AD AD9974BBCZRL

Dual-Channel, 14-Bit, CCD Signal
Processor with Precision Timing Core
AD9974
FEATURES
GENERAL DESCRIPTION
1.8 V analog and digital core supply voltage
Correlated double sampler (CDS) with
−3 dB, 0 dB, +3 dB, and +6 dB gain
6 dB to 42 dB, 10-bit variable gain amplifier (VGA)
14-bit, 65 MHz analog-to-digital converter (ADC)
Black level clamp with variable level control
Complete on-chip timing generator
Precision Timing core with 240 ps resolution @ 65 MHz
On-chip 3 V horizontal and RG drivers
100-lead, 9 mm × 9 mm, 0.8 mm pitch, CSP_BGA package
Internal low dropout (LDO) regulator circuitry
The AD9974 is a highly integrated, dual-channel, chargecoupled device (CCD) signal processor for high speed digital
video camera applications. Each channel is specified at pixel
rates of up to 65 MHz. The AD9974 consists of a complete
analog front end (AFE) with analog-to-digital conversion,
combined with a programmable timing driver. The Precision
Timing™ core allows adjustment of high speed clocks with
approximately 240 ps resolution at 65 MHz operation.
Each AFE includes black level clamping, CDS, VGA, and
a 65 MSPS, 14-bit ADC. The timing driver provides the high
speed CCD clock drivers for the RG_A, RG_B, H1_A to H4_A,
and H1_B to H4_B outputs. A 3-wire serial interface is used to
program each channel of the AD9974.
APPLICATIONS
Professional HDTV camcorders
Professional/high end digital cameras
Broadcast cameras
Industrial high speed cameras
Available in a space-saving, 9 mm × 9 mm, CSP_BGA package,
the AD9974 is specified over an operating temperature range of
−25°C to +85°C.
FUNCTIONAL BLOCK DIAGRAM
REFT_A REFB_A
REFT_B REFB_B
VREF_A
VREF_B
AD9974
CCDINP_A
14
CDS
CCDINM_A
DOUT_A
ADC
VGA
6dB TO 42dB
–3, 0, +3, +6dB
CLAMP
CLAMP
6dB TO 42dB
–3, 0, +3, +6dB
CCDINP_B
CDS
14
ADC
VGA
DOUT_B
CCDINM_B
1.8V OUTPUT
LDO A
1.8V OUTPUT
LDO B
RG_A
RG_B
H1_B TO H4_B
PRECISION
TIMING
CORE
4
4
CLI_A
CLI_B
HORIZONTAL
DRIVERS
SYNC
GENERATOR
INTERNAL
REGISTERS
HD_A VD_A HD_B VD_B
SL_A SDATA_A SL_B SDATA_B
SCK_A
SCK_B
05955-001
H1_A TO H4_A
INTERNAL CLOCKS
Figure 1.
Rev. A
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 that may result from its use. Specifications subject to change without notice. No
license is granted by implication or otherwise under any patent or patent rights of Analog Devices.
Trademarks and registered trademarks are the property of their respective owners.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 781.329.4700
www.analog.com
Fax: 781.461.3113
©2009 Analog Devices, Inc. All rights reserved.
AD9974
TABLE OF CONTENTS
Features .............................................................................................. 1 Precision Timing High Speed Timing Core ............................. 15 Applications ....................................................................................... 1 Horizontal Clamping and Blanking ......................................... 18 General Description ......................................................................... 1 Complete Field—Combining H-Patterns ............................... 25 Functional Block Diagram .............................................................. 1 Mode Registers ........................................................................... 26 Revision History ............................................................................... 2 Horizontal Timing Sequence Example.................................... 28 Specifications..................................................................................... 3 Analog Front End Description and Operation ...................... 29 Channel-to-Channel Specifications ........................................... 3 Applications Information .............................................................. 33 Timing Specifications .................................................................. 4 Recommended Power-Up Sequence ....................................... 33 Digital Specifications ................................................................... 5 Standby Mode Operation .......................................................... 36 Analog Specifications ................................................................... 6 CLI Frequency Change .............................................................. 36 Absolute Maximum Ratings............................................................ 8 Circuit Configuration ................................................................ 37 Thermal Characteristics .............................................................. 8 Grounding and Decoupling Recommendations .................... 37 ESD Caution .................................................................................. 8 3-Wire Serial Interface Timing ..................................................... 39 Pin Configuration and Function Descriptions ............................. 9 Layout of Internal Registers ...................................................... 40 Typical Performance Characteristics ........................................... 11 Updating of Register Values ...................................................... 41 Equivalent Input/Output Circuits ................................................ 12 Complete Register Listing ............................................................. 42 Terminology .................................................................................... 13 Outline Dimensions ....................................................................... 50 Theory of Operation ...................................................................... 14 Ordering Guide .......................................................................... 50 Programmable Timing Generation .............................................. 15 REVISION HISTORY
10/09—Revision A: Initial Version
Changes to Table 1 ............................................................................ 3
Changes to Table 3 ............................................................................ 4
Changes to Pin Function Descriptions Table ............................... 9
Changes to Figure 11 ...................................................................... 12
Changes to Individual HBLK Pattern Section ............................ 20
Changes to Table 14 ........................................................................ 25
Added Example Register Setting for Power-Up Section ........... 34
Added Additional Restrictions Section ....................................... 35
Changes to Table 2 .......................................................................... 36
Changes to 3 V System Compatibility Section ........................... 37
Changes to Grounding and Decoupling
Recommendations Section ............................................................ 37
Changes to Table 30 ........................................................................ 48
Changes to Table 31 ........................................................................ 49
Changes to Ordering Guide .......................................................... 50
Rev. A | Page 2 of 52
AD9974
SPECIFICATIONS
X = A = B, unless otherwise noted.
Table 1.
Parameter
TEMPERATURE RANGE
Operating
Storage
POWER SUPPLY VOLTAGE
AVDD_X (AFE, Timing Core)
RGVDD_X (RG_X Driver)
HVDD_X (H1_X to H4_X Drivers)
DVDD_X (All Other Digital)
DRVDD_X (Parallel Data Output Drivers)
IOVDD_X (I/O Supply Without the Use of LDO)
POWER SUPPLY CURRENTS—65 MHz OPERATION
AVDD_X (1.8 V)
RGVDD_X (3.3 V, 20 pF RG Load)
HVDD_X 1 (3.3 V, 200 pF Total Load on H1 to H4)
DVDD_X (1.8 V)
DRVDD_X (3.0 V)
IOVDD_X (1.8 V)
POWER SUPPLY CURRENTS—STANDBY MODE OPERATION
Reference Standby
Total Shutdown
LDO 2
IOVDD_X (I/O Supply When Using LDO)
Output Voltage
Output Current
CLOCK RATE (CLI)
Min
Typ
−25
−65
1.6
2.7
2.7
1.6
1.6
1.6
8
1.8
3.3
3.3
1.8
3.0
1.8
Max
Unit
+85
+150
°C
°C
2.0
3.6
3.6
2.0
3.6
3.6
V
V
V
V
V
V
55
5
40
15
3
2
mA
mA
mA
mA
mA
mA
10
0.5
mA
mA
3.0
1.85
60
V
V
mA
MHz
100
65
1
The total power dissipated by the HVDD (or RGVDD) supply can be approximated as follows: Total HVDD Power = [CLOAD × HVDD × Pixel Frequency] × HVDD.
Reducing the capacitive load and/or reducing the HVDD supply reduces the power dissipation. CLOAD is the total capacitance seen by all H-outputs.
2
LDO should be used to supply only AVDD and DVDD.
CHANNEL-TO-CHANNEL SPECIFICATIONS
X = A = B, TMIN to TMAX, AVDD_X = DVDD_X = 1.8 V, fCLI = 65 MHz, typical timing specifications, unless otherwise noted.
Table 2.
Parameter
LINEARITY MISMATCH 1
CROSSTALK ERROR
Channel A to Channel B
Channel B to Channel A
1
Min
Typ
<0.5
−82
−82
Max
Unit
%
dB
dB
Test Conditions/Comments
Absolute value above 1⁄16 of maximum output code
CDS = 0 dB
Full-scale step applied to Channel A while measuring response on Channel B
Full-scale step applied to Channel B while measuring response on Channel A
See the Terminology section for further measurement explanation.
Rev. A | Page 3 of 52
AD9974
TIMING SPECIFICATIONS
X = A = B, CL = 20 pF, AVDD_X = DVDD_X = 1.8 V, fCLI = 65 MHz, unless otherwise noted.
Table 3.
Parameter
MASTER CLOCK (CLI)
CLI Clock Period (tCONV)
CLI High/Low Pulse Width (tADC)
Delay from CLI Rising Edge to Internal Pixel Position 0 (tCLIDLY)
AFE
SHP Rising Edge to SHD Rising Edge (tS1)
AFE Pipeline Delay
CLPOB Pulse Width (Programmable) (tCOB) 1
HD Pulse Width
VD Pulse Width
SERIAL INTERFACE
Maximum SCK Frequency (fSCLK)
SL to SCK Setup Time (tLS)
SCK to SL Hold Time (tLH)
SDATA Valid to SCK Rising Edge Setup (tDS)
SCK Rising Edge to SDATA Valid Hold (tDH)
H-COUNTER RESET SPECIFICATIONS
HD Pulse Width
VD Pulse Width
VD Falling Edge to HD Falling Edge(tVDHD)
HD Falling Edge to CLI Rising Edge(tHDCLI)
CLI Rising Edge to SHPLOC (Internal Sample Edge) (tCLISHP)
TIMING CORE SETTING RESTRICTIONS
Inhibited Region for SHP Edge Location (tSHPINH) (See Figure 21) 2
Inhibited Region for SHP or SHD with Respect to H-Clocks
(See Figure 21) 3, 4, 5, 6
RETIME = 0, MASK = 0 (tSHDINH)
RETIME = 0, MASK = 1 (tSHDINH)
RETIME = 1, MASK = 0 (tSHPINH)
RETIME = 1, MASK = 1 (tSHPINH)
Inhibited Region for DOUTPHASE Edge Location (tDOUTINH)
(See Figure 21)
Min
15.38
6.9
6.9
2
tCONV
1 HD period
Typ
Max
Unit
7.7
5
8.9
ns
ns
ns
7.7
16
20
8.5
ns
Cycles
Pixels
ns
ns
Comments
See Figure 17
See Figure 21
See Figure 22
See Figure 52
40
10
10
10
10
MHz
ns
ns
ns
ns
tCONV
1 HD period
0
3
3
VD period − tCONV
tCONV − 2
tCONV − 2
ns
ns
ns
ns
ns
50
64/0
Edge location
H × NEGLOC − 15
H × POSLOC − 15
H × NEGLOC − 15
H × POSLOC − 15
SHDLOC + 0
H × NEGLOC − 0
H × POSLOC − 0
H × NEGLOC − 0
H × POSLOC − 0
SHDLOC + 15
Edge location
Edge location
Edge location
Edge location
Edge location
See Figure 49
1
Minimum CLPOB pulse width is for functional operation only. Wider typical pulses are recommended to achieve good clamp performance.
Only applies to slave mode operation. The inhibited area for SHP is needed to meet the timing requirements for tCLISHP for proper H-counter reset operation.
3
When 0x34[2:0] HxBLKRETIME bits are enabled, the inhibit region for SHD location changes to inhibit region for SHP location.
4
When sequence register 0x09[23:21] HBLK masking registers are set to 0, the H-edge reference becomes H × NEGLOC.
5
The H-clock signals that have SHP/SHD inhibit regions depend on the HCLK mode: Mode 1 = H1, Mode 2 = H1, H2, and Mode 3 = H1, H3.
6
These specifications apply when H1POL, H2POL, RGPOL, and HLPOL are all set to 1 (default setting).
2
Rev. A | Page 4 of 52
AD9974
DIGITAL SPECIFICATIONS
X = A = B, IOVDD_X = 1.6 V to 3.6 V, RGVDD_X = HVDD_X = 2.7 V to 3.6 V, CL = 20 pF, TMIN to TMAX, unless otherwise noted.
Table 4.
Parameter
LOGIC INPUTS
High Level Input Voltage (VIH)
Low Level Input Voltage (VIL)
High Level Input Current (IIH)
Low Level Input Current (IIL)
Input Capacitance (CIN)
LOGIC OUTPUTS
High Level Output Voltage (VOH)
Low Level Output Voltage (VOL)
CLI INPUT (CLI_BIAS = 0)
High Level Input Voltage (VIHCLI)
Low Level Input Voltage (VILCLI)
H-DRIVER OUTPUTS
High Level Output Voltage at Maximum Current (VOH)
Low Level Output Voltage at Maximum Current
Maximum Output Current (Programmable) (VOL)
Maximum Load Capacitance
Min
Typ
Max
IOVDD − 0.6
0.6
10
10
10
IOVDD − 0.5
V
V
IOVDD/2 − 0.5
V
V
HVDD − 0.5
0.5
30
Rev. A | Page 5 of 52
Test Conditions/Comments
V
V
μA
μA
pF
0.5
IOVDD/2 + 0.5
100
Unit
V
V
mA
pF
IOH = 2 mA
IOL = 2 mA
AD9974
ANALOG SPECIFICATIONS
X = A = B, AVDD_X = 1.8 V, fCLI = 65 MHz, typical timing specifications, TMIN to TMAX, unless otherwise noted.
Table 5.
Parameter
CDS 1
Allowable CCD Reset Transient
CDS Gain Accuracy
−3 dB CDS Gain
0 dB CDS Gain (Default)
3 dB CDS Gain
6 dB CDS Gain
Maximum Input Voltage
−3 dB CDS Gain
0 dB CDS Gain (Default)
3 dB CDS Gain
6 dB CDS Gain
Allowable OB Pixel Amplitude
0 dB CDS Gain (Default)
6 dB CDS Gain
VARIABLE GAIN AMPLIFIER (VGA_X)
Gain Control Resolution
Gain Monotonicity
Low Gain Setting (VGA Code 15, Default)
Maximum Gain Setting (VGA Code 1023)
BLACK LEVEL CLAMP
Clamp Level Resolution
Minimum Clamp Level (Code 0)
Maximum Clamp Level (Code 1023)
ADC (CHN_A and CHN_B)
Resolution
Differential Nonlinearity (DNL)
No Missing Codes
Integral Nonlinearity (INL)
Full-Scale Input Voltage
VOLTAGE REFERENCE
Reference Top Voltage (REFT_X)
Reference Bottom Voltage (REFB_X)
SYSTEM PERFORMANCE
VGA Gain Accuracy
Low Gain (Code 15)
Maximum Gain (Code 1023)
Peak Nonlinearity, 500 mV Input Signal
Total Output Noise
Power Supply Rejection (PSR)
1
Min
−3.3
−0.7
2.3
4.9
Typ
Max
Unit
0.5
0.8
V
−2.8
−0.2
2.8
5.4
−2.3
+0.3
3.3
5.9
Test Conditions/Comments
VGA gain = 5.6 dB (Code 15, default value)
1.4
1.0
0.7
0.5
V p-p
V p-p
V p-p
V p-p
−100
−50
14
−1.0
+200
+100
1024
Guaranteed
6
42
dB
dB
1024
0
1023
Steps
LSB
LSB
±0.5
Guaranteed
5
2.0
Steps
+1.2
15
1.4
0.4
5.1
41.3
mV
mV
5.6
41.8
0.1
2
48
Measured at ADC output
Measured at ADC output
Bits
LSB
LSB
V
V
V
6.1
42.3
0.4
Input signal characteristics are defined as shown in Figure 2.
Rev. A | Page 6 of 52
dB
dB
%
LSB rms
dB
Specifications include entire signal chain
0 dB CDS gain (default)
Gain = (0.0359 × code) + 5.1 dB
12 dB total gain applied
AC-grounded input, 6 dB gain applied
Measured with step change on supply
AD9974
MAXIMUM INPUT LIMIT =
LESSER OF 2.2V
OR (AVDD + 0.3V)
+1.8V TYP (AVDD)
800mV
MAXIMUM
200mV MAX
OPTICAL BLACK PIXEL
+1.3V TYP (AVDD – 0.5V)
DC RESTORE VOLTAGE
1V MAXIMUM INPUT
SIGNAL RANGE
(0dB CDS GAIN)
0V (AVSS)
MINIMUM INPUT LIMIT
(AVSS – 0.3V)
Figure 2. Input Signal Characteristics
Rev. A | Page 7 of 52
05957-002
500mV TYP
RESET TRANSIENT
AD9974
ABSOLUTE MAXIMUM RATINGS
Ratings apply to both Channel A and Channel B, unless
otherwise noted.
Table 6.
Parameter
AVDD to AVSS
DVDD to DVSS
DRVDD to DRVSS
IOVDD to DVSS
HVDD to HVSS
RGVDD to RGVSS
Any VSS
RG Output to RGVSS
H1 to H4, HL Output to HVSS
SCK, SL, SDI to DVSS
REFT, REFB, CCDINM, CCDINP to AVSS
Junction Temperature
Lead Temperature (10 sec)
Rating
−0.3 V to +2.2 V
−0.3 V to +2.2 V
−0.3 V to +3.9 V
−0.3 V to +3.9 V
−0.3 V to +3.9 V
−0.3 V to +3.9 V
−0.3 V to +0.3 V
−0.3 V to RGVDD + 0.3 V
−0.3 V to HVDD + 0.3 V
−0.3 V to IOVDD + 0.3 V
−0.2 V to AVDD + 0.2 V
150°C
350°C
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.
THERMAL CHARACTERISTICS
θJA is measured using a 4-layer PCB with the exposed paddle
soldered to the board.
Table 7. Thermal Resistance
Package Type
100-Lead, 9 mm × 9 mm, CSP_BGA
ESD CAUTION
Rev. A | Page 8 of 52
θJA
38.3
Unit
°C/W
AD9974
PIN CONFIGURATION AND FUNCTION DESCRIPTIONS
AD9974
TOP VIEW
(Not to Scale)
A1 CORNER
INDEX AREA
1
2
3
4
5
6
7
8
9 10
A
B
C
D
E
F
G
H
05955-003
J
K
Figure 3. Pin Configuration
Table 8. Pin Function Descriptions
Ball Location
B2
C2
D2
C1
D1
A1
F4
F3
D4
D3
B4
J2
K3
E3
E4
C3
C4
B3
Mnemonic
SL_A
SDATA_A
SCK_A
REFT_A
REFB_A
CCDINM_A
H1_A
H2_A
H3_A
H4_A
RG_A
DRVSS_A
DRVDD_A
HVSS_A
HVDD_A
RGVSS_A
RGVDD_A
IOVDD_A
Type 1
DI
DI
DI
AO
AO
AI
DO
DO
DO
DO
DO
P
P
P
P
P
P
P
A4
B1
A2
F2
F1
E2
E1
B8
C8
A5
A6
D8
C7
D7
A7
F10
F9
CLI_A
AVSS_A
CCDINP_A
DVSS_A
DVDD_A
VD_A
HD_A
SL_B
SDATA_B
LDO_OUT_A
CCDINM_B
SCK_B
REFT_B
REFB_B
CCDINP_B
H1_B
H2_B
DI
P
AI
P
P
DI
DI
DI
DI
P
AI
DI
AO
AO
AI
DO
DO
Description
3-Wire Serial Load for Channel A.
3-Wire Serial Data for Channel A.
3-Wire Serial Clock for Channel A.
Reference with Top Decoupling for Channel A. Decouple with 0.1 μF to AVSS_A.
Reference with Bottom Decoupling for Channel A. Decouple with 0.1 μF to AVSS_A.
Analog Input for Channel A Image Sensor Signal.
CCD Horizontal Clock 1 for Channel A.
CCD Horizontal Clock 2 for Channel A.
CCD Horizontal Clock 3 for Channel A.
CCD Horizontal Clock 4 for Channel A.
CCD Reset Gate Clock for Channel A.
Digital Driver Ground for Channel A.
Digital Driver Supply for Channel A: 1.8 V or 3.0 V.
H1_A to H4_A Driver Ground for Channel A.
H1_A to H4_A Driver Supply for Channel A: 3.0 V.
RG_A Driver Ground for Channel A.
RG_A Driver Supply for Channel A: 3.0 V.
Digital I/O Supply: 1.8 V or 3.0 V (HD, VD, SL, SCK, SDATA) and LDO Input (3.0 V Only)
When LDO Is Used.
Master Clock Input for Channel A.
Analog Ground for Channel A.
Analog Input for Channel A Image Sensor Signal.
Digital Ground for Channel A.
Digital Supply for Channel A: 1.8 V.
Vertical Sync Pulse for Channel A.
Horizontal Sync Pulse for Channel A.
3-Wire Serial Load for Channel B.
3-Wire Serial Data for Channel B.
1.8 V LDO Output from Channel A.
Analog Input for Channel B Image Sensor Signal.
3-Wire Serial Clock for Channel B.
Reference with Top Decoupling for Channel B. Decouple with 0.1 μF to AVSS_B.
Reference with Bottom Decoupling for Channel B. Decouple with 0.1 μF to AVSS_B.
Analog Input for Channel B Image Sensor Signal.
CCD Horizontal Clock 1 for Channel B.
CCD Horizontal Clock 2 for Channel B.
Rev. A | Page 9 of 52
AD9974
Ball Location
D10
D9
B10
J8
K9
E9
E10
C9
C10
B9
Mnemonic
H3_B
H4_B
RG_B
DRVSS_B
DRVDD_B
HVSS_B
HVDD_B
RGVSS_B
RGVDD_B
IOVDD_B
Type 1
DO
DO
DO
P
P
P
P
P
P
P
A10
B7
A8
F8
F7
E8
E7
A3
G1
H1
J1
K1
G2
H2
K2
G3
H3
J3
K4
J4
H4
G4
B5, C5, D5, E5, F5, G5, H5,
J5, K5, B6, C6, D6, E6, F6,
G6, H6, J6, K6
A9
G7
H7
J7
K7
G8
H8
K8
G9
H9
J9
K10
J10
H10
G10
LDO_OUT_B
AVSS_B
AVDD_B
DVSS_B
DVDD_B
VD_B
HD_B
AVDD_A
D0_A
D1_A
D2_A
D3_A
D4_A
D5_A
D6_A
D7_A
D8_A
D9_A
D10_A
D11_A
D12_A
D13_A
GND
P
P
P
P
P
DI
DI
P
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
P
Description
CCD Horizontal Clock 3 for Channel B.
CCD Horizontal Clock 4 for Channel B.
CCD Reset Gate Clock for Channel B.
Digital Driver Ground for Channel B.
Digital Driver Supply for Channel B: 1.8 V or 3.0 V.
H1_B to H4_B Driver Ground for Channel B.
H1_B to H4_B Driver Supply for Channel B: 3.0 V.
RG_B Driver Ground for Channel B.
RG_B Driver Supply for Channel B: 3.0 V.
Digital I/O Supply: 1.8 V or 3.0 V (HD, VD, SL, SCK, SDATA) and LDO Input (3.0 V Only)
When LDO Is Used.
1.8 V LDO Output from Channel B.
Analog Ground for Channel B.
Analog Supply for Channel B: 1.8 V.
Digital Ground for Channel B.
Digital Supply for Channel B: 1.8 V.
Vertical Sync Pulse for Channel B.
Horizontal Sync Pulse for Channel B.
Analog Supply for Channel A: 1.8 V.
Data Outputs Channel A.
Data Outputs Channel A.
Data Outputs Channel A.
Data Outputs Channel A.
Data Outputs Channel A.
Data Outputs Channel A.
Data Outputs Channel A.
Data Outputs Channel A.
Data Outputs Channel A.
Data Outputs Channel A.
Data Outputs Channel A.
Data Outputs Channel A.
Data Outputs Channel A.
Data Outputs Channel A.
Ground Connection.
CLI_B
D0_B
D1_B
D2_B
D3_B
D4_B
D5_B
D6_B
D7_B
D8_B
D9_B
D10_B
D11_B
D12_B
D13_B
DI
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
Master Clock Input for Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
Data Outputs Channel B.
1
AI = analog input, AO = analog output, DI = digital input, DO = digital output, P = power.
Rev. A | Page 10 of 52
AD9974
TYPICAL PERFORMANCE CHARACTERISTICS
10
250
8
200
6
TOTAL POWER
LSB
POWER (mW)
4
150
3.3V SUPPLIES
100
2
0
–2
–4
50
1.8V SUPPLIES
30
35
40
45
50
55
60
65
SAMPLE RATE (MHz)
–8
0
2k
8k
10k
12k
14k
16k
Figure 7. Integral Nonlinearity
180
1.0
160
0.8
140
0.6
120
INL MISMATCH (%)
100
80
60
40
0.4
0.2
0
–0.2
–0.4
–0.6
20
0
5
10
15
20
25
30
35
40
45
VGA GAIN (dB)
05955-006
–0.8
0
Figure 5. RMS Output Noise vs. VGA Gain
0.8
0.6
0.4
0.2
0
–0.2
–0.4
–0.6
–1.0
4k
6k
8k
10k
12k
ADC OUTPUT CODE
14k
16k
05955-007
–0.8
2k
3k
5k
7k
9k
11k
13k
15k
ADC OUTPUT CODE
Figure 8. Linearity Mismatch vs. ADC Output Code
1.0
0
–1.0
1k
Figure 6. Differential Nonlinearity
Rev. A | Page 11 of 52
05955-009
RMS OUTPUT NOISE (LSB)
6k
ADC OUTPUT CODE
Figure 4. Power vs. Sample Rate
LSB
4k
05955-008
25
05955-005
0
20
–6
AD9974
EQUIVALENT INPUT/OUTPUT CIRCUITS
IOVDD
AVDD
330Ω
R
CLI
100kΩ
AVSS
05955-011
AVSS
05955-010
+
AVSS
Figure 11. CLI Input, Register 0x15[0] = 1 Enables the Bias Circuit
Figure 9. CCDIN Input
HVDD OR RGVDD
DATA
IOVDD
OUTPUT
ENABLE
HVSS OR RGVSS
Figure 10. Digital Inputs
Figure 12. H1 to H4 and RG Outputs
Rev. A | Page 12 of 52
05955-013
DVSS
05955-012
330Ω
AD9974
TERMINOLOGY
MAX
Differential Nonlinearity (DNL)
An ideal ADC exhibits code transitions that are exactly 1 LSB
apart. DNL is the deviation from this ideal value. Therefore,
every code must have a finite width. No missing codes guaranteed
to 14-bit resolution indicates that all 16,384 codes, each for its
respective input, must be present over all operating conditions.
Total Output Noise
The rms output noise is measured using histogram techniques.
The standard deviation of the ADC output codes is calculated
in LSB and represents the rms noise level of the total signal
chain at the specified gain setting. The output noise can be
converted to an equivalent voltage using the relationship
1 LSB = (ADC Full Scale/2n Codes)
OUTPUT CODE (LSB)
CHANNEL A
INL A(X)
INL B(X)
OUTPUT (X)
MAX/16
0
0
X
FS
INPUT VOLTAGE
Figure 13. Linearity Mismatch Definition
Power Supply Rejection (PSR)
The PSR is measured with a step change applied to the supply
pins. The PSR specification is calculated from the change in the
data outputs for a given step change in the supply voltage.
Crosstalk
The crosstalk is measured while applying a full-scale step to
one channel and measuring the interference on the opposite
channel.
⎛ Interference (LSB) ⎞
⎟⎟
Crosstalk (dB) = 20 × log ⎜⎜
16,384
⎠
⎝
where n is the bit resolution of the ADC. For the AD9974,
1 LSB is approximately 122.0 μV.
Linearity Mismatch
The linearity mismatch is calculated by taking the difference in
INL of the two channels at Input X, and then expressing the
difference as a percentage of the output code at X. The values
given in Table 2 are obtained over the range of 1⁄16 and
maximum of the output code. The general trend is for the
linearity mismatch to decrease as the output approaches the
maximum code, as shown in Figure 8.
Linearity Mismatch (%) =
IDEAL
05955-004
Peak Nonlinearity
Peak nonlinearity, a full signal chain specification, refers to the
peak deviation of the output of the AD9974 from a true straight
line. The point used as zero scale occurs 0.5 LSB before the first
code transition. Positive full scale is defined as a level 1 LSB
and 0.5 LSB beyond the last code transition. The deviation is
measured from the middle of each particular output code to the
true straight line. The error is then expressed as a percentage
of the 2 V ADC full-scale signal. The input signal is always
appropriately gained up to fill the ADC full-scale range.
CHANNEL B
INL A( X ) − INL B( X )
Output Code( X )
Rev. A | Page 13 of 52
AD9974
THEORY OF OPERATION
All AD9974 clocks are synchronized with VD and HD inputs.
All of the AD9974 horizontal pulses (CLPOB, PBLK, and
HBLK) are programmed and generated internally.
V DRIVER
V1 > Vx, VSG1 > VSGx, SUBCK
H1_A TO H4_A, RG_A
H1_B TO H4_B, RG_B
The H-drivers for H1 to H4 and RG are included in the
AD9974, allowing these clocks to be directly connected to the
CCD. An H-driver voltage of 3 V is supported in the AD9974.
DOUT_A
DOUT_B
CCDINP_A
CCDINM_A
CCD
CCDINP_B
AD9974
INTEGRATED
HD_A, VD_A,
AFE + TD
HD_B, VD_B
DIGITAL IMAGE
PROCESSING
ASIC
CCDINM_B
05955-014
CLI_A, CLI_B
SERIAL
INTERFACE
Figure 15 and Figure 16 show the maximum horizontal and
vertical counter dimensions for the AD9974. All internal horizontal
and vertical clocking is controlled by these counters, which specify
line and pixel locations. Maximum HD length is 8191 pixels per
line, and maximum VD length is 8191 lines per field.
MAXIMUM COUNTER DIMENSIONS
Figure 14. Typical Application
13-BIT HORIZONTAL = 8192 PIXELS MAX
13-BIT VERTICAL = 8192 LINES MAX
05955-015
Figure 14 shows the typical system block diagram for the AD9974.
The charge-coupled device (CCD) output is processed by the
analog front-end (AFE) circuitry of the AD9974, consisting of a
CDS, VGA, black level clamp, and ADC. The digitized pixel
information is sent to the digital image processor chip, which
performs the postprocessing and compression. To operate the
CCD, all CCD timing parameters are programmed into the
AD9974 from the system ASIC through the 3-wire serial
interface. From the system master clock, CLI_X, which is provided
by the image processor or external crystal, the AD9974 generates
the horizontal clocks of the CCD and all internal AFE clocks.
Figure 15. Vertical and Horizontal Counters
MAX VD LENGTH IS 8192 LINES
VD
MAX HD LENGTH IS 8192 PIXELS
HD
05955-016
CLI
Figure 16. Maximum VD/HD Dimensions
Rev. A | Page 14 of 52
AD9974
PROGRAMMABLE TIMING GENERATION
Using a 65 MHz CLI frequency, the edge resolution of the
Precision Timing core is approximately 240 ps. If a 1× system
clock is not available, it is possible to use a 2× reference clock
by programming the CLIDIVIDE register (Address 0x0D). The
AD9974 then internally divides the CLI frequency by 2.
PRECISION TIMING HIGH SPEED TIMING CORE
The AD9974 generates flexible high speed timing signals using
the Precision Timing core. This core, composed of the Reset
Gate RG, Horizontal Driver H1 to Horizontal Driver H4, and
SHP/SHD sample clocks, is the foundation for generating the
timing for both the CCD and the AFE. A unique architecture
makes it routine for the system designer to optimize image
quality by providing precise control over the horizontal CCD
readout and the AFE correlated double sampling.
High Speed Clock Programmability
Figure 18 shows when the high speed clocks, RG, H1 to H4,
SHP, and SHD, are generated. The RG pulse has programmable
rising and falling edges and can be inverted using the polarity
control. The H1 and H2 horizontal clocks have separate programmable rising and falling edges, as well as separate polarity control.
The AD9974 provides additional HCLK-mode programmability,
as described in Table 9.
Timing Resolution
The Precision Timing core uses a master clock input (CLI_X)
as a reference. This clock input should be the same as the CCD
pixel clock frequency. Figure 17 illustrates how the internal timing
core divides the master clock period into 64 steps or edge positions;
therefore, the edge resolution of the Precision Timing core is
(tCLI/64). For more information on using the CLI input, refer to
the Applications Information section.
POSITION
P[0]
P[16]
The edge location registers are each six bits wide, allowing the
selection of all 64 edge locations. Figure 21 shows the default
timing locations for all of the high speed clock signals.
P[32]
P[48]
P[64] = P[0]
CLI
tCLIDLY
1 PIXEL
PERIOD
05955-017
NOTES
1. THE PIXEL CLOCK PERIOD IS DIVIDED INTO 64 POSITIONS, PROVIDING FINE EDGE RESOLUTION FOR HIGH SPEED CLOCKS.
2. THERE IS A FIXED DELAY FROM THE CLI INPUT TO THE INTERNAL PIXEL PERIOD POSITION (tCLIDLY).
Figure 17. High Speed Clock Resolution from CLI Master Clock Input
1
2
CCD
SIGNAL
3
4
RG
5
6
H1, H3
H2, H4
05955-018
PROGRAMMABLE CLOCK POSITIONS:
1SHP SAMPLE LOCATION.
2SHD SAMPLE LOCATION.
3RG RISING EDGE.
4RG FALLING EDGE.
5H1 RISING EDGE.
6H1 FALLING EDGE.
Figure 18. High Speed Clock Programmable Locations (HCLK Mode 1)
Rev. A | Page 15 of 52
AD9974
1
2
H1, H3
4
3
H2, H4
05955-019
H1 TO H4 PROGRAMMABLE LOCATIONS:
1H1 RISING EDGE.
2H1 FALLING EDGE.
3H2 RISING EDGE.
4H2 FALLING EDGE.
Figure 19. HCLK Mode 2 Operation
1
2
H1
H2
3
4
H3
H4
05955-020
H1 TO H4 PROGRAMMABLE LOCATIONS:
1H1 RISING EDGE.
2H1 FALLING EDGE.
3H3 RISING EDGE.
4H3 FALLING EDGE.
Figure 20. HCLK Mode 3 Operation
POSITION
P[0]
P[16]
RGr[0]
RGf[16]
P[32]
P[48]
P[64] = P[0]
CLI
RG
H1r[0]
H1f[32]
H1
H2
tS1
CCD
SIGNAL
SHPLOC[32]
SHP
tSHPINH
SHDLOC[0]
tSHDINH
SHD
DATAPHASEP[32]
tDOUTINH
NOTES
1. ALL SIGNAL EDGES ARE FULLY PROGRAMMABLE TO ANY OF THE 64 POSITIONS WITHIN ONE PIXEL PERIOD.
TYPICAL POSITIONS FOR EACH SIGNAL ARE SHOWN. HCLK MODE 1 IS SHOWN.
2. CERTAIN POSITIONS SHOULD BE AVOIDED FOR EACH SIGNAL, SHOWN ABOVE AS INHIBIT REGIONS.
3. IF A SETTING IN THE INHIBIT REGION IS USED, AN UNSTABLE PIXEL SHIFT CAN OCCUR IN THE HBLK LOCATION OR AFE PIPELINE.
Figure 21. High Speed Timing Default Locations
Rev. A | Page 16 of 52
05955-021
DOUTPHASEP
AD9974
Table 9. HCLK Modes, Selected by HCLKMODE Register (Address 0x23[7:5])
HCLK Mode
Mode 1
Mode 2
Register Value
001
010
Mode 3
100
Invalid Selection
000, 011, 101, 110, 111
Description
H1 edges are programmable, with H3 = H1 and H2 = H4 = inverse of H1.
H1 edges are programmable, with H3 = H1.
H2 edges are programmable, with H4 = H2.
H1 edges are programmable, with H2 = inverse of H1.
H3 edges are programmable, with H4 = inverse of H3.
Invalid register settings.
Table 10. H1, H2, RGCONTROL, DRVCONTROL, and SAMPCONTROL Register Parameters
Parameter
Polarity
Length (Bits)
1
Range
High/low
Positive Edge
Negative Edge
Sample Location
Drive Control
6
6
6
3
0 to 63 edge location
0 to 63 edge location
0 to 63 sample location
0 to 7 current steps
Description
Polarity control for H1/H3 and RG.
0 = no inversion.
1 = inversion.
Positive edge location for H1/H3 and RG.
Negative edge location for H1/H3 and RG.
Sampling location for SHP and SHD.
Drive current for H1 to H4 and RG outputs, 0 to 7 steps of 4.3 mA each.
CLI
tCLIDLY
N
N+1
N+2
N+3
N+4
N+5
N+6
N+7
N–14
N–13
N–12
N–11
N–10
N+8
N+9
N+10
N+11
N+12
N+13
N+14
N+15
N+16
N+17
N–9
N–8
N–7
N–6
N–5
N–4
N–3
N–2
N–1
N
CCDIN
SAMPLE PIXEL N
SHD
(INTERNAL)
ADC OUT
(INTERNAL)
N–17
N–16
N–15
N+1
tDOUTINH
DOUTPHASE
CLK
PIPELINE LATENCY = 16 CYCLES
N–16
N–15
N–14
N–13
N–12
N–11
N–10
N–9
N–8
N–7
N–6
N–5
N–4
N–3
N–2
N–1
NOTES
1. EXAMPLE SHOWN FOR SHDLOC = 0.
2. HIGHER VALUES OF SHD AND/OR DOUTPHASE WILL SHIFT DOUT TRANSITION TO THE RIGHT, WITH RESPECT TO CLI LOCATION.
N
N+1
05955-022
DOUT
N–17
Figure 22. Pipeline Delay of AFE Data Outputs
H-Driver and RG Outputs
Digital Data Outputs
In addition to the programmable timing positions, the AD9974
features on-chip output drivers for the RG and H1 to H4 outputs.
These drivers are powerful enough to drive the CCD inputs
directly. The H-driver and RG-driver current can be adjusted
for optimum rise/fall time into a particular load by using the
drive strength control registers (Address 0x35). Use the register
to adjust the drive strength in 4.3 mA increments. The minimum
setting of 0 is equal to off or three-state, and the maximum setting
of 7 is equal to 30.1 mA.
For maximum system flexibility, the AD9974 uses the
DOUTPHASE registers (Address 0x37[11:0]) to select the
location for the start of each new pixel data value. Any edge
location from 0 to 63 can be programmed. These registers
determine the start location of the data output and the DCLK
rising edge with respect to the master clock input, CLI_X.
The pipeline delay through the AD9974 is shown in Figure 22.
After the CCD input is sampled by SHD, there is a 16-cycle
delay until the data is available.
Rev. A | Page 17 of 52
AD9974
HORIZONTAL CLAMPING AND BLANKING
CLPOB and PBLK Masking Area
The horizontal clamping and blanking pulses of the AD9974 are
fully programmable to suit a variety of applications. Individual
control is provided for CLPOB, PBLK, and HBLK during the
different regions of each field. This allows the dark pixel clamping
and blanking patterns to be changed at each stage of the readout
to accommodate different image transfer timing and high speed
line shifts.
Additionally, the AD9974 allows the CLPOB and PBLK signals
to be disabled during certain lines in the field without changing
any of the existing pattern settings. There are three sets of start
and end registers for both CLPOB and PBLK that allow the
creation of up to three masking areas for each signal.
Individual CLPOB and PBLK Patterns
The AFE horizontal timing consists of CLPOB and PBLK, as
shown in Figure 23. These two signals are programmed independently using the registers in Table 11. The start polarity for
the CLPOB or PBLK signal is CLPOB_POL (PBLK_POL), and the
first and second toggle positions of the pulse are CLPOB_TOG1
(PBLK_TOG1) and CLPOB_TOG2 (PBLK_TOG2). Both signals
are active low and need to be programmed accordingly.
Two separate patterns for CLPOB and PBLK can be programmed
for each H-pattern, CLPOB0, CLPOB1, PBLK0, and PBLK1.
The CLPOB_PAT and PBLK_PAT field registers select which
of the two patterns is used in each field.
For example, to use the CLPOB masking, program the
CLPOBMASKSTART and CLPOBMASKEND registers to
specify the starting and ending lines in the field where the CLPOB
patterns are to be ignored. Figure 24 illustrates this feature.
The masking registers are not specific to a certain H-pattern;
they are always active for any existing field of timing. To disable
the CLPOB and PBLK masking feature, set these registers to the
maximum value of 0x1FFF.
Note that to disable CLPOB and PBLK masking during
power-up, it is recommended that CLPOBMASKSTART
(PBLKMASKSTART) be set to 8191 and CLPOBMASKEND
(PBLKMASKEND) be set to 0. This prevents any accidental
masking caused by different register update events.
Figure 34 shows how the sequence change positions divide the
readout field into different regions. By assigning a different
H-pattern to each region, the CLPOB and PBLK signals can
change with each change in the vertical timing.
HD
2
CLPOB 1
PBLK
3
ACTIVE
ACTIVE
PROGRAMMABLE SETTINGS:
BLANK REGION ARE ACTIVE LOW).
05955-023
1START POLARITY (CLAMP AND
2FIRST TOGGLE POSITION.
3SECOND TOGGLE POSITION.
Figure 23. Clamp and Preblank Pulse Placement
NO CLPOB SIGNAL
FOR LINE 600
NO CLPOB SIGNAL
FOR LINES 6 TO 8
VD
0
1
2
597 598
HD
CLPOBMASKSTART1 = 6
CLPOBMASKEND1 = 8
CLPOBMASKSTART2 = CLPOBMASKEND2 = 600
Figure 24. CLPOB Masking Example
Rev. A | Page 18 of 52
05957-024
CLPOB
AD9974
Table 11. CLPOB and PBLK Pattern Registers
Parameter
CLPOB0_TOG1
CLPOB0_TOG2
CLPOB1_TOG1
CLPOB1_TOG2
CLPOB_POL
CLPOB_PAT
CLPOBMASKSTART
CLPOBMASKEND
PBLK0_TOG1
PBLK0_TOG2
PBLK1_TOG1
PBLK1_TOG2
PBLK_POL
PBLK_PAT
PBLKMASKSTART
PBLKMASKEND
Length (Bits)
13
13
13
13
9
9
13
13
13
13
13
13
9
9
13
13
Range
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
High/low
0 to 9 settings
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
High/low
0 to 9 settings
0 to 8191 pixel location
0 to 8191 pixel location
Description
First CLPOB0 toggle position within the line for each V-sequence.
Second CLPOB0 toggle position within the line for each V-sequence.
First CLPOB1 toggle position within the line for each V-sequence.
Second CLPOB1 toggle position within the line for each V-sequence.
Starting polarity of CLPOB for each V-sequence [8:0] (in field registers).
CLPOB pattern selection for each V-sequence [8:0] (in field registers).
CLPOB mask start position: three values available (in field registers).
CLPOB mask end position: three values available (in field registers).
First PBLK0 toggle position within the line for each V-sequence.
Second PBLK0 toggle position within the line for each V-sequence.
First PBLK1 toggle position within the line for each V-sequence.
Second toggle position within the line for each V-sequence.
Starting polarity of PBLK for each V-sequence [8:0] (in field registers).
PBLK pattern selection for each V-sequence [8:0] (in field registers).
PBLK mask start position: three values available (in field registers).
PBLK mask end position: three values available (in field registers).
HD
HBLK
HBLKTOGE2
BLANK
BLANK
BASIC HBLK PULSE IS GENERATED USING HBLKTOGE1 AND HBLKTOGE2 REGISTERS (HBLKALT = 0).
05955-025
HBLKTOGE1
Figure 25. Typical Horizontal Blanking Pulse Placement (HBLKMODE = 0)
HD
HBLK
H1/H3
THE POLARITY OF H1/H3 DURING BLANKING IS PROGRAMMABLE
(H2/H4 POLARITY IS SEPARATELY PROGRAMMABLE)
05955-026
H1/H3
H2/H4
Figure 26. HBLK Masking Control
Rev. A | Page 19 of 52
AD9974
Individual HBLK Patterns
HBLK Mode 0 Operation
The HBLK programmable timing shown in Figure 25 is similar
to CLPOB and PBLK; however, there is no start polarity control.
Only the toggle positions designate the start and the stop positions
of the blanking period. Additionally, as shown in Figure 26, there
is a polarity control, HBLKMASK, for H1/H3 and H2/H4 that
designates the polarity of the horizontal clock signals during the
blanking period. Setting HBLKMASK_H1 low sets H1 = H3 =
low and HBLKMASK_H2 high sets H2 = H4 = high during the
blanking. As with the CLPOB and PBLK signals, HBLK registers
are available in each H-pattern group, allowing unique blanking
signals to be used with different vertical timing sequences.
There are six toggle positions available for HBLK. Normally,
only two of the toggle positions are used to generate the standard
HBLK interval. However, the additional toggle positions can be
used to generate special HBLK patterns, as shown in Figure 27.
The pattern in this example uses all six toggle positions to generate
two extra groups of pulses during the HBLK interval. By changing
the toggle positions, different patterns are created.
Separate toggle positions are available for even and odd lines.
If alternation is not needed, load the same values into the registers
for even (HBLKTOGE) and odd (HBLKTOGO) lines.
The AD9974 supports three modes of HBLK operation. HBLK
Mode 0 supports basic operation and provides some support for
special HBLK patterns. HBLK Mode 1 supports pixel mixing
HBLK operation. HBLK Mode 2 supports advanced HBLK operation. The following sections describe each mode. Register
parameters are detailed in Table 12.
HBLKTOGE2
HBLKTOGE1
HBLKTOGE4
HBLKTOGE3
HBLKTOGE6
HBLKTOGE5
HBLK
H1/H3
SPECIAL H-BLANK PATTERN IS CREATED USING MULTIPLE HBLK TOGGLE POSITIONS (HBLKALT = 0).
05955-027
H2/H4
Figure 27. Generating Special HBLK Patterns
Table 12. HBLK Pattern Registers
Register
HBLK_MODE
Length (Bits)
2
Range
0 to 2 HBLK modes
HBLKSTART
HBLKEND
HBLKLEN
HBLKREP
HBLKMASK_H1
HBLKMASK_H2
13
13
13
13
1
1
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixels
0 to 8191 repetitions
High/low
High/low
Description
Enables different HBLK toggle position operations.
0 = normal mode. Six toggle positions available for even and odd lines.
If even/odd alternation is not needed, set toggles for even/odd the same.
1 = pixel mixing mode. In addition to six toggle positions, the HBLKSTART,
HBLKEND, HBLKLEN, and HBLKREP registers can be used to generate HBLK
patterns. If even/odd alternation is not needed, set toggles for even/odd
the same.
2 = advanced HBLK mode. Divides HBLK interval into six different repeat
areas. Uses HBLKSTARTA/B/C and RA*H*REPA/B/C registers.
3 = test mode only. Do not access.
Start location for HBLK in HBLK Mode 1 and HBLK Mode 2.
End location for HBLK in HBLK Mode 1 and HBLK Mode 2.
HBLK length in HBLK Mode 1 and HBLK Mode 2.
Number of HBLK repetitions in HBLK Mode 1 and HBLK Mode 2.
Masking polarity for H1 and H3 during HBLK.
Masking polarity for H2 and H4 during HBLK.
Rev. A | Page 20 of 52
AD9974
Register
HBLKTOGO1
HBLKTOGO2
HBLKTOGO3
HBLKTOGO4
HBLKTOGO5
HBLKTOGO6
HBLKTOGE1
HBLKTOGE2
HBLKTOGE3
HBLKTOGE4
HBLKTOGE5
HBLKTOGE6
RA0H1REPA/B/C
Length (Bits)
13
13
13
13
13
13
13
13
13
13
13
13
12
Range
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 15 HCLK pulses
for each A, B, and C
RA1H1REPA/B/C
RA2H1REPA/B/C
RA3H1REPA/B/C
RA4H1REPA/B/C
RA5H1REPA/B/C
RA0H2REPA/B/C
12
12
12
12
12
12
0 to 15 HCLK pulses
0 to 15 HCLK pulses
0 to 15 HCLK pulses
0 to 15 HCLK pulses
0 to 15 HCLK pulses
0 to 15 HCLK pulses
for each A, B, and C
RA1H2REPA/B/C
RA2H2REPA/B/C
RA3H2REPA/B/C
RA4H2REPA/B/C
RA5H2REPA/B/C
HBLKSTARTA
HBLKSTARTB
HBLKSTARTC
HBLKALT_PAT1
12
12
12
12
12
13
13
13
3
0 to 15 HCLK pulses
0 to 15 HCLK pulses
0 to 15 HCLK pulses
0 to 15 HCLK pulses
0 to 15 HCLK pulses
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 5 even repeat area
HBLKALT_PAT2
HBLKALT_PAT3
HBLKALT_PAT4
HBLKALT_PAT5
HBLKALT_PAT6
3
3
3
3
3
0 to 5 even repeat area
0 to 5 even repeat area
0 to 5 even repeat area
0 to 5 even repeat area
0 to 5 even repeat area
Description
First HBLK toggle position for odd lines in HBLK Mode 0 and HBLK Mode 1.
Second HBLK toggle position for odd lines in HBLK Mode 0 and HBLK Mode 1.
Third HBLK toggle position for odd lines in HBLK Mode 0 and HBLK Mode 1.
Fourth HBLK toggle position for odd lines in HBLK Mode 0 and HBLK Mode 1.
Fifth HBLK toggle position for odd lines in HBLK Mode 0 and HBLK Mode 1.
Sixth HBLK toggle position for odd lines in HBLK Mode 0 and HBLK Mode 1.
First HBLK toggle position for even lines in HBLK Mode 0 and HBLK Mode 1.
Second HBLK toggle position for even lines in HBLK Mode 0 and HBLK Mode 1.
Third HBLK toggle position for even lines in HBLK Mode 0 and HBLK Mode 1.
Fourth HBLK toggle position for even lines in HBLK Mode 0 and HBLK Mode 1.
Fifth HBLK toggle position for even lines in HBLK Mode 0 and HBLK Mode 1.
Sixth HBLK toggle position for even lines in HBLK Mode 0 and HBLK Mode 1.
HBLK Repeat Area 0. Number of H1 repetitions for HBLKSTARTA/B/C in
HBLK Mode 2 for even lines; odd lines are defined using HBLKALT_PAT.
[3:0] RA0H1REPA. Number of H1 pulses following HBLKSTARTA.
[7:4] RA0H1REPB. Number of H1 pulses following HBLKSTARTB.
[11:8] RA0H1REPC. Number of H1 pulses following HBLKSTARTC.
HBLK Repeat Area 1. Number of H1 repetitions for HBLKSTARTA/B/C.
HBLK Repeat Area 2. Number of H1 repetitions for HBLKSTARTA/B/C.
HBLK Repeat Area 3. Number of H1 repetitions for HBLKSTARTA/B/C.
HBLK Repeat Area 4. Number of H1 repetitions for HBLKSTARTA/B/C.
HBLK Repeat Area 5. Number of H1 repetitions for HBLKSTARTA/B/C.
HBLK Repeat Area 0. Number of H2 repetitions for HBLKSTARTA/B/C in
HBLK Mode 2 for even lines; odd lines are defined using HBLKALT_PAT.
[3:0] RA0H2REPA. Number of H2 pulses following HBLKSTARTA.
[7:4] RA0H2REPB. Number of H2 pulses following HBLKSTARTB.
[11:8] RA0H2REPC. Number of H2 pulses following HBLKSTARTC.
HBLK Repeat Area 1. Number of H2 repetitions for HBLKSTARTA/B/C.
HBLK Repeat Area 2. Number of H2 repetitions for HBLKSTARTA/B/C.
HBLK Repeat Area 3. Number of H2 repetitions for HBLKSTARTA/B/C.
HBLK Repeat Area 4. Number of H2 repetitions for HBLKSTARTA/B/C.
HBLK Repeat Area 5. Number of H2 repetitions for HBLKSTARTA/B/C.
HBLK Repeat Area Start Position A for HBLK Mode 2.
HBLK Repeat Area Start Position B for HBLK Mode 2.
HBLK Repeat Area Start Position C for HBLK Mode 2.
HBLK Mode 2, Odd Field Repeat Area 0 pattern, selected from even field.
Repeat areas previously defined.
HBLK Mode 2, Odd Field Repeat Area 1 pattern.
HBLK Mode 2, Odd Field Repeat Area 2 pattern.
HBLK Mode 2, Odd Field Repeat Area 3 pattern.
HBLK Mode 2, Odd Field Repeat Area 4 pattern.
HBLK Mode 2, Odd Field Repeat Area 5 pattern.
Rev. A | Page 21 of 52
AD9974
HBLK Mode 1 Operation
Increasing H-Clock Width During HBLK
Enable multiple repeats of the HBLK signal by setting
HBLK_MODE to 1. In this mode, the HBLK pattern can be
generated using a different set of registers: HBLKSTART,
HBLKEND, HBLKLEN, and HBLKREP, along with the six
toggle positions (see Figure 28).
HBLK Mode 0 and HBLK Mode 1 allow the H1 to H4 pulse
width to be increased during the HBLK interval. As shown in
Figure 29, the H-clock frequency can be reduced by a factor of
1/2, 1/4, 1/6, 1/8, 1/10, 1/12, and so on, up to 1/30. To enable this
feature, the HCLK_WIDTH register (Address 0x34[7:4]) is set to
a value between 1 and 15. When this register is set to 0, the wide
HCLK feature is disabled. The reduced frequency occurs only for
H1 to H4 pulses that are located within the HBLK area.
Separate toggle positions are available for even and odd lines.
If alternation is not needed, load the same values into the registers
for even (HBLKTOGE) and odd (HBLKTOGO) lines.
Generating HBLK Line Alternation
HBLK Mode 0 and HBLK Mode 1 provide the ability to
alternate different HBLK toggle positions on even and odd
lines. Separate toggle positions are available for even and odd
lines. If even/odd line alternation is not required, load the same
values into the registers for even (HBLKTOGE) and odd
(HBLKTOGO) lines.
The HCLK_WIDTH register is generally used in conjunction
with special HBLK patterns to generate vertical and horizontal
mixing in the CCD.
Note that the wide HCLK feature is available only in HBLK Mode 0
and HBLK Mode 1, not in HBLK Mode 2.
Table 13. HCLK Width Register
Register
HCLK_WIDTH
Length (Bits)
4
Description
Controls H1 to H4 width during HBLK as a fraction of pixel rate.
0 = same frequency as pixel rate.
1 = 1/2 pixel frequency, that is, doubles the HCLK pulse width.
2 = 1/4 pixel frequency.
3 = 1/6 pixel frequency.
4 = 1/8 pixel frequency.
5 = 1/10 pixel frequency.
…
15 = 1/30 pixel frequency.
HBLKTOGE2
HBLKTOGE4
HBLKSTART HBLKTOGE1
HBLKTOGE3
HBLKEND
HBLK
HBLKLEN
HBLKREP = 3
H1/H3
HBLKREP NUMBER 1
HBLKREP NUMBER 2
05955-028
H2/H4
HBLKREP NUMBER 3
H-BLANK REPEATING PATTERN IS CREATED USING HBLKLEN AND HBLKREP REGISTERS.
Figure 28. HBLK Repeating Pattern Using HBLKMODE = 1
HBLK
H1/H3
1/FPIX
2 × (1/FPIX)
H-CLOCK FREQUENCY CAN BE REDUCED DURING HBLK BY 1/2 (AS SHOWN),
1/4, 1/6, 1/8, 1/10, 1/12, AND SO ON, UP TO 1/30 USING HBLKWIDTH REGISTER.
Figure 29. Generating Wide H-Clock Pulses During HBLK Interval
Rev. A | Page 22 of 52
05955-029
H2/H4
AD9974
Figure 31 shows the following example:
HBLK Mode 2 Operation
RA0H1REPA/RA0H1REPB/RA0H1REPC =
RA0H2REPA/RA0H2REPB/RA0H2REPC =
RA1H1REPA/RA1H1REPB/RA1H1REPC =
RA1H2REPA/RA1H2REPB/RA1H2REPC = 2.
HBLK Mode 2 allows more advanced HBLK pattern operation.
If unevenly spaced HCLK pulses in multiple areas are needed,
HBLK Mode 2 can be used. Using a separate set of registers,
HBLK Mode 2 can divide the HBLK region into up to six repeat
areas (see Table 12). As shown in Figure 31, each repeat area
shares a common group of toggle positions, HBLKSTARTA,
HBLKSTARTB, and HBLKSTARTC. However, the number of
toggles following each start position can be unique in each
repeat area by using the RAH1REP and RAH2REP registers.
As shown in Figure 30, setting the RAH1REPA/RAH1REPB/
RAH1REPC or RAH2REPA/RAH2REPB/RAH2REPC registers
to 0 masks HCLK groups from appearing in a particular repeat
area. Figure 31 shows only two repeat areas being used, although
six are available. It is possible to program a separate number of
repeat area repetitions for H1 and H2, but generally the same
value is used for both H1 and H2.
HD
CREATE UP TO 3 GROUPS OF TOGGLES
A, B, C COMMON IN ALL REPEAT AREAS
A
Furthermore, HBLK Mode 2 allows a different HBLK pattern
on even and odd lines. The HBLKSTARTA, HBLKSTARTB, and
HBLKSTARTC registers, as well as the RAH1REPA/RAH1REPB/
RAH1REPC and RAH2REPA/RAH2REPB/RAH2REPC registers,
define operation for the even lines. For separate control of the
odd lines, the HBLKALT_PAT registers specify up to six repeat
areas on the odd lines by reordering the repeat areas used for the
even lines. New patterns are not available, but the order of the
previously defined repeat areas on the even lines can be changed
for the odd lines to accommodate advanced CCD operation.
CHANGE NUMBER OF A, B, C PULSES IN ANY
REPEAT AREA USING RA*H*REP* REGISTERS
MASK A, B, C PULSES IN ANY REPEAT
AREA BY SETTING RA*H*REP* = 0
B
C
H1
REPEAT AREA 0
REPEAT AREA 1
REPEAT AREA 2
REPEAT AREA 3
REPEAT AREA 4
REPEAT AREA 5
HBLKSTART
HBLKEND
05955-030
H2
Figure 30. HBLK Mode 2 Operation
HD
HBLKLEN
HBLK
HBLKSTARTA
ALL RA*H*REPA, B, C REGISTERS = 2, TO CREATE 2 HCLK PULSES
HBLKSTARTB
HBLKSTARTC
H1
RA0H1REPA RA0H1REPB
RA0H1REPC
RA1H1REPA RA1H1REPB
RA1H1REPC
RA1H2REPA RA1H2REPB
RA1H2REPC
H2
RA0H2REPA RA0H2REPB
RA0H2REPC
REPEAT AREA 1
REPEAT AREA 0
HBLKREP = 2
TO CREATE 2 REPEAT AREAS
Figure 31. HBLK Mode 2 Registers
Rev. A | Page 23 of 52
HBLKEND
05955-031
HBLKSTART
AD9974
For example, if the desired toggle position is 100, CLPOB_TOG
should be set to 88 (that is, 100 − 12). Figure 49 shows the 12-cycle
pipeline delay referenced to the falling edge of HD.
HBLK, PBLK, and CLPOB Toggle Positions
The AD9974 uses an internal horizontal pixel counter to position
the HBLK, PBLK, and CLPOB toggle positions. The horizontal
counter does not reset to 0 until 12 CLI periods after the falling
edge of HD. This 12-cycle pipeline delay must be considered
when determining the register toggle positions. For example, if
CLPOB_TOG1 is 100 and the pipeline delay is not considered,
the final toggle position is applied at 112. To obtain the correct
toggle positions, the toggle position registers must be set to the
desired toggle position minus 12.
PIXEL NO.
Caution
Toggle positions cannot be programmed during the 12-cycle
delay from the HD falling edge until the H-counter has reset.
See Figure 33 for an example of this restriction.
0
60
100 103
112
HD
1
2
H1
3
1.
2.
3.
4.
HBLKTOG1
HBLKTOG2
CLPOB_TOG1
CLPOB_TOG2
DESIRED
TOGGLE
POSITION
ACTUAL
REGISTER
VALUE
60
100
103
112
(60 – 12) = 48
(100 – 12) = 88
(103 – 12) = 91
(112 – 12) = 100
4
05955-032
CLPOB
Figure 32. Example of Register Setting to Obtain Desired Toggle Positions
VD
H-COUNTER
RESET
HD
NO TOGGLE POSITIONS ALLOWED IN THIS AREA
X
X
X
X
N-12 N-11 N-10
N-9
N-8
N-7
N-6
N-5
N-4
N-3
N-2
N-1
N
0
NOTES
1. TOGGLE POSITIONS CANNOT BE PROGRAMMED WITHIN 12 PIXELS OF PIXEL 0 LOCATION.
Figure 33. Restriction for Toggle Position Placement
Rev. A | Page 24 of 52
1
2
05955-033
H-COUNTER
(PIXEL COUNTER)
AD9974
COMPLETE FIELD—COMBINING H-PATTERNS
H-Pattern Selection
After the H-patterns are created, they combine to create different
readout fields. A field consists of up to nine different regions
determined by the SCP registers. Within each region, a different
H-pattern group can be selected up to a maximum of 32 groups.
Registers to control the H-patterns are located in the field registers.
Table 31 describes the field registers.
The H-patterns are stored in the HPAT memory, as described in
Table 20. The user decides how many H-pattern groups are
required, up to a maximum of 32, and then uses the HPAT_SEL
registers to select which H-pattern group is output in each
region of the field. Figure 34 shows how to use the HPAT_SEL
and SCP registers. The SCP registers create the line boundaries
for each region.
SCP 1
SCP 0
SCP 2
SCP 3
SCP 4
SCP 5
SCP 8
VD
REGION 0
REGION 1
REGION 2
REGION 3
REGION 4
REGION 8
HPAT_SEL0
HPAT_SEL1
HPAT_SEL2
HPAT_SEL3
HPAT_SEL4
HPAT_SEL8
H-PATTERNS
FIELD SETTINGS:
1. SEQUENCE CHANGE POSITIONS (SCP0-8) DEFINE EACH OF THE NINE AVAILABLE REGIONS IN THE FIELD.
2. HPAT_SEL SELECTS THE DESIRED H-PATTERN FOR EACH REGION.
05955-034
HD
Figure 34. Complete Field Divided into Regions
Table 14. Field Registers
Register
SCPx
HPAT_SELx
CLPOB_POL
CLPOB_PAT
CLPOBMASKSTARTx,
CLPOBMASKENDx
PBLK_POL
PBLK_PAT
PBLKMASKSTARTx,
PBLKMASKENDx,
Length (Bits)
13
5
9
9
13
Range
0 to 8191 line number
0 to 31 H-patterns
High/low
0 to 9 patterns
Number of lines
Description
Sequence change position for each region. Selects an individual line.
Selected H-pattern for each region of the field.
CLPOB start polarity settings for each region of the field.
CLPOB pattern selector for each region of the field.
CLPOB mask positions for up to three masking configurations.
9
9
13
High/low
0 to 9 patterns
Number of lines
PBLK start polarity settings for each region of the field.
PBLK pattern selector for each region of the field.
PBLK mask positions for up to three masking configurations.
Rev. A | Page 25 of 52
AD9974
MODE REGISTERS
The mode registers contain registers to select the final field timing
of the AD9974. Typically, all of the field and H-pattern group
information is programmed into the AD9974 at startup. During
operation, the mode registers allow the user to select any combination of field timing to meet the current requirements of the
system. The advantage of using the mode registers in conjunction
with preprogrammed timing is that they greatly reduce the system
programming requirements during camera operation. Only a few
register writes are required when the camera operating mode is
changed, rather than having to write in all of the vertical timing
information with each camera mode change.
A basic still camera application can require five fields of horizontal timing: one for draft mode operation, one for autofocusing,
and three for still image readout. With the AD9974, all of the
register timing information for the five fields is loaded at startup.
Then, during camera operation, the mode registers select
which field timing to activate, depending on how the camera
is being used.
The AD9974 supports up to seven field sequences selected from
up to 31 preprogrammed field groups using the FIELD_SEL
registers. When FIELDNUM is greater than 1, the AD9974 starts
with Field 1 and increments to each Field n at the start of each VD.
Figure 35 provides examples of mode configuration settings.
This example assumes to have four field groups, Field Group 0
to Field Group 3, stored in memory.
Table 15. Mode Registers
Register
HPATNUM
FIELDNUM
FIELD_SEL1
FIELD_SEL2
FIELD_SEL3
FIELD_SEL4
FIELD_SEL5
FIELD_SEL6
FIELD_SEL7
Length (Bits)
5
3
5
5
5
5
5
5
5
Range
0 to 31 H-pattern groups
0 to 7 fields
0 to 31 field groups
0 to 31 field groups
0 to 31 field groups
0 to 31 field groups
0 to 31 field groups
0 to 31 field groups
0 to 31 field groups
Description
Total number of H-pattern groups, starting at Address 0x800.
Total number of applied fields. Set to 1 for single-field operation.
Selected first field.
Selected second field.
Selected third field.
Selected fourth field.
Selected fifth field.
Selected sixth field.
Selected seventh field.
Rev. A | Page 26 of 52
AD9974
H-PATTERN MEMORY
FIELD 0
FIELD 1
FIELD 2
FIELD 3
EXAMPLE 1:
TOTAL FIELDS = 3, FIRST FIELD = FIELD 0, SECOND FIELD = FIELD 1, THIRD FIELD = FIELD 2
FIELD_SEL1 = 0
FIELD 0
FIELD_SEL2 = 1
FIELD 1
FIELD_SEL3 = 2
FIELD 2
EXAMPLE 2:
TOTAL FIELDS = 1, FIRST FIELD = FIELD 3
FIELD_SEL1 = 3
FIELD 3
EXAMPLE 3:
TOTAL FIELDS = 4, FIRST FIELD = FIELD 5, SECOND FIELD = FIELD 1, THIRD FIELD = FIELD 4, FOURTH FIELD = FIELD 2
FIELD 5
FIELD_SEL2 = 1
FIELD 1
FIELD_SEL3 = 4
FIELD 4
FIELD_SEL4 = 2
FIELD 2
05955-035
FIELD_SEL1 = 5
Figure 35. Example of Mode Configurations
Rev. A | Page 27 of 52
AD9974
Figure 36 shows an example of a CCD layout. The horizontal
register contains 28 dummy pixels that occur on each line
clocked from the CCD. In the vertical direction, there are
10 optical black (OB) lines at the front of the readout and two
at the back of the readout. The horizontal direction has four
OB pixels in the front and 48 in the back.
Figure 37 shows the basic sequence layout to use during the
effective pixel readout. The 48 OB pixels at the end of each line
are used for the CLPOB signals. PBLK is optional and is often
used to blank the digital outputs during the HBLK time. HBLK
is used during the vertical shift interval.
The CLPOBMASK registers are also useful for disabling the
CLPOB on a few lines without affecting the setup of the
clamping sequences. It is important to use CLPOB only during
valid OB pixels. During other portions on the frame timing,
such as vertical blanking or SG line timing, the CCD does not
output valid OB pixels. Any CLPOB pulse that occurs during
this time causes errors in clamping operation and, therefore,
changes in the black level of the image.
2 VERTICAL
OB LINES
Because PBLK is used to isolate the CDS input (see the Analog
Front End Description and Operation section), do not use the
PBLK signal during CLPOB operation. The change in the offset
behavior that occurs during PBLK impacts the accuracy of the
CLPOB circuitry.
V
EFFECTIVE IMAGE AREA
10 VERTICAL
OB LINES
The HBLK, CLPOB, and PBLK parameters are programmed
in the V-sequence registers. More elaborate clamping schemes,
such as adding a separate sequence to clamp all the shielded
OB lines, can be used. This requires configuring a separate
V-sequence for clocking out the OB lines.
H
48 OB PIXELS
4 OB PIXELS
HORIZONTAL CCD REGISTER
05955-036
HORIZONTAL TIMING SEQUENCE EXAMPLE
28 DUMMY PIXELS
Figure 36. Example CCD Configuration
OPTICAL BLACK
OB
HD
CCD OUTPUT
VERTICAL SHIFT
DUMMY
EFFECTIVE PIXELS
OPTICAL BLACK
VERT. SHIFT
SHP
SHD
H1/H3
H2/H4
HBLK
PBLK
NOTES
1. PBLK ACTIVE (LOW) SHOULD NOT BE USED DURING CLPOB ACTIVE (LOW).
Figure 37. Horizontal Sequence Example
Rev. A | Page 28 of 52
05955-037
CLPOB
AD9974
ANALOG FRONT END DESCRIPTION AND OPERATION
0.1µF 0.1µF
REFB REFT
0.4V
1.4V
AD9974
DC RESTORE
SHP
PBLK (WHEN DCBYP = 1)
SHP
SHD
0.1µF
CCDIN
DOUT PHASE
INTERNAL
VREF
2V FULL SCALE
6dB ~ 42dB
S11
–3dB, 0dB,
+3dB, +6dB
S22
PBLK
1S1
2S2
CDS GAIN
REGISTER
CLI
VGA GAIN
REGISTER
DAC
PRECISION
TIMING
GENERATION
DOUT
OPTICAL BLACK
CLAMP
DIGITAL
FILTER
CLPOB
14
CLPOB PBLK
IS NORMALLY CLOSED.
IS NORMALLY OPEN.
DOUT
SHP SHD PHASE
OUTPUT
DATA
LATCH
14-BIT
ADC
VGA
CDS
BLANK TO
ZERO OR
CLAMP LEVEL
CLAMP LEVEL
REGISTER
PBLK
VD
V-H
TIMING
GENERATION
HD
05955-038
1.2V
Figure 38. Channel A and Channel B Analog Front End Functional Block Diagram
During the PBLK active time, the ADC outputs can be
programmed to output all 0s or the programmed clamp level.
The AD9974 signal processing chain is shown in Figure 38.
Each processing step is essential for achieving a high quality
image from the raw CCD pixel data.
DC Restore
To reduce the large dc offset of the CCD output signal, a dc
restore circuit is used with an external 0.1 μF series coupling
capacitor. This restores the dc level of the CCD signal to approximately 1.2 V, making it compatible with the 1.8 V core supply
voltage of the AD9974. The dc restore switch is active during
the SHP sample pulse time.
The dc restore circuit can be disabled when the optional PBLK
signal is used to isolate large-signal swings from the CCD input
(see the Analog Preblanking section). Bit 6 of Address 0x00
controls whether the dc restore is active during the PBLK interval.
Analog Preblanking
During certain CCD blanking or substrate clocking intervals,
the CCD input signal to the AD9974 may increase in amplitude
beyond the recommended input range. The PBLK signal can be
used to isolate the CDS input from large-signal swings. As shown
in Figure 38, when PBLK is active (low), the CDS input is isolated
from the CCDIN pin (S1 open) and is internally shorted to
ground (S2 closed).
Note that because the CDS input is shorted during PBLK, the
CLPOB pulse should not be used during the same active time as
the PBLK pulse.
Correlated Double Sampler (CDS)
The CDS circuit samples each CCD pixel twice to extract the
video information and to reject low frequency noise. The timing
shown in Figure 21 illustrates how the two internally generated
CDS clocks, SHP and SHD, are used to sample the reference
level and data level of the CCD signal, respectively. The placement
of the SHP and SHD sampling edges is determined by the setting
of the SHPLOC and SHDLOC register located at Address 0x36.
Placement of these two clock signals is critical for achieving the
best performance from the CCD. The CDS gain is variable in
four steps by using the AFE Register Address 0x04: −3 dB, 0 dB
(default), +3 dB, and +6 dB. Improved noise performance results
from using the +3 dB and +6 dB settings, but the input range is
reduced (see Table 5).
Rev. A | Page 29 of 52
AD9974
(N) SIGNAL SAMPLE
Input Configurations
(N) RESET SAMPLE
The CDS circuit samples each CCD pixel twice to extract the video
information and reject low frequency noise (see Figure 39). There
are three possible configurations for the CDS: inverting CDS mode,
noninverting CDS mode, and SHA mode. The CDSMODE
register (Address 0x00[9:8]) selects which configuration is used.
(N + 1) RESET SAMPLE
VDD
RESET LEVEL
(VRST )
05955-041
SHP
SIGNAL LEVEL
(VFS)
CCDINP
SHA1
Figure 41. Traditional Inverting CDS Signal
DIFF
AMP
CCDINM
CDS_OUT
Table 16. Inverting Voltage Levels
Signal Level
Saturation
Reset
Supply Voltage
05955-039
SHA2
SHD
Figure 39. CDS Block Diagram (Conceptual)
Symbol
VFS
VRST
VDD
Min (mV)
VDD − 500
1600
Typ (mV)
1000
VDD − 300
1800
Max (mV)
1400
VDD
2000
Inverting CDS Mode
Noninverting Input
For this configuration, the signal from the CCD is applied to the
positive input of the (CCDINP) CDS system with the minus
side (CCDINM) grounded (see Figure 40). The CDSMODE
register setting for this configuration is 0x00. Traditional CCD
applications use this configuration with the reset level established
below the AVDD supply level by the AD9974 dc restore circuit,
at approximately 1.5 V. The maximum saturation level is 1.0 V
below the reset level, as shown in Figure 41 and Table 16. A maximum saturation voltage of 1.4 V is also possible when using the
minimum CDS gain setting.
If the noninverting input is desired, the reset (or black) level
signal is established at a voltage above ground potential.
Saturation (or white) level is approximately 1 V. Samples are
taken at each signal level. See Figure 42 and Table 17.
SIGNAL LEVEL
(VFS)
(N) RESET SAMPLE
AD9974
GND
(N) SIGNAL SAMPLE
05955-042
RESET LEVEL
(VRST)
CCDINP
IMAGE
SENSOR
(N + 1) RESET SAMPLE
Figure 42. Noninverting CDS Signal
SHA/
CDS
CCDINM
Figure 40. Single Input CDS Configuration
05955-040
Table 17. Noninverting Voltage Levels
NOTES
1. COUPLING CAPACITOR IS NOT REQUIRED FOR CERTAIN
BLACK LEVEL REFERENCE VOLTAGES.
Signal Level
Saturation
Reset
Rev. A | Page 30 of 52
Symbol
VFS
VRST
Min (mV)
0
Typ (mV)
1000
250
Max (mV)
1400
500
AD9974
SHA Mode—Differential Input Configuration
Referring to Figure 46 and Table 19, the CCDINM signal is a
constant dc voltage set at a level above ground potential. The
sensor signal is applied to the other input, and samples are taken
at the signal minimum and at a point of signal maximum. The
resulting differential signal is the difference between the signal
and the reference voltage.
This configuration uses a differential input sample/hold
amplifier (SHA) (see Figure 43).
AD9974
CCDINP
IMAGE
SENSOR
SHA/
CDS
(N + 1) SIGNAL SAMPLE
(N) SIGNAL SAMPLE
05955-043
CCDINM
Figure 43. SHA Mode—Differential Input Configuration
In this configuration, a signal is applied to the CCDINP input
and, simultaneously, an inverse signal is applied to the CCDINM
input. Sampling occurs on both signals at the same time. This
creates the differential output for amplification and the ADC
(see Figure 44 and Table 18).
PEAK SIGNAL
LEVEL (VFS)
INPUT_POS
BLACK SIGNAL LEVEL (VBLK)
05955-046
INPUT_NEG
MINIMUM SIGNAL LEVEL (VMIN)
GND
(N + 1) SIGNAL SAMPLE
Figure 46. SHA Mode—Single–Ended Input Signal (DC-Coupled)
(N) SIGNAL SAMPLE
Table 19. SHA Mode—Single-Ended Input Voltages
Signal Level
Black
Saturation
Minimum
INPUT_POS
BLACK SIGNAL LEVEL (VBLK)
PEAK SIGNAL
LEVEL (VFS)
INPUT_NEG
Figure 44. SHA Mode—Differential Input Signal
Table 18. SHA Mode—Differential Voltage Levels
Signal Level
Black
Saturation
Minimum
Symbol
VBLK
VFS
VMIN
Min (mV)
1000
0
Typ (mV)
0
VDD − 300
1800
Max (mV)
1400
0
Max (mV)
SHA Timing Control
1400
When SHA mode is selected, only the SHPLOC setting is used
to sample the input signal, but the SHDLOC signal should still
be programmed to an edge setting of SHPLOC + 32.
The SHA mode can also be used in a single-ended fashion, with
the signal from the image sensor applied to the CDS/SHA using
a single input, CCDINP. This is similar to the differential configuration, except in this case, the CCDINM line is held at a constant
dc voltage, establishing a reference level that matches the image
sensor reference voltage (see Figure 45).
AD9974
CCDINP
SHA/
CDS
05955-045
CCDINM
NOTES
1. DC VOLTAGE ABOVE GROUND MAYBE USED TO
MATCH THE SENSOR REFERENCE LEVEL.
Typ (mV)
0
1000
The timing shown in Figure 21 illustrates how the two internally
generated CDS clocks, SHP and SHD, are used to sample the
reference level and data level of the CCD signal, respectively. The
placement of the SHP and SHD sampling edges is determined by
the setting of the SHPLOC and SHDLOC register located at
Address 0x36. Placement of these two clock signals is critical in
achieving the best performance from the CCD.
SHA Mode—DC-Coupled, Single–Ended Input
IMAGE
SENSOR
Min (mV)
CDS Timing Control
05955-044
MINIMUM SIGNAL LEVEL (VMIN)
GND
Symbol
VBLK
VFS
VMIN
Figure 45. SHA Mode—Single–Ended Input Configuration, DC-Coupled
Rev. A | Page 31 of 52
AD9974
Variable Gain Amplifier
Optical Black Clamp
The VGA stage provides a gain range of approximately 6 dB to
42 dB, programmable with 10-bit resolution through the serial
digital interface. A gain of 6 dB is needed to match a 1 V input
signal with the ADC full-scale range of 2 V. When compared to
1 V full-scale systems, the equivalent gain range is 0 dB to 36 dB.
The optical black clamp loop is used to remove residual offsets
in the signal chain and track low frequency variations in the
CCD black level. During the optical black (shielded) pixel interval
on each line, the ADC output is compared with a fixed black
level reference, selected by the user in the clamp level register.
The value can be programmed between 0 LSB and 1023 LSB in
1023 steps.
The VGA gain curve follows a linear-in-dB characteristic. The
exact VGA gain is calculated for any gain register value by
The resulting error signal is filtered to reduce noise, and the
correction value is applied to the ADC input through a DAC.
Normally, the optical black clamp loop is turned on once per
horizontal line, but this loop can be updated more slowly to suit
a particular application. If external digital clamping is used during
postprocessing, the AD9974 optical black clamping can be disabled
using Bit 3 in AFE Register Address 0x00. When the loop is
disabled, the clamp level register can still be used to provide
fixed offset adjustment.
Gain (dB) = (0.0359 × Code) + 5.1 dB
where Code is the range of 0 to 1023.
42
VGA GAIN (dB)
36
30
24
Note that if the CLPOB loop is disabled, higher VGA gain
settings reduce the dynamic range because the uncorrected
offset in the signal path is gained up.
18
6
0
127
255
383
511
639
767
VGA GAIN REGISTER CODE
895
1023
05955-047
12
Figure 47. VGA Gain Curve
ADC
The AD9974 uses a high performance ADC architecture optimized for high speed and low power. Differential nonlinearity
(DNL) performance is typically better than 0.5 LSB. The ADC
uses a 2 V input range. See Figure 5, Figure 6, and Figure 7 for
typical noise performance and linearity plots for the AD9974.
The CLPOB pulse should be aligned with the optical black
pixels of the CCD. It is recommended that the CLPOB pulse
duration be at least 20 pixels wide. Shorter pulse widths can be
used, but the ability of the loop to track low frequency variations
in the black level is reduced. See the Horizontal Clamping and
Blanking section for more timing information.
Digital Data Outputs
The AD9974 digital output data is latched using the DOUTPHASE
register value, as shown in Figure 38. Output data timing is shown
in Figure 22. The switching of the data outputs can couple noise
back into the analog signal path. To minimize any switching
noise while using default SHPLOC and SHDLOC, it is
recommended that the DOUTPHASEP register be set to a value
between 36 and 47. Other settings can produce good results, but
experimentation is necessary.
Rev. A | Page 32 of 52
AD9974
APPLICATIONS INFORMATION
RECOMMENDED POWER-UP SEQUENCE
5.
When the AD9974 is powered up, the following sequence is
recommended (see Figure 48 for each step).
6.
1.
2.
3.
4.
Turn on the power supplies for the AD9974 and apply CLI
clock. There is no required sequence for turning on each
supply.
Although the AD9974 contains an on-chip power-on reset,
a software reset of the internal registers is recommended.
Write 1 to the SW_RST register (Address 0x10) to reset all
the internal registers to their default values. This bit is selfclearing and automatically resets to 0.
Write to the desired registers to configure high speed timing
and horizontal timing. Note that all TESTMODE registers
must be written as described in the Complete Register
Listing section.
To place the part into normal power operation, write 0 to the
STANDBY and REFBUF_PWRDN registers (Address 0x00).
The Precision Timing core must be reset by writing 1 to the
TGCORE_RST register (Address 0x14). This starts the
internal timing core operation.
Write 1 to the OUT_CONTROL register (Address 0x11).
The next VD/HD falling edge allows register updates to occur,
including OUT_CONTROL, which enables all clock outputs.
Additional Restrictions
When operating, note the following restrictions:
•
•
The HD falling edge should be located in the same CLI
clock cycle as the VD falling edge or after the VD falling
edge. The HD falling edge should not be located between
one and five cycles prior to the VD falling edge.
If possible, perform all start-up serial writes with VD and
HD disabled. This prevents unknown behavior caused by
partial updating of registers before all information is loaded.
The internal horizontal counter is reset 12 CLI cycles after the
falling edge of HD. See Figure 49 for details on how the internal
counter is reset.
AD9974 SUPPLIES
POWER
SUPPLIES
0V
1
CLI
(INPUT)
2
3
4
5
6
SERIAL
WRITES
1V
VD
(INPUT)
1ST FIELD
1H
HD
(INPUT)
CLOCKS ACTIVE WHEN OUTCONTROL
REGISTER IS UPDATED AT VD/HD EDGE
H-CLOCKS
H2, H4
H1, H3, RG
Figure 48. Recommended Power-Up Sequence
Rev. A | Page 33 of 52
05955-048
HI-Z BY
DEFAULT
AD9974
Example Register Settings for Power-Up
The following settings can be used for basic operation. A single CLPOB pulse is used with only H-pattern and one field. Additional
HPATS and FIELDS can be added, as needed, along with different CLPOB toggle positions.
010
028
800
801
802
803
804
805
806
807
808
809
80a
80b
80c
80d
80e
80f
810
811
812
813
814
815
816
817
818
819
81a
81b
81c
81d
81e
81f
02a
02b
02c
000
014
011
0000001
0000001
0064000
3ffffff
3ffffff
0064000
3ffffff
3ffffff
0000000
0000000
0000000
0000000
0000000
0000000
00dc05a
3ffffff
3ffffff
3ffffff
1000000
1000800
1000800
1000800
0000800
0000000
0000000
0000000
0000001
1000800
1000800
1000800
0000001
1000800
0000000
0000000
0000001
0000000
0000000
0000008
0000001
0000001
//software reset
//total number of H-Pattern groups = 1
//HPAT0 HBLKTOGO1, TOGO2 settings
//unused HBLK odd toggles set to zero or max value
//unused HBLK odd toggles set to zero or max value
//HPAT0 HBLKTOGE1, TOGE2 settings
//unused HBLK Even toggles set to zero or max value
//unused HBLK Even toggles set to zero or max value
//HBLK StartA, B are not used
//HBLK StartC is not used
//HBLK alternation patterns are not used
//HBLKLEN, HBLKREP not used, HBLK masking pol = 0
//HBLKSTART, end not used
//test, set to zero
//CLPOB pat 0 toggles
//CLPOB pat 1 toggles not used, set to max
//PBLK pat 0 toggles not used, set to max
//PBLK pat 1 toggles not used, set to max
//FIELD0 SCP0, SCP1
//SCP2, SCP3 set same as SCP1
//SCP4, SCP5 set same as SCP1
//SCP6, SCP7 set same as SCP1
//SCP8 set same as SCP1
//select HPAT0 for all regions
//select HPAT0 for all regions
//test, set to zero
//CLPOB start polarity = HIGH
//CLPOB masking set to highest SCP value (no mask)
//CLPOB masking set to highest SCP value (no mask)
//CLPOB masking set to highest SCP value (no mask)
//PBLK start polarity = HIGH
//PBLK masking set to highest SCP value (no mask)
//PBLK masking set to highest SCP value (no mask)
//PBLK masking set to highest SCP value (no mask)
//total number of fields = 1
//field select = FIELD0
//field select = FIELD0
//AFE settings
//reset TGCORE
//enable outputs
Rev. A | Page 34 of 52
AD9974
VD
tVDHD
HD
3ns MIN
CLI
3ns MIN
tCLIDLY
SHD
INTERNAL
HD
INTERNAL
H-COUNTER
RESET
11.5 CYCLES
X X X X X X X X X X X X X X 0 1 2
NOTES
1. EXTERNAL HD FALLING EDGE IS LATCHED BY CLI RISING EDGE, THEN LATCHED AGAIN BY SHD INTERNAL FALLING EDGE.
2. INTERNAL H-COUNTER IS ALWAYS RESET 11.5 CLOCK CYCLES AFTER THE INTERNAL HD FALLING EDGE.
3. DEPENDING ON THE VALUE OF SHDLOC, H-COUNTER RESET CAN OCCUR 12 OR 13 CLI CLOCK EDGES AFTER THE EXTERNAL HD FALLING EDGE.
4. SHPLOC = 0 IS SHOWN IN THE ABOVE EXAMPLE. IN THIS CASE, THE H-COUNTER RESET OCCURS 12 CLI RISING EDGES AFTER HD FALLING EDGE.
5. HD FALLING EDGE SHOULD OCCUR COINCIDENT WITH VD FALLING EDGE (WITHIN SAME CLI CYCLE) OR AFTER VD FALLING EDGE. HD FALLING
EDGE SHOULD NOT OCCUR WITHIN 1 AND 5 CLI CYCLES IMMEDIATELY BEFORE VD FALLING EDGE.
05955-049
H-COUNTER
(PIXEL COUNTER)
Figure 49. Horizontal Counter Pipeline Delay
VD
H-COUNTER
RESET
HD
NO TOGGLE POSITIONS ALLOWED IN THIS AREA
X
X
X
X
X
X
N – 11 N – 10 N – 9 N – 8 N – 7 N – 6 N – 5 N – 4 N – 3 N – 2 N – 1
NOTES
1. TOGGLE POSITIONS CANNOT BE PROGRAMMED WITHIN 12 PIXELS OF PIXEL 0 LOCATION.
N
0
1
2
05955-050
H-COUNTER
(PIXEL COUNTER)
Figure 50. No-Toggle Positions
Additional Restrictions
•
When operating, note the following restrictions:
•
The HD falling edge should be located in the same CLI
clock cycle as the VD falling edge or later than the VD
falling edge. The HD falling edge should not be located
within 1 cycle prior to the VD falling edge.
If possible, perform all start-up serial writes with VD and
HD disabled. This prevents unknown behavior caused by
partial updating of registers before all information is loaded.
The internal horizontal counter is reset 12 CLI cycles after the
falling edge of HD. See Figure 49 for details on how the internal
counter is reset.
Rev. A | Page 35 of 52
AD9974
STANDBY MODE OPERATION
The AD9974 contains two standby modes to optimize the
overall power dissipation in a particular application. Bit 1 and
Bit 0 of Address 0x00 control the power-down state of the device.
STANDBY[1:0] = 00 = normal operation (full power)
STANDBY[1:0] = 01 = reference standby mode
STANDBY[1:0] = 10 or 11 = total shutdown mode
(lowest power)
When returning from total shutdown mode to normal operation,
the timing core must be reset at least 100 μs after the STANDBY
register is written to.
There is an additional register to disable the internal voltage
reference buffer (Address 0x00[2]) independently. By default
the buffer is disabled, but it must be enabled for normal operation.
CLI FREQUENCY CHANGE
Table 20 summarizes the operation of each power-down mode.
The OUT_CONTROL register takes priority over the reference
standby mode in determining the digital output states, but total
shutdown mode takes priority over OUT_CONTROL. Total
shutdown mode has the lowest power consumption.
If the input clock, CLI, is interrupted or changes to a different
frequency, the timing core must be reset for proper operation.
After the CLI clock has settled to the new frequency, or the
previous frequency has resumed, write 0 and then 1 to the
TGCORE_RST register (Address 0x14). This guarantees proper
timing core operation.
Table 20. Standby Mode Operation
I/O Block
AFE
Timing Core
H1
H2
H3
H4
HL
RG
DOUT
Total Shutdown (Default) 1, 2
Off
Off
High-Z
High-Z
High-Z
High-Z
High-Z
High-Z
Low 3
OUT_CONTROL = Low2
No change
No change
Low
High
Low
High
Low
Low
Low
1
Reference Standby
Only REFT, REFB on
On
Low (4.3 mA)
High (4.3 mA)
Low (4.3 mA)
High (4.3 mA)
Low (4.3 mA)
Low (4.3 mA)
Low
To exit total shutdown, write 00 to STANDBY (Address 0x00, Bits[1:0]), then reset the timing core after 100 μs to guarantee proper settling.
Total shutdown mode takes priority over OUT_CONTROL for determining the output polarities.
3
The status of the DOUT pins is unknown at power-up. Low status is guaranteed in total shutdown mode after the power-up sequence is completed.
2
Rev. A | Page 36 of 52
AD9974
CIRCUIT CONFIGURATION
The AD9974 recommended circuit configuration is shown in
Figure 51. Achieving good image quality from the AD9974
requires careful attention to PCB layout. All signals should
be routed to maintain low noise performance. The CCD_A
and CCD_B output signals should be directly routed to Pin A1
and Pin A7, respectively, through a 0.1 μF capacitor. The master
clock, CLI_X, should be carefully routed to Pin A3 and Pin A9
to minimize interference with the CCDIN_X, REFT_X, and
REFB_X signals.
The digital outputs and clock inputs should be connected to the
digital ASIC away from the analog and CCD clock signals. Placing
series resistors close to the digital output pins may help reduce
digital code transition noise. If the digital outputs must drive
a load larger than 20 pF, buffering is recommended to minimize
additional noise. If the digital ASIC can accept gray code, the
outputs of the AD9974 can be selected to output data in gray
code format using Register 0x01[2]. Compared with binary
coding, gray coding helps reduce potential digital transition noise.
The H1_X to H4_X and RG_X traces should have low inductance
to avoid excessive distortion of the signals. Heavier traces are
recommended because of the large transient current demand
on H1_X to H4_X from the capacitive load of the CCD. If possible,
physically locating the AD9974 closer to the CCD reduces the
inductance on these lines. As always, the routing path should be
as direct as possible from the AD9974 to the CCD.
The CLI_X and CCDIN_X PCB traces should be carefully matched
in length and impedance to achieve optimal channel-to-channel
matching performance.
3 V System Compatibility
GROUNDING AND DECOUPLING
RECOMMENDATIONS
As shown in Figure 51, a single ground plane is recommended
for the AD9974. This ground plane needs to be as continuous as
possible, particularly around the P-type, AI-type, and A-type
pins to ensure that all analog decoupling capacitors provide the
lowest possible impedance path between the power and bypass
pins and their respective ground pins. All high frequency
decoupling capacitors need to be located as close as possible to
the package pins.
All the supply pins must be decoupled to ground with good
quality, high frequency chip capacitors. There also needs to be
a 4.7 μF or larger bypass capacitor for each main supply, that is,
AVDD, RGVDD, HVDD, and DRVDD, although this is not
necessary for each individual pin. In most applications, it is
easier to share the supply for RGVDD and HVDD, which can
be done as long as the individual supply pins are separately
bypassed. A separate 3 V supply can be used for DRVDD, but
this supply pin still needs to be decoupled to the same ground
plane as the rest of the chip. A separate ground for DRVSS is not
recommended.
The reference bypass pins (REFT, REFB) must be decoupled to
ground as close as possible to their respective pins. The bridge
capacitor between REFT and REFB is recommended for pixel
rates greater than 40 MHz. The analog input capacitor (CCDINM,
CCDINP) also needs to be located close to the pin.
The GND connections should be tied to the lowest impedance
ground plane on the PCB. Performance does not degrade if
several of these GND connections are left unconnected for
routing purposes.
The AD9974 typical circuit connections for a 3 V system are
shown in Figure 51. This application uses an external 3.3 V
supply connected to the IOVDD input of the AD0074, which
also serves as the LDO input. The LDO generates a 1.8 V output
for the AD9974 core supply voltages, AVDD and DVDD. The
LDOOUT pin can then be connected directly to the AVDD and
DVDD pins. In this configuration, the LDOEN pin is tied high
to enable the LDO.
Alternatively, a separate 1.8 V regulated supply voltage may be
used to power the AVDD and DVDD pins. In this case, the
LDOOUT pin needs to be left floating, and the LDOEN pin
needs to be grounded. A typical circuit configuration for a 1.8 V
system is shown in Figure 51.
Rev. A | Page 37 of 52
AD9974
3V IOVDD SUPPLY
1.8V LDO OUTPUT A
RG_A OUTPUT
RG_A DRIVER SUPPLY
A8
CCDINP_B
AVDD_B
A7
RGVDD_B
CCDINM_B
A6
C9
C10
RGVSS_B
0.1µF
0.1µF
RG_B
A10
B10
GND
GND
LDO_OUT_B
B9
B6
C6
GND
GND
GND
D6
E6
E5
GND
GND
D5
C5
IOVDD_B
0.1µF
0.1µF
0.1µF
IOVDD_A
GND
B5
A5
B3
B4
LDO_OUT_A
0.1µF
RG_A
RGVSS_A
RGVDD_A
C4
A1
C3
CCDINM_A
0.1µF
D9
F7
G1
F8
H1
G7
J1
H7
K1
J7
G2
K7
H2
G8
K2
H8
3V
DRIVER
3V
DRIVER
Figure 51. Recommended Circuit Configuration
Rev. A | Page 38 of 52
CLI_B
COMMON MASTER
CLOCK INPUT
REFB_B
REFT_B
0.1µF
0.1µF
0.1µF
SL_B
SDATA_B
SCK_B
HD_B
VD_B
3
HD COMMON INPUT
VD COMMON INPUT
HVDD_B
HVSS_B 0.1µF
H1B
4
H2B
SERIAL
INTERFACE
FOR CHN B
H1B–H4B
DRIVER
SUPPLY
H1B–H4B
OUTPUTS
H3B
H4B
DVDD_B
DVSS_B
1.8V ANALOG
0.1µF SUPPLY
D0_B (LSB)
D1_B
D2_B
D3_B
D4_B
D5_B
D6_B
D7_B
H9
J9
AVSS_B
D8_B
D9_B
K10
D10_B
J10
H10
D11_B
D12_B
G10
K9
D13_B (MSB)
DRVDD_B
J8
DRVSS_B
F6
G6
GND
GND
GND
GND
D10_A
H6
G9
J6
K8
H3
K6
G3
0.1µF
14
D10
F2
D9_A
CHN A DATA
OUTPUTS
CCDINP_A
NOT DRAWN TO SCALE
F9
F1
J3
D8_A
D3
GND
D7_A
AD9974
K5
D6_A
F10
D4
J5
D5_A
F3
GND
D4_A
E9
GND
D3_A
F4
H5
D2_A
E10
GND
D1_A
E8
E3
G5
D0_A (LSB)
E4
F5
DVSS_A
E7
J2
0.1µF
D8
E2
GND
DVDD_A
1.8V ANALOG
SUPPLY
E1
GND
H4A
C8
DRVSS_A
H3A
B8
D2
0.1µF
HVDD_A
0.1µF HVSS_A
H1A
H1A–H4A 4
OUTPUTS
H2A
4.7µF
C2
4.7µF +
VD_A
C7
K3
HD_A
B2
DRVDD_A
H1A–H4A
DRIVER
SUPPLY
HD COMMON INPUT
VD COMMON INPUT
C1
G4
SCK_A
D7
H4
SDATA_A
1.8V ANALOG
SUPPLY
0.1µF
A9
D12_A
SL_A
CCD SIGNAL_B PLUS
B7
D13_A (MSB)
0.1µF
SERIAL
INTERFACE
FOR CHN A
1.8V LDO OUTPUT B
RG_B OUTPUT
RG_B DRIVER
SUPPLY
D1
J4
0.1µF
REFT_A
A4
D11_A
0.1µF
A2
AVSS_A
REFB_A
B1
CLI_A
COMMON MASTER
CLOCK INPUT
AVDD_A
+
A3
4.7µF
K4
CCD SIGNAL_A PLUS
1.8V ANALOG SUPPLY
3V IOVDD SUPPLY
+
4.7µF
14
CHN B DATA
OUTPUTS
AD9974
3-WIRE SERIAL INTERFACE TIMING
All of the internal registers of the AD9974 are accessed through
a 3-wire serial interface. Each register consists of a 12-bit address
and a 28-bit data-word. Both the 12-bit address and 28-bit dataword are written starting with the LSB. To write to each register,
a 40-bit operation is required, as shown in Figure 52. Although
many registers are fewer than 28 bits wide, all 28 bits must be
written for each register. For example, if the register is only 20 bits
wide, the upper eight bits are don’t cares and must be filled with
0s during the serial write operation. If fewer than 28 data bits
are written, the register is not updated with new data.
Figure 53 shows a more efficient way to write to the registers,
using the AD9974 address auto-increment capability. Using this
method, the lowest desired address is written first, followed by
multiple 28-bit data-words. Each new 28-bit data-word is
automatically written to the next highest register address. By
eliminating the need to write each 12-bit address, faster register
loading is achieved. Continuous write operations can be used,
starting with any register location.
12-BIT ADDRESS
SDATA
A0
A1
A2
A3
A4
A5
A6
A7
tDS
SCK
1
2
3
4
5
A8
28-BIT DATA
A9
A10
A11
D0
D1
D2
D3
D25
D26
D27
tDH
6
7
8
9
10
11
12
13
14
15
16
38
tLS
39
40
tLH
SL
05955-052
NOTES
1. SDATA BITS ARE LATCHED ON SCK RISING EDGES. SCK CAN IDLE HIGH OR LOW BETWEEN WRITE OPERATIONS.
2. ALL 40 BITS MUST BE WRITTEN: 12 BITS FOR ADDRESS AND 28 BITS FOR DATA.
3. IF THE REGISTER LENGTH IS <28 BITS, THEN ZEROS MUST BE USED TO COMPLETE THE 28-BIT DATA LENGTH.
4. NEW DATA VALUES ARE UPDATED IN THE SPECIFIED REGISTER LOCATION AT DIFFERENT TIMES, DEPENDING ON THE
PARTICULAR REGISTER WRITTEN TO. SEE THE UPDATING OF NEW REGISTER VALUES SECTION FOR MORE INFORMATION.
Figure 52. Serial Write Operation
DATA FOR STARTING
REGISTER ADDRESS
SDATA
SCK
A0
1
A1
2
A2
3
A3
4
A10
11
A11
12
D0
13
D1
14
D26
39
DATA FOR NEXT
REGISTER ADDRESS
D27
40
D0
D1
41
42
D26
D27
67
68
D0
69
D1
70
D2
71
NOTES
1. MULTIPLE SEQUENTIAL REGISTERS CAN BE LOADED CONTINUOUSLY.
2. THE FIRST (LOWEST ADDRESS) REGISTER ADDRESS IS WRITTEN, FOLLOWED BY MULTIPLE 28-BIT DATA-WORDS.
3. THE ADDRESS AUTOMATICALLY INCREMENTS WITH EACH 28-BIT DATA-WORD (ALL 28 BITS MUST BE WRITTEN).
4. SL IS HELD LOW UNTIL THE LAST DESIRED REGISTER HAS BEEN LOADED.
Figure 53. Continuous Serial Write Operation
Rev. A | Page 39 of 52
05955-053
SL
AD9974
LAYOUT OF INTERNAL REGISTERS
The AD9974 address space is divided into two register areas, as
shown in Figure 54. In the first area, Address 0x00 to Address 0x72
contain the registers for the AFE, miscellaneous functions, VD/HD
parameters, I/O control, mode control, timing core, and update
control functions. The second area of the address space, beginning
at Address 0x800, consists of the registers for the H-pattern groups
and fields. This is a configurable set of register spaces; the user
can decide how many H-patterns and fields are used in a particular
design. The AD9974 supports the use of up to 32 H-patterns.
Register 0x28 specifies the total number of H-pattern groups.
The starting address for the H-pattern groups is always 0x800.
The starting address for the field registers is determined by the
number of H-pattern groups. Each H-pattern group and field
occupies 16 register addresses.
The starting address for the field registers is based on the
number of H-pattern groups and is equal to 0x800 plus the
number of H-pattern groups times 16.
It is important to note that the H-pattern and field registers
must always occupy a continuous block of addresses.
Figure 55 shows an example when three H-pattern groups and
two fields are used. The starting address for the H-pattern groups
is always 0x800. Because HPATNUM is 3, the H-pattern groups
occupy 48 address locations (that is, 16 registers × 3 H-pattern
groups). The starting address of the field registers for this example
is 0x830 (that is, 0x800 + 48 (decimal)). Note that the decimal value
must be converted to a hex number before adding it to 0x800.
The AD9974 address space contains many unused addresses. Any
undefined addresses between Address 0x00 and Address 0x7FF
should not be written to; otherwise, the AD9974 may operate
incorrectly. Continuous register writes should be performed
carefully so that undefined registers are not written to.
CONFIGURABLE REGISTER AREA
FIXED REGISTER AREA
ADDR 0x000
HPAT START 0x800
AFE REGISTERS
MISCELLANEOUS FUNCTION REGISTERS
VD/HD REGISTERS
H-PATTERN GROUPS
I/O REGISTERS
MODE CONTROL REGISTERS
TIMING CORE REGISTERS
FIELD START
TEST REGISTERS
FIELDS
UPDATE CONTROL REGISTERS
INVALID—DO NOT ACCESS
MAX 0xFFF
NOTES
1. THE H-PATTERN AND FIELD REGISTERS MUST ALWAYS OCCUPY A CONTINUOUS BLOCK OF ADDRESSES.
Figure 54. Layout of AD9974 Registers
ADDR 0x800
3 H-PATTERN GROUPS
(16 × 3 = 48 REGISTERS)
ADDR 0x830
2 FIELDS
(16 × 2 = 32 REGISTERS)
UNUSED MEMORY
MAX 0xFFF
Figure 55. Example Register Configuration
Rev. A | Page 40 of 52
05955-055
ADDR 0x850
05955-054
ADDR 0x7FF
AD9974
•
UPDATING OF REGISTER VALUES
The internal registers of the AD9974 are updated at different
times, depending on the particular register.
Table 21 summarizes the three types of register updates. The
tables in the Complete Register Listing section also contain a
column with update type to identify when each register is updated.
•
SCK Updated—Some of the registers are updated immediately
as the 28th data bit (D27) is written. These registers are
used for functions that do not require gating with the next
VD boundary, such as power-up and reset functions.
•
VD Updated—Many of the registers are updated at the
next VD falling edge. By updating these values at the next
VD edge, the current field is not corrupted, and the new
register values are applied to the next field. The VD update
can be further delayed past the VD falling edge by using
UPDATE Register Address 0x17. This delays the VD-updated
register updates to any HD line in the field. Note that the
field registers are not affected by the UPDATE register.
SCP Updated—All of the H-pattern group registers are
updated at the next SCP when they are used.
Table 21. Register Update Types
Update Type
SCK Updated
VD Updated
SCP Updated
Description
Register is immediately updated when the 28th data bit (D27) is clocked in.
Register is updated at the VD falling edge. VD-updated registers can be delayed further by using the UPDATE register at
Address 0x17. Field registers are not affected by the UPDATE register.
Register is updated at the next SCP when the register is used.
Rev. A | Page 41 of 52
AD9974
COMPLETE REGISTER LISTING
All addresses and default values are expressed in hexadecimal. When an address contains less than 28 data bits, all remaining bits must be
written as 0s. All TESTMODE registers must be set to the specified values.
Table 22. AFE Registers
Address
0x00
0x01
0x02
0x03
0x04
0x05
0x06
0x07
0x08
0x09
0x0A
0x0B
Data Bit
Content
[1:0]
Default
Value
3
[2]
1
REFBUF_PWRDN
[3]
1
CLAMPENABLE
[5:4]
[6]
0
0
TESTMODE
PBLK_LVL
[7]
0
DCBYP
[9:8]
0
CDSMODE
[16:10]
[27:17]
[1:0]
[2]
0
TESTMODE
Unused
TESTMODE
GRAYENCODE
[3]
[4]
[27:5]
[0]
[27:1]
[23:0]
[27:24]
[1:0]
[27:2]
[9:0]
[27:10]
[9:0]
[27:10]
[27:0]
[27:0]
[27:0]
[27:0]
[27:0]
0
0
Update
SCK
SCK
0
1
0
SCK
FFFFFF
SCK
1
VD
F
VD
1EC
VD
0
0
0
0
0
Name
STANDBY
TESTMODE
TESTMODE
Unused
TESTMODE
Unused
TESTMODE
Unused
CDSGAIN
Unused
VGAGAIN
Unused
CLAMPLEVEL
Unused
TESTMODE
TESTMODE
TESTMODE
TESTMODE
TESTMODE
Description
Standby Modes.
0 = normal operation.
1= band gap reference in standby.
2, 3 = total power-down.
Reference Buffer for REFT and REFB Power Control.
0 = REFT/REFB internally driven.
1 = REFT/REFB not driven.
Clamp Enable Control.
0 = disable black clamp.
1 = enable black clamp.
Test Operation Only. Set to 0.
PBLK Level Control.
0 = blank to 0.
1 = blank to clamp level.
DC Restore Circuit Control.
0 = enable dc restore circuit during PBLK.
1 = bypass dc restore circuit during PBLK.
CDS Operation.
0 = normal (inverting) CDS mode.
1 = sample and hold (SHA) mode.
2 = positive CDS mode.
3 = invalid, do not use.
Test Operation Only. Set to 0.
Set unused bits to 0.
Test Operation Only. Set to 0.
Gray Coding ADC Outputs.
0 = disable.
1 = enable.
Test Operation Only. Set to 0.
Test Operation Only. Set to 0.
Set unused bits to 0.
Test Operation Only. Set to 0.
Set unused bits to 0.
Test Operation Only. Set to FFFFFF.
Set unused bits to 0.
CDS Gain Setting.
0 = −3 dB.
1 = 0 dB (default).
2 = +3 dB.
3 = +6 dB.
Set unused bits to 0.
VGA Gain. 6 dB to 42 dB (0.035 dB per step).
Set unused bits to 0.
Optical Black Clamp Level. 0 LSB to 1023 LSB (1 LSB per step).
Set unused registers to 0.
Test Operation Only. Set to 0 if this register is accessed.
Test Operation Only. Set to 0 if this register is accessed.
Test Operation Only. Set to 0 if this register is accessed.
Test Operation Only. Set to 0 if this register is accessed.
Test Operation Only. Set to 0 if this register is accessed.
Rev. A | Page 42 of 52
AD9974
Address
0x0C
0x0D
Data Bit
Content
[27:0]
[0]
Default
Value
0
0
0
0x0E
0x0F
[3:1]
[27:4]
[27:0]
[27:0]
Update
VD
VD
Name
TESTMODE
CLIDIVIDE
SCK
SCK
TESTMODE
Unused
Unused
Unused
Description
Test Operation Only. Set to 0 if this register is accessed.
CLI Divide.
1 = divide CLI input frequency by 2.
Test Operation Only. Set to 0.
Set unused bits to 0.
Set unused register to 0 if accessed.
Set unused register to 0 if accessed.
Table 23. Miscellaneous Registers
Address
0x10
Data Bit
Content
[0]
Default
Value
0
Update
SCK
Name
SW_RST
0x11
[27:1]
[0]
0
VD
Unused
OUT_CONTROL
0
SCK
0
SCK
0
SCK
Unused
TESTMODE
Unused
TESTMODE
Unused
TGCORE_RST
0
SCK
Unused
CLI_BIAS
0
SCK
0
SCK
0x14
[27:1]
[1:0]
[27:2]
[0]
[27:1]
[0]
0x15
[27:1]
[0]
0x12
0x13
0x16
0x17
0x18
0x19
0x1A to
0x1F
[27:1]
[0]
[27:1]
[12:0]
Unused
TESTMODE
Unused
UPDATE
[13]
0
PREVENTUP
[27:14]
[27:0]
[27:0]
[27:0]
0
0
Unused
TESTMODE
TESTMODE
Unused
Description
Software Reset. Bit self-clears to 0 when a reset occurs.
1 = reset Address 0x00 to Address 0xFF to default values.
Set unused bits to 0.
Output Control.
0 = make all outputs dc inactive.
1 = enable outputs at next VD edge.
Set unused bits to 0.
Test Operation Only. Set to 0.
Set unused bits to 0.
Test Operation Only. Set to 0.
Set unused bits to 0.
Timing Core Reset Bar.
0 = hold in reset.
1 = resume operation.
Set unused bits to 0.
Enable bias for CLI input (see Figure 11).
0 = disable bias (CLI input is dc-coupled).
1 = enable bias (CLI input is ac-coupled).
Set unused bits to 0.
Test Operation Only. Set to 0.
Set unused bits to 0.
Serial Interface Update Line. Sets the line (HD) within the field to
update the VD-updated registers. Disabled when PREVENTUP = 1.
Prevents normal update of VD-updated registers.
0 = normal update at VD.
1 = prevent update of VD-updated registers.
Set unused bits to 0.
Test Operation Only. Set to 0 if this register is accessed.
Test Operation Only. Set to 0 if this register is accessed.
Set unused bits to 0.
Table 24. VD/HD Registers
0x21
Data Bit
Content
[0]
[27:1]
[0]
0x22
[2:1]
[27:3]
[27:0]
Address
0x20
Default
Value
0
Update
SCK
0
SCK
0
0
Name
TESTMODE
Unused
VDHDPOL
TESTMODE
Unused
TESTMODE
Description
Test Operation Only. Set to 0.
Set unused bits to 0.
VD/HD Active Polarity.
0 = active low.
1 = active high.
Test Operation Only. Set to 0.
Set unused bits to 0.
Test Operation Only. Set to 0 if this register is accessed.
Rev. A | Page 43 of 52
AD9974
Table 25. I/O Control Registers
Address
0x23
0x24
0x25
0x26
0x27
Data Bit
Content
[0]
[1]
[2]
Default
Value
0
0
0
[3]
[4]
[7:5]
[27:8]
[27:0]
[27:0]
[27:0]
[27:0]
0
0
1
Update
SCK
0
0
0
0
Name
TESTMODE
TESTMODE
IO_NVR
DATA_NVR
TESTMODE
HCLKMODE
Unused
TESTMODE
TESTMODE
TESTMODE
TESTMODE
Description
Test Operation Only. Set to 0.
Test Operation Only. Set to 0.
IOVDD Voltage Range for VD, HD, SCK, SDATA, and SL.
0 = 1.8 V.
1 = 3.3 V.
The I/Os are 3 V tolerant, so there is no problem having higher
than 1.8 V inputs at start-up, but this register should be set to 1
at initialization if using higher than 1.8 V supplies.
DRVDD Voltage Range.
Test Operation Only. Set to 0.
Selects HCLK output configuration (see Table 9).
Set unused bits to 0.
Test Operation Only. Set to 0 if this register is accessed.
Test Operation Only. Set to 0 if this register is accessed.
Test Operation Only. Set to 0 if this register is accessed.
Test Operation Only. Set to 0 if this register is accessed.
Name
HPATNUM
Unused
Unused
FIELDNUM
Unused
FIELD_SEL1
FIELD_SEL2
FIELD_SEL3
FIELD_SEL4
FIELD_SEL5
Unused
FIELD_SEL6
FIELD_SEL7
Unused
Unused
Unused
Unused
Description
Total Number of H-Pattern Groups.
Set unused bits to 0.
Set unused register to 0 if accessed.
Total Number of Fields. Set to 1 for single-field operation.
Set unused bits to 0.
Selected First Field.
Selected Second Field.
Selected Third Field.
Selected Fourth Field.
Selected Fifth Field.
Set unused bits to 0.
Selected Sixth Field.
Selected Seventh Field.
Set unused bits to 0.
Set unused register to 0 if this register is accessed.
Set unused register to 0 if this register is accessed.
Set unused register to 0 if this register is accessed.
Table 26. Mode Registers
Address
0x28
0x29
0x2A
0x2B
0x2C
0x2D
0x2E
0x2F
Data Bit
Content
[4:0]
[27:5]
[27:0]
[2:0]
[27:3]
[4:0]
[9:5]
[14:10]
[19:15]
[24:20]
[27:25]
[4:0]
[9:5]
[27:10]
[27:0]
[27:0]
[27:0]
Default
Value
0
Update
VD
0
VD
0
0
0
0
0
VD
VD
0
0
VD
SCK
SCK
SCK
Rev. A | Page 44 of 52
AD9974
Table 27. Timing Core Registers
Address
0x30
0x31
0x32
0x33
0x34
0x35
Data Bit
Content
[5:0]
[7:6]
[13:8]
[15:14]
[16]
[27:17]
[5:0]
[7:6]
[13:8]
[15:14]
[16]
[27:17]
[5:0]
[7:6]
[13:8]
[15:14]
[16]
[27:17]
[5:0]
[7:6]
[13:8]
[15:14]
[16]
Default
Value
0
Update
SCK
20
0
1
0
SCK
20
0
1
0
SCK
20
0
1
0
SCK
10
0
1
Name
H1POSLOC
Unused
H1NEGLOC
TESTMODE
H1POL
Unused
H2POSLOC
Unused
H2NEGLOC
TESTMODE
H2POL
Unused
TESTMODE
Unused
TESTMODE
TESTMODE
TESTMODE
Unused
RGPOSLOC
Unused
RGNEGLOC
TESTMODE
RGPOL
[27:17]
[0]
0
[1]
[2]
[3]
[7:4]
0
0
0
0
H2BLKRETIME
TESTMODE
TESTMODE
HCLK_WIDTH
1
Unused
H1DRV
[27:8]
[2:0]
[3]
[6:4]
[7]
1
SCK
SCK
Unused
H1BLKRETIME
Unused
H2DRV
Unused
Description
H1 Rising Edge Location.
Set unused bits to 0.
H1 Falling Edge Location.
Test Operation Only. Set to 0.
H1 Polarity Control.
0 = inverse of Figure 21.
1 = no inversion.
Set unused bits to 0.
H2 Rising Edge Location.
Set unused bits to 0.
H2 Falling Edge Location.
Test Operation Only. Set to 0.
H2 Polarity Control.
0 = inverse of Figure 21.
1 = no inversion.
Set unused bits to 0.
Test Operation Only. Set to 0.
Set unused bits to 0.
Test Operation Only. Set to 20.
Test Operation Only. Set to 0.
Test Operation Only. Set to 1.
Set unused bits to 0.
RG Rising Edge Location.
Set unused bits to 0.
RG Falling Edge Location.
Test Operation Only. Set to 0.
RG Polarity Control.
0 = inverse of Figure 21.
1 = no inversion.
Set unused bits to 0.
Retime H1 HBLK to Internal Clock.
0 = no retime.
1 = enable retime.
Recommended setting is enable retime. Enabling retime adds one
cycle delay to programmed HBLK positions.
Retime H2 HBLK to Internal Clock.
Test Operation Only. Set to 0
Test Operation Only. Set to 0
Enables wide H-clocks during HBLK interval.
0 = disable (see Table 13).
Set unused bits to 0.
H1 Drive Strength.
0 = off.
1 = 4.3 mA.
2 = 8.6 mA.
3 = 12.9 mA.
4 = 17.2 mA.
5 = 21.5 mA.
6 = 25.8 mA.
7 = 30.1 mA.
Set unused bits to 0.
H2 Drive Strength.
Set unused bits to 0.
Rev. A | Page 45 of 52
AD9974
Address
0x36
0x37
0x38
0x39
0x3A
0x3B
0x3C
0x3D
Data Bit
Content
[10:8]
[11]
[14:12]
[15]
[18:16]
[19]
[22:20]
[27:23]
[5:0]
[11:6]
[17:12]
[27:18]
[5:0]
[11:6]
Default
Value
1
Update
Unused
Unused
Unused
Unused
Unused
Unused
Unused
Description
H3 Drive Strength.
Set unused bits to 0.
H4 Drive Strength.
Set unused bits to 0.
Test Operation Only. Set to 0.
Set unused bits to 0.
RG Drive Strength.
Set unused bits to 0.
SHD Sampling Edge Location.
SHP Sampling Edge Location.
SHP Width. Controls input dc restore switch active time.
Set unused bits to 0.
DOUT Positive Edge Phase Control.
DOUT Negative Edge Phase Control.
Set DOUTPHASEN = DOUTPHASEP + 0x20.
0 = DCLK tracks DOUT phase.
1 = DCLK is CLI post-Schmitt trigger and post-divider when
CLIDIVIDE = 1.
Data Output Clock Selection.
0 = no delay.
1 = ~4 ns.
2 = ~8 ns.
3 = ~12 ns.
0 = no invert.
1 = invert DCLK to output.
Set unused bits to 0.
Set unused register to 0 if this register is accessed.
Set unused register to 0 if this register is accessed.
Set unused register to 0 if this register is accessed.
Set unused register to 0 if this register is accessed.
Set unused register to 0 if this register is accessed.
Set unused register to 0 if this register is accessed.
SCK
Name
TESTMODE
Unused
Unused
TESTMODE
TESTMODE
Unused
Unused
Description
Test Operation Only. Set to 4B020.
Set unused bits to 0.
Set unused register to 0 if these registers are accessed.
Test Operation Only. Set to F if accessed.
Test Operation Only. Set to 0.
Set unused bits to 0.
Set unused register to 0 if these registers are accessed.
SCK
Unused
Set unused register to 0 if these registers are accessed.
1
1
0
0
20
10
0
20
SCK
SCK
Name
H3DRV
Unused
H4DRV
Unused
TESTMODE
Unused
RGDRV
Unused
SHDLOC
SHPLOC
SHPWIDTH
Unused
DOUTPHASEP
DOUTPHASEN
[12]
0
DCLKMODE
[14:13]
2
CLKDATA_SEL
[15]
0
INV_DCLK
[27:16]
[27:0]
[27:0]
[27:0]
[27:0]
[27:0]
[27:0]
Table 28. Test Registers—Do Not Access
Address
0x3E
0x3F
0x40
0x41 to
0x4F
0x50 to
0x5F
Data Bit
Content
[18:0]
[27:19]
[27:0]
[3:0]
[9:4]
[27:10]
[27:0]
[27:0]
Default
Value
4B020
F
0
Update
SCK
SCK
SCK
Rev. A | Page 46 of 52
AD9974
Table 29. Update Control Registers
Address
0x60
Data Bit
Content
[15:0]
0x61
[27:16]
[15:0]
0x62
[27:16]
[15:0]
0x63
[27:16]
[15:0]
0x66
[27:16]
[15:0]
[27:16]
[15:0]
[27:16]
[15:0]
0x67
[27:16]
[15:0]
0x64
0x65
0x68 to
0x72
Default
Value
1803
Update
SCK
Name
AFE_UPDT_SCK
E7FC
SCK
Unused
AFE_UPDT_VD
F8FD
SCK
Unused
MISC_UPDT_SCK
0702
SCK
Unused
MISC_UPDT_VD
FFF9
SCK
0006
SCK
FFFF
SCK
Unused
VDHD_UPDT_SCK
Unused
VDHD_UPDT_VD
Unused
TGCORE_UPDT_SCK
0000
SCK
Unused
TGCORE_UPDT_VD
SCK
Unused
Unused
[27:16]
[27:0]
Description
Enable SCK update of AFE registers. Each bit corresponds to one
address location.
AFE_UPDT_SCK[0] = 1, update Address 0x00 on SL rising edge.
AFE_UPDT_SCK[1] = 1, update Address 0x01 on SL rising edge.
…
AFE_UPDT_SCK[15] = 1, update Address 0x0F on SL rising edge.
Set unused register = 0 if accessed.
Enable VD update of AFE registers. Each bit corresponds to one
address location.
AFE_UPDT_VD[0] = 1, update Address 0x00 on VD rising edge.
AFE_UPDT_VD[1] = 1, update Address 0x01 on VD rising edge.
…
AFE_UPDT_VD[15] = 1, update Address 0x0F on VD rising edge.
Set unused register to 0 if accessed.
Enable SCK update of miscellaneous registers. Address 0x10 to
Address 0x1F.
Set unused register to 0 if accessed.
Enable VD update of miscellaneous registers, Address 0x10 to
Address 0x1F.
Set unused register to 0 if accessed.
Enable SCK update of VDHD Registers, Address 0x20 to Address 0x22.
Set unused register to 0 if accessed.
Enable VD update of VDHD registers, Address 0x20 to Address 0x22.
Set unused register to 0 if accessed.
Enable SCK update of timing core registers, Address 0x30 to
Address 0x37.
Set unused register to 0 if accessed.
Enable VD update of timing core registers, Address 0x30 to
Address 0x37.
Set unused register to 0 if accessed.
Set unused register to 0 if accessed.
Table 30. HPAT Registers (HPAT Registers Always Start at Address 0x800)
Address
0x00
0x01
0x02
0x03
0x04
Data Bit
Content
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
Default
Value
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Update
SCP
SCP
SCP
SCP
SCP
Name
HBLKTOGO1
HBLKTOGO2
Unused
HBLKTOGO3
HBLKTOGO4
Unused
HBLKTOGO5
HBLKTOGO6
Unused
HBLKTOGE1
HBLKTOGE2
Unused
HBLKTOGE3
HBLKTOGE4
Unused
Description
First HBLK Toggle Position for Odd Lines, or RA0H1REPA/B/C.
Second HBLK Toggle Position for Odd Lines, or RA1H1REPA/B/C.
Set unused bits to 0.
Third HBLK Toggle Position for Odd Lines, or RA2H1REPA/B/C.
Fourth HBLK Toggle Position for Odd Lines, or RA3H1REPA/B/C.
Set unused bits to 0.
Fifth HBLK Toggle Position for Odd Lines, or RA4H1REPA/B/C.
Sixth HBLK Toggle Position for Odd Lines, or RA5H1REPA/B/C.
Set unused bits to 0.
First HBLK Toggle Position for Even Lines, or RA0H2REPA/B/C.
Second HBLK Toggle Position for Even Lines, or RA1H2REPA/B/C.
Set unused bits to 0.
Third HBLK Toggle Position for Even Lines, or RA2H2REPA/B/C.
Fourth HBLK Toggle Position for Even Lines, or RA3H2REPA/B/C.
Set unused bits to 0.
Rev. A | Page 47 of 52
AD9974
Address
0x05
0x06
0x07
0x08
0x09
0xA
0xB
0xC
0xD
0xE
0xF
Data Bit
Content
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
[12:0]
[27:13]
[2:0]
[5:3]
[8:6]
[11:9]
[14:12]
[17:15]
[19:18]
Default
Value
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
[20]
[27:21]
[12:0]
[20:13]
[21]
[22]
[27:23]
[12:0]
[25:13]
[27:26]
[27:0]
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Update
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
Name
HBLKTOGE5
HBLKTOGE6
Unused
HBLKSTARTA
HBLKSTARTB
Unused
HBLKSTARTC
Unused
HBLKALT_PAT1
HBLKALT_PAT2
HBLKALT_PAT3
HBLKALT_PAT4
HBLKALT_PAT5
HBLKALT_PAT6
HBLK_MODE
TESTMODE
Unused
HBLKLEN
HBLKREP
HBLKMASK_H1
HBLKMASK_H2
Unused
HBLKSTART
HBLKEND
Unused
TESTMODE
CLPOB0_TOG1
CLPOB0_TOG2
Unused
CLPOB1_TOG1
CLPOB1_TOG2
Unused
PBLK0_TOG1
PBLK0_TOG2
Unused
PBLK1_TOG1
PBLK1_TOG2
Unused
Description
Fifth HBLK Toggle Position for Even Lines, or RA4H2REPA/B/C.
Sixth HBLK Toggle Position for Even Lines, or RA5H2REPA/B/C.
Set unused bits to 0.
HBLK Repeat Area Start Position A. Used during HBLK Mode 2.
HBLK Repeat Area Start Position B. Used during HBLK Mode 2.
Set unused bits to 0.
HBLK Repeat Area Start Position C. Used during HBLK Mode 2.
Set unused bits to 0.
HBLK Pattern 1 Order. Used during pixel mixing mode.
HBLK Pattern 2 Order. Used during pixel mixing mode.
HBLK Pattern 3 Order. Used during pixel mixing mode.
HBLK Pattern 4 Order. Used during pixel mixing mode.
HBLK Pattern 5 Order. Used during pixel mixing mode.
HBLK Pattern 6 Order. Used during pixel mixing mode.
HBLK Mode Selection.
0 = normal HBLK.
1 = pixel mixing mode.
2 = special pixel mixing mode.
3 = not used.
Test Operation Only. Set to 0.
Set unused bits to 0.
HBLK Length in HBLK Alteration Modes.
Number of HBLK Repetitions in HBLK Alternation Modes.
Masking Polarity for H1/H3 During HBLK.
Masking Polarity for H2/H4 During HBLK.
Set unused bits to 0.
HBLK Start Position Used in Pixel Mixing Modes.
HBLK End Position Used in Pixel Mixing Modes.
Set unused bits to 0.
Test Operation Only. Set to 0.
CLPOB0 Toggle Position 1.
CLPOB0 Toggle Position 2.
Set unused bits to 0.
CLPOB1 Toggle Position 1.
CLPOB1 Toggle Position 2.
Set unused bits to 0.
PBLK0 Toggle Position 1.
PBLK0 Toggle Position 2.
Set unused bits to 0.
PBLK1 Toggle Position 1.
PBLK1 Toggle Position 2.
Set unused bits to 0.
Name
SCP0
SCP1
Unused
SCP2
SCP3
Unused
Description
Sequence Change Position 0.
Sequence Change Position 1.
Set unused bits to 0.
Sequence Change Position 2.
Sequence Change Position 3.
Set unused bits to 0.
Table 31. Field Registers
Address
0x00
0x01
Data Bit
Content
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
Default
Value
X
X
X
X
X
X
Update
VD
VD
Rev. A | Page 48 of 52
AD9974
Address
0x02
0x03
0x04
0x05
0x06
0x07
0x08
0x09
0xA
0xB
0xC
0xD
0xE
0xF
Data Bit
Content
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
[12:0]
[27:13]
[4:0]
[9:5]
[14:10]
[19:15]
[24:20]
[27:25]
[4:0]
[9:5]
[14:10]
[19:15]
[27:20]
[27:0]
[8:0]
[17:9]
Default
Value
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
[27:18]
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
[8:0]
[17:9]
X
X
X
X
X
X
X
X
X
X
X
[27:18]
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
[12:0]
[25:13]
[27:26]
X
X
X
X
X
X
X
X
X
X
Update
VD
VD
VD
VD
VD
VD
VD
VD
VD
VD
VD
VD
VD
VD
Name
SCP4
SCP5
Unused
SCP6
SCP7
Unused
SCP8
Unused
HPAT_SEL0
HPAT_SEL1
HPAT_SEL2
HPAT_SEL3
HPAT_SEL4
Unused
HPAT_SEL5
HPAT_SEL6
HPAT_SEL7
HPAT_SEL8
Unused
Unused
CLPOB_POL
CLPOB_PAT
Unused
CLPOBMASKSTART1
CLOBMASKEND1
Unused
CLPOBMASKSTART2
CLOBMASKEND2
Unused
CLPOBMASKSTART3
CLOBMASKEND3
Unused
PBLK_POL
PBLK_PAT
Unused
PBLKMASKSTART1
PBLKMASKEND1
Unused
PBLKMASKSTART2
PBLKMASKEND2
Unused
PBLKMASKSTART3
PBLKMASKEND3
Unused
Description
Sequence Change Position 4.
Sequence Change Position 5.
Set unused bits to 0.
Sequence Change Position 6.
Sequence Change Position 7.
Set unused bits to 0.
Sequence Change Position 8.
Set unused bits to 0.
Selected H-Pattern for First Region in Field.
Selected H-Pattern for Second Region in Field.
Selected H-Pattern for Third Region in Field.
Selected H-Pattern for Fourth Region in Field.
Selected H-pattern for fifth region in field.
Set unused bits to 0.
Selected H-Pattern for Sixth Region in Field.
Selected H-Pattern for Seventh Region in Field.
Selected H-Pattern for Eighth Region in Field.
Selected H-Pattern for Ninth Region in Field.
Set unused bits to 0.
Set unused bits to 0.
CLPOB Start Polarity Settings.
CLPOB Pattern Selector.
0 = CLPOB0_TOG registers are used.
1 = CLPOB1_TOG registers are used.
Set unused bits to 0.
CLPOB Mask 1 Start Position.
CLPOB Mask 1 End Position.
Set unused bits to 0.
CLPOB Mask 2 Start Position.
CLPOB Mask 2 End Position.
Set unused bits to 0.
CLPOB Mask 3 Start Position.
CLPOB Mask 3 End Position.
Set unused bits to 0.
PBLK Start Polarity Settings for Sequence 0 to Sequence 8.
PBLK Pattern Selector.
0 = PBLK0_TOG registers are used.
1 = PBLK1_TOG registers are used.
Set unused bits to 0
PBLK Mask Region 1 Start Position.
PBLK Mask Region 1 End Position.
Set unused bits to 0.
PBLK Mask Region 2 Start Position.
PBLK Mask Region 2 End Position.
Set unused bits to 0.
PBLK Mask Region 3 Start Position.
PBLK Mask Region 3 End Position.
Set unused bits to 0.
Rev. A | Page 49 of 52
AD9974
OUTLINE DIMENSIONS
A1 CORNER
INDEX AREA
9.10
9.00 SQ
8.90
10 9
8 7
6 5
4 3
2 1
A
B
BALL A1
PAD CORNER
TOP VIEW
C
D
7.20
BSC SQ
E
F
G
H
J
K
0.80 BSC
DETAIL A
1.40 MAX
DETAIL A
0.65 MIN
0.25 MIN
SEATING
PLANE
COPLANARITY
0.12
COMPLIANT TO JEDEC STANDARDS MO-205-AB.
012006-0
0.55
0.50
0.45
BALL DIAMETER
Figure 56. 100-Lead Chip Scale Package Ball Grid Array [CSP_BGA]
(BC-100-1)
Dimensions shown in millimeters
ORDERING GUIDE
Model
AD9974BBCZ 1
AD9974BBCZRL1
1
Temperature Range
−25°C to +85°C
−25°C to +85°C
Package Description
100-Lead Chip Scale Package Ball Grid Array [CSP_BGA]
100-Lead Chip Scale Package Ball Grid Array [CSP_BGA]
Z = RoHS Compliant Part.
Rev. A | Page 50 of 52
Package Option
BC-100-1
BC-100-1
AD9974
NOTES
Rev. A | Page 51 of 52
AD9974
NOTES
©2009 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
D05955-0-10/09(A)
Rev. A | Page 52 of 52