-2
±150
52
506
-3
±150
62.5
420
-4.6
±150
a. dB relating to the reference. Reference value is amplitude at 5kHz
Wireless Control Components
4 - 30
Application Note, February 2004
TDA 521X
Functional Description
4.10 Data Filtering
Utilising the on-board voltage follower and the two 100kΩ on-chip resistors a
2nd order Sallen-Key low pass data filter can be constructed by adding 2 external capacitors between pins 19 (SLP) and 22 (FFB) and to pin 21 (OPP) as
depicted in the following figure and described in the following formulas1. As can
be seen from Figure 4-25, the cutoff frequency can be set by the external components C12 and C14. The data filter bandwidth should be set according to the
highest frequency component of the baseband signal received. A wider data filter bandwidth does reduce the sensitivity by passing a wider spectrum of noise
to the data slicer.
C14
C12
FFB
OPP
22
RF1 int
100k
RF2 int
SLP
21
19
100k
Filter_Design.wmf
Figure 4-25
Data Filter Design
C14 =
2Q b
R2πf 3dB
C12 =
b
4QRπf 3dB
RF1int = RF 2 int = R
with
Q=
b
a
Q...quality factor of the poles
Where in case of a Bessel filter a = 1.3617, b = 0.618
and thus Q = 0.577
1. taken from Tietze/Schenk: Halbleiterschaltungstechnik, Springer Berlin, 1999
Wireless Control Components
4 - 31
Application Note, February 2004
TDA 521X
Functional Description
and in case of a Butterworth filter a = 1.414, b = 1
and thus Q = 0.71
Example: Butterworth filter with f3dB = 5kHz and R = 100kΩ:
C14 = 450pF, C12 = 225pF
4.11 Data Slicer
The filtered data signal is fed to the data slicer, which is a one-bit analog-to-digital converter that makes the bit decision and provides a digital data slicer.
There are two different internal analog-to-digital converters. They differ in how
to generate the slice reference.
DataSlicer_adaptive.wmf
Figure 4-26
Data slicer with adaptive slice reference
As can be seen from Figure 4-26, the first circuit is the conventional adaptive
data slicer deriving the threshold by means of a separate low-pass filter. This
data slicer should be used for digital conversion of coded signals with no or only
small DC components. The low-pass filter is designed for a long time constant
in order to derive the average RSSI value (DC component of data) as an adaptive reference for the data slicer. As a design rule, the time constant TA should
be at least 3 times the longest period TL of no signal change within the data
sequence
T A ≥ 3 × TL
TA can be calculated as
TA =
R1⋅ ( RF 3int + RF 4 int )
R1 + RF 3int + RF 4 int
⋅ C13
= R1II ( RF 3 int + RF 4 int ) ⋅ C13
⋅ C13
=
... for ASK
and
TA =
R1⋅ RF 4 int
R1 + RF 3 int + RF 4 int
R1II ( RF 3int + RF 4 int )
v
⋅ C13
... for FSK
R1, RF3 int, RF4 int and C13 see also Figure 4-1
Wireless Control Components
4 - 32
Application Note, February 2004
TDA 521X
Functional Description
This will result in a temporary shift of the reference level by -3.5dB of the data
signal amplitude.
The time constant selected directly affects the data slicer run-in time and as a
result the receiver settling time.
The calculation above yields the TA of the TDA521X, in contrast to the one from
TDA520X, where TA has to be calculated as
TA = R1 × C13
The Relation between TA and the lower cut-off frequency f2 of the ASK/FSKSwitch (see also chapter „4.8.1 FSK Mode“) is shown below.
Considering the formula of f2 given in chapter „4.8.1 FSK Mode“
f 2 = v ⋅ f1
it can be easily seen, that f2 can also be calculated as
f 2 = v ⋅ f1 =
2π ⋅ C13 ⋅
1
R1II ( RF 3int + RF 4 int )
v
Applying additional the formula of TA given above, f2 can be calculated as
f2 =
1
2π ⋅ T A
Naming the ratio between TA / TL as „rt“ and considering, that
f data min =
1
2 ⋅ TL
the ratio between fdata min and f2 is given as
f data min
f2
= π ⋅ rτ
Example:
Choosing TA at least three times higher than TL (rt = 3), as recommended
above, results in
f data min
f2
= 3 ⋅ π ≈ 9,425
which should be sufficiently high for most of the applications.
The applications shown in Figure 4-28, Figure 4-29 and Figure 4-31 are only
usable in ASK mode.
Wireless Control Components
4 - 33
Application Note, February 2004
TDA 521X
Functional Description
As can be seen in Figure 4-19 the ASK/FSK switch is connected via a 300k and
30k resistor with pin 20. That’s the reason why the resistance R2IIR3 “seen” by
pin 20 in Figure 4-27 and Figure 4-28 must not be too high, otherwise a inadmissible ripple caused by the data signal is applied to the negative input of the
data slicer.
Figure 4-27 shows the alternate clamping data slicer which references its bit
decision to the positive peak level of the data signal. This data slicer can be
used with coding schemes employing high DC content. The decay time constant TP of the peak detector should be designed to hold the detector voltage
level well above the decision threshold of the comparator for a time long enough
to bridge the longest time TL with no level change within the data stream. This
is most important for low data levels at the sensitivity threshold of the receiver.
On the other hand TP should be short enough to follow the variations of the
received signal level. As a design rule, the time constant should be
TP = (R2 + R3) × C15 ≥ 20 × TL1
The comparator threshold will be set by the division ratio of the voltage divider
R2 /R3. The voltage drop across R2 acts like an additional offset voltage across
the comparator input. It should be set to a value just big enough for a glitch-free
decoding of a noisy signal at the receiver sensitivity threshold. Setting the
threshold beyond the noise level of the receiver will give an operation similar to
a squelch function. The data output will be held in a high state without a signal
at the receiver input then. There will be some degradation in receiver sensitivity
when applying the clamping data slicer.
C15
SLP
19
R2
R3
PDO
SLN
DATA
26
20
25
LP
Filter
Slicer
Dataslicerclamp.wmf
Figure 4-27
Data slicer with clamping slice reference
The data output DATA, pin25 is a common collector stage utilizing an internal
pull-down resistor of 40kΩ. The bandwidth of this output is limited to approx.
60kHz when loaded by 1MΩ//10pF. This limits the data rate to a maximum of
120kbit/s. Higher capacitive loading at the data output will further reduce the
bandwidth. The bandwidth of the RSSI decoder, the low-pass filter and the data
slicer are internally limited to >100kHz.
Wireless Control Components
4 - 34
Application Note, February 2004
TDA 521X
Functional Description
In case of using the adaptive data slicer the evaluation board has been
designed for
T A = R1× C13 = 100kΩ × 47nF = 4.7ms
The longest uncompensated DC component of the decoded data stream therefore should not exceed 1.5ms. The lower frequency of the data stream is limited
to 330Hz this way.
When using the peak data slicer, TP has been designed for
TP = ( R 2 + R3) × C15 = (100kΩ + 820kΩ) × 47nF = 43ms
The longest uncompensated DC component of the data stream is limited to
2.1ms now. The lower frequency limit of the data slicer has been extended
down to 250Hz.
The disadvantage of the application shown in Figure 4-27 is the dependance of
the duty cycle on the DC level of the data filter output signal and therefore also
on the RF level.
The necessary slicing level for getting a duty cycle of approximately 50% for a
stronger signal is to low for a weak signal. Because of this the sensitivity is virtually reduced.
Coding schemes with long break times between the preamble and the real data,
for instance the shark protocol, cause also problems with such an application.
Even for a large decay time Tp=(R2+R3)*C15 , causing problems after interferer appears, the slicing level will drop below the data filter output voltage
because of discharging C16 against GND, especially for small signals.
To avoid on the one hand the necessity of too large decay time and therefore
large recovery time after appearance of interferer or on the other hand a invalid
data signal the following application shown in Figure 4-28 is useful.
R3
R2
C15
Rdisch
SLP
PDO
26
DATA
25
SLN
19
20
Data
Filter
Slicer
DataSlicerDataValid.wmf
Figure 4-28
Slicing level derived from peak detector1
1.As the circuits or parts of the circuits shown in Figure 4-28, Figure 4-31, Figure 434 and Figure 4-35 are not used in the Infineon Evalboard an index is used (not only
a number).
Wireless Control Components
4 - 35
Application Note, February 2004
TDA 521X
Functional Description
The peak detector capacitor will be discharged mainly against the data filter output and not against GND. The value of Rdisch should be chosen high enough
considering the low current (40µA) of the current source for biasing the emitter
follower on the data filter output. R2+R3 should be at least 3 times Rdisch.
τ
Decay
≈ ( R 2 + R 3 ) // R disch ⋅ C 15
≈ R disch ⋅ C 15
for R 2 + R 3 >> R disch
The steady state voltage of the peak detector output during no transmission of
data is given as
U PDO, steady state = U data filter, Noisefloor ×
R2 + R3
R 2 + R 3 + R disch
The steady state voltage of the slicing level during no data transmission can be
calculated as
U SLN, steady state =
R3
R 2 + R 3 + R disch
Disch. To GND -65dBm1.wmf
Figure 4-29
Wireless Control Components
Discharging to GND; RF-Level = -65dBm
4 - 36
Application Note, February 2004
TDA 521X
Functional Description
Disch. To data f. out -100dBm.wmf
Figure 4-30
Discharging to data filter out; RF-Level = -100dBm
After a sudden increased RF level or a short time period in power down mode
the following application shown in Figure 4-31 is very useful to achieve a very
short settling time of the slicing level. This can be done without the usage of the
precharge circuit, but it has to be mentioned that only ASK mode will work with
this kind of application.
RP
C13
CP
R1
PDO
SLP
26
19
SLN
DATA
20
25
Data
Filter
Slicer
DataslicerFast.wmf
Figure 4-31
Wireless Control Components
Precharging with peak detector
4 - 37
Application Note, February 2004
TDA 521X
Functional Description
For the best functionality the value of Cp should be the same as the value of
C13. To avoid influence on the steady state of the slicing level Rp should be
chosen to a value of at least 10 times R13. A to small Rp value causes also an
additional ripple voltage of the slicing level. An example for various values of Rp
is given in Figure 4-32 and Figure 4-33 below).
Average of Slicing Level
2,5
Uslicing average
2
1,5
1
0,5
0
1000
316,228
100
31,6228
10
3,16228
1
0,31623
0,1
0,03162
ta/tp
Average of Slicing Level
Average of slicing level
Figure 4-32
Average of slicing level
Ripple of Slicing Level
50
45
Uslicing Ripple [mV]
40
35
30
25
20
15
10
5
0
1000
316,228
100
31,6228
10
3,16228
1
0,31623
0,1
0,03162
τ a/ττ p
Ripple of Uslicing level
Ripple of slicing level
Figure 4-33
Wireless Control Components
Ripple of slicing level
4 - 38
Application Note, February 2004
TDA 521X
Functional Description
The best solution for all demands mentioned above, combining the advantages
off the other applications is shown below in Figure 4-34.
D2SL
R1SL
19 SLP
Data Filter
D1SL
Slicing
Level
R2SL
CSL
Slicing Level
Generation
TDA521X
+
20 SLN
-
Data
(Slicer)
Out
Data Slicer
inv_slicing_level.wmf
Figure 4-34
Adapting impedance slicing level generation
R4nSL
T2SL
R3nSL
19 SLP
R1SL
R2SL
Slicing
Level
R3pSL
Data Filter
CSL
R4pSL
TDA521X
T1SL
Slicing Level
Generation
+
-
Data Slicer
20 SLN
Data
(Slicer)
Out
R3nSL,R4nSL, R3pSL and R4pSL "only" optional
inv_slicing_level_trans1.wmf
Figure 4-35
Wireless Control Components
Adapting impedance slicing level generation and buffer
4 - 39
Application Note, February 2004
TDA 521X
Functional Description
The two diodes D1SL and D2SL are working as a voltage controlled resistor. For
a slight deviation of the slicing level from the instantaneous value off the data
filter output signal the impedance of the diode, driven in forward direction, is
higher than for a larger deviation. The higher the deviation the lower the impedance of the forward driven diode. The function of the circuit presented in Figure
4-35 is the same except the using of the transistors instead of the diodes working simultaneously as buffer for the possibility to use a smaller value of R2SL.
The dimensioning of the circuit is not realy critical. The product of R2SL*CSL, the
theoretical lower limit value of the time constant, should be round about ten
times smaller than the inverse value of the highest “data frequency”1 to reach a
very fast settling time.
Depending on the longest possible period of no changing of the logik level during one transmission - time between preamble and data for instance - the
R2SL*CSL product can be chosen more or less smaller. As can be seen in the
measurements below, a R2SL*CSL product of round about 100µs is one the one
hand by far large enough to tide over a brake of 10ms, on the other hand small
enough to reach a proper slicing level within the half period of a 1kHz signal.
But also an R2SL*CSL product of less than 10µs is sufficient to tide over a brake
of 10ms, shown in Figure 4-38, enabling a very short settling time for the slicing
level as can be seen in Figure 4-39. The measurement presented in Figure 439 is done with the circuit shown in Figure 4-35, - slicing level generation and
buffer without the optional resistors R3n, R4n, R3p, R4p - because of the small
value of R2SL.
This application works for FSK as well as for ASK and make the use of precharge circuit described in the next chapter unnecessary.
But please note, that in case of FSK this circuit is not only decisive for the time
constant TA, but also for the cut-off frequencies f1 and f2 (see also Chapter
4.8.1).
The components R1SL, R2SL, D1SL and D2SL decide the resistance value,
which can be used in the calculation of TA, f1 and f2 instead of R1.
The capacitor value CSL can be inserted instead of C13 in those formulas.
1. Fundamental frequency of the data signal, not to mix up with the data rate.
Wireless Control Components
4 - 40
Application Note, February 2004
TDA 521X
Functional Description
T(Data-low)=10ms_Var_b.wmf
Figure 4-36
Measuring circuit with R2*C=127µs with 10ms of low level respectively
RF-off to -80dBm_Var_b.wmf
Figure 4-37
Wireless Control Components
Response time of circuit with R2*C=127µs; Data frequency = 1kHz
4 - 41
Application Note, February 2004
TDA 521X
Functional Description
T(Data-low)=10ms_Var_Tr2.wmf
Figure 4-38
Measuring circuit with R2*C=9.4µs with 10ms of low level respectively
RF-off to -80dBm_Var_Tr2.wmf
Figure 4-39
Wireless Control Components
Response time of circuit with R2*C=9.4µs; Data frequency = 1kHz
4 - 42
Application Note, February 2004
TDA 521X
Functional Description
4.12 Principle of the Precharge Circuit
In case the data slicer threshold shall be generated with an external RC network
as described in Section 4.11 it is necessary to use large values for the capacitor
C attached to the SLN pin (pin 20) in order to achieve long time constants. This
results also from the fact that the choice of the value for R connected between
the SLP and SLN pins (pins 19 and 20) is limited by the 330kΩ resistor appearing in parallel to R as can be seen in Figure 4-19. Apart from this a resistor value
of 100kΩ leads to a voltage offset of 1mv at the comparator input as described
in Section 4.8.1. The resulting startup time constant τ1 can be calculated with:
τ 1 = (R1 || 330kΩ) × C13
In case R1 is chosen to be 100kΩ and C13 is chosen as 47nF this leads to
τ 1 = (100kΩ || 330kΩ ) × 47nF = 77kΩ × 47nF = 3.6ms
When the device is turned on this time constant dominates the time necessary
for the device to be able to demodulate data properly. In the powerdown mode
the capacitor is only discharged by leakage currents.
In order to reduce the turn-on time in the presence of large values of C a precharge circuit was included in the TDA5210 as shown in the following figure.
C18
R4+R5=600k
R5
R4
C13
R1
Uthreshold
3VOUT
THRES
23
Uc>Us
SLN
20
Uc<Us
Iload
SLP
Uc
19
Data Filter
ASK/FSK Switch
-
24
U2
0 / 240uA
+
Us
OTA
+
-
U2<2.4V : I=240uA
U2>2.4V : I=0
20k
+3.1V
+2.4V
precharge.wmf
Figure 4-40
Principle of the precharge circuit
This circuit charges the capacitor C13 with an inrush current Iload of 240µA for
a duration of T2 until the voltage Uc appearing on the capacitor is equal to the
voltage Us at the input of the data filter. This voltage is limited to 2.5V. As soon
as these voltages are equal or the duration T2 is exceeded the precharge circuit
is disabled.
Wireless Control Components
4 - 43
Application Note, February 2004
TDA 521X
Functional Description
τ2 is the time constant of the charging process of C18 which can be calculated
as
τ 2 = 20kΩ × C18
as the sum of R4 and R5 is sufficiently large and thus can be neglected. T2 can
then be calculated according to the following formula:


1
T2 = τ 2 ln
 1 − 2.4V

3V



 = τ 2 × 1.6



The voltage transient during the charging of C18 is shown in the following figure:
U2
3V
2 .4 V
2
T2
e-Fkt1.wmf
Figure 4-41
Voltage appearing on C18 during precharging process
The voltage appearing on the capacitor C13 connected to pin 20 is shown in the
following figure. It can be seen that due to the fact that it is charged by a constant current source it exhibits a linear increase in voltage which is limited to
USmax = 2.5V which is also the approximate operating point of the data filter
input. The time constant appearing in this case can be denoted as T3, which can
be calculated with
T3 =
Wireless Control Components
4 - 44
U S max × C13
2.5V
=
× C13
240µA
240µA
Application Note, February 2004
TDA 521X
Functional Description
Uc
Us
T3
e-Fkt2.wmf
Figure 4-42
Voltage transient on capacitor C13 attached to pin 20
As an example the choice of C18 = 20nF and C13 = 47nF yields
τ2 = 0.4ms
T2 = 0.64ms
T3 = 0.49ms
This means that in this case the inrush current could flow for a duration of
0.64ms but stops already after 0.49ms when the USmax limit has been reached.
T3 should always be chosen to be shorter than T2.
It has to be noted finally that during the turn-on duration T2 the overall device
power consumption is increased by the 240µA needed to charge C13.
The precharge circuit may be disabled if C18 is not equipped. This yields a T2
close to zero. Note that the sum of R4 and R5 has to be 600kΩ in order to produce 3V at the THRES pin as this voltage is internally used also as the reference for the FSK demodulator.
Wireless Control Components
4 - 45
Application Note, February 2004
TDA 521X
Functional Description
4.13 Quiet Data Output during no Transmission
Using one of the above mentioned circuits to generate the slicing level the data
slicer will provide a stochastic signal. For some application like the duty cycle
mode, where a µC is set to idle mode and should only “waken” if valid data are
available, a quite data output or the additional information ”No Data” during no
transmission is required. For such an application one of the the following circuit
is a possible solution. The circuit shown in Figure 4-43 shows a much higher
temperature sensitivity as the circuit shown in Figure 4-44. Please note, that the
temperature sensitivity inherent in the circuit of Figure 4-43 is mainly caused by
the temperature gradient of the base-emitter voltage of T2FND.
TDA
521X
5V
Peak
Detect.
26 PDO
R3FND
220Ω
R2FND
22k
R1FND
2k2
C1FND
100pF
C2FND
4n7
R4FND
470k
R5FND
1k
R6FND
180Ω
T1FND
BC337
5V
5V 5V
R7FND
27k
C3FND
100nF
R8FND
10k
C5FND
100nF
T2FND
BC327
R12FND
R9FND 470k
390Ω
R10FND
56k
R11FND
12k
Data
Detector Out
High = Data
valid
T3FND
BC337
C4FND
47nF
FSK-Noise-Det1.wmf
Figure 4-43
FSK-Noise-Detector (not temperature stablized version)
To compensate the temperature dependence of T2FND only one additional transistor has to be used to get a differential pair as presented in Figure 4-44.
Another advantage of this circuit is the possibility to modify the threshold level1
without changing the gain of the previous stage and subsequently influencing
the tolerances by changing the voltage divider R13FND - R14FND. To avoid toggling of the “Data-Detector Output” for an input signal at the threshold level a
hysteresis can be realized by selecting a proper value of R12FND.
1.Level of the RF-signal where the “Noise-Detector” switches to high (data valid) is
“threshold-high”, level where it switches to low (noise detected) is “threshold-low”.
Wireless Control Components
4 - 46
Application Note, February 2004
TDA 521X
Functional Description
5V
5V
5V
5V
Peak Detect.
26 PDO
R2FND
22k
R3FND
220Ω
C1FND
100pF
R1FND
2k2
C2
4n7
R7FND
33k
R4FND
470k
R6FND
180Ω
T1FND
BC337
C3FND
100nF
R14FND
47k R13FND
100k
1k
T2FND
R5FND
1k
BC557
TDA 521X
R15FND
C5FND
100nF
T3FND
BC557
R12FND
800k
R11FND
12k
Data
Detector Out
High = Data
valid
R16FND
100k
R8FND
10k
T4FND
BC337
R9FND
390Ω
C4FND
47nF
R10FND
56k
FSK-Noise-Det2.wmf
Figure 4-44
FSK-Noise-Detector (Temperature-stabilized version)
A comparison between the circuit shown in Figure 4-43 and the temperature
stabilized circuit of Figure 4-44 is presented in Table 4-11 below, measured in
a 868MHz board.
Table 4-11 Temperature sensitivity
Temperature
Alteration of “threshold-high”
not temperature stabilized
circuit Figure 4-43
Temperature stabilized
circuit in Figure 4-44
[°C]
[dBm]
[dBm]
50
-100.4
-99.7
60
-99.4
-99.7
70
-98.4
-99.7
80
-97.2
-99.7
90
-95.5
-99.7
100
-92.6
-99.7
105
-90.1
-99,7
In point of fact the temperature drift caused by T1FND isn’t covered by the measurement shown in Table 4-11, but it amounts only round about 0.4dB over the
measured temperature range.
Wireless Control Components
4 - 47
Application Note, February 2004
TDA 521X
Functional Description
The hysteresis of the realized temperature stabilized circuit is round about
0.7dB resulting to a “threshold-low” of -100.4dBm in temperature range up to
105°C.
But of course both the hysteresis and the threshold can be modified as
described above.
The data detector output signal can be used as an interrupt signal to waken the
µC for instead. To get a quiet data slicer output during no transmission, the data
(slicer) output ha sto be combined with the data detector output via a AND or
NAND gate. Using a NAND it has to be observed that data are inverted and,
unlike using a AND, the quiescent state is high.
Wireless Control Components
4 - 48
Application Note, February 2004
TDA 521X
Functional Description
4.14
Decoder
For demonstration purposes, a standalone decoder device has been added to
the receiver board. The HCS 512 (Microchip) is a decoder capable of responding to the code-hopping sequences sent by the encoder device HCS 360
(Microchip) at the transmitter side. Both devices, the encoder and the decoder,
must be programmed with an identical customer code. Each decoder can
undergo a learning process with up to 5 individual transmitters. Synchronization
will be established following a defined learning sequence. Closing the “learn”
contact will light up the LED for approx. 2s. Opening the contact after the LED
has turned off sets the decoder to the learn mode. Activating the transmitter
then will transfer the synchronization data. This completes the learning process.
The LED will then light up for 500ms each time a synchronized transmitter signal is received.
4.15 Settling Time
Some receiver applications target an average supply current of 1mA and less.
If an intermittent receiver operation is allowed, the supply current may easily be
reduced from the typical value of 4.6mA to less than 1mA. The pulsed operation of the receiver can be controlled by a signal applied to the PDWN, pin27
input. A high state will enable the receiver. For pulsed receiver operation, a
number of parameters need to be considered, such as receiver settling time,
system response time and on-off duration.
The receiver settling time depends on a number of application parameters:
■
Power up signal slew rate
■
Data slicer design
■
RF settling time
The RF settling time depends on the start-up time of the reference oscillator, the
settling time of the PLL loop and the settling time of the bias of the receiver
blocks. Due to the high excessive gain of the reference oscillator and the wide
bandwidth of the PLL loop filter of 150kHz, the VCO will lock within a time of less
than 1ms. The settling time of the receiver bias depends primarily on the size of
the blocking capacitor C11. The impedance of capacitor C11 has, however,
to be low compared to the input impedance of the IF amplifier at the IF frequency of 10.7MHz. A capacitor of 100pF gives a 1dB loss of IF gain at the
input impedance of 330Ω. Applying C11=10nF will give a bias settling time of
the IF amplifier of only 200µs.
The settling time of the data slicer is given by the low-pass characteristic of the
imparted filter functions.
Wireless Control Components
4 - 49
Application Note, February 2004
TDA 521X
Functional Description
The settling time of the data slicer using the precharge circuit is mainly decided
by C13 and the operating point of the data filter Us as described in chapter 4.12.
The maximum value of Us is round about 2.5V resulting in a maximum settling
time for a valid data signal:
T3 =
U S max × C13
240µA
For a capacitor C13=47nF (Infineon evaluation board):
T3 ≈ 0.49ms
The crystal oscillator settling time is proportional to the ratio of the motional
inductivity (L1)1 of the crystal to the effective negative resistor of the whole circuit.
For a given frequency of the crystal the inductivity and also Q is indirect proportional to the motional capacitance (C1)1 of the crystal.
Assuming a fixed value of the motional capacitance L11 decreases with the
square of the frequency.
Therefore the start up time of the crystal oscillator is decreased proportinal to
an increasing of the motional capacitance and with the square of the frequency.
TSOSC ~
L1
R
To achive a short start up time of the oscillator a step edge of the power supply
is helpful, so the use of unnecessary large “blocking capacitors” should be
avoided.
All the individual settling times have to be considered in finding the total settling
time of the receiver. On the evaluation board they are
■
Bias settling time
0.2ms
■
Start-up time of the reference oscillator
< 0.6ms
■
Power supply slew rate
< 1ms
■
TSA
0.49ms
■
TSP
0.3ms
The settling time of the evaluation receiver will be less than 3ms utilizing the
sliding data slicer together with the precharge circuit.
1.Parameter of a crystal not to be exchanged with the components of the evalboard.
Wireless Control Components
4 - 50
Application Note, February 2004
TDA 521X
Functional Description
4.16 Spurious Radiation
The receiver has to meet the European Telecommunications Standards Institute
ETS 300 220 requirements. Other requirements may apply for other countries.
The product family of short range devices (SRD) is divided into three classes of
equipment, each having its own set of minimum performance criteria. This classification is based upon the impact on persons and/or goods in case the equipment does not operate above the specified minimum performance level under
electromagnetic compatibility (EMC) stress.
Class 1 SRD equipment is a highly reliable communication media; e.g. serving
human life inherent systems, which result in a physical risk to a person. Applications of this class are fire detection, personal identification, telemetry in vehicles, etc.
Class 2 is relevant for medium reliable SRD communication. This is the case
when causing inconvenience to persons, which cannot simply be overcome by
other means. Such as, cargo handling systems, domestic telemetry, car alarms,
vehicle detection and many others.
Class 3 is responsible for standard reliable SRD communication systems, causing inconvenience to persons, which can simply be overcome by other means.
This is relevant for garage door openers, car lock/unlock devices, radio remote
control television and audio, door bell and so on.
According to ETS 300 220, the limits for spurious radiation on the receiver side
are applicable to all receiver classes:
-57 dBm below 1GHz
-47 dBm above 1GHz
Two different types of emission have to be considered, conducted (antenna
port) and board radiation. Conducted emission is defined as the spurious power
level at the antenna port. Radiated emission is defined as the “effective radiated
power” (ERP) emitted by the board.
The most important spurious signal is the LO signal. The low power design of
the VCO and a careful PCB layout keep the spurious radiation well below the
limits. The spurious levels at the relevant frequencies measured at the evaluation board can be summarized as:
Table 4-12 Spurious Radiation Levels
Source of emission
Receiver Frequency
Limit
434 MHz
868 MHz
ETS 300 220
858MHz
-90dBm
-73dBm
-57dBm
VCO/2
423MHz
-102dBm
<-120dBm
-57dBm
Radiated ERP VCO
858MHz
-73dBm
-67dBm
-57dBm
423MHz
<-120dBm
<-100dBm
-57dBm
Antenna port VCO
VCO/2
The TDA 5210 conforms to the ETS 300 220 requirements.
Wireless Control Components
4 - 51
Application Note, February 2004
TDA 521X
Functional Description
4.17 Sensitivity Measurements
4.17.1
Test Setup
The test setup used for the measurements is shown in the following figure. In
case of ASK modulation the Rohde & Schwarz SMIQ generator, which is a vector signal generator, is connected to the I/Q modulation source AMIQ. This
"baseband signal generator" is in turn controlled by the PC based software
WinQSIM via a GPIB interface. The AMIQ generator has a pseudo random
binary sequence (PRBS) generator and a bit error test set built in. The resulting
I/Q signals are applied to the SMIQ to generate a ASK (OOK) spectrum at the
desired RF frequency. In case of FSK modulation the SMIQ is replaced by a
Rohde & Schwarz SME generator.
Data is demodulated by the TDA521X and then sent back to the AMIQ to be
compared with the originally sent data. The bit error rate is calculated by the bit
error rate equipment inside the AMIQ.
Manchester coded data is applied to the I signal as can be seen in the subsequent figure.
Personal Computer
Software
WinQSIM
GPIB/
RS 232
Clock
Marker Output
Rohde & Schwarz
I/Q Modulation Source
AMIQ
I
Q
AMIQ BERT
(Bit Error Rate
Test Set)
Data
Manchester
Encoder
Rohde & Schwarz
(Vector) Signal Generator
SMIQ 03 / SME 03
ASK/FSK User Signal
Manchester
Decoder
RFin
DATAout
DUT
Receiver
Testboard
TDA5210
TestSetup.wmf
Figure 4-45
BER Test Setup
In the following figures the RF power level shown is the average power level.
Wireless Control Components
4 - 52
Application Note, February 2004
TDA 521X
Functional Description
These investigations have been made on an Infineon evaluation board using a
data rate of 4 kBit/s with manchester encoding, a IF filter bandwidth of 280 kHz
and a data filter bandwidth of 5 kHz. This is the standard configuration of our
evaluation boards. All these measurements have been undertaken with several
evaluation boards, so that production scattering and component tolerances are
already included in these results.
4.17.2
Sensitivity of RKE Receivers
The following frequency derivates of the TDA 521x family have been tested:
TDA 5210 working at 434 and 868 MHz
TDA 5211 working at 315 MHz
TDA 5212 working at 915 MHz
The target bit error rate (BER) is specified to a value of 2.10-3. This value is the
criteria to determine the sensitivity.
Frequency Derivates.wmf
Figure 4-46
Sensitivity of RKE Receivers
As can be seen in Figure 4-46 the TDA 5211 shows the best performance, this
is because of minimum noise matching at the front end. Thus it has to be mentioned that the other derivates also show the same behavior, if the matching
network at the front end would be matched for minimum noise.
Wireless Control Components
4 - 53
Application Note, February 2004
TDA 521X
Functional Description
The ASK sensitivity can be up to -113 dBm. In case of FSK -110 dBm are
achievable, using a deviation of 150 kHz peak to peak. All these values are
measured at an ambient temperature of about 20 °C.
4.17.3
Dependence of the ambient temperature
Demonstrating a wide band of application possibilities the temperature behavior
must not be forgotten. In automotive systems the demanded temperature range
is from -40 °C to +85 °C. Showing the receivers very good performance a BER
measurement at the temperature of +105 °C is also documented in the following graph.
Temperature.wmf
Figure 4-47
Temperature Behaviour
Notice that the sensitivity variation in this temperature range of -40 °C to
+105 °C is only about 2.5 to 3 dB.
Wireless Control Components
4 - 54
Application Note, February 2004
TDA 521X
Functional Description
4.17.4
Sensitivity depending on the IF Filter Bandwidth
A significant place to influence the receivers sensitivity is the bandwidth of the
applied IF filter.
IF Filter Bandwidth.wmf
Figure 4-48
Variable IF FIlter Bandwidth
In the case of ASK using a data rate of 4 kBit/s the IF filter bandwidth can be
reduced very dramatically to about 150 kHz resulting in a 1 dB improved sensitivity. A similar situation takes place in the FSK mode, where deviation has to
be taken into account. A very practicable configuration is to set the IF bandwidth
to a value of about 1.5 times the peak to peak deviation. Concerning these
aspects the bandwidth should be chosen small enough. With respect to the
quartz circuitry tolerances, which influence the receiving frequency, a to small
IF filter bandwidth will reduce the sensitivity again. So a compromise has to be
made. For further details on IF bandwidth calculation see also Section 4.7.2.
Wireless Control Components
4 - 55
Application Note, February 2004
TDA 521X
Functional Description
4.17.5
Dependence of Data Filter Bandwidth
Explaining this effect it has to be mentioned that a data rate of 4 kBit/s using
manchester encoding results in a data frequency of 2 kHz to 4 kHz depending
on the occurring data pattern. The test pattern given by the AMIQ is a pseudo
random binary sequency (PRBS9) with a 9 bit shift register. This pattern varies
the resulting data frequency up to 4 kHz.
Data Filter Bandwidth ASK.wmf
Figure 4-49
Wireless Control Components
Variation of Data Filter Bandwidth in case of ASK
4 - 56
Application Note, February 2004
TDA 521X
Functional Description
Data Filter Bandwidth FSK.wmf
Figure 4-50
Variation of Data Filter Bandwidth in case of FSK
As can be seen in the last two figures, a data filter bandwidth of about 5 kHz
shows optimal performance. Both lowering and increasing this bandwidth
results in less sensitivity. Thus the best results can be achieved using a data
filter bandwidth of 1.25 times the maximum of the appearing data frequency.
Wireless Control Components
4 - 57
Application Note, February 2004
5
Reference
Contents of this Chapter
5.1
5.2
5.3
Test Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
Test Board Layouts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-3
Bill of Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-5
TDA 521X
Reference
5.1 Test Circuit
The device performance parameters were measured on an Infineon evaluation
board. This evaluation board can be obtained together with evaluation boards
of the accompanying transmitter device TDA5100/TDA5101 in an evaluation kit
that may be ordered on the INFINEON RKE Webpage www.infineon.com/rke.
In case a matching codeword is received, decoded and accepted by the
decoder the on-board LED will turn on. This signal is also accessible on a 2-pole
pin connector and can be used for simple remote-control applications. More
information on the kit is available on request.
TDA5210_testboard_20_schematicV2.wmf
Figure 5-1
Wireless Control Components
Schematic of the Evaluation Board
5-2
Application Note, February 2004
TDA 521X
Reference
5.2 Test Board Layouts
TDA5210_testboard_20_topV3.wmf
Figure 5-2
Top Side of the Evaluation Board
TDA5210_testboard_20_bot.wmf
Figure 5-3
Wireless Control Components
Bottom Side of the Evaluation Board
5-3
Application Note, February 2004
TDA 521X
Reference
TDA5210_testboard_20_plcV2.wmf
Figure 5-4
Wireless Control Components
Component Placement on the Evaluation Board
5-4
Application Note, February 2004
TDA 521X
Reference
5.3 Bill of Materials
The following components are necessary for evaluation either of the TDA5210
used at 434MHz or 868MHz or of the TDA5211 at 315MHz without use of a
Microchip HCS512 decoder.
Table 5-1 Bill of Materials
Ref
Value
Specification
R1
100kΩ
0805, ± 5%
R2
100kΩ
0805, ± 5%
R3
820kΩ
0805, ± 5%
R4
240kΩ
0805, ± 5%
R5
360kΩ
0805, ± 5%
R6
10kΩ
0805, ± 5%
L1
315MHz: 15nH
434 MHz: 15nH
868 MHz: 3.3nH
Toko, PTL2012-F15N0G
Toko, PTL2012-F15N0G
Toko, PTL2012-F3N3C
L2
315MHz: 12pF
434 MHz: 8.2pF
868 MHz: 3.9nH
0805, COG, ± 1%
0805, COG, ± 0.1pF
Toko, PTL2012-F3N9C
C1
315MHz: 12pF
434MHz: 1pF
868MHz: 1pF
0805, COG, ± 1%
0805, COG, ± 0.1pF
0805, COG, ± 0.1pF
C2
315MHz: 10pF
434 MHz: 4.7pF
868 MHz: 3.9pF
0805, COG, ± 1%
0805, COG, ± 0.1pF
0805, COG, ± 0.1pF
C3
315MHz: 6.8pF
434 MHz: 6.8pF
868 MHz: 5.6pF
0805, COG, ± 0.1pF
0805, COG, ± 0.1pF
0805, COG, ± 0.1pF
C4
100pF
0805, COG, ± 5%
C5
47nF
1206, X7R, ± 10%
C6
315MHz: 15nH
434 MHz: 10nH
868 MHz: 3.9pF
Toko, PTL2012-F10N0G
Toko, PTL2012-F10N0G
0805, COG, ± 0.1pF
C7
100pF
0805, COG, ± 5%
C8
315MHz: 33pF
434 MHz: 33pF
868 MHz: 22pF
0805, COG, ± 5%
0805, COG, ± 5%
0805, COG, ± 5%
C9
100pF
0805, COG, ± 5%
C10
10nF
0805, X7R, ± 10%
C11
10nF
0805, X7R, ± 10%
C12
220pF
0805, COG, ± 5%
C13
47nF
0805, X7R, ± 10%
C14
470pF
0805, COG, ± 5%
Wireless Control Components
5-5
Application Note, February 2004
TDA 521X
Reference
Table 5-1 Bill of materials (continued)
Value
Specification
C15
47nF
0805, X7R, ± 5%
C16
8.2pF
0805, COG, ± 0.1pF
C17
22pF
0805, COG, ± 1%
C18
22nF
0805, X7R, ± 5%
Ref
Q1
315 MHz:(fRF–10.7MHz)/32
434 MHz:(fRF–10.7MHz)/32
868MHz:(fRF–10.7MHz)/64
HC49/U, fundamental mode, CL = 12pF,
e.g. 315MHz: Jauch Q 10,178130-S11-12-10/20
e.g. 434.2MHz: Jauch Q 13,234370-S11-12-10/20
e.g. 868.4MHz: Jauch Q 13,401550-S11-12-10/20
Q2
SFE10.7MA5-A or
SKM107M1-A20-10
Murata
Toko
X2, X3
142-0701-801
Johnson
X1, X4,
S1-S3, S6
2-pole pin connector
S4
3-pole pin connector, or not equipped
IC1
315 MHz: TDA5211
434 MHz:TDA5210
868 MHz: TDA5210
Infineon
Please note that in case of operation at 315MHz and 434 MHz a capacitor has
to be soldered in place of L2 and an inductor in place of C6.
The following components are necessary in addition to the above mentioned
ones for evaluation of the TDA5210 in conjunction with a Microchip HCS512
decoder.
Table 5-2 Bill of Materials Addendum
Ref
Value
Specification
R7
100kΩ
0805, ± 5%
R8
10kΩ
0805, ± 5%
R9
100kΩ
0805, ± 5%
R10
22kΩ
0805, ± 5%
R11
100Ω
0805, ± 5%
R12
100Ω
0805, ± 5%
R13
100Ω
0805, ± 5%
R14
100Ω
0805, ± 5%
R21
22kΩ
0805, ± 5%
R22
10kΩ
0805, ± 5%
R23
22kΩ
0805, ± 5%
R24
820kΩ
0805, ± 5%
R25
560Ω
0805, ± 5%
Wireless Control Components
5-6
Application Note, February 2004
TDA 521X
Reference
Table 5-2 Bill of Materials Addendum (continued)
C19
10pF
0805, COG, ± 5%
C21
100nF
1206, X7R, ± 10%
C22
100nF
1206, X7R, ± 10%
IC2
HCS512
Microchip
S5, X4-X9
2-pole pin connector
T1, T2
BC 847B
Infineon
D1
LS T670-JL
Infineon
Wireless Control Components
5-7
Application Note, February 2004
TDA 521X
List of Figures
6
List of Figures
Figure 2-1
Functional Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2-3
Figure 2-2
P-TSSOP-28-1 package outlines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2-4
Figure 3-1
IC Pin Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-2
Figure 4-1
LNA input matching network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-3
Figure 4-2
LNA voltage gain vs. frequency of 434MHz receiver . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-7
Figure 4-3
Input reflection and relative gain of the LNA vs. frequency
at 869MHz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-7
Figure 4-4
AGC circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-8
Figure 4-5
RSSI voltage as a function of input level for AGC implementation . . . . . . . . . . . . . . . .
4-9
Figure 4-6
AGC time constant; RF-Level=-70dBm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-11
Figure 4-7
Typical VCO tuning curve of TDA5210 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-15
Figure 4-7
Determination of Series Capacitance Value for the Quartz Oscillator . . . . . . . . . . . . . .
4-13
Figure 4-8
Phase noise spectrum of local oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-15
Figure 4-9
Determination of Series Capacitance Value for the Quartz Oscillator . . . . . . . . . . . . . .
4-16
Figure 4-10 Tolerances caused by DL of 20% . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-19
Figure 4-11 Tolerances of the oscillator caused by DCs of 0.1pF . . . . . . . . . . . . . . . . . . . . . . . . . .
4-18
Figure 4-12 Tolerances of the oscillator caused by DCs of 1% . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-20
Figure 4-13 Tolerances of the oscillator caused by DCs and DCs . . . . . . . . . . . . . . . . . . . . . . . . . .
4-20
Figure 4-14 Rs/R1 vs CL/C0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-21
Figure 4-15 Forcing the crystal oscillator without transformer . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-22
Figure 4-16 Forcing the crystal oscillator with transformer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-23
Figure 4-17 Frequency response of the IF amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-24
Figure 4-18 IF frequency response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-25
Figure 4-19 ASK/FSK mode datapath . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-27
Figure 4-20 Frequency characteristic in case of FSK mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-28
Figure 4-21 Frequency characteristic in case of ASK mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-29
Figure 4-22 Limiter demodulator characteristic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-30
Figure 4-23 RSSI Voltage vs. Receiver Input Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-31
Figure 4-24 Block diagram of the FSK-PLL-Demodulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-32
Figure 4-25 Data Filter Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-34
Figure 4-26 Data slicer with adaptive slice reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-36
Figure 4-27 Data slicer with clamping slice reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-37
Figure 4-28 Precharging with peak detector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-38
Figure 4-29 Average of slicing level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-39
Figure 4-30 Ripple of slicing level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-39
Figure 4-31 Slicing level derived from peak detektor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-40
Figure 4-32 Discharging to GND; RF-Level = -65dBm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-41
Figure 4-33 Discharging to data filter out; RF-Level = -100dBm . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-41
Figure 4-34 Adapting impedance slicing level generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-42
Wireless Control Components
List of Figures - i
Application Note, February 2004
TDA 521X
List of Figures
Figure 4-34 Adapting impedance slicing level generation and buffer . . . . . . . . . . . . . . . . . . . . . . . .
4-42
Figure 4-36 Measuring circuit with R2*C=127µs with 10ms of low level respectively . . . . . . . . . . . .
4-43
Figure 4-37 Response time of circuit with R2*C=127µs; Data frequency = 1kHz . . . . . . . . . . . . . . .
4-44
Figure 4-38 Measuring circuit with R2*C=9.4µs with 10ms of low level respectively . . . . . . . . . . . .
4-44
Figure 4-39 Response time of circuit with R2*C=9.4µs; Data frequency = 1kHz . . . . . . . . . . . . . . .
4-45
Figure 4-40 Principle of the precharge circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-46
Figure 4-41 Voltage appearing on C2 during precharging process . . . . . . . . . . . . . . . . . . . . . . . . . .
4-47
Figure 4-42 Voltage transient on capacitor C attached to pin20 . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-48
Figure 4-43 FSK-Noise-Detector (low price version) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-49
Figure 4-44 FSK-Noise-Detector (Temperature-stabilized version). . . . . . . . . . . . . . . . . . . . . . . . . .
4-50
Figure 4-45 BER Test Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-56
Figure 4-46 Sensitivity of RKE Receivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-57
Figure 4-47 Temperature Behaviour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-58
Figure 4-48 Variable IF Filter Bandwidth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-59
Figure 4-49 Variation of Data Filter Bandwidth in case of ASK . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-60
Figure 4-50 Variation of Data Filter Bandwidth in case of FSK . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-61
Figure 5-1
Schematic of the Evaluation Board . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5-2
Figure 5-2
Top Side of the Evaluation Board . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5-3
Figure 5-3
Bottom Side of the Evaluation Board . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5-3
Figure 5-4
Component Placement on the Evaluation Board . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5-4
Wireless Control Components
List of Figures - ii
Application Note, February 2004
TDA 521X
List of Tables
7
List of Tables
Table 2-1
Summary of the Key Parameters of the Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2-3
Table 3-1
Pin Definition and Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-3
Table 4-1
S-Parameters of the LNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-2
Table 4-2
Tolerance Values of the LNA Input Matching Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-4
Table 4-3
Dependence of TL Value on C5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-10
Table 4-4
Measured Mixer Performance at 434 MHz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-12
Table 4-5
Measured Mixer Performance at 869 MHz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-12
Table 4-6
Measured Mixer Performance at 315 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-13
Table 4-7
Measured IIP3 at 434 MHz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-13
Table 4-8
Assessment of the Frequency Error of the Crystal Oscillator . . . . . . . . . . . . . . . . . . . .
4-22
Table 4-9
Design Example: IF Bandwidth Calculation (3σ value) . . . . . . . . . . . . . . . . . . . . . . . . .
4-25
Table 4-10 Frequency response of the FSK-switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-33
Table 4-10 Temperature sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-50
Table 4-11 Spurious Radiation Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-55
Table 5-1
Bill of Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5-5
Table 5-2
Bill of Materials Addendum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5-6
Wireless Control Components
List of Tables - i
Application Note, February 2004