AD9923ABBCZRL [ADI]
CCD Signal Processor with V-Driver and Precision Timing⑩ Generator; CCD信号处理器, V型驱动器和Precision Timing⑩发生器型号: | AD9923ABBCZRL |
厂家: | ADI |
描述: | CCD Signal Processor with V-Driver and Precision Timing⑩ Generator |
文件: | 总88页 (文件大小:1531K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
CCD Signal Processor with V-Driver and
Precision Timing™ Generator
AD9923A
FEATURES
GENERAL DESCRIPTION
Integrated 15-channel V-driver
12-bit, 36 MHz analog-to-digital converter (ADC)
Similar register map to the AD9923
5-field, 10-phase vertical clock support
Complete on-chip timing generator
Precision Timing core with <600 ps resolution
Correlated double sampler (CDS)
The AD9923A is a complete 36 MHz front-end solution for
digital still cameras and other CCD imaging applications.
Similar to the AD9923 product, the AD9923A includes the
analog front end (AFE), a fully programmable timing generator
(TG), and a 15-channel vertical driver (V-driver). A Precision
Timing core allows adjustment of high speed clocks with
approximately 600 ps resolution at 36 MHz operation.
6 dB to 42 dB 10-bit variable gain amplifier (VGA)
Black level clamp with variable level control
On-chip 3 V horizontal and RG drivers
2-phase and 4-phase H-clock modes
Electronic and mechanical shutter support
On-chip driver for external crystal
The on-chip V-driver supports up to 15 channels for use with
5-field, 10-phase CCDs.
The analog front end includes black level clamping, CDS, VGA,
and a 12-bit ADC. The timing generator and V-driver provide
all the necessary CCD clocks: RG, H-clocks, vertical clocks, sensor
gate pulses, substrate clock, and substrate bias control. The
internal registers are programmed using a 3-wire serial
interface.
On-chip sync generator with external sync input
8 mm × 8 mm CSP_BGA package with 0.65 mm pitch
APPLICATIONS
Digital still cameras
Packaged in an 8 mm × 8 mm CSP_BGA, the AD9923A is
specified over an operating temperature range of −25°C to +85°C.
.
FUNCTIONAL BLOCK DIAGRAM
REFT
REFB
–3dB, 0dB, +3dB, +6dB
CDS
+6dB TO +42dB
VGA
AD9923A
VREF
12
12-BIT
ADC
D0 TO D11
DCLK
CCDIN
CLAMP
RG
HL
INTERNAL CLOCKS
HORIZONTAL
DRIVERS
4
H1 TO H4
PRECISION
TIMING
GENERATOR
SL
13
INTERNAL
REGISTERS
SDI
SCK
XV1 TO
XV13
8
V1, V2, V3,
V4, V5A, V5B,
V6, V7A, V7B,
V8, V9, V10,
VERTICAL
TIMING
CONTROL
15
V-DRIVER
SYNC
GENERATOR
XSG1 TO
XSG8
V11, V12, V13
2
XSUBCK,
XSUBCNT
3
SUBCK
VSUB, MSHUT,
STROBE
HD
VD
SYNC CLI
CLO
Figure 1.
Rev. 0
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 registeredtrademarks arethe property of their respective owners.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 781.329.4700
Fax: 781.461.3113
www.analog.com
©2006 Analog Devices, Inc. All rights reserved.
AD9923A
TABLE OF CONTENTS
Features .............................................................................................. 1
Precision Timing High Speed Timing Generation ................. 14
Horizontal Clamping and Blanking......................................... 18
Vertical Timing Generation...................................................... 25
Vertical Timing Example........................................................... 39
Vertical Driver Signal Configuration........................................... 41
Shutter Timing Control............................................................. 45
SUBCK: Normal Operation...................................................... 46
Example of Exposure and Readout of Interlaced Frame....... 53
FG_TRIG Operation.................................................................. 55
Analog Front End Description/Operation ............................. 56
Standby Mode Operation.......................................................... 61
Circuit Layout Information....................................................... 63
Serial Interface Timing.............................................................. 66
Layout of Internal Registers...................................................... 67
Updating New Register Values ................................................. 68
Complete Register Listing......................................................... 69
Outline Dimensions....................................................................... 85
Ordering Guide .......................................................................... 85
Applications....................................................................................... 1
General Description......................................................................... 1
Functional Block Diagram .............................................................. 1
Specifications..................................................................................... 3
Digital Specifications ................................................................... 4
H-Driver Specifications............................................................... 4
Vertical Driver Specifications ..................................................... 4
Analog Specifications................................................................... 5
Timing Specifications .................................................................. 6
Absolute Maximum Ratings............................................................ 7
Package Thermal Characteristics ............................................... 7
ESD Caution.................................................................................. 7
Pin Configuration and Function Descriptions............................. 8
Typical Performance Characteristics ........................................... 10
Equivalent Circuits......................................................................... 11
Terminology .................................................................................... 12
Theory of Operation ...................................................................... 13
REVISION HISTORY
10/06—Revision 0: Initial Version
Rev. 0 | Page 2 of 88
AD9923A
SPECIFICATIONS
Table 1.
Parameter
Conditions/Comments
Min
Typ
Max
Unit
TEMPERATURE RANGE
Operating
Storage
−25
−65
+85
+150 °C
°C
AFETG POWER SUPPLY VOLTAGES
AVDD
AFE analog supply
Timing Core Analog Supply
RG Driver
HL, H1 to H4 Drivers
Data Output Drivers
Digital
2.7
2.7
2.7
2.7
2.7
2.7
3.0
3.0
3.0
3.0
3.0
3.0
3.6
3.6
3.6
3.6
3.6
3.6
V
V
V
V
V
V
TCVDD
RGVDD
HVDD
DRVDD
DVDD
V-DRIVER POWER SUPPLY
VOLTAGES
VDD1, VDD2
VH1, VH2
VL1, VL2
VM1, VM2
VLL
VMM
V-Driver Logic
+2.7
+3.0
+3.6
V
V
V
V
V
V
V-Driver High Supply
V-Driver Low Supply
V-Driver Mid Supply
SUBCK Low Supply
SUBCK Mid Supply
+11.5 +15.0 +16.5
−8.5
−1.5
−8.5
−4.0
−7.5
0.0
−7.5
0.0
−5.5
+1.5
−5.5
+1.5
AFETG POWER DISSIPATION
Total
36 MHz, 3.0 V supply, 400 pF total H-load, 20 pF RG load
335
105
1
mW
mW
mW
mW
mW
mW
mW
mW
mW
mW
Standby 1 Mode
Standby 2 Mode
Standby 3 Mode
Power from HVDD Only1
Power from RGVDD Only
Power from AVDD Only
Power from TCVDD Only
Power from DVDD Only
Power from DRVDD Only
V-DRIVER POWER DISSIPATION2
1
130
10
75
40
75
5
VH1, VH2 = +15 V; VL1, VL2 = −7.5 V; VM1, VM2 = 0 V; VDD1, VDD2 =
3.3 V; all V-driver inputs tied low
VH1, VH2
VL1, VL2
VM1, VM2
VDD1, VDD2
5
2.5
0
mW
mW
mW
mW
MHz
0.5
MAXIMUM CLOCK RATE (CLI)
36
1 The total power dissipated by the HVDD supply can be approximated using the equation
Total HVDD Power = [CLOAD × HVDD × Pixel Frequency] × HVDD
Reducing the H-load and/or using a lower HVDD supply reduces the power dissipation. CLOAD is the total capacitance seen by all H-outputs.
2 V-driver power dissipation depends on the frequency of operation and the load they are driving. All inputs to the V-driver were tied low for the
measurements in Table 1.
Rev. 0 | Page 3 of 88
AD9923A
DIGITAL SPECIFICATIONS
DRVDD = 2.7 V to 3.6 V, CL = 20 pF, TMIN to TMAX, unless otherwise noted.
Table 2.
Parameter
Conditions/Comments
Symbol
Min
Typ
Max
Unit
LOGIC INPUTS
High Level Input Voltage
Low Level Input Voltage
High Level Input Current
Low Level Input Current
Input Capacitance
VIH
VIL
IIH
IIL
CIN
2.1
V
V
μA
μA
pF
0.6
10
10
10
LOGIC OUTPUTS
High Level Output Voltage
Low Level Output Voltage
Powered by DVDD, DRVDD
At IOH = 2 mA
At IOL = 2 mA
VOH
VOL
DVDD − 0.5, DRVDD − 0.5
V
V
0.5
H-DRIVER SPECIFICATIONS
HVDD = RGVDD = 2.7 V to 3.6 V, CL = 20 pF, TMIN to TMAX, unless otherwise noted.
Table 3.
Parameter
Conditions/Comments
RG, HL, and H1 to H4 powered by RGVDD, HVDD
At maximum current
At maximum current
Programmable
Min
Typ Max Unit
RG and H-DRIVER OUTPUTS
High Level Output Voltage
Low Level Output Voltage
Maximum Output Current
Maximum Load Capacitance
RGVDD − 0.5, HVDD − 0.5
V
0.5
V
30
100
mA
pF
For each output
VERTICAL DRIVER SPECIFICATIONS
VDD1 = VDD2 = 3.3 V, VH1 = VH2 = 15 V, VM1 = VM2 = VMM = 0 V, VL1 = VL2 = VLL = −7.5 V, 25°C.
Table 4.
Parameter
Conditions/Comments
Symbol
Min
Typ
Max
Unit
V-DRIVER OUTPUTS
Delay Time
Simplified load conditions, 3000 pF to ground
VL to VM and VM to VH
VM to VL and VH to VM
Rise Time
Rising edges
Falling edges
tPLM, tPMH
tPML, tPHM
35
35
ns
ns
VL to VM
VM to VH
tRLM
tRMH
125
260
ns
ns
Fall Time
VM to VL
VH to VM
tFML
tFHM
220
125
ns
ns
Output Currents
@ −7.25 V
@ −0.25 V
@ +0.25 V
@ +14.75 V
RON
+10
−22
+22
−10
mA
mA
mA
mA
Ω
35
SUBCK OUTPUT
Delay Time
VLL to VH
Simplified load conditions, 1000 pF to ground
tPLH
tPHL
tPLM
25
30
25
ns
ns
ns
VH to VLL
VLL to VMM
Rev. 0 | Page 4 of 88
AD9923A
Parameter
VMM to VH
Conditions/Comments
Symbol
tPMH
Min
Typ
25
Max
Unit
ns
VH to VMM
VMM to VLL
tPHM
tPML
30
25
ns
ns
Rise Time
VLL to VH
VLL to VMM
VMM to VH
tRLH
tRLM
tRMH
40
45
30
ns
ns
ns
Fall Time
VH to VLL
VH to VMM
VMM to VLL
Output Currents
@ −7.25 V
@ −0.25 V
@ +0.25 V
@ +14.75 V
RON
tFHL
tFHM
tFML
40
90
25
ns
ns
ns
20
12
12
20
mA
mA
mA
mA
Ω
35
V-DRIVER
50%
50%
INPUT
tRLM, tRMH, tRLH
tPML
,
tPHM
,
tPHL
90%
90%
V-DRIVER
OUTPUT
tPLM
,
tPMH, tPLH
tFML, tFHM, tFHL
10%
10%
Figure 2. Definition of V-Driver Timing Specifications
ANALOG SPECIFICATIONS
AVDD = 3.0 V, fCLI = 36 MHz, typical timing specifications, TMIN to TMAX, unless otherwise noted.
Table 5.
Parameter
Conditions/Comments
Input characteristics definition1
Min
Typ
Max
Unit
CDS
Allowable CCD Reset Transient
CDS Gain Accuracy
0.5
1.2
V
VGA gain = 6 dB (Code 15, default value)
Default
−3 dB CDS Gain
0 dB CDS Gain
+3 dB CDS Gain
+6 dB CDS Gain
−3
0
+3
+5.5
−2.5
+0.5
+3.5
+6
−2
+1
+4
+6.5
dB
dB
dB
dB
Maximum Input Range Before Saturation
0 dB CDS Gain
−3 dB CDS Gain
Default setting
1.0
1.4
0.5
V p-p
V p-p
V p-p
+6 dB CDS Gain
Maximum CCD Black Pixel Amplitude
0 dB CDS Gain (Default)
+6 dB CDS Gain
Positive offset definition1
−100
−50
+200 mV
+100 mV
VARIABLE GAIN AMPLIFIER (VGA)
Gain Control Resolution
Gain Monotonicity
1024
Guaranteed
Steps
Gain Range
Minimum Gain (VGA Code 15)
Maximum Gain (VGA Code 1023)
6
42
dB
dB
Rev. 0 | Page 5 of 88
AD9923A
Parameter
Conditions/Comments
Min
Typ
Max
Unit
BLACK LEVEL CLAMP
Measured at ADC output
Clamp Level Resolution
Minimum Clamp Level (Code 0)
Maximum Clamp Level (Code 1023)
ANALOG-TO-DIGITAL CONVERTER (ADC)
Resolution
Differential Nonlinearity (DNL)
No Missing Codes
Full-Scale Input Voltage
VOLTAGE REFERENCE
1024
0
255
Steps
LSB
LSB
12
−1.0
Bits
LSB
0.5
Guaranteed
2.0
+1.0
V
Reference Top Voltage (REFT)
Reference Bottom Voltage (REFB)
SYSTEM PERFORMANCE
Gain Accuracy
Low Gain (VGA Code 15)
Maximum Gain (VGA Code 1023)
Peak Nonlinearity, 500 mV Input Signal
Total Output Noise
2.0
1.0
V
V
Includes entire signal chain
Default CDS gain (0 dB)
6.0
42.0
6.5
42.5
0.1
1.0
50
7.0
43.0
dB
dB
%
LSB rms
dB
12 dB gain applied
AC-grounded input, 6 dB gain applied
Measured with step change on supply
Power Supply Rejection (PSR)
1 Input signal characteristics are defined as shown in Figure 3.
1V MAX
INPUT SIGNAL RANGE
(0dB CDS GAIN)
500mV TYP
RESET TRANSIENT
200mV MAX
OPTICAL BLACK PIXEL
Figure 3. Signal Characteristics
TIMING SPECIFICATIONS
CL = 20 pF, AVDD = DVDD = DRVDD = 3.0 V, fCLI = 36 MHz, unless otherwise noted.
Table 6.
Parameter
Conditions/Comments
Symbol Min Typ Max
Unit
MASTER CLOCK, CLI
CLI Clock Period
CLI High/Low Pulse Width
Delay from CLI Rising Edge to Internal Pixel Position 0
AFE CLPOB Pulse Width1, 2
tCONV
27.8
ns
ns
ns
11.2 13.9 16.6
6
2
tCLIDLY
20
Pixels
Allowable Region for HD Falling Edge to CLI Rising Edge
SHP Inhibit Region
AFE SAMPLE LOCATION1
Only valid in slave mode tHDCLI
Only valid in slave mode tSHPINH
4
30
tCONV − 2
39
ns
Edge location
SHP Sample Edge to SHD Sample Edge
DATA OUTPUTS
tS1
11.6 13.9
ns
Output Delay from DCLK Rising Edge1
Inhibited Area for DOUTPHASE Edge Location
Pipeline Delay from SHP/SHD Sampling to Data Output
SERIAL INTERFACE
tOD
8
SHD
16
ns
SHD + 11 Edge location
Cycles
Maximum SCK Frequency
SL to SCK Setup Time
SCK to SL Hold Time
SDATA Valid to SCK Rising Edge Setup
SCK Falling Edge to SDATA Valid Hold
SCK Falling Edge to SDATA Valid Read
fSCLK
tLS
tLH
tDS
tDH
tDV
36
10
10
10
10
10
MHz
ns
ns
ns
ns
ns
1 Parameter is programmable.
2 Minimum CLPOB pulse width is for functional operation only. Wider typical pulses are recommended to achieve good clamp performance.
Rev. 0 | Page 6 of 88
AD9923A
ABSOLUTE MAXIMUM RATINGS
Table 7.
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.
Parameter
To
Rating
AVDD
TCVDD
HVDD
RGVDD
DVDD
DRVDD
AVSS
TCVSS
HVSS
RGVSS
DVSS
DRVSS
−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 +3.9 V
−0.3 V to +3.9 V
VDD1, VDD2
VH1, VH2
VH1, VH2
VL1, VL2
VM1, VM2
VSS1, VSS2 −0.3 V to +6 V
VL1, VL2 −0.3 V to +25 V
VSS1, VSS2 −0.3 V to +17 V
VSS1, VSS2 −17 V to +0.3 V
VSS1, VSS2 −6 V to +6 V
PACKAGE THERMAL CHARACTERISTICS
Thermal Resistance
CSP_BGA package: θJA = 40.3°C/W
ESD CAUTION
VLL
VMM
VSS1, VSS2 −17 V to +0.3 V
VSS1, VSS2 −6 V to + VH
VDR_EN
V1 to V15
VSS1, VSS2 −0.3 V to +6 V
VSS1, VSS2 VL − 0.3 V to VH + 0.3 V
RG Output
H1 to H4 Output
Digital Outputs
Digital Inputs
SCK, SL, SDATA
REFT/REFB, CCDIN
Junction Temperature
Lead Temperature, 10 sec
RGVSS
HVSS
DVSS
DVSS
DVSS
AVSS
−0.3 V to RGVDD + 0.3 V
−0.3 V to HVDD + 0.3 V
−0.3 V to DVDD + 0.3 V
−0.3 V to DVDD + 0.3 V
−0.3 V to DVDD + 0.3 V
−0.3 V to AVDD + 0.3 V
150°C
350°C
Rev. 0 | Page 7 of 88
AD9923A
PIN CONFIGURATION AND FUNCTION DESCRIPTIONS
AD9923A
TOP VIEW
(Not to Scale)
A1 CORNER
INDEX AREA
1
2
3
4
5
6
7
8
9 10 11
A
B
C
D
E
F
G
H
J
K
L
Figure 4. 105-Lead CSPBGA Package Pin Configuration
Table 8. Pin Function Descriptions
Pin No.
A7
Mnemonic
AVDD
AVSS
Type1
P
Description
Analog Supply for AFE.
Analog Ground for AFE.
Analog Supply for Timing Core.
Analog Ground for Timing Core.
Digital Logic Power Supply 1.
Digital Logic Ground 1.
Digital Logic Power Supply 2.
Digital Logic Ground 2.
H1 to H4, HL Driver Supply.
H1 to H4, HL Driver Ground.
RG Driver Supply.
A1, A4, B2, B3, B4, B5, B6, B7
P
B8
B9
E1
F2
K8, L7, L8
K9
D9
D10
B10
A10
L4
L5
J4
K5
TCVDD
TCVSS
DVDD1
DVSS1
DVDD2
DVSS2
HVDD
HVSS
RGVDD
RGVSS
DRVDD
DRVSS
VDD1
VSS1
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
RG Driver Ground.
Data Output Driver Supply.
Data Output Driver Ground.
V-Driver Logic Supply 1.
V-Driver Logic Ground 1.
V-Driver Logic Supply 2.
V-Driver Logic Ground 2.
V-Driver High Supply 1.
V-Driver High Supply 2.
V-Driver Low Supply 1.
V-Driver Low Supply 2.
V-Driver Mid Supply 1.
L10
K10
F9
D1
E9
VDD2
VSS2
VH1
VH2
VL1
P
P
P
P
C1
C9
VL2
VM1
P
D3
F3
E3
A6
A5
A3
A2
C3
C2
B1
G7
E5
VM2
VLL
VMM
CCDIN
CCDGND
REFT
REFB
SL
SCK
SDI
P
P
P
V-Driver Mid Supply 2.
SUBCK Driver Low Supply.
SUBCK Driver Mid Supply.
CCD Signal Input.
AI
AI
AO
AO
DI
DI
DI
DI
DI
CCD Signal Ground.
Voltage Reference Top Bypass.
Voltage Reference Bottom Bypass.
3-Wire Serial Load Pulse.
3-Wire Serial Clock.
3-Wire Serial Data Input.
External System Synchronization Input.
Reset Bar, Active Low Pulse.
SYNC
RSTB
Rev. 0 | Page 8 of 88
AD9923A
Pin No.
A8
A9
F11
E11
D11
C11
B11
C10
K6
Mnemonic
CLI
CLO
H1
H2
H3
H4
HL
RG
VSUB
MSHUT
STROBE
SUBCK
DCLK
D0
D1
D2
D3
D4
Type1
DI
Description
Reference Clock Input (Master Clock).
Clock Output for Crystal.
CCD Horizontal Clock 1.
CCD Horizontal Clock 2.
CCD Horizontal Clock 3.
CCD Horizontal Clock 4.
CCD Last Horizontal Clock.
CCD Reset Gate Clock.
CCD Substrate Bias.
Mechanical Shutter Pulse.
Strobe Pulse.
CCD Substrate Clock (E Shutter).
Data Clock Output.
Data Output (LSB).
Data Output.
Data Output.
Data Output.
Data Output.
Data Output.
Data Output.
Data Output.
Data Output.
Data Output.
Data Output.
Data Output (MSB).
Vertical Sync Pulse. Input in slave mode, output in master mode.
Horizontal Sync Pulse. Input in slave mode, output in master mode.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
CCD Vertical Transfer Clock.
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DO
DIO
DIO
VO3
VO2
VO3
VO2
VO3
VO3
VO2
VO3
VO3
VO2
VO2
VO2
VO3
VO3
VO2
DI
F5
G5
G6
F1
G1
H3
H2
H1
J3
J2
J1
K3
K2
K1
L3
L2
D2
E2
C8
G10
E7
G9
C4
C5
D5
D6
D7
D8
D9
D10
D11
VD
HD
V1
V2
V3
V4
V5A
V5B
V6
V7A
V7B
V8
F10
C6
C7
G11
H11
H10
F6
V9
V10
V11
V12
V13
VDR_EN
TEST0
TEST1
TEST3
NC
F7
E10
K11
J5
J7
J8
V-Driver Output Enable pin.
DI
DI
DI
Test Input. Must be tied to VSS1 or VSS2.
Test Input. Must be tied to VSS1 or VSS2.
Test Input. Must be tied to VDD1 or VDD2.
No Connect.
A11, E6, H9, J6, J9, J10, J11, K4, K7, L1, L6,
L9, L11, G2, G3
1 AI = analog input, AO = analog output, DI = digital input, DO = digital output, DIO = digital input/output, P = power, VO2 = Vertical Driver Output 2 level, VO3 =
Vertical Driver Output 3 level.
Rev. 0 | Page 9 of 88
AD9923A
TYPICAL PERFORMANCE CHARACTERISTICS
450
5
4
400
3.3V
350
3
3.0V
300
2
250
2.7V
1
200
150
100
50
0
–1
–2
–3
0
18
27
36
0
500
1000 1500 2000 2500 3000 3500 4000
CODE
FREQUENCY (MHz)
Figure 5. Power vs. Sample Rate
Figure 7. Typical INL Performance
0.6
55
50
45
40
35
30
25
20
15
10
5
+6dB
0.4
0.2
+3dB
–3dB
0
–0.2
–0.4
–0.6
0dB
0
0
500
1000 1500 2000 2500 3000 3500 4000
CODE
0
100 200 300 400 500 600 700 800 900 1000
GAIN CODE
Figure 6. Typical DNL Performance
Figure 8. Output Noise vs. VGA Gain
Rev. 0 | Page 10 of 88
AD9923A
EQUIVALENT CIRCUITS
HVDD OR RGVDD
RG, HL,
H1 TO H4
AVDD
R
THREE-STATE
OUTPUT
AVSS
AVSS
HVSS OR RGVSS
Figure 12. HL, H1 to H4, and RG Drivers
Figure 9. CCDIN, CCDGND
DVDD
DRVDD
VDVDD
3.5kΩ
DATA
VDR_EN
R
THREE-STATE
D[0:11]
VDVSS
Figure 13. VDR_EN Input
DVSS
Figure 10. Digital Data Outputs
DVDD
DRVSS
330Ω
DVSS
Figure 11. Digital Inputs
Rev. 0 | Page 11 of 88
AD9923A
TERMINOLOGY
Differential Nonlinearity (DNL)
Total Output Noise
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 12-bit resolution indicates that all 4096 codes, respectively,
must be present over all operating conditions.
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)
Integral Nonlinearity (INL)
The deviation of each code measured from a true straight line
between the zero and full-scale values. The point used as zero
scale occurs 0.5 LSB before the first code transition. Positive full
scale is defined as a level 1.5 LSB beyond the last code transition.
The deviation is measured from the middle of each output code
to the true straight line.
where n is the bit resolution of the ADC and 1 LSB is 0.488 mV.
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.
Peak Nonlinearity
Peak nonlinearity, a full signal chain specification, refers to the
peak deviation of the AD9923A output 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 1.5 LSB beyond
the last code transition. The deviation is measured from the
middle of each output code to the true straight line. The error is
expressed as a percentage of the 2 V ADC full-scale signal. The
input signal is always appropriately gained up to fill the full-scale
range of the ADC.
Rev. 0 | Page 12 of 88
AD9923A
THEORY OF OPERATION
Figure 14 shows the typical system block diagram for the
AD9923A in master mode. The CCD output is processed by
the AD9923A AFE circuitry, which consists of a CDS, VGA,
black level clamp, and ADC. The digitized pixel information is
sent to the digital image processor chip that performs the post-
processing and compression. To operate the CCD, CCD timing
parameters are programmed into the AD9923A from the system
microprocessor through the 3-wire serial interface. The AD9923A
generates the CCD horizontal, vertical, and the internal AFE
clocks from the system Master Clock CLI. The CLI is provided
by the image processor or external crystal. External synchroniza-
tion is provided by a sync pulse from the microprocessor, which
resets internal counters and resyncs the VD and HD outputs.
Figure 15 and Figure 16 show the maximum horizontal and
vertical counter dimensions for the AD9923A. Internal hori-
zontal and vertical clocking is controlled by these counters to
specify line and pixel locations. The maximum HD length is
8192 pixels per line, and the maximum VD length is 4096 lines
per field.
V1 TO V13, SUBCK
HL, H1 TO H4, RG, VSUB
D[0:11]
CCDIN
DIGITAL
DCLK
HD, VD
CLI
CCD
AD9923A
AFETG +
V-DRIVER
IMAGE
PROCESSING
ASIC
MSHUT
STROBE
SERIAL
INTERFACE
Alternatively, the AD9923A can be operated in slave mode, in
which the VD and HD are provided externally from the image
processor. In this mode, the AD9923A timing is synchronized
with VD and HD.
MICRO-
PROCESSOR
SYNC
Figure 14. Typical System Block Diagram, Master Mode
MAXIMUM
The H-drivers for HL, H1 to H4, and RG are included in the
AD9923A, allowing these clocks to be directly connected to the
CCD. An H-driver voltage, HVDD, of up to 3.3 V is supported.
An external V-driver is required for the vertical transfer clocks,
the sensor gate pulses, and the substrate clock.
COUNTER
DIMENSIONS
13-BIT HORIZONTAL = 8192 PIXELS MAX
12-BIT VERTICAL = 4096 LINES MAX
The AD9923A also includes programmable MSHUT and
STROBE outputs that can be used to trigger mechanical shutter
and strobe (flash) circuitry.
Figure 15. Vertical and Horizontal Counters
MAX VD LENGTH IS 4096 LINES
VD
MAX HD LENGTH IS 8192 PIXELS
HD
CLI
Figure 16. Maximum VD/HD Dimensions
Rev. 0 | Page 13 of 88
AD9923A
The edge location registers are six bits wide, but there are only
48 valid edge locations available. Therefore, the register values
are mapped into four quadrants, each of which contains 12 edge
locations. Table 10 shows the correct register values for the
corresponding edge locations. Figure 20 shows the default
timing locations for high speed clock signals.
PRECISION TIMING
HIGH SPEED TIMING GENERATION
The AD9923A generates high speed timing signals using the
flexible Precision Timing core. This core is the foundation for
generating the timing used for both the CCD and the AFE. It
consists of the reset gate (RG), horizontal drivers (H1 to H4 and
HL), and sample clocks (SHP and SHD). 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.
H-Driver and RG Outputs
In addition to the programmable timing positions, the
AD9923A features on-chip output drivers for the RG and H1 to
H4 outputs. These drivers are powerful enough to directly drive
the CCD inputs. The H-driver and RG current can be adjusted for
optimum rise/fall times in a particular load by using the H1 to
H4, HL, and RGDRV registers (Address 0x36). The 3-bit drive
setting for each output can be adjusted in 4.1 mA increments,
with the minimum setting of 0 equal to 0 mA or three-state, and
the maximum setting of 7 equal to 30.1 mA.
The high speed timing of the AD9923A operates the same in
master and slave modes. For more information on synchroniza-
tion and pipeline delays, see the Power-Up and Synchronization
in Slave Mode section.
Timing Resolution
The Precision Timing core uses a 1× master clock input (CLI) as
a reference. The frequency of this clock should match the CCD
pixel clock frequency. Figure 17 illustrates how the internal
timing core divides the master clock period into 48 steps, or
edge positions. Using a 36 MHz CLI frequency, the edge
resolution of the Precision Timing core is approximately 0.6 ns.
If a 1× system clock is not available, a 2× reference clock can be
used by programming the CLIDIVIDE register (Address 0x30).
The AD9923A then internally divides the CLI frequency by 2.
As shown in Figure 18, Figure 19, and Figure 20, the H2 and H4
outputs are inverses of H1 and H3 outputs, respectively. The
H1/H2 crossover voltage is approximately 50% of the output
swing. The crossover voltage is not programmable.
Digital Data Outputs
The AD9923A data output and DCLK phase are programmable
using the DOUTPHASE register (Address 0x38, Bits[5:0]). Any
edge from 0 to 47 can be programmed, as shown in Figure 21.
Normally, the DOUT and DCLK signals track in phase, based
on the DOUTPHASE register contents. The DCLK output
phase can also be held fixed with respect to the data outputs by
setting the DCLKMODE register to high (Address 0x38, Bit[8]).
In this mode, the DCLK output remains at a fixed phase equal
to a delayed version of CLI, and the data output phase remains
programmable. For more detail, see the Analog Front End
Description/Operation section.
The AD9923A includes a master clock output (CLO) which is
the inverse of CLI. This output is intended to be used as a
crystal driver. A crystal can be placed between the CLI and
CLO pins to generate the master clock for the AD9923A. For
more information on using a crystal, see Figure 80.
High Speed Clock Programmability
Figure 18 shows how the RG, HL, H1 to H4, SHP, and SHD
high speed clocks are generated. The RG pulse has programmable
rising and falling edges and can be inverted using the polarity
control. The HL, H1, and H3 horizontal clocks have program-
mable rising and falling edges and polarity control. The H2 and
H4 clocks are inverses of the H1 and H3 clocks, respectively.
Table 9 summarizes the high speed timing registers and their
parameters. Figure 19 shows the typical 2-phase, H-clock
operation, in which H3 and H4 are programmed for the same
edge location as H1 and H2.
There is a fixed output delay from the DCLK rising edge to the
DOUT transition, called tOD. This delay can be programmed to
four values between 0 ns and 12 ns, using the DOUTDELAY
register (Address 0x38, Bits[10:9]). The default value is 8 ns.
The pipeline delay through the AD9923A is shown in Figure 22.
After the CCD input is sampled by SHD, there is a 16-cycle
delay before the data is available.
Table 9. Timing Core Register Parameters for HL, H1 to H4, RG, SHP/SHD
Length
(Bits)
Parameter
Polarity
Positive Edge
Negative Edge
Sampling
Location
Range
Description
1
6
6
6
High/low
Polarity control for HL, H1, H3, and RG (0 = no inversion, 1 = inversion)
Positive edge location for HL, H1, H3, and RG (H2/H4 are inverses of H1/H3, respectively)
Negative edge location for HL, H1, H3, and RG (H2/H4 are inverses of H1/H3, respectively)
Sampling location for internal SHP and SHD signals
0 to 47 edge location
0 to 47 edge location
0 to 47 edge location
Drive Strength
3
0 to 7 current steps
Drive current for HL, H1 to H4, and RG outputs (4.1 mA per step)
Rev. 0 | Page 14 of 88
AD9923A
Table 10. Precision Timing Edge Locations
Quadrant
Edge Location (Decimal)
Register Value (Decimal)
0 to 11
16 to 27
32 to 43
48 to 59
Register Value (Binary)
I
II
III
IV
0 to 11
000000 to 001011
010000 to 011011
100000 to 101011
110000 to 111011
12 to 23
24 to 35
36 to 47
POSITION
P[0]
P[12]
P[24]
P[36]
P[48] = P[0]
CLI
tCLIDLY
1 PIXEL
PERIOD
NOTES
1. THE PIXEL CLOCK PERIOD IS DIVIDED INTO 48 POSITIONS, PROVIDING FINE EDGE RESOLUTION FOR HIGH SPEED CLOCK.
2. THERE IS A FIXED DELAY FROM THE CLI INPUT TO THE INTERNAL PIXEL PERIOD POSITION (t = 6ns TYP).
CLIDLY
Figure 17. High Speed Clock Resolution from CLI Master Clock Input
3
4
CCD
SIGNAL
1
5
7
2
RG
HL
6
8
H1
H2
9
10
H3
H4
PROGRAMMABLE CLOCK POSITIONS:
1
RG RISING EDGE.
RG FALLING EDGE.
SHP SAMPLE LOCATION.
SHD SAMPLE LOCATION.
HL RISING EDGE POSITION.
HL FALLING EDGE POSITION.
H1 RISING EDGE POSITION.
H1 FALLING EDGE POSITION (H2 IS INVERSE OF H1).
2
3
4
5
6
7
8
9
H3 RISING EDGE POSITION.
H3 FALLING EDGE POSITION (H4 IS INVERSE OF H3).
10
Figure 18. High Speed Clock Programmable Locations
Rev. 0 | Page 15 of 88
AD9923A
CCD
SIGNAL
RG
HL/H1/H3
H2/H4
NOTES
1. USING THE SAME TOGGLE POSITIONS FOR H1 AND H3 GENERATES STANDARD 2-PHASE H-CLOCKING.
Figure 19. 2-Phase H-Clock Operation
POSITION
P[0]
P[12]
P[24]
P[36]
P[48] = P[0]
PIXEL
PERIOD
RGr[0]
Hr[0]
RGf[12]
RG
Hf[24]
HL/H1/H3
H2/H4
SHP[24]
tS1
CCD
SIGNAL
SHD[48]
NOTES
1. ALL SIGNAL EDGES ARE FULLY PROGRAMMABLE TO ANY OF THE 48 POSITIONS WITHIN ONE PIXEL PERIOD.
2. DEFAULT POSITIONS FOR EACH SIGNAL ARE SHOWN.
Figure 20. High Speed Timing Default Locations
Rev. 0 | Page 16 of 88
AD9923A
P[36]
P[0]
P[48] = P[0]
P[12]
P[24]
PIXEL
PERIOD
DCLK
tOD
DOUT
NOTES
1. DATA OUTPUT (DOUT) AND DCLK PHASE ARE ADJUSTABLE WITH RESPECT TO THE PIXEL PERIOD.
2. WITHIN 1 CLOCK PERIOD, THE DATA TRANSITION CAN BE PROGRAMMED TO 48 DIFFERENT LOCATIONS.
3. OUTPUT DELAY (tOD) FROM DCLK RISING EDGE TO DOUT RISING EDGE IS PROGRAMMABLE.
Figure 21. Digital Output Phase Adjustment
CLI
tCLIDLY
N+12 N+13 N+14 N+15 N+16 N+17
N
N+1
N+4
N+5
N+6
N+7
N+8
N+9 N+10 N+11
N+2
N+3
CCDIN
SAMPLE PIXEL N
SHD
(INTERNAL)
ADC DOUT
(INTERNAL)
N
N–4
N–3
N–2
N–1
N–15 N–14 N–13 N–12 N–11 N–10
tDOUTINH
N–9
N–8
N–7
N–6
N–5
N–17 N–16
DCLK
PIPELINE LATENCY = 16 CYCLES
N–12 N–11 N–10 N–9 N–8 N–7
N
N–4
N–3
N–2
N–1
D[0:11]
N–17 N–16
N–6
N–5
N–14 N–13
N–15
NOTES
1. TIMING VALUES SHOWN ARE SHDLOC = 0, WITH DCLKMODE = 0.
2. HIGHER VALUES OF SHD AND/OR DOUTPHASE SHIFT DOUT TRANSITION TO THE RIGHT WITH RESPECT TO CLI LOCATION.
3. INHIBIT TIME FOR DOUT PHASE IS DEFINED BYtDOUTINH, WHICH IS EQUAL TO SHDLOC PLUS 11 EDGES. IT IS RECOMMENDED THAT
THE 12 EDGE LOCATIONS FOLLOWING SHDLOC NOT BE USED FOR THE DOUTPHASE LOCATION.
4. RECOMMENDED VALUE FOR DOUT PHASE IS TO USE THE SHPLOC EDGE OR THE 11 EDGES FOLLOWING SHPLOC.
5. RECOMMENDED VALUE FOR tOD (DOUT DLY) IS 4ns.
6. THE DOUT LATCH CAN BE BYPASSED USING REGISTER 0x01, BIT [1] = 1 SO THAT THE ADC DATA OUTPUTS APPEAR DIRECTLY AT
THE DATA OUTPUT PINS. THIS CONFIGURATION IS RECOMMENDED IF THE ADJUSTABLE DOUT PHASE IS NOT REQUIRED.
Figure 22. Digital Data Output Pipeline Delay
Rev. 0 | Page 17 of 88
AD9923A
CLPOB and PBLK Masking Area
HORIZONTAL CLAMPING AND BLANKING
The AD9923A allows the CLPOB and/or PBLK signals to be
disabled during certain lines in the field without changing the
existing CLPOB and/or PBLK pattern settings.
The AD9923A horizontal clamping and blanking pulses are
fully programmable to suit a variety of applications. Individual
controls are provided for CLPOB, PBLK, and HBLK during
different regions of each field. This allows 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.
To use CLPOB masking, the CLPMASKSTART and CLPMASKEND
registers are programmed to specify the starting and ending lines
in the field where the CLPOB patterns are ignored. There are three
sets of CLPMASKSTART and CLPMASKEND registers,
allowing up to three CLPOB masking areas to be created.
Individual CLPOB and PBLK Patterns
The AFE horizontal timing consists of CLPOB and PBLK, as
shown in Figure 23. These two signals are independently
programmed using the registers in Table 11. SPOL is the start
polarity for the signal, and TOG1 and TOG2 are the first and
second toggle positions of the pulse. Both signals are active low
and should be programmed accordingly.
CLPOB masking registers are not specific to a given V-sequence;
they are active for any existing field of timing. To disable the
CLPOB masking feature, set these registers to the maximum
value, 0xFFF (default value).
To use PBLK masking, the PBLKMASKSTART and
PBLKMASKEND registers are programmed to specify the
starting and ending lines in the field where the PBLK patterns
are ignored. There are three sets of PBLKMASKSTART and
PBLKMASKEND registers, allowing the creation of up to three
PBLK masking areas.
A separate pattern for CLPOB and PBLK can be programmed
for each V-sequence. As described in the Vertical Timing
Generation section, several V-sequences can be created, each
containing a unique pulse pattern for CLPOB and PBLK.
Figure 46 shows how the sequence change positions divide the
readout field into regions. A different V-sequence can be
assigned to each region, allowing the CLPOB and PBLK signals
to change with each change in the vertical timing. Unused CLPOB
and PBLK toggle positions should be set to 8191.
PBLK masking registers are not specific to a given V-sequence;
they are active for any existing field of timing. To disable the
PBLK masking feature, set these registers to the maximum
value, 0xFFF (default value).
Table 11. CLPOB and PBLK Pattern Registers
Register
Length (Bits) Range
Description
CLPOBPOL
PBLKPOL
1
1
High/low
High/low
Starting polarity of CLPOB for each V-sequence
Starting polarity of PBLK for each V-sequence
CLPOBTOG1
CLPOBTOG2
PBLKTOG1
PBLKBTOG2
CLPMASKSTART
CLPMASKEND
13
13
13
13
12
12
0 to 8191 pixel location First CLPOB toggle position within the line for each V-sequence
0 to 8191 pixel location Second CLPOB toggle position within the line for each V-sequence
0 to 8191 pixel location First PBLK toggle position within the line for each V-sequence
0 to 8191 pixel location Second PBLK toggle position within the line for each V-sequence
0 to 4095 line location
0 to 4095 line location
0 to 4095 line location
0 to 4095 line location
CLPOB masking area—starting line within the field (maximum of three areas)
CLPOB masking area—ending line within the field (maximum of three areas)
PBLK masking area—starting line within the field (maximum of three areas)
PBLK masking area—ending line within the field (maximum of three areas)
PBLKMASKSTART 12
PBLKMASKEND 12
HD
2
3
CLPOB
PBLK
1
ACTIVE
ACTIVE
PROGRAMMABLE SETTINGS:
1
2
3
START POLARITY (CLAMP AND BLANK REGIONS ARE ACTIVE LOW).
FIRST TOGGLE POSITION.
SECOND TOGGLE POSITION.
Figure 23. Clamp and Preblank Pulse Placement
Rev. 0 | Page 18 of 88
AD9923A
NO CLPOB SIGNAL
FOR LINES 6 TO 8
NO CLPOB SIGNAL
FOR LINE 600
VD
0
1
2
597 598
HD
CLPOB
CLPMASKSTART1 = 6
CLPMASKEND1 = 8
CLPMASKSTART2 = CLPMASKEND2 = 600
Figure 24. CLPOB Masking Example
NO PBLK SIGNAL
FOR LINES 6 TO 8
NO PBLK SIGNAL
FOR LINE 703
VD
0
1
2
700 701
HD
PBLK
PBLKMASKSTART1 = 6 PBLKMASKEND1 = 8
PBLKMASKSTART2 = PBLKMASKEND2 = 703
Figure 25. PBLK Masking Example
Rev. 0 | Page 19 of 88
AD9923A
Generating Special HBLK Patterns
Individual HBLK Patterns
There are six toggle positions available for HBLK. Normally,
only two of the toggle positions are used to generate the
standard HBLK interval. However, additional toggle positions
can be used to generate special HBLK patterns, as shown in
Figure 28. 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 can be created.
The HBLK programmable timing shown in Figure 26 is similar
to CLPOB and PBLK; however, there is no start polarity control.
Only the toggle positions are used to designate the start and end
positions of the blanking period. Additionally, there is a polarity
control register, HBLKMASK, that designates the polarity of the
horizontal clock signals during the blanking period. Setting
HBLKMASK high sets H1 = H3 = high and H2 = H4 = low
during blanking, as shown in Figure 27. As with CLPOB and
PBLK registers, HBLK registers are available in each V-sequence,
allowing different blanking signals to be used with different
vertical timing sequences.
Table 12. HBLK Pattern Registers
Length
(Bits)
Register
HBLKMASK
HBLKALT
Range
Description
1
3
High/low
Masking polarity for H1, H3, HL (0 = mask low, 1 = mask high)
Enables different odd/even alternation of HBLK toggle positions
0 to 7 alternation
modes
0: disable alternation (HBLKTOGE1 to HBLKTOGE6 registers are used for each line)
1: TOGE1 and TOGE2 odd lines, TOGE3 to TOGE6 even lines
2: TOGE1 and TOGE2 even lines, TOGE3 to TOGE6 odd lines
3: TOGE1 to TOGE6 even lines, TOGO1 to TOGE6 odd lines (FREEZE/RESUME not
available)
4 to 7: HBLKSTART, HBLKEND, HBLKLEN, and HBLKREP registers are used for each line
HBLK first toggle position (for even lines only when HBLKALT = 3)
HBLKTOGE1 13
HBLKTOGE2 13
HBLKTOGE3 13
HBLKTOGE4 13
HBLKTOGE5 13
HBLKTOGE6 13
0 to 8191 pixel
location
0 to 8191 pixel
location
HBLK second toggle position (for even lines only when HBLKALT = 3)
HBLK third toggle position (for even lines only when HBLKALT = 3)
HBLK fourth toggle position (for even lines only when HBLKALT = 3)
Fifth toggle position, even lines (HBLKSTART when HBLKALT = 4 to 7)
Sixth toggle position, even lines (HBLKEND when HBLKALT = 4 to 7)
0 to 8191 pixel
location
0 to 8191 pixel
location
0 to 8191 pixel
location
0 to 8191 pixel
location
HBLKLEN
HBLKREP
13
8
0 to 8191 pixels
HBLK pattern length, only used when HBLKALT = 4 to 7
0 to 255 repetitions
Number of HBLK pattern repetitions, only used when HBLKALT = 4 to 7
First toggle position for odd lines when HBLKALT = 3 (usually VREPA_3)
HBLKTOGO1 13
HBLKTOGO2 13
HBLKTOGO3 13
HBLKTOGO4 13
HBLKTOGO5 13
HBLKTOGO6 13
0 to 8191 pixel
location
0 to 8191 pixel
location
Second toggle position for odd lines when HBLKALT = 3 (usually VREPA_4)
Third toggle position for odd lines when HBLKALT = 3 (usually FREEZE1)
Fourth toggle position for odd lines when HBLKALT = 3 (usually RESUME1)
Fifth toggle position for odd lines when HBLKALT = 3 (usually FREEZE2)
Sixth toggle position for odd lines when HBLKALT = 3 (usually RESUME2)
0 to 8191 pixel
location
0 to 8191 pixel
location
0 to 8191 pixel
location
0 to 8191 pixel
location
Rev. 0 | Page 20 of 88
AD9923A
HD
HBLKTOGE1
HBLKTOGE2
BLANK
BLANK
HBLK
BASIC HBLK PULSE IS GENERATED USING HBLKTOGE1 AND HBLKTOGE2 REGISTERS (HBLKALT = 0)
Figure 26. Typical Horizontal Blanking (HBLK) Pulse Placement
...
...
HD
HBLK
HL/H1/H3
THE POLARITY OF HL/H1/H3 DURING BLANKING ARE INDEPENDENTLY PROGRAMMABLE
(H2/H4 IS OPPOSITE POLARITY OF H1/H3)
H1/H3
H2/H4
Figure 27. HBLK Masking Polarity Control
HBLKTOGE2
HBLKTOGE4
HBLKTOGE3
HBLKTOGE6
HBLKTOGE5
HBLKTOGE1
HBLK
HL/H1/H3
H2/H4
SPECIAL H-BLANK PATTERN IS CREATED USING MULTIPLE HBLK TOGGLE POSITIONS (HBLKALT = 0)
Figure 28. Using Multiple Toggle Positions for HBLK (HBLKALT = 0)
Rev. 0 | Page 21 of 88
AD9923A
lines. There are also six additional toggle positions, HBLKTOGO1
to HBLKTOGE6, for odd lines. These registers are normally
used for VPAT Group A, VPAT Group B, and freeze/resume
functions, but when HBLKALT = 3, these registers become the
odd line toggle positions for HBLK.
Generating HBLK Line Alternation
The AD9923A can alternate different HBLK toggle positions on
odd and even lines. This feature can be used in conjunction with
V-pattern odd/even alternation, or on its own. When 1 is written
to the HBLKALT register, HBLKTOGE1 and HBLKTOGE2 are
used on odd lines, and HBLKTOGE3 to HBLKTOGE6 are used
on even lines. Writing 2 to the HBLKALT register gives the oppo-
site result: HBLKTOGE1 and HBLKTOGE2 are used on even
lines, and HBLKTOGE3 to HBLKTOGE6 are used on odd lines.
When 3 is written to the HBLKALT register, all six even toggle
positions, HBLKTOGE1 to HBLKTOGE6, are used on even
Another HBLK feature is enabled by writing 4, 5, 6, or 7 to
HBLKALT. In these modes, the HBLK pattern is generated using
a different set of registers—HBLKSTART, HBLKEND, HBLKLEN,
and HBLKREP—along with four toggle positions. This allows
for multiple repeats of the HBLK signal, as shown in Figure 32.
ODD LINE
HD
EVEN LINE
HBLKTOGE4
HBLKTOGE3
HBLKTOGE6
HBLKTOGE5
HBLKTOGE2
HBLKTOGE1
HBLK
HL/H1/H3
H2/H4
ALTERNATING H-BLANK PATTERN USING HBLKALT = 1 MODE
Figure 29. HBLK Odd/Even Alternation Using HBLKALT = 1
ODD LINE
HBLKTOGE4
HBLKTOGE3
EVEN LINE
HD
HBLKTOGE6
HBLKTOGE5
HBLKTOGE2
HBLKTOGE1
HBLK
HL/H1/H3
H2/H4
ALTERNATING H-BLANK PATTERN USING HBLKALT = 2 MODE
Figure 30. HBLK Odd/Even Alternation Using HBLKALT = 2
Rev. 0 | Page 22 of 88
AD9923A
ODD LINE
HBLKTOGO2
HBLKTOGO1
EVEN LINE
HBLKTOGE2
HBLKTOGE1
HD
HBLKTOGO4
HBLKTOGO3
HBLKTOGE4
HBLKTOGE3
HBLK
HL/H1/H3
H2/H4
ALTERNATING H-BLANK PATTERN USING HBLKALT = 3 MODE.
(FREEZE/RESUME FUNCTION NOT AVAILABLE IN THIS MODE.)
Figure 31. HBLK Odd/Even Alternation Using HBLKALT = 3
HBLKTOGE2
HBLKTOGE1 HBLKTOGE3
HBLKTOGE4
HBLKSTART
HBLKEND
HBLK
HBLKLEN
HBLKREP = 3
HL/H1/H3
H2/H4
HBLKREP NUMBER 1
HBLKREP NUMBER 2
HBLKREP NUMBER 3
H-BLANK REPEATING PATTERN IS CREATED USING HBLKLEN AND HBLKREP REGISTERS
Figure 32. HBLK Repeating Pattern Using HBLKALT = 4 to 7
Rev. 0 | Page 23 of 88
AD9923A
be used, such as adding a separate sequence to clamp during the
entire line of OB pixels. This requires configuring a separate
V-sequence for reading the OB lines.
Increasing H-Clock Width During HBLK
The AD9923A allows the H1 to H4 pulse width to be increased
during the HBLK interval. The H-clock pulse width can in-
crease by reducing the H-clock frequency (see Table 13).
The CLPMASKSTART and CLPMASKEND registers can be used
to disable the CLPOB on a few lines without affecting the setup of
the clamp sequences.
The HBLKWIDTH register (Register 0x35, Bits[6:4]) is a 3-bit
register that allows the H-clock frequency to be reduced by 1/2,
1/4, 1/6, 1/8, 1/10, 1/12, or 1/14. The reduced frequency only
occurs for H1 to H4 pulses that are located within the HBLK area.
Horizontal Timing Sequence Example
2 VERTICAL
OB LINES
Figure 33 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 OB pixels in the back.
V
EFFECTIVE IMAGE AREA
10 VERTICAL
OB LINES
Figure 34 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 CLPOB signals. PBLK is optional and it is often
used to blank the digital outputs during the noneffective CCD
pixels. HBLK is used during the vertical shift interval.
H
48 OB PIXELS
4 OB PIXELS
HORIZONTAL CCD REGISTER
28 DUMMY PIXELS
The HBLK, CLPOB, and PBLK parameters are programmed in
the V-sequence registers. More elaborate clamping schemes can
Figure 33. CCD Configuration Example
Table 13. HBLK Width Register
Register
Length (Bits)
Range
Description
HBLKWIDTH
3
1× to 1/14× pixel rate
Controls H1 to H4 width during HBLK as a fraction of pixel rate
0: same frequency as the pixel rate
1: 1/2 pixel frequency, that is, doubles the H1 to H4 pulse width
2: 1/4 pixel frequency
3: 1/6 pixel frequency
4: 1/8 pixel frequency
5: 1/10 pixel frequency
6: 1/12 pixel frequency
7: 1/14 pixel frequency
OPTICAL BLACK
OPTICAL BLACK
HD
CCDIN
VERTICAL SHIFT
DUMMY
EFFECTIVE PIXELS
OPTICAL BLACK
VERT. SHIFT
SHP
SHD
HL/H1/H3
H2/H4
HBLK
PBLK
CLPOB
Figure 34. Horizontal Sequence Example
Rev. 0 | Page 24 of 88
AD9923A
3. Construct the readout for an entire field by dividing the field
into regions and assigning a sequence to each region. Each
field can contain up to nine regions to accommodate different
steps, such as high speed line shifts and unique vertical line
transfers, of the readout. The total number of V-patterns,
V-sequences, and fields are programmable and limited by the
number of registers. High speed line shifts and unique vertical
transfers are examples of the different steps required for
readout.
VERTICAL TIMING GENERATION
The AD9923A provides a very flexible solution for generating
vertical CCD timing; it can support multiple CCDs and different
system architectures. The 13-phase vertical transfer clocks, XV1 to
XV13, are used to shift lines of pixels into the horizontal output
register of the CCD. The AD9923A allows these outputs to be
individually programmed into various readout configurations,
using a four-step process as shown in Figure 35.
1. Use the vertical pattern group registers to create the individual
pulse patterns for XV1 to XV13.
4. Use the MODE register to combine fields in any order for
various readout configurations.
2. Use the V-pattern groups to build the sequences and add more
information.
CREATE THE VERTICAL PATTERN GROUPS,
UP TO FOUR TOGGLE POSITIONS FOR EACH OUTPUT.
BUILD THE V-SEQUENCES BY ADDING START POLARITY,
LINE START POSITION, NUMBER OF REPEATS, ALTERNATION,
GROUP A/B INFORMATION, AND HBLK/CLPOB PULSES.
1
2
XV1
XV2
XV1
XV2
XV3
XV3
VPAT 0
V-SEQUENCE 0
(VPAT0, 1 REP)
XV11
XV12
XV11
XV12
XV1
XV2
XV1
XV2
XV3
V-SEQUENCE 1
(VPAT1, 2 REP)
XV3
VPAT 1
XV11
XV12
XV11
XV12
XV1
XV2
XV3
V-SEQUENCE 2
(VPAT1, N REP)
XV11
XV12
USE THE MODE REGISTER TO CONTROL WHICH FIELDS
ARE USED, AND IN WHAT ORDER (MAXIMUM OF SEVEN
FIELDS MAY BE COMBINED IN ANY ORDER).
BUILD EACH FIELD BY DIVIDING IT INTO DIFFERENT
REGIONS AND ASSIGNING A V-SEQUENCE TO EACH
(MAXIMUM OF NINE REGIONS IN EACH FIELD).
3
4
FIELD 0
FIELD 1
FIELD 4
FIELD 2
FIELD 0
REGION 0: USE V-SEQUENCE 2
REGION 1: USE V-SEQUENCE 0
REGION 2: USE V-SEQUENCE 3
FIELD 3
FIELD 5
REGION 3: USE V-SEQUENCE 0
FIELD 2
FIELD 1
FIELD 4
REGION 4: USE V-SEQUENCE 2
FIELD 1
FIELD 2
Figure 35. Summary of Vertical Timing Generation
Rev. 0 | Page 25 of 88
AD9923A
start polarity for each signal, VSTART specifies the start
position of the VPAT group, and VLEN designates the total
length of the VPAT group, which determines the number of
pixels between each pattern repetition, if repetitions are used.
Vertical Pattern (VPAT) Groups
A vertical pattern (VPAT) group defines the individual pulse
pattern for each XV1 to XV13 output signal. Table 14 summarizes
the registers that are available for generating each VPAT group.
The first, second, third, fourth, fifth, and sixth toggle positions
(XVTOG1, XVTOG2, XVTOG3, XVTOG4, XVTOG5,
XVTOG6) are the pixel locations where the pulse transitions. All
toggle positions are 13-bit values that can be placed anywhere in
the horizontal line.
To achieve the best possible noise performance, ensure that
VSTART + VLEN < the end of the H-blank region.
Toggle positions programmed to either Pixel 0 or Pixel 8191 are
ignored. The toggle positions of unused XV-channels must be
programmed to either Pixel 0 or Pixel 8191. This prevents unpre-
dictable behavior because the default values of the V-pattern
group registers are unknown.
More registers are included in the vertical sequence registers to
specify the output pulses: XV1POL to XV13POL specifies the
Table 14. Vertical Pattern Group Registers
Register
XVTOG1
XVTOG2
XVTOG3
XVTOG4
XVTOG5
XVTOG6
Length (Bits)
Range
Description
13
13
13
13
13
13
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
First toggle position within line for each XV1 to XV12 output
Second toggle position
Third toggle position
Fourth toggle position
Fifth toggle position
Sixth toggle position
START POSITION OF VERTICAL PATTERN GROUP IS PROGRAMMABLE IN VERTICAL SEQUENCE REGISTERS.
HD
4
XV1
XV2
1
1
2
3
2
3
1
XV12
2
3
PROGRAMMABLE SETTINGS:
1
START POLARITY (LOCATED IN V-SEQUENCE REGISTERS).
FIRST TOGGLE POSITION.
SECOND TOGGLE POSITION (A TOTAL OF SIX TOGGLE POSITIONS ALSO AVAILABLE FOR MORE COMPLEX PATTERNS).
TOTAL PATTERN LENGTH FOR ALL VERTICAL OUTPUTS (LOCATED IN VERTICAL SEQUENCE REGISTERS).
2
3
4
Figure 36. Vertical Pattern Group Programmability
Rev. 0 | Page 26 of 88
AD9923A
Generating Line Alternation for V-Sequences and HBLK
Vertical Sequences (VSEQ)
section). The VSTARTA and VSTARTB registers specify the pixel
location where the V-pattern group starts. The VMASK register is
used in conjunction with the FREEZE/RESUME registers to enable
optional masking of the XV outputs. Either or both of the
FREEZE1/RESUME1 and FREEZE2/RESUME2 registers can be
enabled.
A vertical sequence (VSEQ) is created by selecting one of the
V-pattern groups and adding repeats, a start position, and
horizontal clamping and blanking information. Each VSEQ is
programmed using the registers shown in Table 15. Figure 37
shows how each register is used to generate a V-sequence.
The VPATSELA and VPATSELB registers select the V-pattern
group that is used in a given V-sequence. Having two groups
available allows each vertical output to be mapped to a different
V-pattern group. The selected V-pattern group can have
repetitions added for high speed line shifts or line binning by
using the VREP registers for odd and even lines. Generally, the
same number of repetitions is programmed into both registers.
If a different number of repetitions is required on odd and even
lines, separate values can be used for each register (see the
The line length (in pixels) is programmable using the HDLEN
registers. Each V-sequence can have a different line length to
accommodate various image readout techniques. The maximum
number of pixels per line is 8192. Note that the last line of the
field can be programmed separately using the HDLAST register,
located in the field register (see Table 16).
1
HD
2
3
4
4
XV1 TO XV13
VREP 3
V-PATTERN GROUP
VREP 2
5
CLPOB
PBLK
6
HBLK
PROGRAMMABLE SETTINGS FOR EACH VERTICAL SEQUENCE:
1
START POSITION IN THE LINE OF SELECTED V-PATTERN GROUP.
2
HD LINE LENGTH.
3
V-PATTERN SELECT (VPATSEL) TO SELECT ANY V-PATTERN GROUP.
NUMBER OF REPETITIONS OF THE V-PATTERN GROUP (IF NEEDED).
START POLARITY AND TOGGLE POSITIONS FOR CLPOB AND PBLK SIGNALS.
MASKING POLARITY AND TOGGLE POSITIONS FOR HBLK SIGNAL.
4
5
6
Figure 37. V-Sequence Programmability
Rev. 0 | Page 27 of 88
AD9923A
1
Table 15. V-Sequence Registers
Length
(Bits)
Register
HOLD
Range
Description
1
2
On/off
Use in conjunction with VMASK. 1 = hold instead of FREEZE/RESUME.
VMASK
0 to 3 mask mode
Enables the masking of XV1 to XV13 outputs at the locations specified by
the FREEZE/RESUME registers.
0 = no mask.
1 = enable FREEZE1/RESUME1.
2 = enable FREEZE2/RESUME2.
3 = enable both FREEZE1/RESUME1 and FREEZE2/RESUME2.
HD line length in each V-sequence.
Start polarity for each XV1 to XV13 output.
HDLEN
13
1
0 to 8191 pixels
High/low
XV1POL to
XV13POL
GROUPSEL
12
1b for each XV output
Assigns each XV1 to XV13 output to either V-Pattern Group A or
V-Pattern Group B.
0 = assigns to VPATSELA.
1 = assigns to VPATSELB.
TWO_GROUP
VPATSELA
1
5
5
High/low
When high, all XV outputs combine Group A and Group B.
Selected V-pattern for Group A.
0 to 31 V-pattern number
0 to 31 V-pattern number
VPATSELB
Selected V-pattern for Group B. If SPVTP_ENABLE = 1, VPATSELB is used
for second VTP inserted in SPVTP_ACTLINE.
VPATA_MODE
2
0 to 3 repetition mode
Selects alternation repetition mode for Group A only.
0 = disable alternation, use VREPA_1 for all lines.
1 = 2-line. Alternate VREPA_1 and VREPA_2 (same as odd/even).
2 = 3-line. Alternate VREPA_1, VREPA_2, and VREPA_3.
3 = 4-line. Alternate VREPA_1, VREPA_2, VREPA_3, and VREPA_4.
Start position for the selected V-Pattern Group A.
VSTARTA
VSTARTB
13
13
0 to 8191 pixel location
0 to 8191 pixel location
Start position for the selected V-Pattern Group B. If SPVTP_ENABLE = 1,
VSTARTB is used for start position of VPATSELB in SPVTP_ACTLINE.
VLENA
13
13
12
0 to 8191 pixels
0 to 8191 pixels
0 to 4095 repeats
Length of selected V-Pattern Group A.
Length of selected V-Pattern Group B.
VLENB
VREPB_ODD
Number of repetitions for the V-Pattern Group B for odd lines.
If no alternation is required for Group B, set VREPB_ODD equal to
VREPB_EVEN.
VREPB_EVEN
12
0 to 4095 repeats
Number of repetitions for the V-Pattern Group B for even lines.
If no alternation is required for Group B, set VREPB_EVEN equal to
VREPB_ODD.
VREPA_1
12
12
12
12
13
13
13
13
12
1
0 to 4095 repeats
Number of repetitions for the V-Pattern Group A for first lines (odd).
Number of repetitions for the V-Pattern Group A for second lines (even).
Number of repetitions for the V-Pattern Group A for third lines.
Number of repetitions for the V-Pattern Group A for fourth lines.
Pixel location where the XV outputs freeze or hold (see VMASK).
Pixel location where the XV outputs resume operation (see VMASK).
Pixel location where the XV outputs freeze or hold (see VMASK).
Pixel location where the XV outputs resume operation (see VMASK).
Active line for second VTP insertion.
VREPA_2
0 to 4095 repeats
VREPA_3
0 to 4095 repeats
VREPA_4
0 to 4095 repeats
FREEZE1
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 8191 pixel location
0 to 4095 line location
High/low
RESUME1
FREEZE2
RESUME2
SPVTP_ACTLINE
SPVTP_ENABLE
When high, second VTP is inserted into SPVTP_ACTLINE.
1 See Table 11 and Table 12 for CLPOB, PBLK, and HBLK registers.
Rev. 0 | Page 28 of 88
AD9923A
If Group B flexibility is needed, the outputs set to 1 in the
Group A/Group B Selection
GROUPSEL register use the V-pattern group selected by the
VPATSELB register. For example, Figure 38 shows outputs
XV12 and XV13 using a separate V-Pattern Group B to perform
special CCD timing.
The AD9923A has the flexibility to use two V-pattern groups in
a vertical sequence. In general, all vertical outputs use the same
V-pattern group during a sequence, but some outputs can be
assigned to a different V-pattern group. This is useful during
certain CCD readout modes.
Another application of the Group A and Group B registers is to
combine two VPAT groups for more complex patterns. This is
achieved by setting the TWO_GROUP register to 1. Figure 39
shows an example of this timing. When TWO_GROUP = 1,
the Group A and Group B toggle positions are both used. In
addition, length, starting polarity, and number of repetitions
are all determined by the appropriate registers for Group A
when TWO_GROUP = 1. Figure 40 shows the more complex
operation of combining Group A and Group B with repetition.
The GROUPSEL register is used to select Group A or Group B
for each XV output (the LSB is XV1, the MSB is XV13). Setting
each bit to 0 selects Group A; setting each bit to 1 selects Group B.
If only a single V-pattern group is needed for the vertical
outputs, Group A is used by default (GROUPSEL = 0), and the
outputs use the V-pattern group specified by the VPATSELA
register.
HD
OPTIONAL HOLD AREA
FOR GROUP A
XV1 TO XV11 USE
V-PATTERN GROUP A
XV1
XV11
XV12, XV13 USE
V-PATTERN GROUP B
XV12
XV13
Figure 38. Using Separate Group A and Group B Patterns
HD
V-PATTERN GROUP A
V-PATTERN GROUP B
XV1
XV13
Figure 39. Combining Group A and Group B Patterns
HD
V-PATTERN GROUP A V-PATTERN GROUP B
XV1
XV13
GROUP A REP 1
GROUP A REP 2
GROUP A REP 3
Figure 40. Combining Group A and Group B Patterns, with Repetition
Rev. 0 | Page 29 of 88
AD9923A
registers. Group A can also support three-line and four-line
alternation by using the VREPA_3 and VREPA_4 registers.
Generating Line Alternation for V-Sequences and HBLK
During low resolution readout, some CCDs require a different
number of vertical clocks on alternate lines. The AD9923A can
support such CCDs by using different VREP registers. This
allows a different number of VPAT repetitions to be programmed
on odd and even lines.
Additionally, the HBLK signal can be alternated for odd and
even lines. When the HBLKALT = 1, the HBLKTOGE1 and
HBLKTOGE2 positions are used on odd lines, and the
HBLKTOGE3 to HBLKTOGE6 positions are used on even
lines. This allows the HBLK interval to be adjusted on odd
and even lines if needed.
Note that only the number of repeats is different in odd and
even lines, but the VPAT group remains the same. There are
separate controls for the assigned Group A and Group B patterns.
Both Group A and Group B can support odd and even line
alternation. Group A uses the VREPA_1 and VREPA_2 regis-
ters; Group B uses the VREPB_ODD and VREPB_EVEN
Figure 41 shows an example of simultaneous VPAT repetition
alternation and HBLK alternation. Both types of alternation can
be used separately.
HD
VREPA_1 = 2
VREPA_2 = 5
VREPA_1 = 2
(OR VREPB_ODD = 2)
(OR VREPB_EVEN = 5)
(OR VREPB_ODD = 2)
XV1
XV2
XV13
HBLKTOGE1
HBLKTOGE2
HBLKTOGE3
HBLKTOGE4
HBLKTOGE1
HBLKTOGE2
HBLK
NOTES
1. THE NUMBER OF REPEATS FOR V-PATTERN GROUP A OR GROUP B CAN BE ALTERNATED ON ODD AND EVEN LINES.
2. GROUP A ALSO SUPPORTS 3-LINE AND 4-LINE ALTERNATION USING THE ADDITIONAL VREPA_3 AND VREPA_4 REGISTERS.
3. THE HBLK TOGGLE POSITIONS CAN ALSO BE ALTERNATED BETWEEN ODD AND EVEN LINES TO GENERATE DIFFERENT HBLK PATTERNS
FOR ODD/EVEN LINES. SEE THE HORIZONTAL CLAMPING AND BLANKING SECTION FOR MORE INFORMATION ON HBLK.
Figure 41. Odd/Even Line Alternation of VPAT Repetitions and HBLK Toggle Positions
Rev. 0 | Page 30 of 88
AD9923A
which point the signals continue with any remaining toggle
positions.
Masking Using Freeze/Resume Registers
As shown in Figure 42 and Figure 43, the FREEZE/RESUME
registers are used to temporarily mask the XV outputs. The
pixel locations to start (FREEZE) and end (RESUME) the
masking create an area in which the vertical toggle positions are
ignored. At the pixel location specified in the FREEZE register,
the XV outputs are held static at their current dc state, high or
low. The XV outputs are held until the internal pixel counter
reaches the pixel location specified by the RESUME register, at
Two sets of FREEZE/RESUME registers are provided, allowing
the vertical outputs to be interrupted twice in the same line. The
FREEZE and RESUME positions are enabled using the VMASK
register.
It is not recommended to use FREEZE/RESUME at the same
time as the SWEEP function.
HD
XV1
NO MASKING AREA
XV13
Figure 42. Not Using FREEZE/RESUME
MASKING AREA
FOR GROUP A
HD
XV1
FREEZE
RESUME
XV13
NOTES
1. ALL TOGGLE POSITIONS WITHIN THE FREEZE-RESUME MASKING AREA ARE IGNORED. H-COUNTER CONTINUES TO COUNT DURING MASKING.
2. TWO SEPARATE MASKING AREAS ARE AVAILABLE FOR EACH GROUP A, USING FREEZE1/RESUME1 AND FREEZE2/RESUME2 REGISTERS.
Figure 43. Using FREEZE/RESUME
Rev. 0 | Page 31 of 88
AD9923A
because the XV outputs continue from where they stopped (as
opposed to having the pixel counter run continuously), with
any toggle positions that fall between the FREEZE and
RESUME locations being ignored. Signals assigned to Group B
are not affected by the hold area.
Hold Area Using FREEZE/RESUME Registers
The FREEZE/RESUME registers can also be used to create a
hold area, in which the XV outputs are temporarily held and
then later resume at the point where they were held. As shown
in Figure 44, this is different than using the VMASK register,
HD
HOLD AREA
FOR GROUP A
FREEZE
RESUME
XV1
XV11
XV12
XV13
NOTES
1. WHEN HOLD = 1 FOR ANY V-SEQUENCE, THE FREEZE AND RESUME REGISTERS ARE USED TO SPECIFY THE HOLD AREA FOR GROUP A.
2. ABOVE EXAMPLE: ALL XV-OUTPUTS ARE ASSIGNED TO GROUP A.
3. H-COUNTER FOR GROUP A (XV1 TO XV13) STOPS DURING HOLD AREA.
Figure 44. Hold Area for Group A
HD
XV1
HOLD AREA
FOR GROUP A
FREEZE
RESUME
XV11
XV12
XV13
NO HOLD
AREA FOR
GROUP B
NOTES
1. ABOVE EXAMPLES: XV12 AND XV13 ARE ASSIGNED TO GROUP B.
2. GROUP B DOES NOT USE HOLD AREA.
Figure 45. Group B Does Not Use Hold Area
Rev. 0 | Page 32 of 88
AD9923A
register specifies the number of pixels in the last line of the
Complete Field: Combining V-Sequences
field. HDLEN, VDLEN, HDLAST registers are ignored when
the part is in slave mode. The VPATSECOND register is used to
add a second V-pattern group to the XV1 to X12 outputs during
the sensor gate (VSG) line.
After the V-sequences are created, they are combined to create
different readout fields. A field consists of up to nine regions.
Within each region, a different V-sequence can be selected.
Figure 46 shows how the sequence change position (SCP)
registers designate the line boundary for each region and how
the VSEQSEL registers select the V-sequence for each region.
Registers to control the VSG outputs are also included in the
field registers. Table 16 summarizes the registers used to create
the different fields.
The SGMASK register is used to enable or disable each VSG
output. There are two bits for each VSG output to enable
separate masking during SGACTLINE1 and SGACTLINE2.
Setting a masking bit high disables, or masks, the output; setting it
low enables the output. The SGPATSEL register assigns one of the
eight SG patterns to each VSG output. Each SG pattern is created
separately using the SG pattern registers. The SGACTLINE1
register specifies which line in the field contains the VSG
outputs. The optional SGACTLINE2 register allows the same
VSG pulses to repeat on a different line, although separate
masking is available for SGACTLINE1 and SGACTLINE2.
The VSEQSEL registers, one for each region, select which
V-sequences are active during each region. The SWEEP
registers can enable the sweep mode during any region.
The MULTI registers are used to enable the multiplier mode
during any region. The SCP registers create the line boundaries
for each region. The VDLEN register specifies the total number
of lines in the field. The total number of pixels per line (HDLEN)
is specified in the V-sequence registers, and the HDLAST
Table 16. Field Registers
Length
(Bits)
Register
Range
Description
VSEQSEL
5
0 to 31 V-sequence
number
Selected V-sequence for each region in the field.
SWEEP
MULTI
SCP
VDLEN
HDLAST
1
1
12
12
13
High/low
High/low
Enables sweep mode for each region when set high.
Enables multiplier mode for each region when set high.
0 to 4095 line number Sequence change position (SCP) for each region.
0 to 4095 lines
0 to 8191 pixels
0 to 8191 pixels
0 to 31 V-pattern
group number
Total number of lines in each field.
Length in pixels of the last HD line in each field.
Start position of the second V-pattern group applied during VSG line.
Selected V-pattern group for the second pattern applied during VSG line.
VSTARTSECOND 13
VPATSECOND
5
SGMASK
16
High/low, each VSG
Set high to mask each VSG output. Two bits for each VSG output: one for SGLINE1,
and one for SGLINE2.
[0] Masking for VSG1 on SGLINE1.
[1] Masking for VSG1 on SGLINE2.
[2] Masking for VSG2 on SGLINE1.
[3] Masking for VSG2 on SGLINE2.
[15] Masking for VSG8 on SGLINE1.
[16] Masking for VSG8 on SGLINE2.
SGPATSEL
24
0 to 7 pattern
number, each VSG
Selects the VSG pattern number for each VSG output. VSG1[2:0], VSG2[5:3],
VSG3[8:6], VSG4[11:9], VSG5[14:12], VSG6[17:15], VSG7[20:18], VSG8[23:21].
SGACTLINE1
SGACTLINE2
12
12
0 to 4095 line number Selects the line in the field where the VSG is active.
0 to 4095 line number Selects a second line in the field to repeat the VSG signals.
Rev. 0 | Page 33 of 88
AD9923A
SCP 0
SCP 1
SCP 2
SCP 3
SCP 4
SCP 5
SCP 8
VD
HD
REGION 0
REGION 1
REGION 2
REGION 3
REGION 4
REGION 8
XV1 TO XV13
VSG
VSEQSEL0
VSEQSEL1
SGACTLINE
VSEQSEL8
VSEQSEL2
VSEQSEL3
VSEQSEL4
FIELD SETTINGS:
1. SEQUENCE CHANGE POSITIONS (SCP1 TO SCP8) DEFINE EACH OF THE NINE AVAILABLE REGIONS IN THE FIELD.
2. VSEQSEL0 TO VSEQSEL8 SELECTS THE DESIRED V-SEQUENCE FOR EACH REGION.
3. SGLINE1 REGISTER SELECTS WHICH HD LINE IN THE FIELD CONTAINS THE SENSOR-GATE PULSE(S).
Figure 46. Complete Field Is Divided into Regions
Second V-Pattern Group During VSG Active Line and
Special V-Pattern Insertion
Table 17. Special Second V-Pattern Insertion
Length
Register
(Bits)
Range
Description
Most CCDs require additional vertical timing during the sensor
gate line. The AD9923A can output a second V-pattern group
for XV1 to XV13 during the line when the VSG1 to VSG8
sensor gates are active. Figure 47 shows a typical VSG line,
which includes two sets of V-pattern groups for XV1 to XV13.
At the start of the VSG line, the V-pattern group is selected
using the appropriate VSEQSEL register. The second V-pattern
group, unique to the VSG line, is selected using the VPATSECOND
register, located in the field registers. The start position of the
second VPAT group uses the VSTARTSECOND register. For
more information, see Table 16.
SPXV_EN
1
0 or 1
0 = off, 1= enable special
second V-pattern insertion.
Active line for special
SPXV_ACT 12
Line 0 to
Line 4095
0 to 31
V-pattern
number
0 to 8191
pixel
location
second V-pattern insertion.
VPATSELB
VSTARTB
5
Selected V-pattern for
special second V-pattern
insertion if SPXV_EN = 1.
Start position for selected
V-pattern for special
second V-pattern
13
insertion if SPXV_EN = 1.
In addition to inserting a second V-pattern into the VSG line,
the AD9923A can insert a second V-pattern into any other
single line in each sequence. To enable this function in a par-
ticular sequence, set the SPXV_EN register in the appropriate
set of sequence registers to 1. The SPXV_ACT register determines
the active line for the special second V-pattern. The VPATSELB
and VSTARTB registers control both the V-pattern used and the
starting pixel location of the special second V-pattern. For more
information, see Table 17.
Sweep Mode Operation
The AD9923A contains an additional mode of vertical timing
operation called sweep mode. This mode is used to generate a
large number of repetitive pulses that span across multiple HD
lines. Normally, the vertical timing of the AD9923A must be
contained within one HD line length, but when sweep mode is
enabled, the HD boundaries are ignored until the region is
finished. This is useful, for example, in CCD readout operations.
Depending on the vertical resolution of the CCD, up to 3000 clock
cycles, spanning across several HD line lengths, can be required
to shift charge out of the vertical interline CCD registers. These
registers must be free of all charge at the end of the image
exposure before the image is transferred. This can be accom-
plished in sweep mode by quickly shifting out any charge using
a long series of pulses from the XV1 to XV13 outputs. To enable
sweep mode in any region, program the appropriate SWEEP
register to high.
To avoid undesired behavior, do not use the special second
V-pattern in the VSG line; use the existing VPATSECOND and
VSTARTSECOND registers to insert a second V-pattern into
the VSG line. It is recommended that VPATSECOND and
VSTARTSECOND registers are used to create complex timing in
the sensor gate line and not the GROUPB registers. Additionally,
given that the special second V-pattern insertion uses some of the
Group B registers, the user cannot use the special second V-pattern
insertion function and Group B in the same sequence.
Rev. 0 | Page 34 of 88
AD9923A
START POSITION FOR SECOND VPAT GROUP
USES VSTARTSECOND REGISTER
HD
VSG
XV1
XV2
XV13
SECOND VPAT GROUP
Figure 47. Example of Second VPAT Group During Sensor Gate Line
VD
HD
SCP 1
SCP 2
LINE 0
LINE 1
LINE
2
LINE 24
LINE 25
XV1 TO XV13
REGION 0
REGION 1: SWEEP REGION
REGION 2
Figure 48. Example of Sweep Region for High Speed Vertical Shift
Figure 48 shows an example of sweep mode operation. The
number of required vertical pulses depends on the vertical
resolution of the CCD. The XV1 to XV13 output signals are
generated using the V-pattern registers (shown in Table 14).
A single pulse is created using the polarity and toggle position
registers. The number of repetitions is then programmed to
match the number of vertical shifts required by the CCD.
Repetitions are programmed in the V-sequence registers using
the VREP registers. This produces a pulse train of the appropriate
length. Normally, the pulse train is truncated at the end of the
HD line length, but with sweep mode enabled, the HD
boundaries are ignored. In Figure 48, the sweep region occupies
23 HD lines. After the sweep mode region is complete, normal
sequence operation resumes in the next region. When using
sweep mode, set the region boundaries, using the sequence
change position registers, to the appropriate lines to prevent the
sweep operation from overlapping with the next V-sequence.
The start polarity and toggle positions are used in the same
manner as the standard VPAT group programming, but the
VLEN register is used differently. Instead of using the pixel
counter (HD counter) to specify the toggle position locations
(XVTOG1, XVTOG2, XVTOG3, XVTOG4, XVTOG5, and
XVTOG6) of the VPAT group, the VLEN value is multiplied by
the XVTOG value to allow very long pulses to be generated. To
calculate the exact toggle position, counted in pixels after the
start position, use the following equation:
Multiplier Mode Toggle Position = XVTOG × VLEN
Because the XVTOG value is multiplied by the VLEN value, the
resolution of the toggle position placement is reduced.
If VLEN = 4, the toggle position accuracy is reduced to four
pixel steps, instead of single pixel steps. Table 18 summarizes
how the VPAT group registers are used in multiplier mode
operation. In multiplier mode, the VREP registers should be
programmed to the value of the highest toggle position.
Multiplier Mode
To generate very wide vertical timing pulses, a vertical region
can be configured into a multiplier region. This mode uses the
V-pattern registers in a slightly different manner. Multiplier mode
can be used to support unusual CCD timing requirements, such
as vertical pulses that are wider than the 13-bit V-pattern toggle
position counter.
The example shown in Figure 49 illustrates this operation. The
first toggle position is 2, and the second toggle position is 9. In
nonmultiplier mode, this causes the V-sequence to toggle at
Rev. 0 | Page 35 of 88
AD9923A
Pixel 2 and Pixel 9 within a single HD line. However, in
multiplier mode the toggle positions are multiplied by
VLEN = 4; therefore, the first toggle occurs at pixel count = 8,
and the second toggle occurs at pixel count = 36. Sweep mode is
also enabled to allow the toggle positions to cross the HD line
boundaries.
The MULTI function only applies to signals assigned to Group A.
It cannot be used at the same time as the TWOGROUP function or
if any signals are assigned to Group B.
Table 18. Multiplier Mode Register Parameters
Length
(Bits)
Register
MULTI
XVPOL
XVTOG
VLEN
Range
Description
1
1
13
13
12
High/low
High/low
0 to 8191 pixel location
0 to 8191 pixels
0 to 4095
High enables multiplier mode
Starting polarity of XV1 to XV13 signals in each VPAT group
Toggle positions for XV1 to XV13 signals in each VPAT group
Used as multiplier factor for toggle position counter
VREPE/VREPO should be set to the value of the highest XVTOG value
VREP
START POSITION OF VPAT GROUP IS STILL PROGRAMMED IN THE V-SEQUENCE REGISTERS
HD
5
5
3
VLEN
1
1
2
2
3
3
4
4
1
5
2
6
3
7
4
8
1
9
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
PIXEL
NUMBER
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
4
4
XV1 TO XV13
1
2
2
MULTIPLIER MODE V-PATTERN GROUP PROPERTIES:
1
START POLARITY (ABOVE: STARTPOL = 0).
FIRST AND SECOND TOGGLE POSITIONS (ABOVE: XVTOG1 = 2, XVTOG2 = 9).
LENGTH OF VPAT COUNTER (ABOVE: VPATLEN = 4); THIS IS THE MINIMUM RESOLUTION FOR TOGGLE POSITION CHANGES.
TOGGLE POSITIONS OCCUR AT LOCATION EQUAL TO (XVTOG × VPATLEN).
2
3
4
5
IF SWEEP REGION IS ENABLED, THE V-PULSES MAY ALSO CROSS THE HD BOUNDRIES, AS SHOWN ABOVE.
Figure 49. Example of Multiplier Region for Wide Vertical Pulse Timing
Rev. 0 | Page 36 of 88
AD9923A
and SGACTLINE2 registers. Additionally, any of the VSG1 to
VSG8 pulses can be individually disabled using the SGMASK
register. The individual masking allows all SG patterns to be
preprogrammed, and the appropriate pulses for each field can
be separately enabled. For maximum flexibility, the SGPATSEL,
SGMASK, and SGACTLINE registers are separately programmable
for each field. More detail is given in the Complete Field:
Combining V-Sequences section.
Vertical Sensor Gate (Shift Gate) Patterns
In an interline CCD, the vertical sensor gates (VSG) are used to
transfer the pixel charges from the light sensitive image area
into the light shielded vertical registers. From the light shielded
vertical registers, the image is then read line-by-line using the
XV1 to XV13 vertical transfer pulses in conjunction with the
high speed horizontal clocks.
Table 19 summarizes the VSG pattern registers. The AD9923A
has eight VSG outputs, VSG1 to VSG8. Each output can be
assigned to one of eight programmed patterns by using the
SGPATSEL register. Each pattern is generated in a similar
manner as the V-pattern groups, with a programmable start
polarity (SGPOL), first toggle position (SGTOG1), and second
toggle position (SGTOG2). The active line where the VSG1 to
VSG8 pulses occur is programmable using the SGACTLINE1
Additionally, there is the SGMASK_BYP register (Address 0x59)
that overrides SG masking in the field registers. The SGMASK_BYP
register allows sensor gate masking to be changed without
modifying the field register values. The SGMASK_BYP register
is SCK updated; therefore, the new SG-masking values update
immediately.
Table 19. VSG Pattern Registers1
Length
(Bits)
Register
SGPOL
Range
Description
Sensor gate starting polarity for SG patterns 0 to 7.
1
High/low
SGTOG1
13
0 to 8191 pixel First toggle position for SG patterns 0 to 7.
location
0 to 8191 pixel
location
High/low for
each VSG
SGTOG2
13
8
Second toggle position for SG patterns 0 to 7.
SGMASK_BYP
SGMASK Bypass. This register overrides the SGMASK values in each field register. One bit
for each output, where Bit[0] is for VSG1 output and Bit 7 is for VSG8 output.
0 = active.
1 = mask output.
SGMASK_BYP_EN
1
0 or 1
1: enables SGMASK bypass.
1 See field registers in Table 16.
VD
4
HD
1
2
3
VSG PATTERNS
PROGRAMMABLE SETTINGS FOR EACH PATTERN:
1
START POLARITY OF PULSE.
2
FIRST TOGGLE POSITION.
3
SECOND TOGGLE POSITION.
4
ACTIVE LINE FOR VSG PULSES WITHIN THE FIELD (PROGRAMMABLE IN THE FIELD REGISTER, NOT FOR EACH PATTERN).
Figure 50. Vertical Sensor Gate Pulse Placement
Rev. 0 | Page 37 of 88
AD9923A
Table 20 shows how the MODE register bits are used. Unlike
other registers, the MODE register uses 10 address bits as data
bits to increase the total register size to 38 bits. The address
MSBs, A11 and A10, are 1 and 0, respectively, and are used to
specify the MODE register write. The three MSBs, D37, D36,
and D35 are used to specify the number of fields used. A value
from 1 to 7 can be selected using these three bits. The remaining
register bits are divided into five-bit sections to select which
programmed fields are used and in which order. Up to seven
fields can be used in a single MODE write. The AD9923A starts
with the field timing specified by the first field bit, and switches
to the timing specified by the second field bit on the next VD,
and so on.
MODE Register
The MODE register is a single register that selects the field timing
of the AD9923A. Typically, all field, V-sequence, and V-pattern
group information is programmed into the AD9923A at startup.
During operation, the MODE register allows the user to select
any combination of field timing to meet the current requirements
of the system. Using the MODE register in conjunction with
preprogrammed timing greatly reduces 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 rewrite the vertical timing information with each
camera mode change.
A basic still camera application can require five fields of vertical
timing—one for draft mode operation, one for autofocusing, and
three for still image readout. The register timing information for
the five fields is loaded at startup. Depending on how the camera
is being used, the MODE register selects which field timing is
active during camera operation.
After completing the number of fields specified in Bit D37 to
Bit D35, the timing generator of the AD9923A repeats itself by
starting at the first field. This continues until a new write to the
MODE register occurs. Figure 51 shows MODE register settings
for various field configurations.
Table 20. Mode Register Contents—VD Updated
Data Bits
Address (Binary)
Default Value
Description
12b10_xx_xxxx_xxxx
[37:0]
[37:35]
[34:30]
[29:25]
[24:20]
[19:15]
[14:10]
[9:5]
0
A11, A10 must be set to 0x10; remaining A9:A0 bits used for D37:D28
Number of fields (maximum of seven)
Selected field for Field 7
Selected field for Field 5
Selected field for Field 6
Selected field for Field 4
Selected field for Field 3
Selected field for Field 2
[4:0]
Selected field for Field 1
EXAMPLE 1:
TOTAL FIELDS = 3, FIRST FIELD = FIELD 0, SECOND FIELD = FIELD 1, THIRD FIELD = FIELD 2
MODE REGISTER CONTENTS = 0x9800000820
FIELD 1
FIELD 2
FIELD 0
EXAMPLE 2:
TOTAL FIELDS = 1, FIRST FIELD = FIELD 3
MODE REGISTER CONTENTS = 0x8800000003
FIELD 3
EXAMPLE 3:
TOTAL FIELDS = 4, FIRST FIELD = FIELD 5, SECOND FIELD = FIELD 1, THIRD FIELD = FIELD 4, FOURTH FIELD = FIELD 2
MODE REGISTER CONTENTS = 0xA000011025
FIELD 2
FIELD 1
FIELD 4
FIELD 5
Figure 51. Using the Mode Register to Select Field Timing
Rev. 0 | Page 38 of 88
AD9923A
number of high speed vertical pulses needed to clear any charge
from the vertical registers of the CCD.
VERTICAL TIMING EXAMPLE
To better understand how the AD9923A vertical timing
generation is used, consider the example CCD timing chart
in Figure 52. It illustrates a CCD using a general three-field
readout technique. As described in the Complete Field:
Combining V-Sequences section, each readout field should
be divided into separate regions to perform each step of the
readout. The sequence change position (SCP) registers deter-
mine the line boundaries for each region. Then, the VSEQSEL
registers assign a V-sequence to each region. Each V-sequence
contains specific timing information required for each region:
XV1 to XV6 pulses (using VPAT groups), HBLK/CLPOB timing,
and VSG patterns for the SG active lines.
Region 1
Region 1 consists of two lines and uses standard, single line,
vertical shift timing. The timing of this region is the same as the
timing of Region 3.
Region 2
Region 2 is the sensor gate line, where the VSG pulses transfer
the image into the vertical CCD registers. This region might
require use of the second V-pattern group for the SG active line.
Region 3
Region 3 also uses the standard, single line, vertical shift timing,
the same timing used in Region 1. In summary, four regions are
required in each of the three fields.
The example shown in Figure 52 requires four regions, labeled
Region 0, Region 1, Region 2, and Region 3, for each of the
three fields. Because the AD9923A allows many individual
fields to be programmed, Field 0, Field 1, and Field 2 can be
created to meet the requirements of this timing example. In this
example, the four regions for each field are very similar, but the
individual registers for each field allow flexibility to accommodate
more complex timing requirements.
The timing for Region 1 and Region 3 is essentially the same,
reducing the complexity of the register programming. Other
registers, such as the MODE register, shutter control registers
(that is, TRIGGER, and the registers to control the SUBCK,
VSUB, MSHUT, and STROBE outputs), and the AFE gain
registers, VGAGAIN and CDSGAIN, must be used during the
readout operation. These registers are explained in the MODE
Register and Variable Gain Amplifier sections.
Region 0
Region 0 is a high speed vertical shift region. Sweep mode can
be used to generate this timing operation, with the desired
Rev. 0 | Page 39 of 88
AD9923A
1
0 5 5 - 4 1 0 6
N
3
N –
2 1
1 8
1 5
1 2
9
6
3
1
4
N –
N –
2 0
1 7
1 4
1 1
8
5
2
2
5
N –
N –
1 6
1 3
1 0
7
4
1
Figure 52. CCD Timing Example—Dividing Each Field into Regions
Rev. 0 | Page 40 of 88
AD9923A
VERTICAL DRIVER SIGNAL CONFIGURATION
VDR_EN = low, V1 to V13 are forced to VM and SUBCK is
forced to VLL. The VDR_EN pin takes priority over the XV and
VSG signals coming from the timing generator.
As shown in Figure 53, XV1 to XV13, VSG1 to VSG8, and
XSUBCK are outputs from the internal AD9923A timing
generator, and V1 to V13 and SUBCK are the resulting outputs
from the AD9923A vertical driver. When VDR_EN = high, the
vertical driver mixes the XV and VSG pulses and amplifies
them to the high voltages required for driving the CCD. Table 21
through Table 36 describe the output polarities for these signals
vs. their input levels. Refer to these tables when determining the
register settings for the desired output levels. Note that when
The VDR_EN pin can be driven either with an external 3 V
logic signal or by one of the AD9923A shutter outputs
(MSHUT, VSUB, STROBE). To make the AD9923A compatible
with existing AD9923 designs, drive the VDR_EN pin with a
diode to either an external 3 V logic signal or to one of the
shutter outputs.
AD9923A
V-DRIVER
VH VL
3V
XV1
VSG1
XV11
VSG2
XV3
V1
V11
V3
VSG3
XV12
VSG4
3-LEVEL OUTPUTS
V12
VSG5
XV5
V5A
V5B
V7A
V7B
VSG6
INTERNAL
TIMING
GENERATOR
VSG7
XV7
VSG8
XV2
XV4
V2
V4
XV6
V6
XV8
V8
2-LEVEL OUTPUTS
XV9
V9
XV10
XV13
V10
V13
XSUBCK
SUBCK
XSUBCNT
VDR_EN
Figure 53. Internal Vertical Driver Input Signals
Rev. 0 | Page 41 of 88
AD9923A
Table 21. V1 Output Polarity
Table 27. V11 Output Polarity
Vertical Driver Input
Vertical Driver Input
V1 Output
V11 Output
XV1
L
L
H
H
VSG1
XV11
VSG2
L
H
L
VH
VM
VL
L
L
H
H
L
H
L
VH
VM
VL
H
VL
H
VL
Table 22. V3 Output Polarity
Table 28. V12 Output Polarity
Vertical Driver Input
Vertical Driver Input
V3 Output
V12 Output
XV3
L
L
H
H
VSG3
XV12
VSG4
L
H
L
VH
VM
VL
L
L
H
H
L
H
L
VH
VM
VL
H
VL
H
VL
Table 23. V5A Output Polarity
Table 29. V2 Output Polarity
Vertical Driver Input
Vertical Driver Input XV2
V2 Output
V5A Output
L
H
VM
VL
XV5
L
L
H
H
VSG5
L
H
L
VH
VM
VL
Table 30. V4 Output Polarity
Vertical Driver Input XV4
H
VL
V4 Output
L
H
VM
VL
Table 24. V5B Output Polarity
Vertical Driver Input
V5B Output
XV5
L
L
H
H
VSG6
Table 31. V6 Output Polarity
Vertical Driver Input XV6
L
H
L
H
L
VH
VM
VL
V6 Output
VM
VL
H
VL
Table 32. V8 Output Polarity
Vertical Driver Input XV8
Table 25. V7A Output Polarity
V8 Output
Vertical Driver Input
L
H
VM
VL
V7A Output
XV7
L
VSG7
L
VH
VM
VL
L
H
H
L
Table 33. V9 Output Polarity
H
H
VL
Vertical Driver Input XV9
V9 Output
L
H
VM
VL
Table 26. V7B Output Polarity
Vertical Driver Input
V7B Output
XV7
L
L
VSG8
L
H
L
VH
VM
VL
H
H
H
VL
Rev. 0 | Page 42 of 88
AD9923A
Table 34. V10 Output Polarity
Table 36. SUBCK Output Polarity
Vertical Driver Input XV10
V10 Output
Vertical Driver Input
SUBCK Output
L
H
VM
VL
XSUBCK
XSUBCNT
L
L
VH
L
H
H
H
L
H
VH
VMM
VLL
Table 35. V13 Output Polarity
Vertical Driver Input XV13
V13 Output
L
H
VM
VL
XV1
VSG1
VH
V1
VM
VL
Figure 54. XV1, VSG1, and V1 Output Polarities
XV11
VSG2
VH
VM
VL
V11
Figure 55. XV11, VSG2, and V11 Output Polarities
XV3
VSG3
VH
VM
VL
V3
Figure 56. XV3, VSG3, and V3 Output Polarities
Rev. 0 | Page 43 of 88
AD9923A
XV12
VSG4
VH
V12
VM
VL
Figure 57. XV12, VSG4, and V12 Output Polarities
XV5
VSG5
VH
VM
VL
V5A
Figure 58. XV5, VSG5, and V5A Output Polarities
XV5
VSG6
VH
VM
VL
V5B
Figure 59. XV5, VSG6, and V5B Output Polarities
XV7
VSG7
VH
VM
VL
V7A
Figure 60. XV7, VSG7, and V7A Output Polarities
Rev. 0 | Page 44 of 88
AD9923A
XV7
VSG8
VH
VM
VL
V7B
Figure 61. XV7, VSG8, and V7B Output Polarities
XV2, XV4, XV6, XV8
XV9, XV10, XV13
VM
V2, V4, V6, V8
V9, V10, V13
VL
Figure 62. XV2, XV4, XV6, XV8, XV9, XV10, XV13 and V2, V4, V6, V8, V9, V10, V13 Output Polarities
XSUBCNT
XSUBCK
VH
SUBCK
VMM
VLL
Figure 63. XSUBCNT, XSUBCK, and SUBCK Output Polarities
SHUTTER TIMING CONTROL
shutter and VSUB pulses with a logic XOR operation (symbolized
by ^) to generate more complex timing (up to four toggle posi-
tions per line) for MSHUT, STROBE and VSUB: SHUT0 ^ VSUB0,
SHUT0 ^ VSUB1, SHUT0 ^ SHUT1, and SHUT0 ^ SHUT2.
The CCD image exposure time is controlled by the substrate clock
signal (SUBCK) that pulses the CCD substrate to clear out
accumulated charge. The AD9923A supports three types of
electronic shuttering: normal, high precision, and low speed.
Together with the SUBCK pulse placement, the AD9923A can
accommodate different readout configurations to further
suppress the SUBCK pulses during multiple field readouts. The
AD9923A also provides programmable outputs to control an
external mechanical shutter (MSHUT), strobe/flash (STROBE),
and CCD bias select signal (VSUB). Up to four general shutter
pulses (SHUT0 to SHUT3) and two VSUB pulses (VSUB0 and
VSUB1) can be programmed and assigned to any of the three
shutter output pins. The user can also combine the following
SUBCK: Three-Level Output
The AD9923A supports a three-level output from the SUBCK
buffer: VH, VMM, and VLL. The VH power supply is shared
with the V-driver outputs, but VMM and VLL are dedicated
mid and low supplies for the SUBCK buffer. There are two in-
puts to the SUBCK buffer, XSUBCK and XSUBCNT. XSUBCNT
is created by an internal multiplexer that selects from XV1 to
XV13, VSG1 to VSG8, MSHUT, STROBE, VSUB, SHUT0 to
SHUT3, FG_TRIG, high and low.
Rev. 0 | Page 45 of 88
AD9923A
SUBCK: Normal Operation
Table 37. XSUBCNT Multiplexer
By default, the AD9923A operates in a normal SUBCK configu-
ration with the SUBCK signal pulsing in every VD field (see
Figure 64). The SUBCK pulse occurs once per line, and the total
number of repetitions within the field determines the exposure
time. The SUBCK pulse polarity and toggle positions within a
line are programmable, using the SUBCKPOL and SUBCK1TOG
registers (see Table 38). The number of SUBCK pulses per field
is programmed in the SUBCKNUM register (Address 0x64).
Length
(Bits)
Register
Range
Description
XSUBCNT_MUX
5
0 to 31
Selects internal signal to
be used for XSUBCNT
0: XV6
1: XV8
2: XV9
3: XV10
4: VSG5
As shown in Figure 64, the SUBCK pulses always begin in the
line following the SG active line (specified in the SGACTLINE
registers for each field). The SUBCKPOL, SUBCK1TOG,
SUBCK2TOG, SUBCKNUM, and SUBCKSUPPRESS registers
are updated at the start of the line after the sensor gate line, as
described in the Updating New Register Values section.
5: VSG6
6: VSG7
7: VSG8
8: VSG2
9: VSG3
10: VSG4
11: VSG1
12: XV13
13: VSUB
14: MSHUT
15: STROBE
16: XV1
17: XV2
18: XV3
19: XV4
20: XV5
SUBCK: High Precision Operation
High precision shuttering is used in the same manner as normal
shuttering, but an additional register is used to control the last
SUBCK pulse. In this mode, the SUBCK pulses once per line,
but the last SUBCK in the field has an additional SUBCK pulse,
whose location is determined by the SUBCK2TOG register, as
shown in Figure 65. Finer resolution of the exposure time is
possible using this mode. Leaving the SUBCK2TOG register set
to its maximum value (0xFFFFFF) disables the last SUBCK
pulse (default setting).
21: XV7
22: XV11
23: XV12
24: SHUT0
25: SHUT1
26: SHUT2
27: SHUT3
28: FG_TRIG
29: invalid setting
30: high
31: low
SUBCK: Low Speed Operation
Normal and high precision shutter operations are used when
the exposure time is less than one field long. For exposure times
longer than one field interval, low speed shutter operation is
used. The AD9923A uses a separate exposure counter to
achieve long exposure times. The number of fields for the low
speed shutter operation is specified in the EXPOSURENUM
register (Address 0x63). As shown in Figure 66, this shutter
mode suppresses the SUBCK and VSG outputs from 0 fields up
to 4095 fields (VD periods). The VD and HD outputs can be
suppressed during the exposure period by programming the
VDHDOFF register to 1.
To generate a low speed shutter operation, trigger a long ex-
posure by writing to the TRIGGER register, Bit D3. When this
bit is set high, the AD9923A begins an exposure operation at
the next VD edge. If a value greater than 0 is specified in the
EXPOSURENUM register, the AD9923A suppresses the
SUBCK output on subsequent fields.
If the exposure is generated using the TRIGGER register and
the EXPOSURENUM register is set to 0, the behavior of the
SUBCK is the same as during normal shutter or high precision
shutter operations, in which the TRIGGER register is not used.
Rev. 0 | Page 46 of 88
AD9923A
VD
HD
VSG
tEXP
tEXP
SUBCK
SUBCK PROGRAMMABLE SETTINGS:
1. PULSE POLARITY USING THE SUBCKPOL REGISTER.
2. NUMBER OF PULSES WITHIN THE FIELD USING THE SUBCKNUM REGISTER (SUBNUM = 3 IN THE ABOVE EXAMPLE).
3. PIXEL LOCATION OF PULSE WITHIN THE LINE AND PULSE WIDTH PROGRAMMED USING THE SUBCK1TOG REGISTER.
Figure 64. Normal Shutter Mode
VD
HD
VSG
tEXP
tEXP
SUBCK
NOTES
1. SECOND SUBCK PULSE IS ADDED IN THE LAST SUBCK LINE.
2. LOCATION OF SECOND PULSE IS FULLY PROGRAMMABLE USING THE SUBCK2TOG REGISTER.
Figure 65. High Precision Shutter Mode
TRIGGER
EXPOSURE
VD
VSG
tEXP
SUBCK
NOTES
1. SUBCK CAN BE SUPPRESSED FOR MULTIPLE FIELDS BY PROGRAMMING THE EXPOSURE REGISTER TO BE GREATER THAN 0.
2. ABOVE EXAMPLE USES EXPOSURE = 1.
3. TRIGGER REGISTER MUST ALSO BE USED TO START THE LOW SPEED EXPOSURE.
4. VD/HD OUTPUTS CAN ALSO BE SUPPRESSED USING THE VDHDOFF REGISTER = 1.
Figure 66. Low Speed Shutter Mode Using EXPOSURE Register
Rev. 0 | Page 47 of 88
AD9923A
users to select the internal SHUT3 signal and create a custom
SUBCK masking pattern that spans several fields.
SUBCK: Suppression
Normally, the SUBCK begins pulsing on the line following the
sensor gate line (VSG). Some CCDs require suppressing the
SUBCK pulse for one or more lines following the VSG line. The
SUBCKSUPPRESS register enables such suppression.
When generating an exposure by using the TRIGGER register,
as previously described in the Readout After Exposure section,
the AD9923A outputs the SUBCK and VSG signals on every
field by default. This works well for continuous, single field
exposure and readout operations, such as those in the CCD live
preview mode. However, if the CCD requires a longer exposure
time, or if multiple readout fields are needed, the TRIGGER
register is needed to initiate specific exposure and readout
sequences.
Readout After Exposure
After the exposure, the readout of the CCD data occurs, beginning
with the sensor gate (VSG) operation. By default, the AD9923A
generates VSG pulses in every field. When only a single exposure
and readout frame are needed, as is the case in the CCD preview
mode, the VSG and SUBCK pulses can operate in every field.
Typically, the exposure and readout bits in the TRIGGER register
are used together. This initiates a complete exposure-plus-readout
operation. After the exposure, the readout occurs automatically.
The values in the EXPOSURE and READOUTNUM registers
determine the length of each operation.
However, often during readout, the SUBCK output must be
suppressed until the readout is complete. The READOUTNUM
register specifies the number of additional fields after the exposure
to continue the suppression of SUBCK. READOUTNUM can be
programmed for 0 to 7 fields, and should be preprogrammed at
startup, not at the same time as the exposure write. A typical
interlaced CCD frame readout mode generally requires two fields
of SUBCK suppression (READOUTNUM = 2) during readout.
A three-field, six-phase CCD requires three fields of SUBCK
suppression after the readout begins (READOUTNUM = 3).
It is possible to independently trigger the readout operation
without triggering the exposure operation. This causes the readout
to occur at the next VD, and the SUBCK output is suppressed
according to the value set in the READOUTNUM register.
The TRIGGER register also controls the SHUT and VSUB signals.
Each signal is individually controlled, but dependent on the
triggering of the exposure and readout operations. See Figure 71
for a complete example of triggering the exposure and readout
operations.
If SUBCK output is required to initiate backup during the last field
of readout, program the READOUTNUM register to one less
than the total number of CCD readout fields. Similar to the
exposure operation, the readout operation must be triggered
using the TRIGGER register.
Alternatively, it is possible to manually control the exposure and
readout operations by carefully updating the SUBCKSUPPRESS
and VSG masking registers upon every VD field. As described in
the following sections, it is possible to have partial or full manual
control of the shutter signals. This allows greater flexibility in
generating custom exposure/readout/shutter signal timing.
SUBCK: Additional Masking
The SUBCKMASK register (Address 0x65) allows more complex
SUBCK masking. If SUBCKMASK = 1, it starts masking the
SUBCK at the next VD edge. If SUBCKMASK = 2, it enables
Rev. 0 | Page 48 of 88
AD9923A
Table 38. SUBCK and TRIGGER Register Parameters
Length
(Bits)
Register
Range
Description
TRIGGER
8
On/off for eight signals
0: triggers SHUT0 signal.
1: triggers SHUT1 signal.
2: triggers SHUT2 signal.
3: triggers SHUT3 signal.
4: triggers VSUB0 signal.
5: triggers VSUB1 signal.
6: triggers EXPOSURE operation.
7: triggers READOUT operation.
Number of fields to suppress SUBCK after exposure.
READOUTNUM
EXPOSURENUM
3
0 to 7 fields
12
0 to 4095 fields
Number of fields to suppress to SUBCK and VSG during exposure time (low
speed shutter).
VDHDOFF
1
On/off
Disable VD/HD output during exposure.
1 = disable VD.
0 = enable VD.
SUBCKPOL1
1
High/low
SUBCK start polarity for SUBCK1 and SUBCK2.
SUBCK1TOG11
SUBCK1TOG21
SUBCK2TOG11
SUBCK2TOG21
SUBCKNUM1
SUBCKSUPPRESS1 12
SUBCKMASK1
12
12
12
12
12
0 to 4095 pixel locations First toggle positions for first SUBCK pulse (normal shutter).
0 to 4095 pixel locations Second toggle positions for first SUBCK pulse (normal shutter).
0 to 4095 pixel locations First toggle positions for second SUBCK pulse in last line (high precision).
0 to 4095 pixel locations Second toggle positions for second SUBCK pulse in last line (high precision).
1 to 4095 pulses
Total number of SUBCKs per field, at one pulse per line.
Number of pulses, after the VSG line, to suppress SUBCK.
Additional masking of SUBCK output.
0 = no additional mask.
0 to 4095 pulses
0 to 3 masking mode
2
1 = start mask at VD edge.
2 = use internal SHUT3 signal to mask.
1 Register is not VD updated, but updated at the start of the line after the sensor gate line.
Shutter Outputs
VSUB Signal Operation
The AD9923A contains three shutter output pins: VSUB,
MSHUT, and STROBE. Internally, there are six possible shutter
signals available: VSUB0, VSUB1, SHUT0, SHUT1, SHUT2,
and SHUT3. Any of these signals, and the following combinations:
SHUT0 ^ VSUB0, SHUT0 ^ VSUB1, SHUT0 ^ SHUT1,
SHUT0 ^ SHUT2, can be mapped to any of the output pins
using the VSUB_CTRL, MSHUT_CTRL, and STROBE_CTRL
registers.
The CCD readout bias (VSUB) can be programmed to accom-
modate different CCDs. Figure 67 shows two available modes.
In Mode 0, VSUB goes active when the exposure begins during
the field of the last SUBCK. The on position (rising edge in
Figure 67) is programmable to any line within the field. VSUB
remains active until the end of the image readout. In Mode 1,
the VSUB is not activated until the start of the readout.
A function called VSUB_KEEPON is also available. When the
appropriate VSUB_KEEPON bit is set high, the VSUB output
remains active, even after the readout has finished. To disable
the VSUB at a later time, return this bit to low.
The VSUB signals behave differently than the SHUT signals,
and are generally used for the VSUB output pin. If a more generic
approach is desired for the shutter signals, the SHUT signals can
be used for the VSUB output pin.
The AD9923A contains two programmable VSUB signals,
VSUB0 and VSUB1. Either of these signals can be mapped to
the VSUB output pin, the MSHUT pin, or the STROBE pin.
It is also possible to configure the SYNC pin as an output and
send one of the internal shutter signals, or the combinations listed
above, to the SYNC pin using the TESTO_CTRL register
function. This provides the flexibility of outputting up to four
shutter outputs if the external SYNC input function is not needed.
Rev. 0 | Page 49 of 88
AD9923A
SHUT Signal Operation
readout operation. Also, single trigger operation is useful when
the exposure or readout operation is manually generated without
using the TRIGGER register, and the SUBCK and VSG masking
are manually controlled.
SHUT signal operation is shown in Figure 68 through Figure 71.
Table 39 shows the register parameters for controlling the
SHUT signals. There are three different ways to use the SHUT
signals: automatic trigger, single trigger, and manual control.
Note that single trigger operation cannot occur if an exposure
operation has been triggered. SHUT signals behave in automatic
trigger mode if they, and an exposure operation, have been
triggered.
Automatic Trigger
Generally, SHUT signals are triggered together with an expo-
sure or readout operation, using the TRIGGER register. The
SHUT_ON and SHUT_OFF positions are fully programmable
to anywhere within the exposure period, using the field
Manual Control
Any SHUT signal can be controlled in manual control mode,
instead of using the TRIGGER register to activate it. In this
mode, the individual on and off lines and pixel positions are
used separately, depending on the status of the manual signal
control register. Note that only a single toggle position, either
off or on, can be used in a VD interval.
(SHUT_ON_FD/SHUT_OFF_FD), line
(SHUT_ON_LN/SHUT_OFF_LN), and pixel
(SHUT_ON_PX/SHUT_OFF_PX) registers.
The field registers define the field in which the line and pixel
values are used, with respect to the value of the exposure counter.
The on and off positions can occur as soon as the field contains
the last SUBCK (Exposure Field 0), or as late as the final exposure
field before the readout begins. Separate field registers allow the
on and off positions to occur in different exposure fields.
As with single trigger operation, when manual control is
enabled, the SHUT_ON_FD/SHUT_OFF_FD register values
are ignored.
Because there is a separate bit to enable manual control on
SHUT signals, this operation can be used regardless of the
status of a triggered exposure operation.
Single Trigger
SHUT signals can be triggered without triggering an exposure
or readout operation. In this case, SHUT signals are triggered
using the TRIGGER register, but the exposure bit is not triggered.
Both the SHUT on and off positions occur in the next field, and
the SHUT_ON_FD/SHUT_OFF_FD register values are ignored.
Single trigger operation is useful if a pulse is required immedi-
ately in the next field without the occurrence of an exposure or
Note that manual control can be used in conjunction with
automatic or single trigger operations. If a SHUT signal is turned
on using manual control, and then manual control is disabled,
the SHUT signal remains on. If a subsequent trigger operation
occurs, the on position toggle is ignored, because the signal is
already on. In this case, only the off position can be triggered.
TRIGGER
VSUB
VD
VSG1
tEXP
READOUT
SUBCK
4
2
2
MODE 0
VSUB
1
3
MODE 1
VSUB OPERATION:
1
2
3
4
ACTIVE POLARITY IS DEFINED BY VSUBPOL (ABOVE EXAMPLE IS VSUB ACTIVE HIGH).
ON POSITION IS PROGRAMMABLE, MODE 0 TURNS ON AT THE START OF EXPOSURE, MODE 1 TURNS ON AT THE START OF READOUT.
OFF POSITION OCCURS AT END OF READOUT.
OPTIONAL VSUB KEEP-ON MODE LEAVES THE VSUB ACTIVE AT THE END OF THE READOUT.
Figure 67. VSUB0, VSUB1 Signal Programmability
Rev. 0 | Page 50 of 88
AD9923A
SINGLE TRIGGER
WRITE
VD
VSG
ON STATE
SHUT0 TO SHUT3
1
OFF STATE
2
3
SHUT PROGRAMMABLE SETTINGS:
1
ACTIVE POLARITY. DEFINES THE LOGIC LEVEL DURING ON TIME. ABOVE EXAMPLE USES ACTIVE POLARITY = 1.
ON POSITION IS PROGRAMMABLE TO ANY LINE/PIXEL IN FIELD IMMEDIATELY FOLLOWING SINGLE TRIGGER WRITE.
OFF POSITION IS PROGRAMMABLE TO ANY LINE/PIXEL IN FIELD IMMEDIATELY FOLLOWING SINGLE TRIGGER WRITE.
2
3
Figure 68. SHUT0 to SHUT3 Signal Programmability
SHUT_ON = 1
TRIGGER
EXPOSURE
SHUT_ON = 0
(OFF)
(ON)
VD
EXPOSURE FIELD 0
EXPOSURE FIELD 1
EXPOSURE FIELD 2
VSG
tEXP
SUBCK
ON STATE
SHUT0 TO SHUT3
1
OFF STATE
2
3
SHUT PROGRAMMABLE SETTINGS:
1
ACTIVE POLARITY. DEFINES THE LOGIC LEVEL DURING ON TIME. ABOVE EXAMPLE USES ACTIVE POLARITY = 1.
ON POSITION IS PROGRAMMABLE DURING ANY EXPOSURE FIELD. ABOVE EXAMPLE USES SHUTON_FD = 1.
OFF POSITION IS PROGRAMMABLE DURING ANY EXPOSURE FIELD. ABOVE EXAMPLE USES SHUTOFF_FD = 2.
2
3
Figure 69. Manual Control of SHUT0 to SHUT3 Signals
TRIGGER
EXPOSURE
AND SHUT
VD
EXPOSURE FIELD 0
EXPOSURE FIELD 1
EXPOSURE FIELD 2
VSG
tEXP
SUBCK
ON STATE
SHUT0 TO SHUT3
1
OFF STATE
2
3
SHUT PROGRAMMABLE SETTINGS:
1
ACTIVE POLARITY. DEFINES THE LOGIC LEVEL DURING ON TIME. ABOVE EXAMPLE USES ACTIVE POLARITY = 1.
ON POSITION IS PROGRAMMABLE DURING ANY EXPOSURE FIELD. ABOVE EXAMPLE USES SHUTON_FD = 1.
OFF POSITION IS PROGRAMMABLE DURING ANY EXPOSURE FIELD. ABOVE EXAMPLE USES SHUTOFF_FD = 2.
2
3
Figure 70. Single Trigger Control of SHUT0 to SHUT3 Signals
Rev. 0 | Page 51 of 88
AD9923A
Table 39. VSUB0 to VSUB1 and SHUT0 to SHUT3 Register Parameters
Length
(Bits)
Register
Range
Description
VSUB_CTRL
3
0 to 7
Selects which internal shutter signal is mapped to the VSUB pin.
0: SHUT0.
1: SHUT1.
2: SHUT2.
3: SHUT3.
4: use VSUB0_MUX output.
5: use VSUB1_MUX output.
6: invalid setting.
7: use SHUT1_SHUT2_MUX output.
MSHUT_CTRL
STROBE_CTRL
TESTO_CTRL
3
3
3
0 to 7
0 to 7
0 to 7
Selects which internal shutter signal is mapped to the MSHUT pin.
0: SHUT0.
1: SHUT1.
2: SHUT2.
3: SHUT3.
4: use VSUB0_MUX output.
5: use VSUB1_MUX output.
6: invalid setting.
7: use SHUT1_SHUT2_MUX output.
Selects which internal shutter signal is mapped to the STROBE pin.
0: SHUT0.
1: SHUT1.
2: SHUT2.
3: SHUT3.
4: use VSUB0_MUX output.
5: use VSUB1_MUX output.
6: invalid setting.
7: use SHUT1_SHUT2_MUX output.
Selects which internal shutter signal is mapped to the TEST0 signal.
0: SHUT0.
1: SHUT1.
2: SHUT2.
3: SHUT3.
4: use VSUB0_MUX output.
5: use VSUB1_MUX output.
6: invalid setting.
7: use SHUT1_SHUT2_MUX output.
0 = use VSUB0.
1= use SHUT0 ^ VSUB0.
0 = use VSUB1.
1= use SHUT0 ^ VSUB 1.
0 = use SHUT0 ^ SHUT1.
1 = use SHUT0 ^ SHUT2.
VSUB mode. See Figure 67.
0 = Mode 0.
VSUB0_MUX
1
High/low
High/low
High/low
High/low
VSUB1_MUX
1
SHUT1_SHUT2_MUX
VSUB_MODE
1
1b
1 = Mode 1.
VSUB_KEEPON
VSUB_ON
1
High/low
VSUB keep-on mode. VSUB stays active after readout when set high.
VSUB on position. Can turn on at any line in the field.
VSUB start polarity. When VSUB is triggered on.
SHUT manual control.
12
1
0 to 4095 line location
High/low
VSUBPOL
SHUT_ON
1
On/off
0 = SHUT off.
1 = SHUT on.
Rev. 0 | Page 52 of 88
AD9923A
Length
(Bits)
Register
Range
Description
SHUTPOL
SHUT_MAN
1
1
High/low
SHUT active polarity.
Enable/disable
Enables SHUT manual control mode.
0 = disable.
1 = enable.
SHUT_ON_FD
SHUT_ON_LN
SHUT_ON_PX
SHUT_OFF_FD
SHUT_OFF_LN
SHUT_OFF_PX
12
12
13
12
12
13
0 to 4095 field location
0 to 4095 line location
0 to 8191 pixel location
0 to 4095 field location
0 to 4095 line location
0 to 8191 pixel location
Field location to switch on MSHUT. Inactive, or closed.
Line position to switch on MSHUT. Inactive, or closed.
Pixel position to switch on MSHUT. Inactive, or closed.
Field location to switch off MSHUT. Inactive, or closed.
Line position to switch off MSHUT. Inactive, or closed.
Pixel position to switch off MSHUT. Inactive, or closed.
Explanation of Figure 71
4. STROBE output turns on and off at the location specified
in the SHUT0_ON/SHUT0_OFF registers (Address 0x6D/
Address 0x71).
The numbers in this section, Explanation of Figure 71,
correspond precisely to the numbers embedded in Figure 71.
5. MSHUT output turns off at the location specified in the
SHUT1_OFF_FD, SHUT1_OFF_LN, and SHUT1_OFF_PX
registers (Address 0x75 and Address 0x76). The SHUT1 on
position is ignored, because the SHUT1 signal is already on
from a previous manual operation (see Step 10).
1. Write to the READOUTNUM register (Address 0x62) to
specify the number of fields to suppress SUBCK during
readout of CCD data. In this example, READOUTNUM = 3.
Write to the EXPOSURENUM register (Address 0x63) to
specify the number of fields to suppress SUBCK and VSG out-
puts during exposure. In this example, EXPOSURENUM = 1.
6. The next VD falling edge automatically starts the first
readout field.
Write to the TRIGGER register (Address 0x61) to trigger
the SHUT0 (STROBE), SHUT1 (MSHUT), and VSUB0
(VSUB) signals, and to start the exposure-plus-readout
operation. To trigger these events (see Figure 71), set the
register TRIGGER = 0xD3. Readout automatically occurs
after the exposure period finishes.
7. The next VD falling edge automatically starts the second
readout field.
8. The next VD falling edge automatically starts the third
readout field.
9. Write to the MODE register to reconfigure the single draft
mode field timing.
Write to the MODE register to configure the next five
fields. The first two fields during exposure are the same as
the current draft mode fields, and the next three fields are
the still frame readout fields. The register settings for the
draft mode field and the three readout fields are previously
programmed.
Write a 1 to the SHUT1_MAN and SHUT1_ON registers
(Address 0x72) to turn the MSHUT output back manually.
10. VD/HD falling edge updates the serial writes from 9. VSG
outputs return to draft mode timing. SUBCK output
resumes operation.
2. VD/HD falling edge updates the serial writes from 1.
3. If VSUB0 MODE = 0 (Address 0x69), VSUB output
turns on at the line specified in the VSUB0_ON
register (Address 0x6A).
MSHUT output returns to the on position (active or open).
Be sure to disable manual control of SHUT1 before another
automatic trigger of the SHUT1 signal is needed.
VSUB output returns to the off position (inactive).
Rev. 0 | Page 53 of 88
AD9923A
EXAMPLE OF EXPOSURE AND READOUT OF INTERLACED FRAME
6 0 9 5 - 4 1 0 6
Figure 71. Example of Exposure and Still Image Readout Using Shutter Signals and MODE Register
Rev. 0 | Page 54 of 88
AD9923A
and pixel resolution. The field registers for SHUT1 are ignored
because the field placement of the FG_TRIG pulse is matched
to the field count specified by the MODE register operation.
The FG_TRIGEN register contains a three-bit value that specifies
which field count contains the FG_TRIG pulse. Figure 72 shows
how the FG_TRIG pulse is generated using these registers.
FG_TRIG OPERATION
The AD9923A contains one additional signal that can be used
in conjunction with shutter operation or general system
operation. The FG_TRIG signal is an internally generated pulse
that can be output on the SYNC pins for shutter or other system
functions. A unique feature of the FG_TRIG signal is that it is
output with respect to the MODE register field status.
After the FG_TRIG signal is specified, it can be enabled using
Bit 3 of the FG_TRIGEN register. The FG_TRIG signal is
mapped to the SYNC output if the SYNC pin is configured as
an output (SYNCENABLE = 0).
The FG_TRIG signal is generated using the SHUT1 start
polarity and toggle position registers, programmable with line
Table 40. FG_TRIG Operation Registers
Register
Address Bit Location Description
SYNCENABLE
0x12
[0]
0 = configures SYNC pin as an output. By default, the FG_TRIG signal is output on the SYNC pin.
1 = SYNC pin is an external synchronization input.
[2:0] selects the field count for the pulse based on the mode field counter.
[3] = 1 to enable FG_TRIG signal output.
FG_TRIGEN
0xF1
[3:0]
SHUT1POL
SHUT1_ON_LN
SHUT1_ON_PX
0x72
0x74
0x74
[1]
[1] FG_TRIG start polarity.
FG_TRIG first toggle, line location.
[11:0]
[25:13]
[11:0]
[25:13]
FG_TRIG first toggle, pixel location.
SHUT1_OFF_LN 0x76
SHUT1_OFF_PX 0x76
FG_TRIG second toggle, line location.
FG_TRIG second toggle, pixel location.
VD
MODE REGISTER
FIELD COUNT
FIELD 0
FIELD 1
FIELD 2
FIELD 0
FIELD 1
4
4
FG_TRIG
1
2
3
FG_TRIG PROGRAMMABLE SETTINGS:
1
ACTIVE POLARITY.
2
FIRST TOGGLE POSITION, LINE AND PIXEL LOCATION.
3
4
SECOND TOGGLE POSITION, LINE AND PIXEL LOCATION.
FIELD PLACEMENT BASED ON MODE REGISTER FIELD COUNT.
Figure 72. FG_TRIG Signal Generation
Rev. 0 | Page 55 of 88
AD9923A
0.1µF 0.1µF
REFB REFT
1.0V 2.0V
FIXED
DELAY
1
0
DC RESTORE
1.5V
CLI
DCLK
DOUT
INTERNAL
DOUT PHASE
V
REF
DOUT
DLY
SHP
DCLK
MODE
SHD
2V FULL SCALE
6dB ~ 42dB
VGA
0.1µF
CCDIN
12
12-BIT
ADC
OUTPUT
DATA
LATCH
CDS
OPTICAL BLACK
CLAMP
DAC
VGA GAIN
REGISTER
CLPOB PBLK
DIGITAL
FILTER
DOUT
PHASE
SHP SHD
CLPOB PBLK
8
CLAMP LEVEL
REGISTER
PRECISION
V-H
TIMING
TIMING
CLI
GENERATION
GENERATION
AD9923A
Figure 73. Analog Front End Functional Block Diagram
signal with an ADC full-scale range of 2 V. When compared to 1 V
full-scale systems, the equivalent range of gain is 0 dB to 36 dB.
ANALOG FRONT END DESCRIPTION/OPERATION
The AD9923A signal processing chain is shown in Figure 73.
Each step is essential to achieve a high quality image from the
raw CCD pixel data.
The VGA gain curve follows a linear-in-dB characteristic. The
exact VGA gain can be calculated for any gain register value
using the following equation
DC Restore
Gain (dB) = (0.0358 × Code) + 5.5 dB
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 1.5 V
so that it is compatible with the 3 V supply voltage of the AD9923A.
where the code range is 0 to 1023.
42
Correlated Double Sampler
36
30
24
18
12
6
The CDS circuit samples each CCD pixel twice to extract video
information and reject low frequency noise. The timing shown in
Figure 20 illustrates how the two internally generated CDS
clocks, SHP and SHD, are used to sample the reference and data
levels of the CCD signal, respectively. The placement of the SHP
and SHD sampling edges is determined by the setting of the
SHPLOC and SHDLOC registers located at Address 0x37.
Placement of these clock signals is critical to achieve the best
CCD performance.
0
127
255
383
511
639
767
895
1023
The CDS gain can be set to −3 dB, 0 dB (default), +3 dB, or +6 dB
in the CDSGAIN register, Address 0x04. The +3 dB and +6 dB
settings improve noise performance, but reduce the input range
(see Figure 8).
VGA GAIN REGISTER CODE
Figure 74. VGA Gain Curve
ADC
The AD9923A uses a high performance ADC architecture
optimized for high speed and low power. Differential nonlin-
earity (DNL) performance is typically better than 1 LSB. The
ADC uses a 2 V input range. See Figure 6 and Figure 8 for
typical linearity and noise performance plots.
Variable Gain Amplifier
The VGA stage provides gain in the range of 6 dB to 42 dB,
programmable with 10-bit resolution through the serial digital
interface. A minimum gain of 6 dB is needed to match a 1 V input
Rev. 0 | Page 56 of 88
AD9923A
Optical Black Clamp
immediately valid. Programming the DOUTLATCH register,
Bit D1 to 1 sets the output latches transparent. The data outputs
can also be disabled (three-stated) by setting the DOUTDISABLE
Register 0x01, Bit D0 to 1.
The optical black clamp loop removes residual offsets in the
signal chain and tracks 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 CLAMPLEVEL register.
The value can be programmed between 0 LSB and 255 LSB in
1023 steps. 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 AD9923A optical black clamping
can be disabled using the CLPENABLE register (Address 0x00,
Bit D2). Even though the loop is disabled, the CLAMPLEVEL
register can still be used to provide programmable offset
adjustment.
The DCLK output can be used for external latching of the data
outputs. By default, the DCLK output tracks the value of the
DOUTPHASE register. By changing the DCLKMODE register,
the DCLK output can be held at a fixed phase, and the
DOUTPHASE register value is ignored.
To optimize the delay between the DCLK rising edge and the
data output transition, the DOUTDELAY register is used. By
default, there is approximately 8 ns of delay from the rising edge
of DCLK to the transition of the data outputs. See the High
Speed Timing Generation section for more information.
Switching the data outputs can couple noise into the analog
signal path. To minimize switching noise, set the DOUTPHASE
register to the same edge as the SHP sampling location, or up to
11 edges after the SHP sampling location. Other settings can
produce good results, but require experimentation. It is
recommended that the DOUTPHASE location not occur
between the SHD sampling location and 11 edges after the SHD
location. For example, if SHDLOC = 0, set DOUTPHASE to an
edge location of 12 or greater. If adjustable phase is not required
for the data outputs, the output latch can be left transparent
using Register 0x01, Bit D1.
The CLPOB pulse should be placed during the CCD optical
black pixels. It is recommended that the CLPOB pulse duration
is at least 20 pixels wide to minimize clamping noise. Shorter
pulse widths can be used, but clamping noise might increase,
reducing the ability to track low frequency variations in the
black level. See the Horizontal Clamping and Blanking section
for timing examples.
Digital Data Outputs
The digital output data is latched using the DOUTPHASE
register value, as shown in Figure 73. Output data timing is shown
in Figure 21 and Figure 22. It is also possible to leave the output
latches transparent, so that the data outputs from the ADC are
Data output coding is normally straight binary, but can be
changed to gray coding by setting the GRAYEN Register 0x01,
Bit D2 to 1.
Rev. 0 | Page 57 of 88
AD9923A
CLI input, set the CLIDIVIDE register (Address 0x30) to
1 before resetting the timing core. It is important to wait
at least 500 ꢀs after starting the master clock (CLI) before
resetting the timing core, especially if using a crystal or
crystal oscillator.
Recommended Power-Up Sequence for Master Mode
When the AD9923A is powered up, the following sequence is
recommended (see Figure 75):
1.
2.
3.
4.
Turn on the +3 V power supplies for the AD9923A, and
start the master clock (CLI).
7.
8.
Configure the AD9923A for master mode timing by
writing 1 to the MASTER register (Address 0x20).
Turn on the V-driver supplies (VH and VL). There are no
restrictions on the order in which VH and VL are turned on.
Bring the VDR_EN signal high to +3 V to enable the
V-driver outputs. If VDR_EN = 0 V, all V-driver outputs =
VM, and SUBCK = VLL.
Reset the internal AD9923A registers by writing 1 to the
SW_RST register (Address 0x10).
9.
Write 1 to the OUTCONTROL register (Address 0x11).
This allows the outputs to become active after the next
SYNC rising edge.
Load the required registers to configure the required
VPAT group, V-sequence, field timing information, high
speed timing, horizontal timing, and shutter timing
information.
10. Generate a SYNC event. If SYNC is high at power-up,
bring SYNC input low for a minimum of 100 ns. Then,
bring SYNC high. This causes the internal counters to
reset and starts a VD/HD operation. The first VD/HD
edge allows VD register updates to occur, including
OUTCONTROL to enable all outputs. If an external
SYNC pulse is not available, generate an internal SYNC
pulse by writing to the SYNCPOL register as described in
the Generating Software Sync Without External Sync
Signal section.
5.
6.
To place the part into normal power operation, write 0x04
to the AFE STANDBY register (Bits[1:0], Address 0x00)
and 0x60 to TEST3 Register 0xEA. If the CLO output is
being used to drive a crystal, also power up the CLO
oscillator by writing 1 to Register 0x16.
By default, the internal timing core is held in a reset state
with TGCORE_RSTB register = 0. Write 1 to the
TGCORE_RSTB register (Address 0x15) to start the
internal timing core operation. If a 2× clock is used for the
VH SUPPLY
1
2
+3V SUPPLIES
VL SUPPLY
0V
POWER
SUPPLIES
5
CLI
(INPUT)
3
4
5
6
7
9
SERIAL
WRITES
10
SYNC
(INPUT)
tSYNC
1V
(HI-Z BY DEFAULT)
(HI-Z BY DEFAULT)
VD
FIRST FIELD
(OUTPUT)
1H
HD
(OUTPUT)
DIGITAL
OUTPUTS
H1/H3, RG, DCLK, STROBE, MSHUT, VSUB
(AND INTERNAL XV1 TO XV13, VSG1 TO VSG8, XSBUCK, XSUBCNT)
CLOCKS ACTIVE WHEN OUTCONTROL
REGISTER IS UPDATED AT VD/HD EDGE.
+3V
8
VDR_EN
0V
VH
VM
V1 TO V13
VM
VL
Figure 75. Recommended Power-Up Sequence and Synchronization, Master Mode
Rev. 0 | Page 58 of 88
AD9923A
Table 41. Power-Up Register Write Sequence
Register
Address
Data
Description
SW_RST
0x10
0x20 to 0xFFF
0x00
0xEA
0x16
0x01
User defined
0x04
0x60
0x01
Resets all registers to default values
Horizontal, vertical, shutter timing
Powers up the AFE
Set TEST3 register to required value
Resets crystal oscillator circuit
Resets internal timing core
Configures master mode
STANDBY
TEST3
OSC_RST
TGCORE_RSTB
MASTER
0x15
0x20
0x01
0x01
OUTCONTROL
SYNCPOL
0x11
0x13
0x01
0x01
Enables all outputs after SYNC
SYNC active polarity (for software SYNC only)
SYNC
VD
SUSPEND
HD
HL, H1 TO H4, RG,
XV1 TO XV13,
VSG1 TO VSG8, SUBCK
NOTES
1. THE SYNC RISING EDGE RESETS VD/HD AND COUNTERS TO 0.
2. SYNC POLARITY IS PROGRAMMABLE USING SYNCPOL REGISTER (ADDR 0x13).
3. DURING SYNC LOW, ALL INTERNAL COUNTERS ARE RESET AND VD/HD CAN BE SUSPENDED USING THE SYNCSUSPEND REGISTER (ADDR 0x14).
4. IF SYNCSUSPEND = 1, VERTICAL CLOCKS, H1 TO H4, AND RG ARE HELD AT THE SAME POLARITY SPECIFIED BY OUTCONTROL = LOW.
5. IF SYNCSUSPEND = 0, ALL CLOCK OUTPUTS CONTINUE TO OPERATE NORMALLY UNTIL SYNC RESET EDGE.
Figure 76. SYNC Timing to Synchronize AD9923A with External Timing
Generating Software Sync Without External Sync Signal
Power-Up and Synchronization in Slave Mode
If an external sync pulse is not available, it is possible to generate
an internal sync pulse by writing to the SYNCPOL register
(Address 0x13). If the software SYNC option is used, the SYNC
input (Pin 35) should be low (VSS) during the power-up proce-
dure. After the power-up procedure is complete, the SYNC pin
can be used as an output by setting the SYNCENABLE register
low (Address 0x12).
The power-up procedure for slave mode operation is the same
as the procedure described for master mode operation, with
two exceptions:
•
Eliminate Step 8. Do not configure the part for master
mode timing.
•
No sync pulse is required in slave mode. Substitute Step 10
with starting the external VD and HD signals. This
synchronizes the part, allows the register updates, and
starts the timing operation.
After power-up, follow Step 1 to Step 9 of the procedure in the
Recommended Power-Up Sequence for Master Mode section.
For Step 10, instead of using the external sync pulse, write 1 to
the SYNCPOL register to generate an internal sync pulse and
begin the timing operation.
Note that DCLK does not begin to transition until Step 7 is
complete.
SYNC During Master Mode Operation
When the AD9923A is in slave mode, the VD/HD inputs
synchronize the internal counters. After a falling edge of VD,
there is a latency of 34 master clock edges (CLI) after the falling
edge of HD until the internal H-counter is reset. The reset
operation is shown in Figure 77.
The SYNC input can be used anytime during master mode
operation to synchronize the AD9923A counters with external
timing, as shown in Figure 76.
To suspend operation of the digital outputs during the SYNC
operation, set the SYNCSUSPEND register (Address 0x14) to 1. If
SYNCSUSPEND = 1, the polarities of the outputs are held at the
same state as when OUTCONTROL = low, as shown in Table 42
and Table 43.
Note that if SHDLOC is set so that the 3 ns minimum delay
between the rising edge of SLI and the falling edge of the
internal SHD signal is not met, the internal H-counter can reset
after only 33 master clock edges (CLI).
Rev. 0 | Page 59 of 88
AD9923A
VD
HD
3ns MIN
CLI
3ns MIN
tCLIDLY
SHD
INTERNAL
HD
INTERNAL
H-COUNTER
RESET
32.5 CYCLES
H-COUNTER
(PIXEL COUNTER)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0
1
2
NOTES
1. INTERNAL HD FALLING EDGE IS LATCHED BY CLI RISING EDGE, THEN LATCHED AGAIN BY SHD INTERNAL FALLING EDGE.
2. INTERNAL H-COUNTER IS ALWAYS RESET 32.5 CLOCK CYCLES AFTER THE INTERNAL HD FALLING EDGE.
3. DEPENDING ON THE VALUE OF SHDLOC, H-COUNTER RESET CAN OCCUR 33 OR 34 CLI CLOCK EDGES AFTER THE EXTERNAL HD FALLING EDGE.
4. SHDLOC = 0 IS SHOWN IN ABOVE EXAMPLE. IN THIS CASE, THE H-COUNTER RESET OCCURS 34 CLI RISING EDGES AFTER HD FALLING EDGE.
Figure 77. External VD/HD and Internal H-Counter Synchronization, Slave Mode
VD
H-COUNTER
RESET
HD
NO TOGGLE POSITIONS ALLOWED IN THIS AREA
H-COUNTER
N-28 N-27 N-26 N-25 N-24 N-23 N-22 N-21 N-20 N-19 N-18 N-17 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
N
0
1
2
3
4
(PIXEL COUNTER)
NOTES
1. TOGGLE POSITIONS CANNOT BE PROGRAMMED WITHIN 28 PIXELS OF PIXEL 0 LOCATION.
Figure 78. Toggle Position Inhibited Area—Master Mode
VD
HD
H-COUNTER
RESET
NO TOGGLE POSITIONS ALLOWED IN THIS AREA
H-COUNTER
(PIXEL COUNTER)
N-33 N-32 N-31 N-30 N-29 N-28 N-27 N-26 N-25 N-24 N-23 N-22 N-21 N-20 N-19 N-18 N-17 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
N
0
1
2
NOTES
1. TOGGLE POSITIONS CANNOT BE PROGRAMMED WITHIN 28 PIXELS OF PIXEL 0 LOCATION.
Figure 79. Toggle Position Inhibited Area—Slave Mode
Vertical Toggle Position Placement Near Counter Reset
Figure 79 shows the same example for slave mode. The same
restriction applies—the last 28 pixels before the counters are
reset cannot be used. However, the counter reset is delayed with
respect to VD/HD placement; therefore, the inhibited area is
different than it is in master mode.
One additional consideration during the reset of the internal
counters is the vertical toggle position placement. Prior to the
internal counters being reset, there is a region of 28 pixels
during which no toggle positions can be programmed.
It is also recommended that Pixel Location 0 is not used for
toggle positions for the VSG and SUBCK pulses.
As shown in Figure 78, in master mode, the last 28 pixels before
the HD falling edge should not be used for toggle position place-
ment of the XV, VSG, SUBCK, HBLK, PBLK, or CLPOB pulses.
Rev. 0 | Page 60 of 88
AD9923A
to hold a specific value during the Standby 3 mode using
STANDBY MODE OPERATION
Register 0xE2, as detailed in Table 43. The vertical outputs can be
programmed to hold a specific value when OUTCONTROL =
low, or when in Standby 1 or Standby 2 mode, by using
Register 0xF3. The following list provides guidelines for the
mapping of the bits in these registers to the various vertical and
shutter outputs when the device is in one of the three standby
modes, or when OUTCONTROL = low.
The AD9923A contains three standby modes to optimize the
overall power dissipation in various applications. Bits[1:0] of
Register 0x00 control the power-down state of the device:
STANDBY[1:0] = 00 = normal operation (full power)
STANDBY[1:0] = 01 = Standby 1 mode
STANDBY[1:0] = 2 = Standby 2 mode
•
•
Standby 3 mode takes priority over OUTCONTROL for
determining the output polarities.
These polarities assume OUTCONTROL = high, because
OUTCONTROL = low takes priority over Standby 1 and
Standby 2.
Standby 1 and Standby 2 set H and RG drive strength to
their minimum values (4.3 mA).
VD and HD default to High-Z status when in slave mode
regardless of standby mode or OUTCONTROL status.
STANDBY[1:0] = 3 = Standby 3 mode (lowest power)
Table 42 and Table 43 summarize the operation of each power-
down mode. Note that when OUTCONTROL = LO, it takes
priority over the Standby 1 and Standby 2 modes in determining
the digital output states, but Standby 3 mode takes priority over
OUTCONTROL. Standby 3 has the lowest power consumption,
and can shut down the crystal oscillator circuit between CLI
and CLO. If CLI and CLO are being used with a crystal to
generate the master clock, this circuit is powered down and
there is no clock signal. When returning the device from
Standby 3 mode to normal operation, reset the timing core at
least 500 μs after writing to the STANDBY register (Bits[1:0],
Address 0x00). This allows sufficient time for the crystal circuit
to settle. The vertical and shutter outputs can be programmed
•
•
This feature is useful during power-up if different polarities are
required by the V-driver and CCD to prevent damage.
It is important to note that when VDR_EN = 0 V, V1 to V13 are
at VM, and SUBCK is at VLL regardless of the state of the value
of the STANDBY and OUTCONTROL registers.
Table 42. Standby Mode Operation
I/O Block
Standby 3 (Default)1, 2
OUTCONTROL = LOW2
No change
No change
No change
Running
Low
Low
High
Low
High
Standby 23, 4
Off
Off
On
Standby 13, 4
Only REFT, REFB on
On
On
Running
Low (4.3 mA)
Low (4.3 mA)
High (4.3 mA)
Low (4.3 mA)
High (4.3 mA)
Low (4.3 mA)
Running
AFE
Off
Timing Core
Off
CLO Oscillator
Off
CLO
HL
H1
H2
H3
H4
RG
VD5
High
High-Z
High-Z
High-Z
High-Z
High-Z
High-Z
Low
Running
Low (4.3 mA)
Low (4.3 mA)
High (4.3 mA)
Low (4.3 mA)
High (4.3 mA)
Low (4.3 mA)
VDHDPOL value
VDHDPOL value
Low
Low
VDHDPOL value
VDHDPOL value
Running
Low
HD
Low
Running
DCLK
D0 to D11
Low
Low
Running
Low
Low
1 To exit Standby 3, write 00 to STANDBY (Bits[1:0], Address 0x00), then reset the timing core after 500 μs to guarantee proper settling of the oscillator.
2 Standby 3 mode takes priority over OUTCONTROL for determining the output polarities.
3 These polarities assume OUTCONTROL = high, because OUTCONTROL = low takes priority over Standby 1 and Standby 2.
4 Standby 1 and Standby 2 set H and RG drive strength to their minimum values (4.3 mA).
5
VD and HD default to High-Z status when in slave mode regardless of Standby mode or OUTCONTROL status.
Rev. 0 | Page 61 of 88
AD9923A
Table 43. Standby Mode Operation—Vertical and Shutter Outputs
Output
Standby 3 (Default)1, 2
OUTCONTROL = Low
Standby 23, 4
Low
Standby 13, 4
Low
XV1
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
High
High
High
High
High
High
High
High
High
Low
Low
Low
XV2
XV3
XV4
XV5
XV6
XV7
XV8
XV9
XV10
XV11
XV12
XV13
VSG1
VSG2
VSG3
VSG4
VSG5
VSG6
VSG7
VSG8
XSUBCK
VSUB5
MSHUT5
STROBE5
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
High
High
High
High
High
High
High
High
High
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
High
High
High
High
High
High
High
High
High
Low
Low
Low
Low
Low
1 Polarities for vertical and shutter outputs when the AD9923 is in Standby 3 mode are programmable using the STANDBY3POL register, Address 0xE2 (default register
value = 0x000000).
2 Bit assignments for the STANDBY3POL[23:0] register (Address 0xE2): (MSB) STROBE, MSHUT, VSUB, XSUBCK, VSG8, VSG7, VSG6, VSG3, VSG5, VSG4, VSG2, VSG1, XV13,
XV12, XV11, XV10, XV9, XV8, XV7, XV6, XV5, XV4, XV3, XV2, and XV1 (LSB).
3 Polarities for vertical outputs when the AD9923 is in Standby 1, Standby 2, or if OUTCONTROL = low, are programmable using the STANDBY12POL register,
Address 0xF3 (default register value = 0x3FE000)
4 Bit assignments for the STANDBY12POL[20:0] register (Address 0xF3): (MSB) XSUBCK, VSG8, VSG7, VSG6, VSG3, VSG5, VSG4, VSG2, VSG1, XV13, XV12, XV11, XV10, XV9,
XV8, XV7, XV6, XV5, XV4, XV3, XV2, and XV1 (LSB).
5 VSUB, MSHUT, and STROBE polarities for Standby 1, Standby 2, or if OUTCONTROL = low are controlled by STANDBY3POL.
Rev. 0 | Page 62 of 88
AD9923A
mutual inductance, route the complementary signals, H1 and
H2, as symmetrically and close together as possible. The same
should be done for the H3 and H4 signals. Heavier PCB traces
are recommended because of the large transient current
demand placed by the CCD on HL and H1 to H4. If possible,
physically locating the AD9923A closer to the CCD reduces the
inductance on these lines. The routing path should be as direct
as possible from the AD9923A to the CCD.
CIRCUIT LAYOUT INFORMATION
The AD9923A typical circuit connections are shown in Figure 82.
The PCB layout is critical for achieving good image quality from
the AD9923A. All supply pins, particularly the pins for the AVDD,
TCVDD, RGVDD, and HVDD supplies, must be decoupled to
ground with quality, high frequency chip capacitors.
The decoupling capacitors should be as close as possible to the
supply pins and have a very low impedance path to a continuous
ground plane. There should be a bypass capacitor of at least
4.7 μF for each main supply—AVDD, HVDD, and DRVDD—
but this is not necessary for each individual pin. In most
applications, it is easier to share the supply for RGVDD and
HVDD; this requires bypassing each supply pin separately. A
separate 3 V supply can also be used for DRVDD, but it should
be decoupled to the same ground plane as the rest of the chip. A
separate ground for DRVSS is not recommended.
The AD9923A also contains an on-chip oscillator for driving an
external crystal. The maximum crystal frequency that the
AD9923A can support is 36 MHz. Figure 80 shows an example
application using a typical 24 MHz crystal. For the exact values
of the external resistors and capacitors, see the crystal
manufacturer’s data sheet.
~2MΩ
AD9923A
The analog bypass pins (REFT and REFB) should be carefully
decoupled to ground, as close as possible to their respective
pins. The analog input (CCDIN) capacitor should also be
located close to the pin.
~375Ω
CLI (H12)
CLO (J12)
10pF~20pF
10pF~20pF
24MHz
XTAL
To avoid excessive distortion of the signals, design the HL, H1
to H4, and RG traces to have low inductance. To minimize
Figure 80. Crystal Driver Application
VH
V5V
VDR_EN
VDD
0V
VL
Figure 81. AD9923A Recommended Power up Sequence
Rev. 0 | Page 63 of 88
AD9923A
3
SERIAL INTERFACE (FROM ASIC/DSP)
SYNC (FROM ASIC/DSP)
VERTICAL SYNC (TO/FROM ASIC/DSP)
HORIZONTAL SYNC (TO/FROM ASIC/DSP)
RESETB (FROM ASIC/DSP)
MASTER CLOCK INPUT
OPTIONAL CLOCK OSCILLATOR OUTPUT
(FOR CRYSTAL APPLICATION)
DCLK OUTPUT
12
VSUB OUTPUT (TO CCD BIAS CIRCUIT)
DATA OUTPUTS
STROBE CONTROL OUTPUT
MECHANICAL SHUTTER CONTROL OUTPUT
0.1µF
0.1µF
REFB
REFT
A2
A3
DRVDD
DVDD1
+3V SUPPLY
0.1µF
L4
E1
0.1µF
0.1µF
CCDGND
CCDIN
A5
A6
0.1µF
CCD SIGNAL INPUT
DVDD2
DVDD2
DVDD2
K8
+3V SUPPLY
RG
HL
H4
H3
H2
H1
C10
B11
C11
D11
E11
L8
L7
TEST3
VDR_EN
NC
+3V SUPPLY
J8
V-DRIVER ENABLE
(FROM ASIC/DSP)
K11
K4
F11
6
H, RG OUTPUTS
(TO CCD)
AVSS
AVSS
AVSS
A1
B2
B3
B4
B5
A4
TEST1
J7
J5
TEST0
DRVSS
DVSS1
DVSS2
AD9923ABBCZ
(Not to Scale)
L5
AVSS
AVSS
AVSS
AVSS
F2
K9
A11
E6
G2
G3
H9
NC
B6
B7
B9
NC
NC
AVSS
TCVSS
NC
NC
RGVSS
HVSS
A10
D10
+3V ANALOG
SUPPLY
NC
NC
NC
NC
NC
NC
NC
NC
NC
J6
AVDD
J9
A7
B8
J10
J11
K7
L1
L6
L9
4.7µF
6.3V
0.1µF
TCVDD
0.1µF
+3V H,
RG SUPPLY
RGVDD
HVDD
B10
0.1µF
0.1µF
+3V H,
RG SUPPLY
L11
D9
NC = NOT INTERNALLY
CONNECTED
4.7µF
6.3V
VH SUPPLY
VL SUPPLY
+3V SUPPLY
0.1µF
2.2µF
25V
25V
4.7µF
10V
0.1µF
10V
0.1µF
15
VERTICAL OUTPUTS (TO CCD)
SUBCK OUTPUT (TO CCD)
Figure 82. AD9923ABBCZ Typical Circuit Configuration using External Hardware Sync
Rev. 0 | Page 64 of 88
AD9923A
3
SERIAL INTERFACE (FROM ASIC/DSP)
VERTICAL SYNC (TO/FROM ASIC/DSP)
HORIZONTAL SYNC (TO/FROM ASIC/DSP)
RESETB (FROM ASIC/DSP)
MASTER CLOCK INPUT
OPTIONAL CLOCK OSCILLATOR OUTPUT
(FOR CRYSTAL APPLICATION)
DCLK OUTPUT
12
VSUB OUTPUT (TO CCD BIAS CIRCUIT)
DATA OUTPUTS
STROBE CONTROL OUTPUT
MECHANICAL SHUTTER CONTROL OUTPUT
0.1µF
0.1µF
REFB
REFT
A2
A3
DRVDD
DVDD1
+3V SUPPLY
0.1µF
L4
E1
0.1µF
0.1µF
CCDGND
CCDIN
A5
A6
0.1µF
CCD SIGNAL INPUT
DVDD2
DVDD2
DVDD2
K8
+3V SUPPLY
RG
HL
H4
H3
H2
H1
C10
B11
C11
D11
E11
L8
L7
TEST3
VDR_EN
NC
+3V SUPPLY
J8
V-DRIVER ENABLE
(FROM ASIC/DSP)
K11
K4
F11
6
H, RG OUTPUTS
(TO CCD)
AVSS
AVSS
AVSS
A1
B2
B3
B4
B5
A4
TEST1
J7
J5
TEST0
DRVSS
DVSS1
DVSS2
AD9923ABBCZ
(Not to Scale)
L5
AVSS
AVSS
AVSS
AVSS
F2
K9
A11
E6
G2
G3
H9
NC
B6
B7
B9
NC
NC
AVSS
TCVSS
NC
NC
RGVSS
HVSS
A10
D10
+3V ANALOG
SUPPLY
NC
NC
NC
NC
NC
NC
NC
NC
NC
J6
AVDD
J9
A7
B8
J10
J11
K7
L1
L6
L9
4.7µF
6.3V
0.1µF
TCVDD
0.1µF
+3V H,
RG SUPPLY
RGVDD
HVDD
B10
0.1µF
0.1µF
+3V H,
RG SUPPLY
L11
D9
NC = NOT INTERNALLY
CONNECTED
4.7µF
6.3V
VH SUPPLY
VL SUPPLY
+3V SUPPLY
0.1µF
2.2µF
25V
25V
4.7µF
10V
0.1µF
10V
0.1µF
15
VERTICAL OUTPUTS (TO CCD)
SUBCK OUTPUT (TO CCD)
Figure 83. Typical Circuit Configuration When Using Software Sync Function
Rev. 0 | Page 65 of 88
AD9923A
with 0s during the serial write operation. If fewer than 28 data
bits are written, the register is not updated with new data.
SERIAL INTERFACE TIMING
All of the AD9923A internal registers are accessed through a
3-wire serial interface. Each register consists of a 12-bit address
and a 28-bit data-word. Both the address and data-word are
written by starting with the LSB. To write to each register, a 40-bit
operation is required, as shown in Figure 84. Although many
data-words are fewer than 28 bits wide, all 28 bits must be
written for each register. For example, if the data-word is only
20 bits wide, the upper 8 bits are don’t cares and must be filled
Figure 85 shows a more efficient way to write to the registers,
using the AD9923A address auto-increment capability. Using
this method, the lowest desired address is written first, followed
by multiple 28-bit data-words. Each data-word is automatically
written to the address of the next highest register. By eliminating
the need to write each address, faster register loading is achieved.
Continuous write operations can start with any register location.
12-BIT ADDRESS
28-BIT DATA
SDI
A0
A1
A2
A3
A4
A5
A6
A8
A9 A10 A11 D0
D1
D2
D3
D25 D26 D27
A7
tDS
tDH
SCK
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
38
39
40
tLS
tLH
SL
NOTES
1. SDATA BITS ARE LATCHED ON SCK RISING EDGES. SCK MAY IDLE HIGH OR LOW BETWEEN WRITE OPERATIONS.
2. ALL 40 BITS MUST BE WRITTEN: 12 BITS FOR ADDRESS AND 28 BITS FOR DATA-WORD.
3. IF THE DATA-WORD IS <28 BITS, 0s MUST BE USED TO COMPLETE THE 28-BIT DATA-WORD 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 84. Serial Write Operation
DATA FOR STARTING
REGISTER ADDRESS
DATA FOR NEXT
REGISTER ADDRESS
SDI
A0
A1
A2
A10 A11 D0
D1
D26 D27 D0
D1
D26 D27 D0
A3
D1
D2
SCK
SL
70
71
1
2
3
4
11
12
13
14
39
40
41
42
67
68
69
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 IS LOADED.
Figure 85. Continuous Serial Write Operation
Rev. 0 | Page 66 of 88
AD9923A
number of V-sequences. Each V-sequence occupies 20 register
addresses, and each field occupies 12 register addresses.
LAYOUT OF INTERNAL REGISTERS
The AD9923A address space is divided into two register areas,
as illustrated in Figure 86. In the first area, Address 0x00 to
Address 0x91 contain the registers for the AFE, miscellaneous
functions, VD/HD parameters, timing core, CLPOB masking,
SG patterns, shutter functions, and memory configuration. The
second area of the address space, beginning at Address 0x400,
consists of the registers for the V-pattern groups, V-sequences,
and fields. This is a configurable set of registers; the user can
decide how many V-pattern groups, V-sequences, and fields are
used in a particular design. Therefore, the addresses for these
registers vary, depending on the number of V-patterns and
V-sequences chosen.
The starting address for a V-sequence is equal to 0x400 plus the
number of V-pattern groups times 40. The starting address for a
field is equal to the starting address of a V-sequence plus the
number of V-sequences times 20. The VPAT, VSEQ, and field
registers must occupy a continuous block of addresses.
Figure 87 shows an example with three V-pattern groups, four
V-sequences, and two fields. The starting address for the V-pattern
groups is 0x400. Because VPAT_NUM = 3, the V-pattern groups
occupy 120 address locations. The start of the V-sequence register
is 0x400 + 120 = 0x478. With VSEQ_NUM = 3, the V-sequences
occupy 60 address locations. Therefore, the field registers begin at
0x448 + 60 = 0x4B4.
Register 0x90 (VPAT_NUM) and Register 0x91 (VSEQ_NUM)
specify the total number of V-pattern groups and V-sequences
used. The starting address for the V-pattern groups is 0x400.
The starting address for a V-sequence is based on the number
of V-pattern groups used, with each V-pattern group occupying
40 register addresses. The starting address for a field register
depends on both the number of V-pattern groups and the
The AD9923A address space contains many unused addresses.
Undefined addresses between Address 0x00 and Address 0x399
should not be written to, or the AD9923A might operate
incorrectly. Continuous register writes should be performed
carefully to avoid writing to undefined registers.
CONFIGURABLE REGISTER DATA
FIXED REGISTER AREA
ADDR 0x00
ADDR 0x400
AFE REGISTERS
ADDR 0x10
MISCELLANEOUS REGISTERS
V-PATTERN GROUPS
(EACH GROUP USES 40 REGISTERS)
ADDR 0x20
VD/HD REGISTERS
ADDR 0x30
TIMING CORE REGISTERS
ADDR 0x40
VSEQ START
FIELD START
MAX 0x7FF
CLPOB MASK REGISTERS
ADDR 0x50
V-SEQUENCES
(EACH VSEQ USES 20 REGISTERS)
SG PATTERN REGISTERS
ADDR 0x60
SHUTTER CONTROL REGISTERS
ADDR 0x90
FIELDS
CONFIGURE MEMORY USING
VPAT_NUM AND VSEQ_NUM
ADDR 0x92
(EACH FIELD USES 12 REGISTERS)
INVALID DO NOT ACCESS
ADDR 0xFF
Figure 86. Layout of AD9923A Registers
ADDR 0x400
3 V-PATTERN GROUPS
(40 × 3 = 120 REGISTERS)
ADDR 0x478
ADDR 0x4B4
4 V-SEQUENCES
(20 × 3 = 60 REGISTERS)
2 FIELDS
(12 × 2 = 24 REGISTERS)
ADDR 0x4CC
MAX 0x7FF
UNUSED MEMORY
Figure 87. Example of Register Configuration
Rev. 0 | Page 67 of 88
AD9923A
•
SCP Updated—All V-pattern and V-sequence registers are
updated at the next SCP where they will be used. For
example, in Figure 88, this field has selected Region 1 to
use V-Sequence 3 for the vertical outputs; therefore, a write
to a V-Sequence 3 or V-pattern group register, which is
referenced by V-Sequence 3, is updated at SCP 1. If there
are multiple writes to the same register, only the last one
before SCP1 is updated. Likewise, a register write to a
V-Sequence 5 register is updated at SCP 2, and a register
write to a V-Sequence 8 register is updated at SCP 3.
UPDATING NEW REGISTER VALUES
The AD9923A internal registers are updated at different times,
depending on the particular register. Table 44 summarizes the
four types of register updates. The register listing (Table 45
through Table 57) also contain a column with update type to
identify when each register is updated:
•
SCK Updated—Some registers are updated when the 28th
data bit (D27) is written. These registers are used for
functions, such as power-up and reset, that do not require
gating with the next VD boundary.
Table 44. Register Update Locations
•
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
the UPDATE register (Address 0x18). This delays the
VD-updated register updates to any desired HD line in the
field. Note that the field registers are not affected by the
UPDATE register.
Update
Type
Description
SCK
Register is immediately updated when the 28th data
bit (D27) is written.
VD
Register is updated at the VD falling edge. VD
updated registers can be delayed further by using the
UPDATE register at Address 0x18. Field registers are
not affected by the UPDATE register.
Register is updated at the HD falling edge at the end
of the SG active line.
SG
SCP
Register is updated at the next SCP when the register
is used.
•
SG Updated—A few shutter registers are updated at the
HD falling edge at the end of an SG active line. These
registers control the SUBCK signal; therefore, the SUBCK
output is not updated until the SG line is complete.
SCK
VD
SG
UPDATED
SCP
UPDATED
UPDATED UPDATED
SERIAL
WRITE
VD
HD
SGLINE
VSG
USE VSEQ2
USE VSEQ3
REGION 1
USE VSEQ5
REGION 2
USE VSEQ8
REGION 3
V1A TO V10
REGION 0
SCP 0
SCP 0
SCP 1
SCP 2
SCP 3
Figure 88. Register Update Locations (See Table 44 for Definitions)
Rev. 0 | Page 68 of 88
AD9923A
COMPLETE REGISTER LISTING
When an address contains less than 28 data bits, all remaining bits must be written as 0s.
Table 45. AFE Registers
Address Data
Default
Value
Update
Type
(Hex)
Bits
Name
Description
00
[1:0]
3
SCK
STANDBY
Standby modes.
0: normal operation.
1: Standby 1 mode.
2: Standby 2 mode.
3: Standby 3 mode.
[2]
1
0
0
0
0
0
CLPENABLE
CLPSPEED
FASTUPDATE
PBLK_LVL
DCBYP
0: disable OB clamp.
1: enable OB clamp.
[3]
0: select normal OB clamp settling.
1: select fast OB clamp settling.
0: ignore VGA update.
1: very fast clamping when VGA is updated.
0: blank data outputs to 0 during PBLK.
1: blank data outputs to programmed clamp level during PBLK.
0: enable input dc-restore circuit during PBLK.
1: disable input dc-restore circuit during PBLK.
[4]
[5]
[6]
[15:11]
XSUBCNT_MUX Selects which internal signal is used for the XSUBCNT signal.
0: assign XV6 to XSUBCNT.
1: assign XV8 to XSUBCNT.
2: assign XV9 to XSUBCNT.
3: assign XV10 to XSUBCNT.
4: assign VSG5 to XSUBCNT.
5: assign VSG6 to XSUBCNT.
6: assign VSG7 to XSUBCNT.
7: assign VSG8 to XSUBCNT.
8: assign VSG2 to XSUBCNT.
9: assign VSG3 to XSUBCNT.
10: assign VSG4 to XSUBCNT.
11: assign VSG1 to XSUBCNT.
12: assign XV13 to XSUBCNT.
13: assign VSUB to XSUBCNT.
14: assign MSHUT to XSUBCNT.
15: assign STROBE to XSUBCNT.
16: assign XV1 to XSUBCNT.
17: assign XV2 to XSUBCNT.
18: assign XV3 to XSUBCNT.
19: assign XV4 to XSUBCNT.
20: assign XV5 to XSUBCNT.
21: assign XV7 to XSUBCNT.
22: assign XV11 to XSUBCNT.
23: assign XV12 to XSUBCNT.
24: assign SHUT0 to XSUBCNT.
25: assign SHUT1 to XSUBCNT.
26: assign SHUT2 to XSUBCNT.
27: assign SHUT3 to XSUBCNT.
28: assign FG_TRIG to XSUBCNT.
29: invalid setting.
30: tie XSUBCNT high.
31: tie XSUBCNT low.
Rev. 0 | Page 69 of 88
AD9923A
Address Data
Default
Value
Update
Type
(Hex)
Bits
Name
Description
01
[0]
0
0
0
SCK
DOUTDISABLE
0: data outputs are driven.
1: data outputs are three-stated.
0: latch data outputs using DOUT PHASE register setting.
1: output latch is transparent.
[1]
[2]
DOUTLATCH
GRAYEN
0: straight binary encoding of ADC digital output data.
1: enable gray encoding of ADC digital output data.
[3]
1
1
TEST
Set to 1.
04
[1:0]
VD
CDSGAIN
CDS gain setting.
0: −3 dB.
1: 0 dB.
2: +3 dB.
3: +6 dB.
05
06
[9:0]
[9:0]
F
VD
VD
VGAGAIN
VGA gain. 6 dB to 42 dB (0.035 dB per step).
Optical black clamp level. 0 LSB to 256 LSB (0.25 LSB per step).
1EC
CLAMPLEVEL
Rev. 0 | Page 70 of 88
AD9923A
Table 46. Miscellaneous Registers
Address Data
Default
Value
Update
Type
(Hex)
Bits
Name
Description
10
[0]
0
0
1
SCK
SW_RST
Software reset. Bit resets to 0.
1: reset Register 0x00 to Register 0x91 to default values.
0: make all outputs dc inactive.
11
12
[0]
[0]
VD
OUTCONTROL
SYNCENABLE
1: enable outputs at next VD edge.
SCK
0: configure Ball G7 as an output signal, determined by
Register 0x12, Bits[9:8].
1: external synchronization enable (configure Ball G7 as SYNC input).
Test mode only. Must be set to 0.
When SYNCENABLE = 0, selects which signal is output on the SYNC pin.
0: CLPOB.
[7:1]
[9:8]
0
0
TEST
OUTPUTPBLK
1: PBLK.
2: GPO (from Register 0x1A).
3: TESTOUT (from shutter registers).
SYNC active polarity.
0: active low.
1: active high.
13
14
15
16
17
[0]
[0]
[0]
[0]
0
0
0
0
SCK
SCK
SCK
SCK
SCK
SYNCPOL
SYNCSUSPEND
TGCORE_RSTB
OSC_RST
Suspends clocks during SYNC active pulse.
0: don’t suspend.
1: suspend.
Timing core reset bar.
0: reset TG core.
1: resume operation.
CLO oscillator reset.
0: oscillator in power-down state.
1: resume oscillator operation.
Test mode only. Must be set to 0.
Test mode only. Must be set to 0.
[7:0]
[8]
0
0
0
TEST1
TEST2
18
19
[11:0]
VD
UPDATE
Serial update line. Sets the HD line within the field to update the VD
updated registers.
[0]
[0]
0
SCK
PREVENTUP
GPO
Prevents the updating of the VD updated registers.
0: normal update.
1: prevent update of VD updated registers.
1A
0
VD
General-purpose output (GPO) value when SYNCENABLE = 0 and
OUTPUTPBLK = 2.
0: GPO is low at next VD edge.
1: GPO is high at next VD edge.
Table 47. VD/HD Registers
Address
(Hex)
Data
Bits
Default
Value
Update
Type
Name
Description
20
21
22
[0]
0
SCK
SCK
VD
MASTER
VD/HD master or slave mode.
0: slave mode.
1: master mode.
[0]
0
VDHDPOL
VD/HD active polarity.
0: low.
1: high.
[12:0]
[24:13]
0
0
HDRISE
VDRISE
Rising edge location for HD.
Rising edge location for VD.
Rev. 0 | Page 71 of 88
AD9923A
Table 48. Timing Core Registers
Address
(Hex)
Data
Bits
Default
Value
Update
Type
Name
Description
30
[0]
0
SCK
CLIDIVIDE
Divide CLI input frequency by 2.
0: no divide.
1: divide by 2.
31
[5:0]
[13:8]
[16]
0
20
1
SCK
H1POSLOC
H1NEGLOC
H1H2POL
H1 rising edge location.
H1 falling edge location.
H1/H2 polarity control.
0: inverse of convention in Figure 18.
1: no inversion.
32
33
34
35
[5:0]
[13:8]
[16]
0
20
1
SCK
SCK
SCK
VD
H3POSLOC
H3NEGLOC
H3H4POL
H3 rising edge location.
H3 falling edge location.
H3/H4 polarity control.
0: inverse of convention in Figure 18.
1: no inversion.
[5:0]
[13:8]
[16]
0
20
1
HLPOSLOC
HLNEGLOC
HLPOL
HL rising edge location.
HL falling edge location.
HL polarity control.
0: inverse of convention in Figure 18.
1: no inversion.
[5:0]
[13:8]
[16]
0
10
1
RGPOSLOC
RGNEGLOC
RGPOL
RG rising edge location.
RG falling edge location.
RG polarity control.
0: inverse of convention in Figure 18.
1: no inversion.
[0]
0
H1H2RETIME Retime HBLK for H1/H2 to the internal H1 clock. The preferred setting is 1,
which adds one cycle of delay to the HBLK toggle positions.
0: no retime.
1: retime.
[1]
[2]
[3]
0
0
0
H3H4RETIME Retime HBLK for H3/H4 to the internal H3 clock.
HLRETIME
HLHBLKEN
Retime HBLK for HL to the internal HL clock.
Enable HBLK for HL output.
0: disable.
1: enable.
[6:4]
0
HBLKWIDTH
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, it doubles the H1 to H4 pulse width.
2: 1/4 pixel frequency.
3: 1/6 pixel frequency.
4: 1/8 pixel frequency.
5: 1/10 pixel frequency.
6: 1/12 pixel frequency.
7: 1/14 pixel frequency.
H1 drive strength.
0: off.
36
[3:0]
1
SCK
H1DRV
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.
Rev. 0 | Page 72 of 88
AD9923A
Address
(Hex)
Data
Bits
Default
Value
Update
Type
Name
Description
[7:4]
1
1
1
1
1
H2DRV
H3DRV
H4DRV
HLDRV
RGDRV
SHPLOC
SHDLOC
H2 drive strength.
H3 drive strength.
H4 drive strength.
HL drive strength.
RG drive strength.
SHP sample location.
SHD sample location.
[11:8]
[15:12]
[19:16]
[23:20]
[5:0]
37
38
24
0
SCK
SCK
[13:8]
[5:0]
0
DOUTPHASE DOUT (internal signal) phase control.
[7:6]
0
Unused
Must be set to 0.
[8]
0
DCLKMODE
DCLK mode.,
0: DCLK tracks DOUT phase.
1: DCLK phase is fixed.
[10:9]
[11]
2
0
DOUTDELAY Data output delay (tOD) with respect to DCLK rising edge.
0: no delay.
1: ~4 ns.
2: ~8 ns.
3: ~12 ns.
DCLKINV
Invert DCLK output.
0: no inversion.
1: inversion of DCLK.
Table 49. CLPOB and PBLK Masking Registers
Address
(Hex)
Data
Bits
Update
Type
Default Value
Name
Description
40
41
42
43
44
45
[11:0]
[12]
[24:13]
[11:0]
[12]
[24:13]
[11:0]
[12]
[24:13]
[11:0]
[12]
[24:13]
[11:0]
[12]
[24:13]
[11:0]
[12]
FFF
0
FFF
FFF
0
FFF
FFF
0
FFF
FFF
0
FFF
FFF
0
FFF
FFF
0
VD
VD
VD
VD
VD
VD
CLPOBMASKSTART1
Unused
CLPOBMASKEND1
CLPOBMASKSTART2
Unused
CLPOBMASKEND2
CLPOBMASKSTART3
Unused
CLPOBMASKEND3
PBLKMASKSTART1
Unused
PBLKMASKEND1
PBLKMASKSTART12
Unused
PBLKMASKEND2
PBLKMASKSTART3
Unused
CLPOB Masking Start Line 1.
Must be set to 0.
CLPOB Masking End Line 1.
CLPOB Masking Start Line 2.
Must be set to 0.
CLPOB Masking End Line 2.
CLPOB Masking Start Line 3.
Must be set to 0.
CLPOB Masking End Line 3.
PBLK Masking Start Line 1.
Must be set to 0.
PBLK Masking End Line 1.
PBLK Masking Start Line 2.
Must be set to 0.
PBLK Masking End Line 2.
PBLK Masking Start Line 3.
Must be set to 0.
[24:13]
FFF
PBLKMASKEND3
PBLK Masking End Line 3.
Rev. 0 | Page 73 of 88
AD9923A
Table 50. SG Pattern Registers
Address Data
Default Update
(Hex)
Bits
Value
Type
Name
Description
50
[0]
1
VD
SGPOL_0
Start polarity for SGPattern 0.
0: low.
1: high.
[1]
[2]
[3]
[4]
[5]
[6]
[7]
1
1
1
1
1
1
1
SGPOL_1
SGPOL_2
SGPOL_3
SGPOL_4
SGPOL_5
SGPOL_6
SGPOL_7
Start polarity for SGPattern 1.
Start polarity for SGPattern 2.
Start polarity for SGPattern 3.
Start polarity for SGPattern 4.
Start polarity for SGPattern 5.
Start polarity for SGPattern 6.
Start polarity for SGPattern 7.
Pattern 0. Toggle Position 1.
Pattern 0. Toggle Position 2.
Pattern 1. Toggle Position 1.
Pattern 1. Toggle Position 2.
Pattern 2. Toggle Position 1.
Pattern 2. Toggle Position 2.
Pattern 3. Toggle Position 1.
Pattern 3. Toggle Position 2.
Pattern 4. Toggle Position 1.
Pattern 4. Toggle Position 2.
Pattern 5. Toggle Position 1.
Pattern 5. Toggle Position 2.
Pattern 6. Toggle Position 1.
Pattern 6. Toggle Position 2.
Pattern 7. Toggle Position 1.
Pattern 7. Toggle Position 2.
51
52
53
54
55
56
57
58
59
[12:0]
1FFF
VD
VD
VD
VD
VD
VD
VD
VD
SGTOG1_0
SGTOG2_0
SGTOG1_1
SGTOG2_1
SGTOG1_2
SGTOG2_2
SGTOG1_3
SGTOG2_3
SGTOG1_4
SGTOG2_4
SGTOG1_5
SGTOG2_5
SGTOG1_6
SGTOG2_6
SGTOG1_7
SGTOG2_7
SGMASK_BYP
[25:13] 1FFF
[12:0] 1FFF
[25:13] 1FFF
[12:0] 1FFF
[25:13] 1FFF
[12:0] 1FFF
[25:13] 1FFF
[12:0] 1FFF
[25:13] 1FFF
[12:0] 1FFF
[25:13] 1FFF
[12:0] 1FFF
[25:13] 1FFF
[12:0] 1FFF
[25:13] 1FFF
[7:0]
0
SCK
SCK
SGMASK override. These values override the VSG mask value located in the
field registers.
[8]
0
SGMASK_BYP_EN SGMASK override enable. Must be set to 1 to enable override.
Rev. 0 | Page 74 of 88
AD9923A
Table 51. Shutter Control Registers
Address Data
Default
Value
Update
Type
(Hex)
Bits
Name
Description
60
[2:0]
0
1
2
3
0
VD
VSUB_CTRL
Selects which internal signal is used for the VSUB output pin.
0: use SHUT0 parameters (Register 0x06D to Register 0x071).
1: use SHUT1 parameters (Register 0x072 to Register 0x076).
2: use SHUT2 parameters (Register 0x077 to Register 0x07B).
3: use SHUT3 parameters (Register 0x07C to Register 0x080).
4: use VSUB0_MUX output.
5: use VSUB1_MUX output.
6: invalid setting.
7: use SHUT1_SHUT2_MUX output.
See Register 0xEB, Bits[15,13:12] for VSUB0_MUX, VSUB1_MUX,
and SHUT1_SHUT2_MUX.
[5:3]
[8:6]
[11:9]
[7:0]
MSHUT_CTRL
STROBE_CTRL
TESTO_CTRL
TRIGGER
Selects which internal signal is used for the MSHUT output pin.
0: use SHUT0 parameters.
1: use SHUT1 parameters.
2: use SHUT2 parameters.
3: use SHUT3 parameters.
4: use VSUB0_MUX output.
5: use VSUB1_MUX output.
6: invalid setting.
7: use SHUT1_SHUT2_MUX output.
See Register 0xEB, Bits[15,13:12] for VSUB0_MUX, VSUB1_MUX, and
SHUT1_SHUT2_MUX.
Selects which internal signal is used for the STROBE output pin.
0: use SHUT0 parameters.
1: use SHUT1 parameters.
2: use SHUT2 parameters.
3: use SHUT3 parameters.
4: use VSUB0_MUX output.
5: use VSUB1_MUX output.
6: invalid setting.
7: use SHUT1_SHUT2_MUX output.
See Register 0xEB, Bits[15,13:12] for VSUB0_MUX, VSUB1_MUX, and
SHUT1_SHUT2_MUX.
Selects which internal signal is used for the TESTO signal.
0: use SHUT0 parameters.
1: use SHUT1 parameters.
2: use SHUT2 parameters.
3: use SHUT3 parameters.
4: use VSUB0_MUX output.
5: use VSUB1_MUX output.
6: invalid setting.
7: use SHUT1_SHUT2_MUX output.
See Register 0xEB, Bits[15,13:12] for VSUB0_MUX, VSUB1_MUX, and
SHUT1_SHUT2_MUX.
61
VD
Trigger for exposure/readout operation. Set bits high to trigger.
[0]: SHUT0.
[1]: SHUT1.
[2]: SHUT2.
[3]: SHUT3.
[4]: VSUB0.
[5]: VSUB1.
Rev. 0 | Page 75 of 88
AD9923A
Address Data
Default
Value
Update
Type
(Hex)
Bits
Name
Description
[6]: EXPOSURE.
[7]: READOUT.
Note that if EXPOSURE and READOUT are triggered together,
READOUT occurs immediately after the exposure is complete.
62
63
[2:0]
2
0
VD
VD
READOUTNUM
EXPOSURENUM
Number of fields to suppress the SUBCK pulses during READOUT.
[11:0]
Number of fields to suppress the SUBCK and VSG pulses during
exposure.
[12]
0
VDHDOFF
Disable VD and HD during exposure.
0: enable.
1: disable.
64
65
[11:0]
[23:12]
[1:0]
0
0
0
SG
SG
SUBSUPPRESS
SUBCKNUM
SUBCKMASK
Number of SUBCK pulses to suppress after VSG line.
Number of SUBCK pulses per field.
Additional masking of SUBCK output.
0: no mask.
1: begin mask on VD edge.
2: mask using internal SHUT3 signal.
3: same as 1 and 2 (1 has priority).
SUBCK pulse start polarity.
First SUBCK Pulse Toggle Position 1.
First SUBCK Pulse Toggle Position 2.
Second SUBCK Pulse Toggle Position 1.
Second SUBCK Pulse Toggle Position 2.
VSUB0 readout mode.
66
67
[0]
1
SG
SG
SUBCKPOL
[12:0]
[25:13]
[12:0]
[25:13]
[0]
1FFF
1FFF
1FFF
1FFF
0
SUBCK1TOG1
SUBCK1TOG2
SUBCK2TOG1
SUBCK2TOG2
VSUB0_MODE
68
69
SG
VD
0: Mode 0.
1: Mode 1.
[1]
0
VSUB0_KEEPON VSUB0 keep-on mode.
0: turn VUB0 off after READOUT or at next VD.
1: keep VSUB0 active beyond READOUT, until reset to 0.
6A
6B
[11:0]
[12]
[13]
[0]
0
0
1
0
VD
VD
VSUB0_ON
Unused
VSUB0POL
VSUB1_MODE
VSUB0 on position.
Must be set to 0.
VSUB0 start polarity.
VSUB1 readout mode.
0: Mode 0.
1: Mode 1.
[1]
0
VSUB1_KEEPON VSUB1 keep on mode.
1: keep VSUB1 active beyond readout.
VSUB1 on position.
6C
6D
[11:0]
[12]
[13]
[0]
0
0
1
0
VD
VD
VSUB1_ON
Unused
Must be set to 0.
VSUB1POL
SHUT0_ON
VSUB1 start polarity.
SHUT0 manual control of signal.
0: off.
1: on.
[1]
[2]
1
0
SHUT0POL
SHUT0_MAN
SHUT0 active polarity. 1: on state produces high output.
SHUT0 manual control enable.
0: disable.
1: enable manual control.
6E
6F
[11:0]
[11:0]
[12]
[25:13]
[11:0]
0
0
0
0
0
VD
VD
SHUT0_ON_FD
SHUT0_ON_LN
Unused
SHUT0_ON_PX
SHUT0_OFF_FD
SHUT0 field on position. Ignored during manual or nonshutter mode.
SHUT0 line on position.
Must be set to 0.
VD
VD
SHUT0 pixel on position.
70
SHUT0 field off position. Ignored during manual or nonshutter mode.
Rev. 0 | Page 76 of 88
AD9923A
Address Data
Default
Value
Update
Type
(Hex)
Bits
Name
Description
71
[11:0]
[12]
[25:13]
[0]
0
0
0
0
VD
SHUT0_OFF_LN
Unused
SHUT0_OFF_PX
SHUT1_ON
SHUT0 line off position.
Must be set to 0.
SHUT0 pixel off position.
SHUT1 manual control of signal.
0: off.
VD
VD
72
1: on.
[1]
[2]
1
0
SHUT1POL
SHUT1_MAN
SHUT1 active polarity. 1 = on state produces high output.
SHUT1 manual control enable.
0: disable.
1: enable manual control.
73
74
[11:0]
[11:0]
[12]
[25:13]
[11:0]
[11:0]
[12]
0
0
0
0
0
0
0
0
0
VD
VD
SHUT1_ON_FD
SHUT1_ON_LN
Unused
SHUT1_ON_PX
SHUT1_OFF_FD
SHUT1_OFF_LN
Unused
SHUT1 field on position. Ignored during manual or nonshutter mode.
SHUT1 line on position.
Must be set to 0.
SHUT1 pixel on position.
VD
VD
VD
75
76
SHUT1 field off position. Ignored during manual or nonshutter mode.
SHUT1 line off position.
Must be set to 0.
SHUT1 pixel off position.
[25:13]
[0]
VD
VD
SHUT1_OFF_PX
SHUT2_ON
77
SHUT2 manual control of signal.
0: off.
1: on.
[1]
[2]
1
0
SHUT2POL
SHUT2_MAN
SHUT2 active polarity. 1: on state produces high output.
SHUT2 manual control enable.
0: disable.
1: enable manual control.
78
79
[11:0]
[11:0]
[12]
[25:3]
[11:0]
[11:0]
[12]
0
0
0
0
0
0
0
0
0
VD
VD
SHUT2_ON_FD
SHUT2_ON_LN
Unused
SHUT2_ON_PX
SHUT2_OFF_FD
SHUT2_OFF_LN
Unused
SHUT2 field on position. Ignored during manual or nonshutter mode.
SHUT2 line on position.
Must be set to 0.
SHUT2 pixel on position.
VD
VD
VD
7A
7B
SHUT2 field off position. Ignored during manual or nonshutter mode.
SHUT2 line off position.
Must be set to 0.
SHUT2 pixel off position.
[25:13]
[0]
VD
VD
SHUT2_OFF_PX
SHUT3_ON
7C
SHUT3 manual control of signal.
0: off.
1: on.
[1]
[2]
1
0
SHUT3POL
SHUT3_MAN
SHUT3 active polarity. 1: on state produces high output.
SHUT3 manual control enable.
0: disable.
1: enable manual control.
7D
7E
[11:0]
[11:0]
[12]
[25:13]
[11:0]
[11:0]
[12]
0
0
0
0
0
0
0
0
VD
VD
SHUT3_ON_FD
SHUT3_ON_LN
Unused
SHUT3_ON_PX
SHUT3_OFF_FD
SHUT3_OFF_LN
Unused
SHUT3 field on position. Ignored during manual or nonshutter mode.
SHUT3 line on position.
Must be set to 0.
VD
VD
VD
SHUT3 pixel on position.
7F
80
SHUT3 field off position. Ignored during manual or nonshutter mode.
SHUT3 line off position.
Must be set to 0.
[25:13]
VD
SHUT3_OFF_PX
SHUT3 pixel off position.
Rev. 0 | Page 77 of 88
AD9923A
Table 52. Memory Configuration Registers
Address Data
Default
Value
(Hex)
Bits
[4:0]
[4:0]
Update Name
Description
90
0
0
VD
VD
VPAT_NUM
VSEQ_NUM
Total number of V-pattern groups.
Total number of V-sequences.
91
Table 53. Standby Polarity, Shutter Mux, and FG_TRIG Registers
Address Data
Default
Value
(Hex)
Bits
Update Name
SCK STANDBY3POL
Description
E2
[24:0]
0
Programmable polarities for vertical and shutter outputs during Standby 3.
[0] = XV1 polarity.
[1] = XV2 polarity.
[2] = XV3 polarity.
[3] = XV4 polarity.
[4] = XV5 polarity.
[5] = XV6 polarity.
[6] = XV7 polarity.
[7] = XV8 polarity.
[8] = XV9 polarity.
[9] = XV10 polarity.
[10] = XV11 polarity.
[11] = XV12 polarity.
[12] = XV13 polarity.
[13] = VSG1 polarity.
[14] = VSG2 polarity.
[15] = VSG3 polarity.
[16] = VSG4 polarity.
[17] = VSG5 polarity.
[18] = VSG6 polarity.
[19] = VSG7 polarity.
[20] = VSG8 polarity.
[21] = XSUBCK polarity.
[22] = VSUB polarity.
Note: controls polarity for Standby 1, Standby 2, Standby 3, or if
OUTCONTROL = low.
[23] = MSHUT polarity.
Note: controls polarity for Standby 1, Standby 2, Standby 3, or if
OUTCONTROL = low.
[24] = STROBE polarity.
Note: controls polarity for Standby 1, Standby 2, Standby 3, or if
OUTCONTROL = low.
E6
[0]
0
SCK
VCNT_RUN
0: counters behave the same as AD9923 in sweep region.
1: enables additional toggles after last repeat of sweep region.
Required start-up register; must be set to 0x60
Test register.
0: use VSUB0, 1: Use SHUT0 ^ VSUB0.
0: use VSUB1, 1: Use SHUT0 ^ VSUB1.
Test register. Set to 0.
EA
EB
[9:0]
[11:0]
[12]
[13]
[14]
[15]
0
SCK
SCK
SCK
SCK
SCK
SCK
TEST3
300
0
0
0
0
TEST4
VSUB0_MUX
VSUB1_MUX
TEST5
SHUT1_SHUT2_MUX 0: use SHUT0 ^ SHUT1.
1: Use SHUT0 ^ SHUT2.
F1
[3:0]
0
SCK
FG_TRIGEN
FG_TRIG operation enable and field count selection.
[2:0] Selects field count for pulse (based on mode field counter).
[3] = 1 to enable FG_TRIG signal output.
Rev. 0 | Page 78 of 88
AD9923A
Address Data
Default
Value
(Hex)
Bits
Update Name
Description
F3
[21:0]
3FE000 SCK
STANDBY12POL
Programmable polarities for V-outputs and XSUBCK during Standby 1,
Standby 2, or if OUTCONTROL = low.
[0] = XV1 polarity.
[1] = XV2 polarity.
[2] = XV3 polarity.
[3] = XV4 polarity.
[4] = XV5 polarity.
[5] = XV6 polarity.
[6] = XV7 polarity.
[7] = XV8 polarity.
[8] = XV9 polarity.
[9] = XV10 polarity.
[10] = XV11 polarity.
[11] = XV12 polarity.
[12] = XV13 polarity.
[13] = VSG1 polarity.
[14] = VSG2 polarity.
[15] = VSG3 polarity.
[16] = VSG4 polarity.
[17] = VSG5 polarity.
[18] = VSG6 polarity.
[19] = VSG7 polarity.
[20] = VSG8 polarity.
[21] = XSUBCK polarity.
Table 54. Mode Register: VD Updated
Address (Binary)
12b10_xx_xxxx_xxxx
(Set A11, A10 = 10)
Data Bits
[37:0]
Default Value
Description
0
A11, A10 set to 10, remaining A9 to A0 bits used for D37:D28.
Number of fields (maximum of seven).
Selected field for Field 7.
Selected field for Field 6.
Selected field for Field 5.
Selected field for Field 4.
Selected field for Field 3.
Selected field for Field 2.
[37:35]
[34:30]
[29:25]
[24:20]
[19:15]
[14:10]
[9:5]
[4:0]
Selected field for Field 1.
Rev. 0 | Page 79 of 88
AD9923A
Unused XV-channels must have toggle positions programmed to maximum values. For example, if XV1 to XV8 are used, XV9 to XV12
must have all toggle positions set to maximum values. This prevents unpredictable behavior because the default values are unknown.
Table 55. V-Pattern Group 0 (VPAT0) Registers
Address
(Hex)
Default
Value
Update
Type
Data Bits
[12:0]
Name
Description
00
01
02
03
04
05
06
07
08
09
0A
0B
0C
0D
0E
0F
10
11
12
13
14
15
16
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
XV1TOG1
XV1TOG2
XV1TOG3
XV1TOG4
XV2TOG1
XV2TOG2
XV2TOG3
XV2TOG4
XV3TOG1
XV3TOG2
XV3TOG3
XV3TOG4
XV4TOG1
XV4TOG2
XV4TOG3
XV4TOG4
XV5TOG1
XV5TOG2
XV5TOG3
XV5TOG4
XV6TOG1
XV6TOG2
XV6TOG3
XV6TOG4
XV7TOG1
XV7TOG2
XV7TOG3
XV7TOG4
XV8TOG1
XV8TOG2
XV8TOG3
XV8TOG4
XV9TOG1
XV9TOG2
XV9TOG3
XV9TOG4
XV10TOG1
XV10TOG2
XV10TOG3
XV10TOG4
XV11TOG1
XV11TOG2
XV11TOG3
XV11TOG4
XV12TOG1
XV12TOG2
XV1 Toggle Position 1.
XV1 Toggle Position 2.
XV1 Toggle Position 3.
XV1 Toggle Position 4.
XV2 Toggle Position 1.
XV2 Toggle Position 2.
XV2 Toggle Position 3.
XV2 Toggle Position 4.
XV3 Toggle Position 1.
XV3 Toggle Position 2.
XV3 Toggle Position 3.
XV3 Toggle Position 4.
XV4 Toggle Position 1.
XV4 Toggle Position 2.
XV4 Toggle Position 3.
XV4 Toggle Position 4.
XV5 Toggle Position 1.
XV5 Toggle Position 2.
XV5 Toggle Position 3.
XV5 Toggle Position 4.
XV6 Toggle Position 1.
XV6 Toggle Position 2.
XV6 Toggle Position 3.
XV6 Toggle Position 4.
XV7 Toggle Position 1.
XV7 Toggle Position 2.
XV7 Toggle Position 3.
XV7 Toggle Position 4.
XV8 Toggle Position 1.
XV8 Toggle Position 2.
XV8 Toggle Position 3.
XV8 Toggle Position 4.
XV9 Toggle Position 1.
XV9 Toggle Position 2.
XV9 Toggle Position 3.
XV9 Toggle Position 4.
XV10 Toggle Position 1.
XV10 Toggle Position 2.
XV10 Toggle Position 3.
XV10 Toggle Position 4.
XV11 Toggle Position 1.
XV11 Toggle Position 2.
XV11 Toggle Position 3.
XV11 Toggle Position 4.
XV12 Toggle Position 1.
XV12 Toggle Position 2.
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
Rev. 0 | Page 80 of 88
AD9923A
Address
(Hex)
Default
Value
Update
Type
Data Bits
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
Name
Description
17
18
19
1A
1B
1C
1D
1E
1F
20
21
22
23
24
25
26
27
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
SCP
SCP
SCP
SCP
XV12TOG3
XV12TOG4
XV1TOG5
XV1TOG6
XV2TOG5
XV2TOG6
XV3TOG5
XV3TOG6
XV4TOG5
XV4TOG6
XV5TOG5
XV5TOG6
XV6TOG5
XV6TOG6
XV7TOG5
XV7TOG6
XV8TOG5
XV8TOG6
XV9TOG5
XV9TOG6
XV10TOG5
XV10TOG6
XV11TOG5
XV11TOG6
XV12TOG5
XV12TOG6
XV13TOG1
XV13TOG2
XV13TOG3
XV13TOG4
XV13TOG5
XV13TOG6
Unused
XV12 Toggle Position 3.
XV12 Toggle Position 4.
XV1 Toggle Position 5.
XV1 Toggle Position 6.
XV2 Toggle Position 5.
XV2 Toggle Position 6.
XV3 Toggle Position 5.
XV3 Toggle Position 6.
XV4 Toggle Position 5.
XV4 Toggle Position 6.
XV5 Toggle Position 5.
XV5 Toggle Position 6.
XV6 Toggle Position 5.
XV6 Toggle Position 6.
XV7 Toggle Position 5.
XV7 Toggle Position 6.
XV8 Toggle Position 5.
XV8 Toggle Position 6.
XV9 Toggle Position 5.
XV9 Toggle Position 6.
XV10 Toggle Position 5.
XV10 Toggle Position 6.
XV11 Toggle Position 5.
XV11 Toggle Position 6.
XV12 Toggle Position 5.
XV12 Toggle Position 6.
XV13 Toggle Position 1.
XV13 Toggle Position 2.
XV13 Toggle Position 3.
XV13 Toggle Position 4.
XV13 Toggle Position 5.
XV13 Toggle Position 6.
Must be set to 0.
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
SCP
[25:13]
[25:0]
Table 56. V-Sequence Registers
Address
(Hex)
Default
Value
Update
Type
Data Bits
[0]
[1]
[2]
[4:3]
Name
Description
00
Undefined
Undefined
Undefined
Undefined
SCP
CLPOBPOL
PBLKPOL
HOLD
CLPOB start polarity.
PBLK start polarity.
HOLD function.
Enable masking of V-outputs.
0: no mask.
VMASK
1: enable Freeze1/Resume1.
2: enable Freeze2/Resume2.
3: enable both Freeze1/Resume1 and Freeze2/Resume2.
Enable HBLK alternation.
Must be set to 0.
[7:5]
[12:8]
[25:13]
Undefined
Undefined
Undefined
HBLKALT
Unused
HDLEN
HD line length (number of pixels in the line).
Rev. 0 | Page 81 of 88
AD9923A
Address
(Hex)
Default
Value
Update
Type
Data Bits
Name
Description
01
[0]
[1]
[2]
[3]
[4]
[5]
[6]
[7]
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
SCP
XV1POL
XV2POL
XV3POL
XV4POL
XV5POL
XV6POL
XV7POL
XV8POL
XV1 start polarity.
XV2 start polarity.
XV3 start polarity.
XV4 start polarity.
XV5 start polarity.
XV6 start polarity.
XV7 start polarity.
XV8 start polarity.
[8]
[9]
XV9POL
XV9 start polarity.
XV10POL
XV11POL
XV12POL
XV13POL
XV1POL2
XV2POL2
XV3POL2
XV4POL2
XV5POL2
XV6POL2
XV7POL2
XV8POL2
XV9POL2
XV10POL2
XV11POL2
XV12POL2
XV13POL13
GROUPSEL
TWO_GROUP
XV10 start polarity.
XV11 start polarity.
XV12 start polarity.
XV13 start polarity.
XV1 second polarity.
XV2 second polarity.
XV3 second polarity.
XV4 second polarity.
XV5 second polarity.
XV6 second polarity.
XV7 second polarity.
XV8 second polarity.
XV9 second polarity.
XV10 second polarity.
XV11 second polarity.
XV12 second polarity.
XV13 second polarity.
Select between Group A and Group B. 0: Group A, 1: Group B.
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[12:0]
[13]
02
SCP
1: use all Group A and Group B toggle positions for single V-
pattern.
[18:14]
Undefined
VPATSELB
Selected V-pattern Group B or Special V-pattern Second
position.
[23:19]
[25:24]
Undefined
Undefined
VPATSELA
VPATA_MODE
Selected V-pattern Group A.
Number of alternation repeats.
0: disable alternation, use VREPA_1 for all lines.
1: 2-line alternation.
2: 3-line alternation.
3: 4-line alternation.
03
[12:0]
Undefined
SCP
VSTARTB
Start position of selected V-pattern Group B, or start
position of special V-pattern.
[25:13]
[12:0]
[25:13]
[11:0]
[12]
[24:13]
[11:0]
[12]
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
VLENB
Length of selected V-pattern Group B.
Start position of selected V-pattern Group A.
Length of selected V-pattern Group A.
Number of repetitions for V-pattern Group B for odd lines.
Must be set to 0.
Number of repetitions for V-pattern Group B for even lines.
Number of repetitions for V-pattern Group A for first lines.
Must be set to 0.
04
05
SCP
SCP
VSTARTA
VLENA
VREPB_ODD
Unused
VREPB_EVEN
VREPA_1
Unused
06
SCP
[24:13]
VREPA_2
Number of repetitions for V-pattern Group A for second
lines.
Rev. 0 | Page 82 of 88
AD9923A
Address
(Hex)
Default
Value
Update
Type
Data Bits
Name
Description
07
08
09
[12:0]
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
SCP
SCP
SCP
VREPA_3
Number of repetitions for V-pattern Group A for third lines,
or first HBLK toggle position for odd lines.
Number of repetitions for V-pattern Group A for fourth
lines, or second HBLK toggle position for odd lines.
[25:13]
[12:0]
VEPA_4
FREEZE1
RESUME1
FREEZE2
RESUME2
Holds the XV1 to XV13 outputs at their current levels, or
third HBLK toggle position for odd lines.
Resumes the operation of XV1 to XV13 outputs to finish the
pattern, or fourth HBLK toggle position for odd lines.
[25:13]
[12:0]
Holds the XV1 to XV13 outputs at their current levels, or
fifth HBLK toggle position for odd lines.
Resumes the operation of XV1 to XV13 outputs to finish the
pattern, or sixth HBLK toggle position for odd lines.
[25:13]
0A
0B
0C
[12:0]
[25:13]
[12:0]
[25:13]
[12:0]
Undefined
Undefined
Undefined
Undefined
Undefined
SCP
SCP
SCP
HBLKTOGE1
HBLKTOGE2
HBLKTOGE3
HBLKTOGE4
HBLKSTART
First HBLK toggle position for even lines.
Second HBLK toggle position for even lines.
Third HBLK toggle position for even lines.
Fourth HBLK toggle position for even lines.
Start location for HBLK in Alternation Mode 4 to Alternation
Mode 7, or fifth HBLK toggle position for even lines.
[25:13]
[12:0]
Undefined
Undefined
Undefined
HBLKEND
HBLKLEN
HBLKREP
End location for HBLK in Alternation Mode 4 to Alternation
Mode 7, or sixth HBLK toggle position for even lines.
0D
SCP
HBLK length in HBLK Alternation Mode 4 to Alternation
Mode 7.
Number of HBLK repetitions in HBLK Alternation Mode 4 to
Alternation Mode 7.
[20:13]
[21]
[22]
[23]
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
HBLKMASK_H1
HBLKMASK_H3
HBLKMASK_HL
CLPOBTOG1
CLPOBTOG2
PBLKTOG1
PBLKTOG2
UNUSED
Masking polarity for H1 during HBLK.
Masking polarity for H3 during HBLK.
Masking polarity for HL during HBLK.
CLPOB Toggle Position 1.
CLPOB Toggle Position 2.
PBLK Toggle Position 1.
0E
0F
[12:0]
[25:13]
[12:0]
[25:13]
[25:0]
[11:0]
[12]
SCP
SCP
PBLK Toggle Position 2.
10
11
SCP
SCP
Must be set to 0.
SPXV_ACT
UNUSED
SPXV_EN
Special XV-pattern active line.
Must be set to 0.
Special XV-pattern enable (active high).
Must be set to 0.
[13]
12
13
[25:0]
[25:0]
SCP
SCP
UNUSED
UNUSED
Must be set to 0.
Rev. 0 | Page 83 of 88
AD9923A
Table 57. Field Registers
Address
(Hex)
Default
Value
Update
Type
Data Bits
[4:0]
[9:5]
[14:10]
[19:15]
[24:20]
[4:0]
Name
SEQ0
SEQ1
SEQ2
SEQ3
Description
00
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
VD
Selected V-sequence for first region in the field.
Selected V-sequence for second region in the field.
Selected V-sequence for third region in the field .
Selected V-sequence for fourth region in the field.
Selected V-sequence for fifth region in the field.
Selected V-sequence for sixth region in the field.
Selected V-sequence for seventh region in the field.
Selected V-sequence for eighth region in the field.
Selected V-sequence for ninth region in the field.
Enables multiplier mode and/or sweep mode for Region 0.
0: multiplier off/sweep off.
SEQ4
01
02
VD
VD
SEQ5
SEQ6
SEQ7
SEQ8
[9:5]
[14:10]
[19:15]
[1:0]
MULT_SWEEP0
1: multiplier off/sweep on.
2: multiplier on/sweep off.
3: multiplier on/sweep on.
[3:2]
[5:4]
[7:6]
[9:8]
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
Undefined
MULT_SWEEP1
MULT_SWEEP2
MULT_SWEEP3
MULT_SWEEP4
MULT_SWEEP5
MULT_SWEEP6
MULT_SWEEP7
MULT_SWEEP8
SCP0
Enables multiplier mode and/or sweep mode for Region 1.
Enables multiplier mode and/or sweep mode for Region 2.
Enables multiplier mode and/or sweep mode for Region 3.
Enables multiplier mode and/or sweep mode for Region 4.
Enables multiplier mode and/or sweep mode for Region 5.
Enables multiplier mode and/or sweep mode for Region 6.
Enables multiplier mode and/or sweep mode for Region 7.
Enables multiplier mode and/or sweep mode for Region 8.
V-Sequence Change Position 0.
[11:10]
[13:12]
[15:14]
[17:16]
[11:0]
[12]
[24:13]
[11:0]
[12]
[24:13]
[11:0]
[12]
[24:13]
[11:0]
[12]
[24:13]
[11:0]
[12]
[24:13]
[12:0]
[25:13]
[4:0]
[20:5]
[23:0]
[11:0]
[12]
03
04
05
06
07
VD
VD
VD
VD
VD
Unused
SCP1
Must be set to 0.
V-Sequence Change Position 1.
SCP2
Unused
SCP3
V-Sequence Change Position 2.
Must be set to 0.
V-Sequence Change Position 3.
SCP4
Unused
SCP5
V-Sequence Change Position 4.
Must be set to 0.
V-Sequence Change Position 5.
SCP6
Unused
SCP7
V-Sequence Change Position 6.
Must be set to 0.
V-Sequence Change Position 7.
SCP8
Unused
VDLEN
V-Sequence Change Position 8.
Must be set to 0.
VD field length (number of lines in the field).
HD last line length. Line length of last line in the field.
Start position for second V-pattern on SG active line.
Selected second V-pattern group for SG active line.
Masking of VSG outputs during SG active line.
Selection of VSG patterns for each VSG output.
SG Active Line 1.
08
09
VD
VD
HDLAST
VSTARTSECOND
VPATSECOND
SGMASK
0A
0B
VD
VD
SGPATSEL
SGACTLINE1
Unused
Must be set to 0.
SG Active Line 2.
[24:13]
SGACTLINE2
Rev. 0 | Page 84 of 88
AD9923A
OUTLINE DIMENSIONS
A1 CORNER
INDEX AREA
8.00
BSC SQ
11 10
9
8
7
6
5
4
3
2
1
A
B
C
D
E
F
A1 BALL
CORNER
6.50
BSC SQ
TOP VIEW
G
H
J
K
L
BOTTOM
VIEW
0.65 BSC
DETAIL A
*1.40
1.31
1.16
DETAIL A
0.91 MIN
0.25 MIN
0.10 MAX
COPLANARITY
0.45
0.40
0.35
SEATING
PLANE
BALL DIAMETER
*
COMPLIANT TO JEDEC STANDARDS MO-225
WITH THE EXCEPTION TO PACKAGE HEIGHT.
Figure 89. 105-Lead Chip Scale Package Ball Grid Array [CSP_BGA]
(BC-105)
Dimensions shown in millimeters
ORDERING GUIDE
Model
Temperature Range
Package Description
Package Option
AD9923ABBCZ1
AD9923ABBCZRL1
−25°C to +85°C
−25°C to +85°C
105-Lead Chip Scale Package Ball Grid Array [CSP_BGA]
105-Lead Chip Scale Package Ball Grid Array [CSP_BGA]
BC-105
BC-105
1 Z = Pb-free part.
Rev. 0 | Page 85 of 88
AD9923A
NOTES
Rev. 0 | Page 86 of 88
AD9923A
NOTES
Rev. 0 | Page 87 of 88
AD9923A
NOTES
©2006 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
D06415-0-10/06(0)
Rev. 0 | Page 88 of 88
相关型号:
AD9928BBCZ
Dual-Channel, 14-Bit CCD Signal Processor with V-Driver and Precision Timing Generator
ADI
AD9928BBCZRL
Dual-Channel, 14-Bit CCD Signal ocessor with V-Driver and Precision Timing Generator
ADI
AD9929BBCZRL
SPECIALTY CONSUMER CIRCUIT, PBGA64, 9 X 9 MM, LEAD FREE, PLASTIC, MO-205-AB, CSBGA-64
ROCHESTER
©2020 ICPDF网 联系我们和版权申明