AD AD6122ACPRL

a
CDMA 3 V Transmitter IF Subsystem
with Integrated Voltage Regulator
AD6122
FEATURES
Fully Compliant with IS98A and PCS Specifications
Linear IF Amplifier
–63 dB to +34 dB
Linear-in-dB Gain Control
Temperature-Compensated Gain Control
Quadrature Modulator
Modulates IFs from 50 MHz to 350 MHz
Integral Low Dropout Regulator
Accepts 2.9 V to 4.2 V Input from Battery
Low Power
10.4 mA at Midgain
<10 ␮A Sleep Mode Operation
Companion Receiver IF Chip Available (AD6121)
APPLICATIONS
CDMA, W-CDMA, AMPS and TACS Operation
QPSK Transmitters
GENERAL DESCRIPTION
The AD6122 is a low power IF transmitter subsystem, specifically designed for CDMA applications. It consists of an I and Q
modulator, a divide-by-two quadrature generator, high dynamic
range IF amplifiers with voltage-controlled gain and a powerdown control input. An integral low dropout regulator allows
operation from battery voltages from 2.9 V to 4.2 V.
The gain control input accepts an external gain control voltage
input from a DAC. It provides 97 dB of gain control with a
nominal 75 dB/V scale factor. Either an internal or an external
reference may be used to set the gain-control scale factor.
The I and Q modulator accepts differential quadrature baseband inputs from a CDMA baseband converter. The local oscillator is injected at twice the IF frequency. A divide-by-two
quadrature generator followed by dual polyphase filters ensures
± 1° quadrature accuracy.
The modulator provides a common-mode reference output to
bias the transmit DACs in the baseband converter to the same
common-mode voltage as the modulator inputs, allowing dc
coupling between the two ICs and thus eliminating the need to
charge and discharge coupling capacitors. This allows the fastest
power-up and power-down times for the AD6122 and CDMA
baseband ICs.
The AD6122 is fabricated using a 25 GHz f t silicon BiCMOS
process and is packaged in a 28-lead SSOP and a 32-leadless
LPCC chip scale package (5 mm × 5 mm).
FUNCTIONAL BLOCK DIAGRAM
VCC
QUADRATURE
MODULATOR
OUTPUT
ATTENUATOR
IF AMPLIFIER
INPUT
QUADRATURE MODULATOR
I INPUT
IF AMPLIFIERS
LOCAL
OSCILLATOR
INPUT
TRANSMIT
OUTPUT
ⴜ2
AD6122
Q INPUT
COMMON-MODE
REFERENCE
OUTPUT
VPOS
LOW
DROPOUT
REGULATOR
POWER- POWERDOWN 1 DOWN 2
VREG
GAIN
CONTROL
SCALE
FACTOR
TEMPERATURE
COMPENSATION
1.23 V
GAIN CONTROL GAIN CONTROL
VOLTAGE
REFERENCE REFERENCE
INPUT
OUTPUT
VOLTAGE
INPUT
REV. B
Information furnished by Analog Devices is believed to be accurate and
reliable. However, no responsibility is assumed by Analog Devices for its
use, nor for any infringements of patents or other rights of third parties
which may result from its use. No license is granted by implication or
otherwise under any patent or patent rights of Analog Devices.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 781/329-4700
World Wide Web Site: http://www.analog.com
Fax: 781/326-8703
© Analog Devices, Inc., 2000
(T = +25ⴗC, V
AD6122–SPECIFICATIONS
noted) NOTE: All powers shown in dBm are referred to 1 k⍀.
A
CC
= +3.0 V, LO = 2 ⴛ IF, REFIN = 1.23 V, LDO Enabled, unless otherwise
Specification
Conditions
MODULATOR
LO = 260.76 MHz (2 × IF), 100 mV p-p
500 mV p-p Differential I and Q Inputs;
Output Level Referred to a 1 kΩ Differential Load
Output Level
Output Third Order Harmonic
I/Q Inputs
Differential Input Voltage
Bandwidth
Resistance
Quadrature Accuracy
Amplitude Balance
Output Referred Noise
Modulator Common-Mode Reference
LO Input Resistance
LO Input Capacitance
LO Carrier Leakage
IF AMPLIFIER
Noise Figure
Input 1 dB Compression Point
Input Third-Order Intercept
Gain Flatness
Input Capacitance
Differential IF Input Resistance
Differential IF Output Resistance
Differential IF Output Capacitance
GAIN CONTROL INTERFACE
Gain Scaling
Gain Scaling Linearity
Minimum Gain
Maximum Gain
Gain Control Response Time
Input Resistance at REFIN
Input Resistance at VGAIN
POWER-DOWN INTERFACE
Logic Threshold High
Logic Threshold Low
Input Current for Logical High
Turn-On Response Time
Turn-Off Response Time
LOW DROPOUT REGULATOR
Min
Max
Unit
–21
–50
dBm
dBc
500
Differential Input at 260.38 MHz
Differential Input at 260.38 MHz
Bias I/Q Using MODCMREF
30
±1
± 0.1
–169
1.408
1.2
2.4
–40
mV p-p
MHz
kΩ
°
dB
dBm/Hz
V
kΩ
pF
dBc
FIF = 130.38 MHz
VGAIN = 2.5 V, 1 kΩ Differential Load
VGAIN = 2.5 V
VGAIN = 2.5 V
IF ± 630 kHz
Shunt Equivalent Model at 130.38 MHz
Shunt Equivalent Model at 130.38 MHz
Per Pin at 130.38 MHz
Per Pin at 130.38 MHz
10
–32
–24
± 0.25
2.3
680
4.2
2.0
dB
dBm
dBm
dB
pF
Ω
kΩ
pF
Using Internal Reference
For a Typical Dynamic Range of 92 dB
VGAIN = 0.5 V
VGAIN = 2.5 V
90 dB Gain Change, Min Gain to Max Gain
75
±3
–63
+34
0.7
10
109
dB/V
dB/V
dB
dB
µs
MΩ
kΩ
Power-Up on Logical High
1.34
1.30
0.1
23
187
V
V
µA
µs
ns
2.9–4.2
2.70
200
1.23
V
V
mV
V
2.7–5.0
10.4
7.8
V
mA
µA
Differential
–3 dB
20
0.9 MHz to 5.0 MHz Offsets
Measure to Settling of AGC from Standby Mode
To 200 µA Supply Current
External PNP Pass Transistor, VCE SAT = –0.4 V
Max, hFE = 100/300 Min/Max
Input Range
Nominal Output
Dropout Voltage
Reference Output
POWER SUPPLY
Supply Range Bypassing Internal LDO
Supply Current
Standby Current
Typ
VGAIN = 1.5 V (Unity Gain)
OPERATING TEMPERATURE
TMIN to TMAX
–40
+85
°C
Specifications subject to change without notice.
–2–
REV. B
AD6122
ABSOLUTE MAXIMUM RATINGS 1
NOTES
1
Stresses above those listed under Absolute Maximum Ratings may cause permanent damage to the device. This is a stress rating only; functional operation of the
device at these or any other conditions above those indicated in the operational
section of this specification is not implied. Exposure to absolute maximum rating
conditions for extended periods may affect device reliability.
2
Thermal Characteristics: 28-lead SSOP Package: θJA = 115.25°C/W.
Supply Voltage DVCC, IFVCC, TXVCC to DGND,
IFGND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +5 V
Internal Power Dissipation2 . . . . . . . . . . . . . . . . . . . 600 mW
Operating Temperature Range . . . . . . . . . . . –40°C to +85°C
Storage Temperature Range . . . . . . . . . . . . –65°C to +150°C
Lead Temperature Range (Soldering 60 sec) . . . . . . . . +300°C
PIN CONFIGURATIONS
LDOB 4
25 IFVCC
LDOC 5
24 IFGND
LDOC 2
DGND 7
AD6122
TOP VIEW
NC
IFVCC
REFIN
REFOUT
32 31 30 29 28 27 26 25
26 REFOUT
LDOB 1
LDOGND 6
VGAIN
27 REFIN
LDOE 3
PD1
28 VGAIN
PD2 2
PD2
PD1 1
LPCC Package
LDOE
SSOP Package
23 IIPP
LDOGND 3
22 IIPN
LDOGND 4
24 IFGND
23 IFGND
22 IIPP
21 IIPN
AD6122 Top View
(Not to Scale)
LOIPP
8 (Not to Scale) 21 MODCMREF
LOIPN
9
20 QIPN
LOIPP 6
19 QIPN
DVCC 10
19 QIPP
LOIPN 7
18 QIPP
IFINP
MODOPN
15 IFINN
IFINN
16 IFINP
IFGND 14
10 11 12 13 14 15 16
IFGND
TXVCC 13
17 MODOPP
9
IFGND
17 MODOPN
TXVCC
TXOPN 12
DVCC 8
TXOPP
18 MODOPP
20 MODCMREF
TXOPN
TXOPP 11
DGND 5
NC = NO CONNECT
ORDERING GUIDE
Model
Temperature
Range
Package Description
AD6122ARS
AD6122ARSRL
AD6122ACP
AD6122ACPRL
–40°C to +85°C
–40°C to +85°C
–40°C to +85°C
–40°C to +85°C
Shrink Small Outline Package (SSOP)
28-Lead SSOP on Tape-and-Reel
Chip Scale Package (LPCC)
32-Leadless LPCC on Tape-and-Reel
CAUTION
ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily
accumulate on the human body and test equipment and can discharge without detection.
Although the AD6122 features proprietary ESD protection circuitry, permanent damage may
occur on devices subjected to high-energy electrostatic discharges. Therefore, proper ESD
precautions are recommended to avoid performance degradation or loss of functionality.
REV. B
–3–
Package
Option
RS-28
CP-32
WARNING!
ESD SENSITIVE DEVICE
AD6122
PIN FUNCTION DESCRIPTIONS
SSOP
Pin #
LPCC
Pin #
Pin Label
Description
Function
1
30
PD1
Power-Down 1
2
31
PD2
Power-Down 2
3
32
LDOE
4
1
LDOB
5
2
LDOC
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
3, 4
5
6
7
8
9
10
11
12, 13
14
15
16
17
18
19
20
LDOGND
DGND
LOIPP
LOIPN
DVCC
TXOPP
TXOPN
TXVCC
IFGND
IFINN
IFINP
MODOPN
MODOPP
QIPP
QIPN
MODCMREF
22
23
24
25
26
21
22
23, 24
25
26
27
IIPN
IIPP
IFGND
NC
IFVCC
REFOUT
Low Dropout Regulator Pass
Transistor Emitter Connection
Low Dropout Regulator Pass
Transistor Base
Low Dropout Regulator Pass
Transistor Collector
Low Dropout Regulator Ground
Digital Ground
Local Oscillator “Positive” Input
Local Oscillator “Negative” Input
Digital VCC
Transmit Output “Positive”
Transmit Output “Negative”
Transmit Output VCC
IF Ground
IF Input “Negative”
IF Input “Positive”
Modulator “Negative” If Output
Modulator “Positive” Output
Q Input “Positive”
Q Input “Negative”
Modulator Common-Mode
Reference Out
I Input “Negative”
I Input “Positive”
Ground
No Connect
IF VCC
Gain Control Reference Output
IF Amplifier Power-Down Control Input; CMOS Compatible; HIGH = Entire IC Powers Down, LOW = IF
Amplifiers On.
Modulator Power-Down Control Input; CMOS Compatible; HIGH = Modulator Off , LOW = Modulator On.
Connects to Emitter of External PNP Pass Transistor
and VCC.
Connects to Base of External PNP Pass Transistor.
27
28
REFIN
Gain Control Reference Input
28
29
VGAIN
Gain Control Voltage Input
–4–
Connects to Collector of External PNP Pass Transistor.
Ground.
Ground.
Connects to Local Oscillator; AC Coupled.
Connects to Ground via Decoupling Capacitor.
Connects to Digital Supply.
Connects to Output Filter; AC Coupled.
Connects to Output Filter; AC Coupled.
Connects to LDO Output via Decoupling Network.
Ground.
IF “Negative” Input from LC Roofing Filter.
IF “Positive” Input from LC Roofing Filter.
Output Modulator Output to LC Roofing Filter.
Modulator Output to LC Roofing Filter.
Connects to Q “Positive” Output of Baseband IC.
Connects to Q “Negative” Output of Baseband IC.
Connects to CDMA Baseband Converter Tx DAC
Common-Mode Reference Input.
Connects to I “Negative” Output of Baseband IC.
Connects to I “Positive” Output of Baseband IC.
Connects to IF Ground.
Connects to Decoupled Output of LDO Regulator.
Provides 1.23 V Voltage Reference Output for DAC in
CDMA Baseband Converter and REFIN.
Accepts 1.23 V Reference Input from REFOUT or
External Reference.
Accepts Gain Control Input Voltage from External DAC.
Max Gain = 2.5 V; Min Gain = 0.5 V.
REV. B
AD6122
Test Figures
0.1␮F
+15V
8
VP
1 X1
2 X2 V–1
MUST BE EQUAL
LENGTHS
MODCMREF
OUT
A=1
3 Y1
7
IIPP
50⍀
4 Y2 V–1
AD830
VN
–15V
I DATA
AD6122
5
0.1␮F
50⍀
0.1␮F
+15V
1 X1
8
VP
2 X2 V–1
MODCMREF
OUT
A=1
3 Y1
7
IIPN
50⍀
4 Y2 V–1
VREG OUT
AD830
VN
–15V
5
0.1␮F
0.1␮F
450⍀
MODOPP
0.1␮F
+15V
1 X1
MUST BE EQUAL
LENGTHS
OUT
A=1
3 Y1
7
10nF
450⍀
QIPP
0.1␮F
VREG OUT
50⍀
4 Y2 V–1
205⍀
MODOPN
8
VP
2 X2 V–1
MODCMREF
10nF
AD830
VN
–15V
Q DATA
5
0.1␮F
50⍀
0.1␮F
+15V
8
VP
1 X1
2 X2 V–1
MODCMREF
MODCMREF
OUT
A=1
3 Y1
7
QIPN
50⍀
4 Y2 V–1
AD830
VN
–15V
5
0.1␮F
LOIPP
LOIPN
LO INPUT
Figure 1. Quadrature Modulator’s Characterization Input and Output Impedance Matches
REV. B
–5–
MOD_OUT
AD6122
VREG OUT
PULL-UP INDUCTORS CHOSEN
FOR PEAK RESPONSE AT THE
TEST FREQUENCY.
1:8
RF SOURCE
383⍀
511⍀
IFINP
453⍀
TXOPP
10nF
0.1␮F
10nF
205⍀
10nF
453⍀
TO
SPECTRUM
ANALYZER
4:1
IFINN
383⍀
TXOPN
10nF
AD6122
0.1␮F
VREG OUT
Figure 2. IF Amplifier’s Characterization Input and Output Impedance Matches
NOTE: RF CABLES FOR I AND Q PATHS MUST BE OF EQUAL LENGTH
TEST BED MOTHERBOARD
I DATA
TEKTRONIX
AFG2002
500mVp-p DIFFERENTIAL
MOD OUT
Q DATA
R&S
SMT03
I CHANNEL
R&S FSEA20/30
SPECTRUM
RF
ANALYZER
INPUT
Q CHANNEL
AUX MEAS
PORT
IFTX OUT
LO INPUT
RF
IF IN
RF SOURCE 1
TO RF SWITCHES
R&S
SMT03
HP34970A
DATA ACQUISITION
& SWITCH CONTROL
RF
RF SOURCE 2
HPE3610
POWER SUPPLY
DC MEASUREMENTS
& CONTROL BITS
Figure 3. General Test Set
–6–
REV. B
AD6122
VREG OUT
PULL-UP INDUCTORS CHOSEN
FOR PEAK RESPONSE AT THE
TEST FREQUENCY.
REACTIVE
CONJUGATE
MATCH
NOISE
SOURCE
1:8
IFINP
0.1␮F
453⍀
TXOPP
10nF
10nF
205⍀
10nF
453⍀
4:1
TO NOISE
FILTER
METER
IFINN
TXOPN
10nF
AD6122
0.1␮F
VREG OUT
Figure 4. IF Amplifier’s Noise Figure Test Set
HP8116A
FUNCTION GEN.
4 kHz, 0.5V TO 2.5V
SQ. WAVE
AGC
HP8116A
FUNCTION GEN.
ROHDE & SCHWARZ
SMT03
4kHz, 0V TO 2.7V
SQ. WAVE
100MHz, –30dBm
IFIN
PD1,
PD2
TEKTRONIX TDS 744A
ROHDE & SCHWARZ
SMT03
100kHz, –30dBm
IFIN
IFOUT
AD6122 TEST BED
CH 2 WITH COAX CABLE
50⍀
IFOUT
AD6122 TEST BED
CH 2 WITH COAX CABLE
50⍀
b. Response Time from PD1 and PD2 Control to IF Output
a. Response Time from Gain Control to IF Output
Figure 5. Response Time Setup
REV. B
TEKTRONIX TDS 744A
CH 1 WITH X10 PROBE
CH 1 WITH X10 PROBE
–7–
AD6122 –Typical Performance Characteristics
1
–50
–60
30kHz
100kHz
2s
UNIT
dBm
–49.18dBm
1 (T1)
130.67458918MHz
CH PWR –33.92dBm
–77.32dB
ACP UP
77.46dB
AVE LOW
–30
A
UNDESIRED SIDEBAND – dBc
RBW
VBW
SWT
REF LEV
–40dBm
–40
POWER – dBm
–70
–80
–90
–100
–110
CL1
–120
CU1
–35
–40
–130
–45
50
–140
CENTER 130.38MHz
519kHz/DIV
SPAN 5.19MHz
MODULATOR OUTPUT – dBm REFERRED
TO A 1k⍀ DIFFERENTIAL LOAD
LO LEAKAGE – dBc
300
350
–10
–35
–40
–45
100
150
200
250
FREQUENCY – MHz
300
–15
–20
–25
–30
–35
–14.0
350
Figure 7. Modulator LO Leakage vs. Output Frequency
–12.0
–10.0
–8.0
–6.0
MODULATOR, I = Q – dBV
–4.0
–2.0
Figure 10. Modulator Gain: Input (dBV) vs. Output (dBm)
–45
–15
–20
THIRD HARMONIC – dBc
OUTPUT DESIRED SIDEBAND LEVEL –
dBm REFERRED TO 1k⍀
150
200
250
OUTPUT FREQUENCY – MHz
Figure 9. Modulator Output Undesired Sideband vs.
Output Frequency
Figure 6. Spectral Plot at Modulator Outputs: ACPR
–50
50
100
–25
–30
–50
–55
–60
–35
–40
50
100
150
200
250
OUTPUT FREQUENCY – MHz
300
–65
50
350
100
150
200
250
OUTPUT FREQUENCY – MHz
300
350
Figure 11. Modulator Third Harmonic
Figure 8. Modulator Output Desired Sideband vs.
Output Frequency Without Roofing Filter
–8–
REV. B
AD6122
40
–24
–25
IIP3 – dBm Referred to 1k⍀
GAIN – dB With a 1k⍀ Load
20
0
TA = –40ⴗC
–20
TA = +85ⴗC
–40
–60
–26
–27
TA = +25ⴗC
–80
0.5
1.0
1.5
VGAIN – V
2.0
–28
2.5
2.5
Figure 12. IF Amplifier Response Curve: Gain vs.
VGAIN, TA = –40 °C, +25 °C, +85 °C
45
15
3.0
GAIN ERROR
GAIN – dB
5
2.0
1.0
–5
–15
0
–25
–1.0
–2.0
–35
GAIN
–45
–3.0
–55
–4.0
–65
–5.0
–75
0.5
0.9
1.3
1.7
VGAIN – V
2.1
IIP3 – dBm Referred to 1k⍀
5.0
4.0
150
200
250
FREQUENCY – MHz
100
300
350
30.0
0
25.0
NOISE FIGURE – dB
IIP3 – dBm Referred to 1k⍀
–25
Figure 16. IF Amplifier Input IP3 vs. Frequency
5.0
–5.0
–10.0
–15.0
20.0
133MHz
313MHz
15.0
238MHz
10.0
–20.0
0.9
1.3
1.7
VGAIN – V
2.1
5.0
–10.0
2.5
0
10.0
20.0
GAIN – dB
30.0
40.0
Figure 17. IF Amplifier Noise Figure vs. Gain
Figure 14. IF Amplifier Input IP3 vs. VGAIN
REV. B
3.7
–24
–26
50
–6.0
2.5
Figure 13. IF Amplifier Gain and Error vs. VGAIN
–25.0
0.5
3.5
–23
ERROR FROM PREDICTED VALVE – dB
35
2.9
3.1
3.3
SUPPLY VOLTAGE – V
Figure 15. IF Amplifier Input IP3 vs. Supply Voltage
6.0
25
2.7
–9–
AD6122
18.0
TOTAL CURRENT CONSUMPTION – mA
40
VGAIN = 2.5V
20
VGAIN = 2.0V
GAIN – dB
0
–20
VGAIN = 1.5V
–40
VGAIN = 1.0V
–60
16.0
14.0
12.0
10.0
VGAIN = 0.5V
–80
50
100
150
200
250
FREQUENCY – MHz
300
8.0
0.5
350
Figure 18. IF Amplifier Gain vs. Frequency for
VGAIN = 2.5 V, 2.0 V, 1.5 V, 1.0 V
1.5
VGAIN – V
2.0
2.5
Figure 19. Total Current Consumption vs. VGAIN
RBW
VBW
SWT
REF LEV
–30dBm
–30
–40
1
30kHz
300kHz
2s
UNIT
dBm
–46.78dBm
1 (T1)
130.38000000MHz
CH PWR –31.93dBm
–66.95dB
ACP UP
AVE LOW –68.95dB
–0.28 dB
1 (T1)
330.66132265kHz
1
–50
–60
POWER – dBm
1.0
A
–70
–80
–90
–100
–110
CL1
CO
CO
CU1
–120
–130
CENTER 130.38MHz
600kHz/DIV
SPAN 6MHz
Figure 20. ACPR of Cascaded Modulator, 20 dB Pad and IF
Amplifier: Spectral Plot
–10–
REV. B
AD6122
VCC
QUADRATURE
MODULATOR
OUTPUT
ATTENUATOR
IF AMPLIFIER
INPUT
QUADRATURE MODULATOR
I INPUT
IF AMPLIFIERS
LOCAL
OSCILLATOR
INPUT
TRANSMIT
OUTPUT
ⴜ2
AD6122
Q INPUT
COMMON-MODE
REFERENCE
OUTPUT
VPOS
LOW
DROPOUT
REGULATOR
POWERDOWN 1
GAIN
CONTROL
SCALE
FACTOR
VREG
POWERDOWN 2
1.23 V
REFERENCE
OUTPUT
GAIN CONTROL
REFERENCE
VOLTAGE
INPUT
TEMPERATURE
COMPENSATION
GAIN CONTROL
VOLTAGE
INPUT
Figure 21. Block Diagram
THEORY OF OPERATION
IF Amplifiers and Gain Control
The CDMA Transmitter IF Subsystem (Figure 21) consists of
an I and Q modulator with a divide-by-two quadrature generator, high dynamic range IF amplifiers with voltage-controlled
gain, a low dropout regulator and power-down control inputs.
The IF amplifiers provide an 86 dB linear in dB gain control
range. The input stage uses a differential, continuously variable
attenuator based on Analog Devices’ patented X-AMP™ topology. This low noise attenuator consists of a differential R-2R
ladder network, linear interpolator and a fixed gain amplifier.
The IF amplifier’s input impedance is 1 kΩ differential. Similar
to the I and Q modulator’s output, the IF amplifier’s output is a
differential current, which will vary depending upon the gain
control voltage. In order to achieve the specified gain, the output of the IF amplifiers should be loaded with a 1 kΩ differential load.
I and Q Modulator
The I and Q modulator accepts differential quadrature baseband
inputs from CDMA baseband converters. The LO is injected at
twice the IF frequency. A divide-by-two quadrature generator
followed by dual polyphase filters ensures ± 1° quadrature accuracy (Figure 22).
For 500 mV p-p differential I and Q input signals, the output
power of the modulator will be –21 dBm referred to 1 kΩ when
the output of the modulator is loaded with a 1 kΩ differential
load. With the maximum input conditions stated above, the
modulator outputs are a 225 µA p-p differential current; consequently, the output load will greatly affect the output power of
the modulator.
2 ⴛ IF
LO INPUT
ⴜ2
I
I
POLYPHASE
FILTERS
180ⴗ
ⴜ2
The gain control circuits contain both temperature compensation circuitry and a choice of internal or external reference for
adjusting the gain scale factor. The gain control input accepts
an external gain control voltage input from a DAC. It provides
97 dB of gain control range with a nominal 75 dB/V scale factor.
Q
QUADRATURE
OUTPUT TO
MODULATOR
Q
Figure 22. Simplified Quadrature Generator Circuit
The I and Q modulator also provides a common mode reference
signal at the MODCMREF pin. This voltage is a dc voltage set
to 1.408 V when a 2.7 V supply is used. It is used to dc bias
the output of the DAC that provides I and Q inputs to the
modulator.
The external gain control input signal should be a clean signal.
It is recommended to filter this signal in order to eliminate the
noise that results from the DAC. If a noisy signal is used for the
gain control voltage, VGAIN inband and adjacent channel noise
peaking can occur at the output of the AD6122. A simple RC
filter can be employed, but care should be taken with its design.
If too big a resistor is used, a large voltage drop may occur
across the resistor, resulting in lower gain than expected (as a
result of a lower voltage reaching the AD6122). An RC filter
with a 20 kHz bandwidth, employing a 1 kΩ resistor is appropriate. This results in an 8.2 nF capacitor. The resulting circuit
is shown in Figure 23. Note that the input resistance at the
VGAIN pin is approximately 100 kΩ.
AD6122
1k⍀
8.2nF
VGAIN
109k⍀
Figure 23. Gain Voltage Filtering
X-AMP is a trademark of Analog Devices, Inc.
REV. B
FROM
BASEBAND
CONVERTER
–11–
AD6122
The AD6122’s overall gain, expressed in decibels, is linear in
dB with respect to the automatic gain control (AGC) voltage,
VGAIN. Either REFOUT or an external reference voltage connected to REFIN may be used to set the voltage range for VGAIN.
When the internal 1.23 V reference, REFOUT, is connected to
REFIN , VGAIN will control the entire AGC range when it is
typically set between 0.5 V and 2.5 V. Minimum gain occurs at
minimum voltage on VGAIN and maximum gain occurs at maximum voltage on VGAIN. The maximum and minimum gain
will not change with a change in voltage at REFIN. Rather, the
slope of the gain curve will change as a result of a change in the
required range for VGAIN. Figure 24 shows the piecewise linear
approximation of the gain curve for the AD6122.
GAIN – V/V
MAXIMUM
GAIN
MINIMUM
GAIN
Figure 24. Piecewise Linear Approximation for the
AD6122 Gain Curve
Because the minimum and maximum gain from the AD6122
are constant, we can approximate the VGAIN range for a
given REFIN voltage by using Equation 1.
(GAIN – MinGain) × 1.6REFIN
+ 0.4 REFIN
MaxGain – MinGain
Table I. Operating Modes
PD1
PD2
IF Amp
Modulator
0
0
1
1
0
1
0
1
ON
ON
INVALID STATE
OFF
ON
OFF
INVALID STATE
OFF
Low Dropout Regulator
The AD6122 incorporates an integrated low dropout regulator.
The regulator accepts inputs from 2.9 V to 4.2 V and supplies a
constant 2.7 V reference output at LDOC. The 2.7 V signal can
be used to provide the dc voltages required for the DVCC,
TXVCC and IFVCC dc supplies. In order to configure the low
dropout regulator, an external pass transistor is required. A pnp
bipolar junction transistor with a minimum hFE of 100 and a
maximum hFE of 300 and a VCESAT of –0.4 V is required. In
order to use the low dropout regulator, configure the transistor as
shown in Figure 25. The 18 pF capacitor in Figure 25 is used for
decoupling the 2.7 V dc signal.
In addition to the low dropout regulator, a band-gap voltage
reference produces a 1.23 V reference voltage at REFOUT.
This reference voltage will be present whenever a 2.7 V dc signal is present on pin LDOC. This 1.23 V reference voltage can
then be used to provide the gain reference signal required for
REFIN and the reference voltage for the transmit DACs in a
baseband converter.
VGAIN – V
VGAIN =
up. The control is provided via two control pins, PD1 and PD2.
Table I shows the operating modes of the AD6122.
AD6122
(1)
LDOE
2.9V – 4.2V
Where MaxGain is the maximum gain (+34 dB) in dB, MinGain
is the minimum gain (–63 dB) in dB, REFIN is the reference
input voltage, in volts, VGAIN is the gain control voltage input,
in volts, and GAIN is the particular gain, in dB, we would have
for a given REFIN and VGAIN. Consequently, for any REFIN
we choose, we can calculate the VGAIN range by solving
Equation 1 for VGAIN. For example, in order to determine the
VGAIN value for the maximum gain condition, given a 1.23 V
REFIN, we can solve Equation 1 for VGAIN by substituting
+34 dB for GAIN and MaxGain, –63 dB for MinGain and 1.23 V
for REFIN. VGAIN can then be calculated to be 2.46 V, or
approximately 2.5 V. For the minimum gain condition, we can
determine the VGAIN value by substituting 34 dB for MaxGain,
–63 dB for GAIN and MinGain and 1.23 V for REFIN. VGAIN
can then be calculated to be 0.492 V or approximately 0.5 V.
Power-Down Control
PASS
TRANSISTOR
LDOB
LDOC
2.7V
18pF
REFOUT
1.23V
Figure 25. Configuring the Low Dropout Regulator
It is possible to bypass the low dropout regulator on the AD6122
and use an external regulator instead. In order to bypass the
integrated low dropout regulator, connect pins LDOE, LDOB
and LDOC together and then connect them all to the 2.7 V
external regulator voltage. This configuration is shown in
Figure 26. Even when the low dropout regulator is bypassed,
the 1.23 V reference voltage at pin REFOUT is still present.
The AD6122 can be operated with the IF amplifiers and quadrature modulator both powered up, both powered down or with
the IF amplifiers powered up and the modulator powered down.
The AD6122 cannot operate with only the modulator powered
–12–
REV. B
AD6122
The attenuator is discussed in the next section entitled Measuring Adjacent Channel Protection Ratio (ACPR).
AD6122
In order to confirm whether the roofing filter has been correctly
designed, sweep the LO frequency and view the output of the IF
amplifier on a spectrum analyzer. The signal should peak at the
IF frequency if the inductor value is correct. The Q of the filter
should be low enough so that variations in the parasitic capacitances should be negligible.
LDOE
FROM EXTERNAL
VOLTAGE REGULATOR
LDOB
LDOC
REFOUT
1.23V
The value of inductor required will be a function of the IF frequency at which we are operating. The values of inductors used
during characterization at Analog Devices are shown in Table
II. Because the exact value will also be a function of printed
circuit board layout, we will have to vary the value from those in
Table II to those required for our board.
Figure 26. Configuration for Bypassing the Low Dropout
Regulator
ROOFING FILTER
Because the outputs of the AD6122 modulator are open collector, the parasitic capacitances seen at the output of the modulator, and inputs of the IF amplifiers, are high enough to create a
low-pass filter, which may attenuate the IF signal. Consequently,
the parasitic capacitance must be cancelled by using external
inductors to form a parallel resonant circuit. The external inductors and the internal parasitic capacitors form what is known
as the roofing filter, with the resonant frequency given by
Equation 2.
f0 =
1
(2)
2 π LCPAR
where f0 is the IF frequency, in Hertz, CPAR is the total parasitic
capacitance in Farads, and L is the value of external inductors,
in henrys.
The roofing filter may be composed of the pull-up inductors
required on the open collector outputs of the I and Q modulator. This configuration is shown in Figure 27. The 10 nF capacitors are used for ac coupling.
AD6122
MODOPP
2CPAR
L/2
2CPAR
L/2
MODOPN
VCC
10nF
PARALLEL
RESONANT
CIRCUIT
10nF
IFINN
IFINP
10nF
ATTENUATOR
Figure 27. Roofing Filter Configuration
REV. B
Table II. Roofing Filter Inductor Values
IF Frequency (MHz)
Value of Roofing Filter
Inductor (nH)
50–125
126–200
201–275
276–350
470
150
68
27
It should be noted that the roofing filter is only required when
cascading the output from the I/Q modulator to the input of the
IF amplifiers. If we are driving into the IF amplifiers directly, no
roofing filter is required, however, pull-up inductors are required
in order to set the dc voltage of the open collector modulator
outputs.
MEASURING ADJACENT CHANNEL POWER RATIO
(ACPR)
At maximum IF gain and specified input conditions (500 mV
p-p baseband inputs), the output of the I/Q modulator is 11 dB
greater than the P1 dB (one dB compression point) of the IF
amplifiers. This configuration maximizes the ratio of signal to
LO feedthrough and also maximizes the signal to noise ratio.
Once these ratios are maximized, we can attenuate the noise,
signal and LO feedthrough without affecting the ratios. Therefore, attenuation is required between the I/Q modulator and the
IF amplifiers.
In order to determine exactly how much attenuation is required,
we must recognize that ACPR is a function of the attenuation
from the modulator outputs to the IF amplifier inputs. As a
result, in order to determine how much attenuation is required,
we must first know how good an ACPR performance is desired.
If too much attenuation is applied, the ACPR will be very good,
but, the IF amplifier’s output power level will be low, possibly
resulting in poor signal to noise ratio and possibly requiring
additional amplification external to the AD6122.
An appropriate method that can be used to provide the correct
amount of attenuation between the modulator outputs and the
IF amplifier inputs is a simple differential voltage divider. The
topology and its design equations are shown in Figure 28 and
Equations 3 and 4. The input impedance of the IF amplifiers is
typically 1 kΩ. As a result, if we design resistor R2 to be much
less than 1 kΩ, we can neglect the effects of the IF amplifier’s
input impedance on the attenuator.
–13–
AD6122
This circuit is very sensitive to parasitic capacitances. As a result, extra care should be taken to ensure minimum and equal
printed circuit board transmission lines. We should also try to
keep R2 small in order to minimize the effects of printed circuit
board parasitic capacitance on loading the output of the pad.
AD6122
MODOPP
R1
MODOPN
R1
IFINP
R2
ZIN
In conclusion, we have to develop a system-level ACPR budget
for our radio, and from that budget determine how much ACPR
performance we desire from the AD6122. We then need to implement the appropriate attenuation network to get that ACPR
performance.
RSHUNT >>R2
IFINN
Figure 28. Pad Topology
LEVEL DIAGRAM


1


1
R
L = 20 log 
1
1 

 +
 R1 R2 / 2
(3)
Z IN = 2R1+ R2
(4)
where L is the transducer loss (or loss through the pad) in dB
and ZIN is the desired input resistance in ohms. Using these
equations, we can design the attenuator circuit to provide whatever amount of attenuation we require.
Figure 29 is provided to better understand the different voltage
levels you can expect to see at different points of the AD6122. It
represents the voltage and power levels expected for a maximum
input condition of 500 mV p-p at the I and Q modulator and
maximum gain in the IF amplifiers. When trying to make these
measurements, a high impedance (10 MΩ) active FET probe
(for example, the Tek P6204, from Tektronix) should be used to
minimize the effects of loading the circuit with the probe.
In order to produce these results, the attenuator is designed to
have a 1 kΩ input impedance and the output of the IF amplifiers
are loaded with 1 kΩ. The roofing filter is designed to resonate
the parasitic capacitance at the IF frequency.
MODULATORS
I
500mV p-p
DIFFERENTIAL
LO
ⴜ2
100mV p-p
DIFFERENTIAL
–21dBm
(REFERRED TO 1k⍀)
252.1mV p-p DIFFERENTIAL
–41dBm
(REFERRED TO 1k⍀)
25.21mV p-p
IF AMPLIFIERS
DIFFERENTIAL
MODOP
TRANSMIT
OUTPUT
IFIN
VCC
Q
500mV p-p
DIFFERENTIAL
–7dBm
(REFERRED TO 1k⍀)
1.263V p-p DIFFERENTIAL
VGAIN = 2.5V
GAIN = +34dB
1k⍀
20dB
ATTENUATOR
ZOUT = 1k⍀
ZIN = 1k⍀
Figure 29. Level Diagram
–14–
REV. B
AD6122
INPUT INTERFACES
The AD6122 interfaces to CDMA baseband converters providing either IF or baseband outputs. The baseband input is provided by direct connection of the baseband converter’s baseband
output to the baseband input of the AD6122 (Figure 30). The
IF amplifier’s gain control is provided by connection of the
transmit AGC DAC’s output on the baseband converter, through a
low-pass filter to the VGAIN pin on the AD6122.
PD1
PD2
VCC
VGAIN
TEMPERATURE
COMPENSATION
GAIN
CONTROL
SCALE
FACTOR
TX AGC DAC
REFIN
EXT REF IN
LDOE
LDOB
LDOC
REFOUT
LOW
DROPOUT
REGULATOR
IFVCC
IFGND
LDOGND
IIPP
DGND
I OUTPUT
Q
LOIPP
LOIPN
IIPN
I OUTPUT
MODCMREF
ⴜ2
I
VCM REF IN
QIPN
Q OUTPUT
Q OUTPUT
QIPP
DVCC
MODOPP
MODOPN
AD6122
VCC
VCC
CDMA
BASEBAND
IC
IFINP
TXOPP
TXOPN
IFINN
TXVCC
IFGND
Figure 30. Typical Connections to Baseband IC Using I and Q Inputs with SSOP Package
REV. B
–15–
AD6122
The IF output port impedance match used during characterization at Analog Devices is as follows:
AD6122 Evaluation Board
The AD6122 Evaluation Board consists of an AD6122, I/O connectors, a 20-pin dual header, 2-pin headers and four AD830
high speed video difference amplifiers. It allows the user to
evaluate the AD6122’s IF amplifier and modulator together or
separately. Because the AD6122 may be used at any IF from 50
MHz to 350 MHz, pads are provided on the LOIPP input,
TXOP output, MODOP output and IFIP inputs to allow the
user to add matching networks. The board is configured for an
IF frequency of 130.38 MHz when shipped. There is no difference between the configuration of the boards with the SSOP or
LPCC package.
AD6122
AD6122
1:8
383⍀
IFINP
511⍀
SIGNAL
GENERATOR
IFINN
383⍀
1k⍀
Figure 31. IF Input Port Impedance Match Used During
Characterization at ADI
This is a broadband lossy match used for characterization over
the 50 MHz to 350 MHz frequency range. All dBm references
in the characterization data collected using this match are referenced to 1 kΩ. Note that the 1:8 ratio in Figure 31 is an impedance ratio and not a voltage ratio.
50⍀
453⍀
SPECTRUM
ANALYZER
1k⍀
Figure 32. IF Output Port Impedance Match Used During
Characterization at ADI
This is a broadband lossy output match for the 50 MHz to
350 MHz frequency range. The 4:1 ratio in Figure 32 is an
impedance ratio and not a voltage ratio.
In order to test the power-down modes of the AD6122, locate
the two pin headers on the AD6122 evaluation boards labeled
PD1 and PD2. By open-circuiting the pins labeled PD1, the IF
amplifiers power down. By open-circuiting the pins labeled
PD2, the modulator powers down. Note that the IF amplifiers
and modulator are powered down unless the pins on the two pin
headers, PD1 and PD2, are short circuited.
50⍀
4:1
205⍀
TXOPN
The AD830s are used to provide single-ended to differential
conversion and the appropriate phase shift for the I and Q data
input pins. As a result, a single-ended signal generator can be
used to generate these signals.
The IF input port impedance match used during characterization of the AD6122 at Analog Devices is as follows:
453⍀
TXOPP
As shipped, the board is configured as follows:
1. J1 is open and J2 is shorted. This enables the LDO regulator.
The external PNP transistor should remain in place even
when the regulator is bypassed (the Pin LDOB is pulled up
by the transistor).
2. X11, X25, X18 and X26 are shorted and X12, X14, X19
and X21 are opened in order to connect the output of the
modulator to the input of the IF amplifiers.
3. L4 and L5, the roofing filter components are optimized for
an IF frequency of 130.38 MHz.
4. R14, R15 and R16 set the attenuation between the modulator outputs and the IF amplifier inputs to 20 dB.
5. PD1 and PD2 are pulled low by the jumpers on the two pin
headers. To power down the chip, set PD1 and PD2 high by
removing the jumpers.
In order to look at the modulator and IF amplifiers separately,
disconnect the output of the modulator from the input of the IF
amplifiers. This is accomplished by short circuiting X12, X14,
X19 and X20 and open circuiting X11, X18, X25 and X26.
–16–
REV. B
AD6122
Table III describes the high frequency signal connectors on the
AD6122 customer sample boards.
Table IV lists the connections for the 20-pin power-supply
connector.
Table III. Evaluation Board SMA Signal Connector
Description
Connector
Description
I CH
I Modulator Input. 250 mV p-p into 50 Ω
termination, dc coupled. The level shifting and
phase splitting is done on board by the AD830
amplifiers.
Table IV. 20-Pin Power Supply Connection Information
Pin #
Function
1
VPOS for AD6122; 2.9 V to 4.2 V using regulator; 2.7 V
to 4.2 V bypassing regulator.
VPOS for AD6122; 2.9 V to 4.2 V using regulator; 2.7 V
to 3.6 V bypassing regulator.
Ground.
Ground.
Ground.
Regulated Output or Input Voltage; Connects to Pin 5
on AD6122.
Ground.
Ground.
Ground.
Ground.
Ground.
PD1; Power-Down 1 Input.
Ground.
1.23 V Reference Voltage from AD6122.
Ground.
VGAIN; Gain Control Voltage Input.
–15 V Supply for AD830 Differential Amplifier.
+15 V Supply for AD830 Differential Amplifier.
MODCMREF; common-mode reference output for
baseband converter common-mode reference input.
PD2; Power-Down 2 Input.
2
3
4
5
6
Q CH
Q Modulator Input. 250 mV p-p into 50 Ω
termination, ac coupled. The level shifting and
phase splitting is done on board by the AD830
amplifiers.
MODOP
Modulator Output. The differential-to-single
ended conversion is performed by a balun on
the board. Impedance matched to 50 Ω for
130.38 MHz IF frequency.
IFIP
IF Amplifier Input. Single-ended-to-differential
conversion performed by a balun on board.
Impedance matched to 50 Ω for 130.38 MHz IF
frequency.
TXOP
IF Amplifier Output. Differential-to-singleended conversion performed by a balun on
board. Impedance matched to 50 Ω for 130.38
MHz IF frequency.
LOIPP
Local oscillator positive input at 2 × IF
frequency.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
A schematic diagram of the evaluation board is on the next two
pages.
REV. B
–17–
AD6122
AD6122
PD1
PD2
J2
0⍀
VPOS
2.9V – 4.2V
R12
0⍀
PD1
VGAIN
PD2
REFIN
C28
10nF
VGAIN
FMMT4403CT-ND
LDOE
REFOUT
REFOUT
C29
10nF
Q1
J1
VREG OUT
C23
18pF
LDOB
IFVCC
LDOC
IFGND
IFVCC
LDOGND
IIPP
IIPP
DGND
IIPN
IIPN
LOIPP
MODCMREF
LOIPN
QIPN
QIPN
DVCC
QIPP
QIPP
VREG OUT
VREG OUT
C1
10nF
X2
0⍀
LOIPP
X1
X3
X5
100nH
TXOP
X4
C24
0.1␮F
1:8
X6
3pF
X8 C3
0⍀ 10nF
X7
T1
VCC
C2
10nF
DVCC
L2
220nH
TXOPP
MODCMREF
X10 C4
0⍀ 10nF
VCC
TXVCC
L3
220nH
C25
10nF
C26
10nF
L4
180nH
L5
180nH
8:1
MODOPP
C8
10nF
X9
TXOPN
C27
10nF
MODOPN
TXVCC
IFINP
IFGND
IFINN
C10
10nF
X11
0⍀
X13
X17
X14
L6
MODOP
X15
4pF
T2
X25
0⍀
C30
C11
10nF
R14 = 442⍀
R15 = 100⍀
R16 = 442⍀
C9
10nF
X12
X16
100nH
R13
R14
R15
X18
0⍀
R16
X19
X26
0⍀
8:1
X23
27nH
IFIP
X22
56nH
X20
X24
X21
T3
Figure 33. Schematic Diagram of the Evaluation Board
–18–
REV. B
AD6122
C15
0.1␮F
C19
0.1␮F
+15V
+15V
8
1
2
MODCMREF
SOIC PACKAGE
U2
V–1
A=1
3
4
V–1
5
7
R7
50⍀
V–1
4
AD830
5
+15V
A=1
3
V–1
8
1
U3
V–1
7
R8
50⍀
2
TO
IIPN
U5
V–1
A=1
3
MODCMREF
V–1
4
AD830
5
AD830
–15V
C22
0.1␮F
C18
0.1␮F
TO
DVCC
TO
IFVCC
C6
18pF
C5
18pF
C7
18pF
R1
10⍀
C13
0.01␮F
P1
VREG OUT
R2
10⍀
C12
0.01␮F
R3
10⍀
C14
0.01␮F
VPOS
P2
1
2
3
4
5
6
L1
470nH
7
8
9
10
11
12
13
14
REFOUT
R4
10k⍀
R5
10k⍀
PD1
15
16
VGAIN
–15V
17
18
+15V
MODCMREF
19
20
PD2
Figure 34. Schematic Diagram of the Evaluation Board
–19–
FROM VPOS
2.9V–4.2V
VREG OUT
NOTES:
1. TO USE THE LDO REGULATOR, SHORT J2 AND OPEN J1.
2. TO BYPASS THE REGULATOR, SHORT J1 AND OPEN J2
3. TO CONNECT THE OUTPUT OF THE MODULATOR TO THE
INPUT OF THE IF AMP, SHORT J5 AND J6.
TO TEST THE MODULATOR AND THE IF AMP SEPARATELY,
OPEN J5 AND J6.
4.
INDICATES A 50⍀ TRACE.
REV. B
TO
QIPN
5
–15V
TO
TXVCC
R11
50⍀
C21
0.1␮F
8
1
4
7
TO
QIPP
C20
0.1␮F
R9
50⍀
+15V
MODCMREF
R10
50⍀
AD830
C17
0.1␮F
2
7
–15V
QCH
C16
0.1␮F
R6
50⍀
A=1
3
MODCMREF
–15V
ICH
U4
V–1
2
TO
IIPP
SOIC PACKAGE
8
1
PD1
PD2
AD6122
OUTLINE DIMENSIONS
Dimensions shown in inches and (mm).
C00946a–.5–6/00 (rev. B)
28-Lead SSOP
(RS-28)
28
15
1
14
0.212 (5.38)
0.205 (5.21)
0.311 (7.9)
0.301 (7.64)
0.407 (10.34)
0.397 (10.08)
0.07 (1.79)
0.066 (1.67)
0.078 (1.98) PIN 1
0.068 (1.73)
0.008 (0.203) 0.0256
(0.65)
0.002 (0.050) BSC
0.03 (0.762)
0.022 (0.558)
8°
0.015 (0.38)
SEATING 0.009 (0.229) 0°
0.010 (0.25)
PLANE
0.005 (0.127)
32-Leadless Chip Scale Package (LPCC)
(CP-32)
0.205 (5.20)
0.197 (5.00) SQ
0.189 (4.80)
0.128 (3.25)
0.106 (2.70) SQ
0.049 (1.25)
25
32
1
24
0.015 (0.38)
0.012 (0.30)
0.009 (0.23)
BOTTOM
VIEW
17
PIN 1
INDICATOR
8
9
16
0.138 (3.50) BSC
0.018 (0.45)
0.016 (0.40)
0.014 (0.35)
0.039 (1.00)
0.035 (0.90)
0.031 (0.80)
0.010
(0.25)
REF
0.020 (0.50)
BSC
0.002 (0.05)
0.001 (0.02)
0.000 (0.00)
PRINTED IN U.S.A.
CONTROLLING DIMENSIONS ARE IN MILLIMETERS
DIMENSIONS MEET JEDEC MO-220-VHHD-2
–20–
REV. B