Datasheet AD9948 Datasheet (Analog Devices)

10-Bit CCD Signal Processor with

FEATURES
Correlated Double Sampler (CDS)
0 dB to 18 dB Pixel Gain Amplifier (

PxGA

®

)
6 dB to 42 dB 10-Bit Variable Gain Amplifier (VGA)
10-Bit 25 MSPS A/D Converter
Black Level Clamp with Variable Level Control
Complete On-Chip Timing Driver
Precision Timing Core with 800 ps Resolution
On-Chip 3 V Horizontal and RG Drivers
40-Lead LFCSP Package

APPLICATIONS
Digital Still Cameras
High Speed Digital Imaging Applications

Precision Timing

™

Core

AD9948

GENERAL DESCRIPTION

The AD9948 is a highly integrated CCD signal processor for
digital still camera applications. Specified at pixel rates of up to
25 MHz, the AD9948 consists of a complete analog front end
with A/D conversion, combined with a programmable timing
driver. The Precision Timing core allows adjustment of high
speed clocks with 800 ps resolution.

The analog front end includes black level clamping, CDS, PxGA,
VGA, and a 25 MHz 10-bit A/D converter. The timing driver
provides the high speed CCD clock drivers for RG and H1–H4.
Operation is programmed using a 3-wire serial interface.

Packaged in a space-saving 40-lead LFCSP package, the
AD9948 is specified over an operating temperature range of
–20°C to +85°C.

CCDIN

RG

H1–H4

4

AD9948

CDS

HORIZONTAL

DRIVERS

FUNCTIONAL BLOCK DIAGRAM

REFB

REFT

V

0dB TO 18dB

PxGA

6dB TO 42dB

VGA

INTERNAL

CLOCKS

PRECISION

TIMING

CORE

SYNC

GENERATOR

HD VD

REF

CLAMP

SL

10-BIT

ADC

INTERNAL

REGISTERS

10

DOUT

HBLK

CLP/PBLK

CLI

SDATASCK

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. No license is granted by implication or otherwise
under any patent or patent rights of Analog Devices. Trademarks and
registered trademarks are the property of their respective companies.

One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: [phone] www.analog.com
Fax: [phone] © 2003 Analog Devices, Inc. All rights reserved.

AD9948–SPECIFICATIONS

GENERAL SPECIFICATIONS

Parameter Min Typ Max Unit

TEMPERATURE RANGE

Operating –20 +85 °C
Storage –65 +150 °C

MAXIMUM CLOCK RATE 25 MHz

POWER SUPPLY VOLTAGE

AVDD, TCVDD (AFE, Timing Core) 2.7 3.0 3.6 V
HVDD (H1–H4 Drivers) 2.7 3.0 3.6 V
RGVDD (RG Driver) 2.7 3.0 3.6 V
DRVDD (D0–D9 Drivers) 2.7 3.0 3.6 V
DVDD (All Other Digital) 2.7 3.0 3.6 V

POWER DISSIPATION

25 MHz, HVDD = RGVDD = 3 V, 100 pF H1–H4 Loading* 220 mW
Total Shutdown Mode 1 mW

*The total power dissipated by the HVDD supply may be approximated using the equation

Total HVDD Power C HVDD Pixel Frequency HVDD Number of H Outputs Used

Reducing the H-loading, using only two of the outputs, and/or using a lower HVDD supply will reduce the power dissipation.

Specifications subject to change without notice.

=×× ×× −()()

LOAD

DIGITAL SPECIFICATIONS

(T

to T

MIN

, AVDD = DVDD = DRVDD = HVDD = RGVDD = 2.7 V, CL = 20 pF, unless otherwise noted.)

MAX

Parameter Symbol Min Typ Max Unit

LOGIC INPUTS

High Level Input Voltage V
Low Level Input Voltage V
High Level Input Current I
Low Level Input Current I
Input Capacitance C

IH

IL

IH

IL

IN

2.1 V

0.6 V
10 µA
10 µA
10 pF

LOGIC OUTPUTS

High Level Output Voltage, IOH = 2 mA V
Low Level Output Voltage, IOL = 2 mA V

OH

OL

2.2 V

0.5 V

CLI INPUT

High Level Input Voltage

(TCVDD/2 + 0.5 V) V

Low Level Input Voltage V

IH–CLI

IL–CLI

1.85 V

0.85 V

RG AND H-DRIVER OUTPUTS

High Level Output Voltage

(RGVDD – 0.5 V and HVDD – 0.5 V) V

Low Level Output Voltage V

OH

OL

2.2 V

0.5 V

Maximum Output Current (Programmable) 30 mA
Maximum Load Capacitance 100 pF

Specifications subject to change without notice.

REV. 0–2–

AD9948

(T

to T

, AVDD = DVDD = 3.0 V, f

MAX

ANALOG SPECIFICATIONS

MIN

unless otherwise noted.)

Parameter Min Typ Max Unit Notes

CDS

Gain 0 dB
Allowable CCD Reset Transient* 500 mV
Max Input Range before Saturation* 1.0 V p-p
Max CCD Black Pixel Amplitude* ±50 mV

PIXEL GAIN AMPLIFIER (PxGA)

Gain Control Resolution 256 Steps
Gain Monotonicity

Min Gain 0 dB
Max Gain 18 dB

VARIABLE GAIN AMPLIFIER (VGA)

Max Input Range 1.0 V p-p
Max Output Range 2.0 V p-p
Gain Control Resolution 1024 Steps
Gain Monotonicity Guaranteed
Gain Range

Min Gain (VGA Code 0) 6 dB
Max Gain (VGA Code 1023) 42 dB

BLACK LEVEL CLAMP

Clamp Level Resolution 256 Steps
Clamp Level Measured at ADC output

Min Clamp Level (0) 0 LSB
Max Clamp Level (255) 63.75 LSB

A/D CONVERTER

Resolution 10 Bits
Differential Nonlinearity (DNL) –1.0 ± 0.5 +1.0 LSB
No Missing Codes Guaranteed
Full-Scale Input Voltage 2.0 V

VOLTAGE REFERENCE

Reference Top Voltage (REFT) 2.0 V
Reference Bottom Voltage (REFB) 1.0 V

SYSTEM PERFORMANCE Specifications include entire

VGA Gain Accuracy

Min Gain (Code 0) 5.0 5.5 6.0 dB

Max Gain (Code 1023) 40.5 41.5 42.5 dB
Peak Nonlinearity, 500 mV Input Signal 0.2 % 12 dB gain applied
Total Output Noise 0.25 LSB rms AC grounded input, 6 dB

Power Supply Rejection (PSR) 50 dB Measured with step change

*Input signal characteristics defined as follows:

= 25 MHz, Typical Timing Specifications,

CLI

signal chain

gain applied

on supply

500mV TYP

RESET TRANSIENT

Specifications subject to change without notice.

50mV MAX

OPTICAL BLACK PIXEL

1V MAX

INPUT SIGNAL RANGE

REV. 0

–3–

AD9948

TIMING SPECIFICATIONS

(CL = 20 pF, f

= 25 MHz, Serial Timing in Figure 3, unless otherwise noted.)

CLI

Parameter Symbol Min Typ Max Unit

MASTER CLOCK (CLI) (See Figure 4)

CLI Clock Period t
CLI High/Low Pulsewidth t

CLI

ADC

40 ns
16 20 24 ns

Delay from CLI to Internal Pixel

Period Position t

CLPOB Pulsewidth (Programmable)* t

CLIDLY

COB

220 Pixels

6ns

SAMPLE CLOCKS (See Figure 6)

SHP Rising Edge to SHD Rising Edge t

S1

17 20 ns

DATA OUTPUTS (See Figures 7a and 7b)

Output Delay From Programmed Edge t

OD

6ns

Pipeline Delay 11 Cycles

SERIAL INTERFACE

Maximum SCK Frequency f
SL to SCK Setup Time t
SCK to SL Hold Time t
SDATA Valid to SCK Rising Edge Setup t
SCK Falling Edge to SDATA Valid Hold t
SCK Falling Edge to SDATA Valid Read t

*Minimum CLPOB pulsewidth is for functional operation only. Wider typical pulses are recommended to achieve low noise clamp reference.

Specifications subject to change without notice.

SCLK

LS

LH

DS

DH

DV

10 MHz
10 ns
10 ns
10 ns
10 ns
10 ns

ABSOLUTE MAXIMUM RATINGS*

With

Parameter Respect To Min Max Unit

AVDD, TCVDD AVSS –0.3 +3.9 V
HVDD, RGVDD HVSS, RGVSS –0.3 +3.9 V
DVDD, DRVDD DVSS, DRVSS –0.3 +3.9 V
Any VSS Any VSS –0.3 +0.3 V
Digital Outputs DRVSS –0.3 DRVDD + 0.3 V
CLPOB/PBLK, HBLK DVSS –0.3 DVDD + 0.3 V
SCK, SL, SDATA DVSS –0.3 DVDD + 0.3 V
RG RGVSS –0.3 RGVDD + 0.3 V
H1–H4 HVSS –0.3 HVDD + 0.3 V
REFT, REFB, CCDIN AVSS –0.3 AVDD + 0.3 V
Junction Temperature 150 °C
Lead Temperature (10 sec) 300 °C

*Stresses above those listed under Absolute Maximum Ratings may cause permanent damage to the device. This is a stress

rating only; functional operation of the device at these or any other conditions above those listed in the operational sections
of this specification is not implied. Exposure to absolute maximum rating conditions for extended periods may affect
device reliability.

ORDERING GUIDE

Temperature Package Package

Model Range Description Option

AD9948KCP –20°C to +85°C LFCSP CP-40
AD9948KCPRL –20°C to +85°C LFCSP CP-40

THERMAL CHARACTERISTICS

Thermal Resistance

40-Lead LFCSP Package

␪

= 27°C/W*

JA

*␪

is measured using a 4-layer PCB with the exposed paddle

JA

soldered to the board.

AD9948KCPZ* –20°C to +85°C LFCSP CP-40
AD9948KCPZRL* –20°C to +85°C LFCSP CP-40

*This is a lead free product.

CAUTION

ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily
accumulate on the human body and test equipment and can discharge without detection. Although
the AD9948 features proprietary ESD protection circuitry, permanent damage may occur on
devices subjected to high energy electrostatic discharges. Therefore, proper ESD precautions are
recommended to avoid performance degradation or loss of functionality.

REV. 0–4–

PIN CONFIGURATION

40 NC

38 HBLK

37 DVDD

36 DVSS

35 HD

34 VD

32 SDI

39 CLP/PBLK

33 SCK

31 SL

AD9948

NC 1

(LSB) D0 2

D1 3

D2 4

DRVSS 5

DRVDD 6

D3 7

D4 8

D5 9

D6 10

PIN 1
IDENTIFIER

D7 11

D8 12

AD9948

TOP VIEW

H1 14

H2 15

HVSS 16

(MSB) D9 13

H3 18

HVDD 17

30 REFB

29 REFT

28 AVSS
27 CCDIN

26 AVDD

25 CLI

24 TCVDD

23 TCVSS

22 RGVDD

21 RG

H4 19

RGVSS 20

PIN FUNCTION DESCRIPTIONS

Pin No. Mnemonic Type* Description

2–4 D0–D2 DO Data Outputs (D0 is LSB)
5 DRVSS P Digital Driver Ground
6 DRVDD P Digital Driver Supply
7–13 D3–D9 DO Data Outputs (D9 is MSB)
14 H1 DO CCD Horizontal Clock 1
15 H2 DO CCD Horizontal Clock 2
16 HVSS P H1–H4 Driver Ground
17 HVDD P H1–H4 Driver Supply
18 H3 DO CCD Horizontal Clock 3
19 H4 DO CCD Horizontal Clock 4
20 RGVSS P RG Driver Ground
21 RG DO CCD Reset Gate Clock
22 RGVDD P RG Driver Supply
23 TCVSS P Analog Ground for Timing Core
24 TCVDD P Analog Supply for Timing Core
25 CLI DI Master Clock Input
26 AVDD P Analog Supply for AFE
27 CCDIN AI Analog Input for CCD Signal (Connect through Series 0.1 µF Capacitor)
28 AVSS P Analog Ground for AFE
29 REFT AO Reference Top Decoupling (Decouple with 1.0 µF to AVSS)
30 REFB AO Reference Bottom Decoupling (Decouple with 1.0 µF to AVSS)
31 SL DI 3-Wire Serial Load
32 SDI DI 3-Wire Serial Data Input
33 SCK DI 3-Wire Serial Clock
34 VD DI Vertical Sync Pulse
35 HD DI Horizontal Sync Pulse
36 DVSS P Digital Ground
37 DVDD P Digital Supply
38 HBLK DI Optional HBLK Input
39 CLP/PBLK DO CLPOB or PBLK Output
1, 40 NC Not Internally Connected

*Type: AI = Analog Input, AO = Analog Output, DI = Digital Input, DO = Digital Output, P = Power.

REV. 0

–5–

AD9948

TERMINOLOGY
Differential Nonlinearity (DNL)

An ideal ADC exhibits code transitions that are exactly 1 LSB
apart. DNL is the deviation from this ideal value. Thus every
code must have a finite width. No missing codes guaranteed to
10-bit resolution indicates that all 1024 codes, respectively,
must be present over all operating conditions.

Peak Nonlinearity

Peak nonlinearity, a full signal chain specification, refers to the
peak deviation of the output of the AD9948 from a true straight
line. The point used as zero scale occurs 0.5 LSB before the
first code transition. Positive full scale is defined as a level 1 LSB
and 0.5 LSB beyond the last code transition. The deviation is
measured from the middle of each particular output code to the
true straight line. The error is then expressed as a percentage of
the 2 V ADC full-scale signal. The input signal is always appro­priately gained up to fill the ADC’s full-scale range.

EQUIVALENT CIRCUITS

AVDD

R

Total Output Noise

The rms output noise is measured using histogram techniques.
The standard deviation of the ADC output codes is calculated in
LSB, and represents the rms noise level of the total signal chain
at the specified gain setting. The output noise can be converted
to an equivalent voltage, using the relationship

12

LSB (ADC full scale/ codes)=

n

where n is the bit resolution of the ADC. For the AD9948,
1LSB is approximately 1.95 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.

DVD D

330

AVSS

AVSS

Circuit 1. CCDIN (Pin 27)

AVDD

330

CLI

25k

1.4V

AVSS

Circuit 2. CLI (Pin 25)

DVSS DRVDD

DATA

THREE-

STATE

DVSS DRVSS

Circuit 3. Data Outputs D0–D9 (Pins 2–4, 7–13)

DOUT

DVSS

Circuit 4. Digital Inputs (Pins 31–35, 38)

HVDD or RGVDD

DATA

ENABLE

HVSS or RGVSS

Circuit 5. H1–H4 and RG (Pins 14, 15, 18, 19, 21)

OUTPUT

REV. 0–6–

1.0

0.5

0

DNL (LSB)

–0.5

Typical Performance Characteristics–AD9948

–1.0

0

200 600 800

400

ADC OUTPUT CODE

TPC 1. Typical DNL

10

7.5

5.0

OUTPUT NOISE (LSB)

2.5

0

0

200

400

VGA GAIN CODE (LSB)

600 800

TPC 2. Output Noise vs. VGA Gain

275

1000

1000

REV. 0

250

225

200

175

150

POWER DISSIPATION (mW)

125

100

10

VDD = 3.3V

VDD = 3.0V

VDD = 2.7V

15

SAMPLE RATE (MHz)

20

TPC 3. Power Curves

25

–7–

AD9948

SYSTEM OVERVIEW

V-DRIVER

V1–Vx, VSG1–VSGx, SUBCK

H1–H4, RG

DOUT

CCD

CCDIN

INTEGRATED

SERIAL

INTERFACE

AD9948

AFE + TD

HD, VD

CLI

DIGITAL IMAGE

PROCESSING

ASIC

Figure 1. Typical Application

Figure 1 shows the typical system application diagram for the
AD9948. The CCD output is processed by the AD9948’s AFE
circuitry, which consists of a CDS, a PxGA, a VGA, a black level
clamp, and an A/D converter. The digitized pixel information is
sent to the digital image processor chip, where all postprocessing
and compression occurs. To operate the CCD, CCD timing
parameters are programmed into the AD9948 from the image
processor through the 3-wire serial interface. From the system
master clock, CLI, provided by the image processor, the AD9948

generates the high speed CCD clocks and all internal AFE clocks.
All AD9948 clocks are synchronized with VD and HD. All of
the AD9948’s horizontal pulses (CLPOB, PBLK, and HBLK)
are programmed and generated internally.

The H-drivers for H1–H4 and RG are included in the AD9948,
allowing these clocks to be connected directly to the CCD.
H-drive voltage of 3 V is supported in the AD9948.

Figure 2a shows the horizontal and vertical counter dimensions
for the AD9948. All internal horizontal clocking is programmed
using these dimensions to specify line and pixel locations.

MAXIMUM FIELD DIMENSIONS

12-BIT HORIZONTAL = 4096 PIXELS MAX

12-BIT VERTICAL = 4096 LINES MAX

Figure 2a. Vertical and Horizontal Counters

VD

HD

CLI

MAX VD LENGTH IS 4095 LINES

MAX HD LENGTH IS 4095 PIXELS

Figure 2b. Maximum VD/HD Dimensions

REV. 0–8–

AD9948

SERIAL INTERFACE TIMING

All of the internal registers of the AD9948 are accessed through
a 3-wire serial interface. Each register consists of an 8-bit address
and a 24-bit data-word. Both the 8-bit address and 24-bit data­word are written starting with the LSB. To write to each register,
a 32-bit operation is required, as shown in Figure 3a. Although
many registers are less than 24 bits wide, all 24 bits must be
written for each register. If the register is only 16 bits wide, then
the upper eight bits are don’t cares and may be filled with zeros
during the serial write operation. If fewer than 24 bits are written,
the register will not be updated with new data.

Figure 3b shows a more efficient way to write to the registers by
using the AD9948’s address auto-increment capability. Using
this method, the lowest desired address is written first, followed
by multiple 24-bit data-words. Each new 24-bit data-word will
be written automatically to the next highest register address. By
eliminating the need to write each 8-bit address, faster register
loading is achieved. Address auto-increment may be used start­ing with any register location, and may be used to write to as
few as two registers or as many as the entire register space.

8-BIT ADDRESS

COMPLETE REGISTER LISTING

All addresses and default values are expressed in hexadecimal.

All registers are VD/HD updated as shown in Figure 3a, except
for the registers indicated in Table I, which are SL updated.

Table I. SL-Updated Registers

Register Description

OPRMODE AFE Operation Modes
CTLMODE AFE Control Modes
SW_RESET Software Reset Bit
TGCORE _RSTB Reset Bar Signal for Internal TG Core
PREVENTUPDATE Prevents Update of Registers
VDHDEDGE VD/HD Active Edge
FIELDVAL Resets Internal Field Pulse
HBLKRETIME Retimes the HBLK to Internal Clock
CLPBLKOUT CLP/BLK Output Pin Select
CLPBLKEN Enables CLP/BLK Output Pin
H1CONTROL H1/H2 Polarity Control
RGCONTROL H1 Positive Edge Location
DRVCONTROL H1 Negative Edge Location
SAMPCONTROL H1 Drive Current
DOUTPHASE H2 Drive Current

24-BIT DATA

SDATA

SCK

SDATA

SCK

A0 A1 A2 A4 A5 A6 A7

t

DS

1 32234567891011 12 30 31

t

LS

SL

VD

HD

NOTES

1. INDIVIDUAL SDATA BITS ARE LATCHED ON SCK RISING EDGES.

2. ALL 32 BITS MUST BE WRITTEN: 8 BITS FOR ADDRESS AND 24 BITS FOR DATA.

3. IF THE REGISTER LENGTH IS <24 BITS, THEN DON’T CARE BITS MUST BE USED TO COMPLETE THE 24-BIT DATA LENGTH.

4. NEW DATA IS UPDATED AT EITHER THE SL RISING EDGE OR AT THE HD FALLING EDGE AFTER THE NEXT VD FALLING EDGE.

5. VD/HD UPDATE POSITION MAY BE DELAYED TO ANY HD FALLING EDGE IN THE FIELD USING THE UPDATE REGISTER.

A3

t

DH

D1 D2 D3 D21 D22 D23

D0

...

...

...
...

...

t

LH

SL UPDATED

Figure 3a. Serial Write Operation

DATA FOR STARTING

REGISTER ADDRESS

A0 A1 A2 A4 A5 A6 A7 D0 D1 D22 D23

1 322345678910 31

A3

...

...

DATA FOR NEXT

REGISTER ADDRESS

D0 D1 D22 D23

...

...

3433 5655

VD/HD UPDATED

D0

D2D1

585759

...

...

REV. 0

SL

NOTES

1. MULTIPLE SEQUENTIAL REGISTERS MAY BE LOADED CONTINUOUSLY.

2. THE FIRST (LOWEST ADDRESS) REGISTER ADDRESS IS WRITTEN, FOLLOWED BY MULTIPLE 24-BIT DATA-WORDS.

3. THE ADDRESS WILL AUTOMATICALLY INCREMENT WITH EACH 24-BIT DATA-WORD (ALL 24 BITS MUST BE WRITTEN).

4. SL IS HELD LOW UNTIL THE LAST DESIRED REGISTER HAS BEEN LOADED.

5. NEW DATA IS UPDATED AT EITHER THE SL RISING EDGE OR AT THE HD FALLING EDGE AFTER THE NEXT VD FALLING EDGE.

...

...

Figure 3b. Continuous Serial Write Operation

–9–

...

AD9948

Table II. AFE Register Map

Data Bit Default

Address Content Value Name Description

00 [11:0] 4 OPRMODE AFE Operation Modes. (See Table VIII.)

01 [9:0] 0 VGAGAIN VGA Gain.

02 [7:0] 80 CLAMP LEVEL Optical Black Clamp Level.

03 [11:0] 4 CTLMODE AFE Control Modes. (See Table IX.)

04 [17:0] 0 PxGA GAIN01 PxGA Gain Registers for Color 0 [8:0] and Color 1 [17:9].

05 [17:0] 0 PxGA GAIN23 PxGA Gain Registers for Color 2 [8:0] and Color 3 [17:9].

Table III. Miscellaneous Register Map

Data Bit Default

Address Content Value Name Description

10 [0] 0 SW_RST Software Reset.

1 = Reset all registers to default, then self-clear back to 0.

11 [0] 0 OUT_CONTROL Output Control.

0 = Make all dc outputs inactive.

12 [0] 0 TGCORE_RSTB Timing Core Reset Bar.

0 = Reset TG core.
1 = Resume operation.

13 [11:0] 0 UPDATE Serial Update.

Sets the line (HD) within the field to update serial data.

14 [0] 0 PREVENTUPDATE Prevents the update of the VD-Updated Registers.

1 = Prevent update.

15 [0] 0 VDHDEDGE VD/HD Active Edge.

0 = Falling edge triggered.
1 = Rising edge triggered.

16 [1:0] 0 FIELDVAL Field Value Sync.

0 = Next Field 0.
1 = Next Field 1.
2/3 = Next Field 2.

17 [0] 0 HBLKRETIME Retime HBLK to Internal H1 Clock.

Preferred setting is 1. Setting to 1 will add one cycle delay to HBLK
toggle positions.

18 [1:0] 0 CLPBLKOUT CLP/BLK Pin Output Select.

0 = CLPOB.
1 = PBLK.
2 = HBLK.
3 = Low.

19 [0] 1 CLPBLKEN Enable CLP/BLK Output.

1 = Enable.

1A [0] 0 TEST MODE Internal Test Mode.

Should always be set low.

REV. 0–10–

AD9948

Table IV. CLPOB Register Map

Data Bit Default

Address Content Value (Hex) Name Description

20 [3:0] F CLPOBPOL Start Polarities for CLPOB Sequences 0, 1, 2, and 3.

21 [23:0] FFFFFF CLPOBTOG_0 Sequence 0. Toggle Position 1 [11:0] and Toggle Position 2 [23:12].

22 [23:0] FFFFFF CLPOBTOG_1 Sequence 1. Toggle Position 1 [11:0] and Toggle Position 2 [23:12].

23 [23:0] FFFFFF CLPOBTOG_2 Sequence 2. Toggle Position 1 [11:0] and Toggle Position 2 [23:12].

24 [23:0] FFFFFF CLPOBTOG_3 Sequence 3. Toggle Position 1 [11:0] and Toggle Position 2 [23:12].

0 CLPOBSCP0 CLPOB Sequence-Change-Position 0 (Hard-Coded to 0).

25 [7:0] 0 CLPOBSPTR CLPOB Sequence Pointers for Region 0 [1:0], 1 [3:2], 2 [5:4], 3 [7:6].

26 [11:0] FFF CLPOBSCP1 CLPOB Sequence-Change-Position 1.

27 [11:0] FFF CLPOBSCP2 CLPOB Sequence-Change-Position 2.

28 [11:0] FFF CLPOBSCP3 CLPOB Sequence-Change-Position 3.

Table V. PBLK Register Map

Data Bit Default

Address Content Value (Hex) Name Description

30 [3:0] F PBLKPOL Start Polarities for PBLK Sequences 0, 1, 2, and 3.

31 [23:0] FFFFFF PBLKTOG_0 Sequence 0. Toggle Position 1 [11:0] and Toggle Position 2 [23:12].

32 [23:0] FFFFFF PBLKTOG_1 Sequence 1. Toggle Position 1 [11:0] and Toggle Position 2 [23:12].

33 [23:0] FFFFFF PBLKTOG_2 Sequence 2. Toggle Position 1 [11:0] and Toggle Position 2 [23:12].

34 [23:0] FFFFFF PBLKTOG_3 Sequence 3. Toggle Position 1 [11:0] and Toggle Position 2 [23:12].

0 PBLKSCP0 PBLK Sequence-Change-Position 0 (Hard-Coded to 0).

35 [7:0] 0 PBLKSPTR PBLK Sequence Pointers for Region 0 [1:0], 1 [3:2], 2 [5:4], 3 [7:6].

36 [11:0] FFF PBLKSCP1 PBLK Sequence-Change-Position 1.

37 [11:0] FFF PBLKSCP2 PBLK Sequence-Change-Position 2.

38 [11:0] FFF PBLKSCP3 PBLK Sequence-Change-Position 3.

REV. 0

–11–

AD9948

Table VI. HBLK Register Map

Data Bit Default

Address Content Value (Hex) Name Description

40 [0] 0 HBLKDIR HBLK Internal/External.

0 = Internal.
1 = External.

41 [0] 0 HBLKPOL HBLK External Active Polarity.

0 = Active Low.
1 = Active High.

42 [0] 1 HBLKEXTMASK HBLK External Masking Polarity.

0 = Mask H1 Low.
1 = Mask H1High.

43 [3:0] F HBLKMASK HBLK Internal Masking Polarity.

0 = Mask H1 Low.
1 = Mask H1 High.

44 [23:0] FFFFFF HBLKTOG12_0 Sequence 0. Toggle Position 1 [11:0] and Toggle Position 2 [23:12].

45 [23:0] FFFFFF HBLKTOG34_0 Sequence 0. Toggle Position 3 [11:0] and Toggle Position 4 [23:12].

46 [23:0] FFFFFF HBLKTOG56_0 Sequence 0. Toggle Position 5 [11:0] and Toggle Position 6 [23:12].

47 [23:0] FFFFFF HBLKTOG12_1 Sequence 1. Toggle Position 1 [11:0] and Toggle Position 2 [23:12].

48 [23:0] FFFFFF HBLKTOG34_1 Sequence 1. Toggle Position 3 [11:0] and Toggle Position 4 [23:12].

49 [23:0] FFFFFF HBLKTOG56_1 Sequence 1. Toggle Position 5 [11:0] and Toggle Position 6 [23:12].

4A [23:0] FFFFFF HBLKTOG12_2 Sequence 2. Toggle Position 1 [11:0] and Toggle Position 2 [23:12].

4B [23:0] FFFFFF HBLKTOG34_2 Sequence 2. Toggle Position 3 [11:0] and Toggle Position 4 [23:12].

4C [23:0] FFFFFF HBLKTOG56_2 Sequence 2. Toggle Position 5 [11:0] and Toggle Position 6 [23:12].

4D [23:0] FFFFFF HBLKTOG12_3 Sequence 3. Toggle Position 1 [11:0] and Toggle Position 2 [23:12].

4E [23:0] FFFFFF HBLKTOG34_3 Sequence 3. Toggle Position 3 [11:0] and Toggle Position 4 [23:12].

4F [23:0] FFFFFF HBLKTOG56_3 Sequence 3. Toggle Position 5 [11:0] and Toggle Position 6 [23:12].

0 HBLKSCP0 HBLK Sequence-Change-Position 0 (Hard-coded to 0).

50 [7:0] 0 HBLKSPTR HBLK Sequence Pointers for Region 0 [1:0], 1 [3:2], 2 [5:4], 3 [7:6].

51 [11:0] FFF HBLKSCP1 HBLK Sequence-Change-Position 1.

52 [11:0] FFF HBLKSCP2 HBLK Sequence-Change-Position 2.

53 [11:0] FFF HBLKSCP3 HBLK Sequence-Change-Position 3.

Table VII. H1–H2, RG, SHP, SHD Register Map

Data Bit Default

Address Content Value Name Description

60 [12:0] 01001 H1CONTROL H1 Signal Control. Polarity [0] (0 = Inversion, 1 = No Inversion).

H1 Positive Edge Location [6:1].
H1 Negative Edge Location [12:7].

61 [12:0] 00801 RGCONTROL RG Signal Control. Polarity [0] (0 = Inversion, 1 = No Inversion).

RG Positive Edge Location [6:1].
RG Negative Edge Location [12:7].

62 [14:0] 0 DRVCONTROL Drive Strength Control for H1 [2:0], H2 [5:3], H3 [8:6], H4 [11:9], and

RG [14:12].
Drive Current Values: 0 = Off, 1 = 4.3 mA, 2 = 8.6 mA, 3 = 12.9 mA,
4 = 17.2 mA, 5 = 21.5 mA, 6 = 25.8 mA, 7 = 30.1 mA.

63 [11:0] 00024 SAMPCONTROL SHP/SHD Sample Control. SHP Sampling Location [5:0].

SHD Sampling Location [11:6].

64 [5:0] 0 DOUTPHASE DOUT Phase Control.

REV. 0–12–

Table VIII. AFE Operation Register Detail

Data Bit Default

Address Content Value Name Description

00 [1:0] 0 PWRDOWN 0 = Normal Operation.

1 = Reference Standby.
2/3 = Total Power-Down.

[2] 1 CLPENABLE 0 = Disable OB Clamp.

1 = Enable OB Clamp.

[3] 0 CLPSPEED 0 = Select Normal OB Clamp Settling.

1 = Select Fast OB Clamp Settling.

[4] 0 FASTUPDATE 0 = Ignore VGA Update.

1 = Very Fast Clamping when VGA Is Updated.

[5] 0 PBLK_LVL DOUT Value during PBLK.

0 = Blank to Zero.

1 = Blank to Clamp Level.
[7:6] 0 TEST MODE Test Operation Only. Set to zero.
[8] 0 DCBYP 0 = Enable DC Restore Circuit.

1 = Bypass DC Restore Circuit during PBLK.
[9] 0 TESTMODE Test Operation Only. Set to zero.
[11:10] 0 CDSGAIN Adjustment of CDS Gain.

0 = 0 dB.

01= –2 dB.

10 = –4 dB.

11 = 0 dB.

AD9948

Table IX. AFE Control Register Detail

Data Bit Default

Address Content Value Name Description

04 [1:0] 0 COLORSTEER 0 = Off.

1 = Progressive.

2 = Interlaced.

3 = Three Field.
[2] 1 PXGAENABLE 0 = Disable PxGA.

1 = Enable PxGA.
[3] 0 DOUTDISABLE 0 = Data Outputs Are Driven.

1 = Data Outputs Are Three-Stated.
[4] 0 DOUTLATCH 0 = Latch Data Outputs with DOUT Phase.

1 = Output Latch Transparent.
[5] 0 GRAYENCODE 0 = Binary Encode Data Outputs.

1= Gray Encode Data Outputs.

REV. 0

–13–

AD9948

PRECISION TIMING HIGH SPEED TIMING GENERATION

The AD9948 generates flexible high speed timing signals using
the Precision Timing core. This core is the foundation for gener­ating the timing used for both the CCD and the AFE; the reset
gate RG, horizontal drivers H1–H4, and the SHP/SHD sample
clocks. 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.

Timing Resolution

The Precision Timing core uses a 1× master clock input (CLI)
as a reference. This clock should be the same as the CCD pixel
clock frequency. Figure 4 illustrates how the internal timing core
divides the master clock period into 48 steps or edge positions.
Therefore, the edge resolution of the Precision Timing core is

/48). For more information on using the CLI input, refer to

(t

CLI

the Applications Information section.

POSITION

CLI

1 PIXEL
PERIOD

t

CLIDLY

P[0] P[48] = P[0]

P[12] P[24] P[36]

...

High Speed Clock Programmability

Figure 5 shows how the high speed clocks, RG, H1–H4, SHP,
and SHD, are generated. The RG pulse has programmable rising
and falling edges, and may be inverted using the polarity control.
The horizontal clocks H1 and H3 have programmable rising and
falling edges, and polarity control. The H2 and H4 clocks are
always inverses of H1 and H3, respectively. Table X summarizes
the high speed timing registers and their parameters.

Each edge location setting is 6 bits wide, but only 48 valid edge
locations are available. Therefore, the register values are mapped
into four quadrants, with each quadrant containing 12 edge
locations. Table XI shows the correct register values for the
corresponding edge locations.

...

NOTES

1. PIXEL CLOCK PERIOD IS DIVIDED INTO 48 POSITIONS, PROVIDING FINE EDGE RESOLUTION FOR HIGH SPEED CLOCKS.

2. THERE IS A FIXED DELAY FROM THE CLI INPUT TO THE INTERNAL PIXEL PERIOD POSITIONS (

t

CLIDLY

= 6 ns TYP).

Figure 4. High Speed Clock Resolution From CLI Master Clock Input

(3)

CCD SIGNAL

RG

H1/H3

H2/H4

(1) (2)

(5) (6)

PROGRAMMABLE CLOCK POSITIONS:

1. RG RISING EDGE

2. RG FALLING EDGE

3. SHP SAMPLE LOCATION

4. SHD SAMPLE LOCATION

5. H1/H3 RISING EDGE POSITION

6. H1/H3 FALLING EDGE POSITION (H2/H4 ARE INVERSE OF H1/H3)

(4)

Figure 5. High Speed Clock Programmable Locations

Table X. H1CONTROL, RGCONTROL, DRVCONTROL, and SAMPCONTROL Register Parameters

Parameter Length Range Description

Polarity 1b High/Low Polarity Control for H1/H3 and RG (0 = No Inversion, 1 = Inversion).
Positive Edge 6b 0–47 Edge Location Positive Edge Location for H1/H3 and RG.
Negative Edge 6b 0–47 Edge Location Negative Edge Location for H1/H3 and RG.
Sample Location 6b 0–47 Sample Location Sampling Location for SHP and SHD.
Drive Control 3b 0–7 Current Steps Drive Current for H1–H4 and RG Outputs, 0–7 Steps of 4.1 mA Each.
DOUT Phase 6b 0–47 Edge Location Phase Location of Data Outputs with Respect to Pixel Period.

REV. 0–14–

AD9948

Table XI. Precision Timing Edge Locations

Quadrant Edge Location (Decimal) Register Value (Decimal) Register Value (Binary)

I0 to 11 0 to 11 000000 to 001011
II 12 to 23 16 to 27 010000 to 011011
III 24 to 35 32 to 43 100000 to 101011
IV 36 to 47 48 to 59 110000 to 111011

H-Driver and RG Outputs

In addition to the programmable timing positions, the AD9948
features on-chip output drivers for the RG and H1–H4 outputs.
These drivers are powerful enough to directly drive the CCD
inputs. The H-driver and RG driver current can be adjusted
for optimum rise/fall time into a particular load by using the
DRVCONTROL register (Address x062). The DRVCONTROL
register is divided into five different 3-bit values, each one being
adjustable in 4.1 mA increments. The minimum setting of 0 is
equal to OFF or three-state, and the maximum setting of 7 is
equal to 30.1 mA.

t

H1/H3

H2/H4

RISE

<<

tPDt

RISE

FIXED CROSSOVER VOLTAGE

Figure 6. H-Clock Inverse Phase Relationship

As shown in Figure 6, the H2/H4 outputs are inverses of H1/H3.
The internal propagation delay resulting from the signal inversion
is less than l ns, which is significantly less than the typical rise
time driving the CCD load. This results in a H1/H2 crossover
voltage at approximately 50% of the output swing. The crossover
voltage is not programmable.

Digital Data Outputs

The AD9948 data output phase is programmable using the
DOUTPHASE register (Address x064). Any edge from 0 to 47
may be programmed, as shown in Figure 7a. The pipeline delay
for the digital data output is shown in Figure 7b.

t

PD

H1/H3 H2/H4

REV. 0

CLI

CCDIN

SHD

(INTERNAL)

DOUT

1 PIXEL PERIOD

CLI

DOUT

P[0]

t

OD

NOTES

1. DIGITAL OUTPUT DATA (DOUT) PHASE IS ADJUSTABLE WITH RESPECT TO THE PIXEL PERIOD.

2. WITHIN ONE CLOCK PERIOD, THE DATA TRANSITION CAN BE PROGRAMMED TO ANY OF THE 48 LOCATIONS.

P[12]

P[24]

P[36]

P[48] = P[0]

Figure 7a. Digital Output Phase Adjustment

t

CLIDLY

N–1

N–13

NOTES
DEFAULT TIMING VALUES ARE SHOWN: SHDLOC = 0, DOUT PHASE = 0.
HIGHER VALUES OF SHD AND/OR DOUTPHASE WILL SHIFT DOUT TRANSITION TO THE RIGHT, WITH RESPECT TO CLI LOCATION.

N N+1

SAMPLE PIXEL N

N–12

N+2

N+3

PIPELINE LATENCY = 11 CYCLES

N+12N+11N+10N+9N+8N+7N+6N+5N+4

N–3N–4N–5N–6N–7N–8N–9N–10N–11

N–2

N–1

N

Figure 7b. Pipeline Delay for Digital Data Output

–15–

N+13

N+1

AD9948

HORIZONTAL CLAMPING AND BLANKING

The AD9948’s horizontal clamping and blanking pulses are
fully programmable to suit a variety of applications. Individual
sequences are defined for each signal, which are then organized
into multiple regions during image readout. This allows the dark
pixel clamping and blanking patterns to be changed at each
stage of the readout to accommodate different image transfer
timing and high speed line shifts.

Individual CLPOB and PBLK Sequences

The AFE horizontal timing consists of CLPOB and PBLK, as
shown in Figure 8. These two signals are independently pro­grammed using the parameters shown in Table XII. The start
polarity, first toggle position, and second toggle position are

HD

CLPOB

(1)

PBLK

PROGRAMMABLE SETTINGS:

1. START POLARITY (CLAMP AND BLANK REGION ARE ACTIVE LOW)

2. FIRST TOGGLE POSITION

3. SECOND TOGGLE POSITION

(3)(2)

ACTIVE

Figure 8. Clamp and Preblank Pulse Placement

fully programmable for each signal. The CLPOB and PBLK
signals are active low, and should be programmed accordingly.
Up to four individual sequences can be created for each signal.

Individual HBLK Sequences

The HBLK programmable timing shown in Figure 9 is similar
to CLPOB and PBLK. However, there is no start polarity con­trol. Only the toggle positions are used to designate the start and
the stop positions of the blanking period. Additionally, there is a
polarity control, HBLKMASK, which designates the polarity of
the horizontal clock signals H1–H4 during the blanking period.
Setting HBLKMASK high will set H1 = H3 = low and H2 =
H4 = high during the blanking, as shown in Figure 10. Up to
four individual sequences are available for HBLK.

. . .

. . .

ACTIVE

. . .

HD

(1)

HBLK

PROGRAMMABLE SETTINGS:

1. FIRST TOGGLE POSITION = START OF BLANKING

2. SECOND TOGGLE POSITION = END OF BLANKING

Parameter Length Range Description

Polarity 1b High/Low Starting Polarity of Clamp and PBLK Pulses for Sequences 0–3.
Toggle Position 1 12b 0–4095 Pixel Location First Toggle Position within the Line for Sequences 0–3.
Toggle Position 2 12b 0–4095 Pixel Location Second Toggle Position within the Line for Sequences 0–3.

Parameter Length Range Description

HBLKMASK 1b High/Low Masking Polarity for H1 for Sequences 0–3 (0 = H1 Low, 1 = H1 High).
Toggle Position 1 12b 0–4095 Pixel Location First Toggle Position within the Line for Sequences 0–3.
Toggle Position 2 12b 0–4095 Pixel Location Second Toggle Position within the Line for Sequences 0–3.
Toggle Position 3 12b 0–4095 Pixel Location Third Toggle Position within the Line for Sequences 0–3.
Toggle Position 4 12b 0–4095 Pixel Location Fourth Toggle Position within the Line for Sequences 0–3.
Toggle Position 5 12b 0–4095 Pixel Location Fifth Toggle Position within the Line for Sequences 0–3.
Toggle Position 6 12b 0–4095 Pixel Location Sixth Toggle Position within the Line for Sequences 0–3.

(2)

BLANK

BLANK

Figure 9. Horizontal Blanking (HBLK) Pulse Placement

Table XII. CLPOB and PBLK Individual Sequence Parameters

Table XIII. HBLK Individual Sequence Parameters

. . .

REV. 0–16–

HD

HBLK

H1/H3

AD9948

. . .

. . .

THE POLARITY OF H1 DURING BLANKING IS PROGRAMMABLE (H2 IS OPPOSITE POLARITY OF H1).

. . .

H1/H3

H2/H4

. . .

Figure 10. HBLK Masking Control

TOG1

HBLK

H1/H3

H2/H4

SPECIAL H-BLANK PATTERN IS CREATED USING MULTIPLE HBLK TOGGLE POSITIONS.

TOG2 TOG3

TOG4 TOG5 TOG6

Figure 11. Generating Special HBLK Patterns

Table XIV. Horizontal Sequence Control Parameters for CLPOB, PBLK, and HBLK

Register Length Range Description

SCP 12b 0–4095 Line Number CLOB/PBLK/HBLK SCP to Define Horizontal Regions 0–3.

SPTR 2b 0–3 Sequence Number Sequence Pointer for Horizontal Regions 0–3.

GENERATING SPECIAL HBLK PATTERNS

Six toggle positions are available for HBLK. Normally, only
two of the toggle positions are used to generate the standard
HBLK interval. However, the additional toggle positions may be
used to generate special HBLK patterns, as shown in Figure 11.
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.

Horizontal Sequence Control

The AD9948 uses sequence change positions (SCPs) and sequence
pointers (SPTRs) to organize the individual horizontal sequences.
Up to four SCPs are available to divide the readout into four
separate regions, as shown in Figure 12. The SCP 0 is always hard­coded to Line 0, and SCP1–SCP3 are register programmable.
During each region bounded by the SCP, the SPTR registers

REV. 0

–17–

designate which sequence is used by each signal. CLPOB, PBLK,
and HBLK each have a separate set of SCPs. For example,
CLPOBSCP1 will define Region 0 for CLPOB, and in that region,
any of the four individual CLPOB sequences may be selected with
the CLPOBSPTR register. The next SCP defines a new region,
and in that region each signal can be assigned to a different
individual sequence. The sequence control registers are sum­marized in Table XIV.

External HBLK Signal

The AD9948 can also be used with an external HBLK signal. Set­ting the HBLKDIR register (Address x040) to high will disable the
internal HBLK signal generation. The polarity of the external
signal is specified using the HBLKPOL register, and the mask­ing polarity of H1 is specified using the HBLKMASK register.
Table XV summarizes the register values when using an external
HBLK signal.

AD9948

SEQUENCE CHANGE OF POSITION 0

SEQUENCE CHANGE OF POSITION 1

SEQUENCE CHANGE OF POSITION 2

SEQUENCE CHANGE OF POSITION 3

(V-COUNTER = 0)

UP TO FOUR INDIVIDUAL HORIZONTAL CLAMP AND BLANKING REGIONS MAY BE
PROGRAMMED WITHIN A SINGLE FIELD, USING THE SEQUENCE CHANGE POSITIONS.

SINGLE FIELD (1 VD INTERVAL)

CLAMP AND PBLK SEQUENCE REGION 0

CLAMP AND PBLK SEQUENCE REGION 1

CLAMP AND PBLK SEQUENCE REGION 2

CLAMP AND PBLK SEQUENCE REGION 3

Figure 12. Clamp and Blanking Sequence Flexibility

Table XV. External HBLK Register Parameters

Register Length Range Description

HBLKDIR 1b High/Low Specifies HBLK Internally Generated or Externally Supplied.

1 = External.

HBLKPOL 1b High/Low External HBLK Active Polarity.

0 = Active Low.
1 = Active High.

HBLKEXTMASK 1b High/Low External HBLK Masking Polarity.

0 = Mask H1 Low.
1 = Mask H1 High.

VD

H-COUNTER

(PIXEL COUNTER)

PxGA GAIN

REGISTER

HD

CLI

XXXXXXX

X

X

XXXXXXX

NOTES

1. INTERNAL H-COUNTER IS RESET SEVEN CLI CYCLES AFTER THE HD FALLING EDGE (WHEN USING VDHDEDGE = 0).

2. TYPICAL TIMING RELATIONSHIP: CLI RISING EDGE COINCIDES WITH HD FALLING EDGE.

3. PxGA STEERING IS SYNCHRONIZED WITH THE RESET OF THE INTERNAL H-COUNTER (MOSAIC SEPARATE MODE IS SHOWN).

H-COUNTER

RESET

X

01234567891011121415012 3

XXX

000 1 12111 0 031100

Figure 13. H-Counter Synchronization

H-COUNTER SYNCHRONIZATION

The H-Counter reset occurs seven CLI cycles following the HD
falling edge. The PxGA steering is synchronized with the reset
of the internal H-Counter (see Figure 13).

023

REV. 0–18–

POWER-UP PROCEDURE

VDD

(INPUT)

CLI

(INPUT)

SERIAL

WRITES

VD

(OUTPUT)

HD

(OUTPUT)

DIGITAL

OUTPUTS

t

PWR

...

ODD FIELD EVEN FIELD

1 H

...

H2/H4

H1/H3, RG

CLOCKS ACTIVE WHEN OUT_CONTROL REGISTER IS
UPDATED AT VD/HD EDGE

Figure 14. Recommended Power-Up Sequence

AD9948

1V

...

...

Recommended Power-Up Sequence

When the AD9948 is powered up, the following sequence is
recommended (refer to Figure 14 for each step):

1. Turn on the power supplies for the AD9948.

2. Apply the master clock input, CLI, VD, and HD.

3. Although the AD9948 contains an on-chip power-on reset, a
software reset of the internal registers is recommended. Write
a 1 to the SW_RST register (Address x010), which will reset
all the internal registers to their default values. This bit is
self-clearing and will automatically be reset back to 0.

4. The Precision Timing core must be reset by writing a 0 to the
TGCORE_RSTB register (Address x012) followed by writing
a l to the TGCORE_RSTB register. This will start the internal
timing core operation.

5. Write a 1 to the PREVENTUPDATE register (Address x014).
This will prevent the updating of the serial register data.

6. Write to the desired registers to configure high speed timing
and horizontal timing.

7. Write a 1 to the OUT_CONTROL register (Address x011).
This will allow the outputs to become active after the next
VD/HD rising edge.

8. Write a 0 to the PREVENTUPDATE register (Address x014).
This will allow the serial information to be updated at next
VD/HD falling edge.

The next VD/HD falling edge allows register updates to occur,
including OUT_CONTROL, which enables all clock outputs.

REV. 0

–19–

AD9948

1.0F
CCDIN

DC RESTORE

1.5V

SHP

0dB, –2dB, –4dB

CDS

SHD

SHP

PxGA GAIN

REGISTERS

SHD

0dB ~ 18dB

PxGA

DOUT

PHASE

CLPOB

6dB ~ 42dB

VGA

VGA GAIN

REGISTER

PBLK

DAC

1.0F 1.0F

1.0V 2.0V

INTERNAL

V

REF

10-BIT

ADC

OPTICAL BLACK

CLAMP

DIGITAL

FILTER

REFTREFB

2V FULL SCALE

CLAMP LEVEL

REGISTER

CLPOB

8

AD9948

OUTPUT

DATA

LATCH

PBLK

DOUT

PHASE

10

DOUT

PRECISION

TIMING

GENERATION

V- H

TIMING

GENERATION

Figure 15. Analog Front End Functional Block Diagram

ANALOG FRONT END DESCRIPTION AND OPERATION

The AD9948 signal processing chain is shown in Figure 15. Each
processing step is essential in achieving a high quality image from
the raw CCD pixel data.

DC Restore

To reduce the large dc offset of the CCD output signal,
adcrestore 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 to be compatible with the 3 V supply voltage
of the AD9948.

Correlated Double Sampler

The CDS circuit samples each CCD pixel twice to extract the
video information and reject low frequency noise. The timing
shown in Figure 5 illustrates how the two internally generated
CDS clocks, SHP and SHD, are used to sample the reference
level and the CCD signal level, respectively. The placement of
the SHP and SHD sampling edges is determined by the setting
of the SAMPCONTROL register located at Address 0x63.
Placement of these two clock signals is critical in achieving the
best performance from the CCD.

The gain in the CDS is fixed at 0 dB by default. Using Bits D10
and D11 in the AFE operation register, the gain may be reduced
to –2 dB or –4 dB. This will allow the AD9948 to accept an input
signal of greater than 1 V p-p. See Table VIII for register details.

PxGA

The PxGA provides separate gain adjustment for the individual
color pixels. A programmable gain amplifier with four separate
values, the PxGA has the capability to multiplex its gain value on
a pixel-to-pixel basis (see Figure 16). This allows lower output
color pixels to be gained up to match higher output color pixels.
Also, the PxGA may be used to adjust the colors for white bal­ance, reducing the amount of digital processing that is needed.
The four different gain values are switched according to the color
steering circuitry. Three different color steering modes for differ­ent types of CCD color filter arrays are programmable in the AFE
CTLMODE register at Address 0x03 (see Figures 18a to 18c for
timing examples). For example, progressive steering mode accom­modates the popular Bayer arrangement of red, green, and blue
filters (see Figure 17a).

VD

HD

SHP/SHD

8

COLOR

STEERING

CONTROL

2

4:1

MUX

PxGA STEERING

3

SELECTION

GAIN0

GAIN1

GAIN2

GAIN3

MODE

PxGA GAIN
REGISTERS

CONTROL
REGISTER
BITS D0–D1

Table XVI. Adjustable CDS Gain

Operation Register Bits
D11 D10 CDS Gain Max CDS Input

00 0 dB 1.0 V p-p
01 –2 dB 1.2 V p-p
10 –4 dB 1.6 V p-p
11 0 dB 1.0 V p-p

CDS

PxGA

VGA

Figure 16. PxGA Block Diagram

REV. 0–20–

AD9948

CCD: PROGRESSIVE BAYER

RR

Gr

Gb B

RR

Gr

Gb B

Gr

Gb B

Gr

Gb B

COLOR STEERING MODE:
PROGRESSIVE

LINE0 GAIN0, GAIN1, GAIN0, GAIN1, ...

LINE1

LINE2

GAIN2, GAIN3, GAIN2, GAIN3, ...

GAIN0, GAIN1, GAIN0, GAIN1, ...

Figure 17a. CCD Color Filter Example—Progressive Scan

The same Bayer pattern can also be interlaced, and the interlaced
mode should be used with this type of CCD (see Figure 17b).
The color steering performs the proper multiplexing of the R,
G, and B gain values (loaded into the PxGA gain registers), and
is synchronized by the user with vertical (VD) and horizontal
(HD) sync pulses. For timing information, see Figure 18b.

CCD: INTERLACED BAYER

EVEN FIELD

RR

Gr

RR

Gr

RR

Gr

RR

Gr

ODD FIELD

Gb GbBB

Gb GbBB

Gb GbBB

Gr

Gr

Gr

Gr

COLOR STEERING MODE:
INTERLACED

LINE0 GAIN0, GAIN1, GAIN0, GAIN1, ...

LINE1

LINE2

LINE0 GAIN2, GAIN3, GAIN2, GAIN3, ...

LINE1

LINE2

GAIN0, GAIN1, GAIN0, GAIN1, ...

GAIN0, GAIN1, GAIN0, GAIN1, ...

GAIN2, GAIN3, GAIN2, GAIN3, ...

GAIN2, GAIN3, GAIN2, GAIN3, ...

A third type of readout uses the Bayer pattern divided into three
different readout fields. The three-field mode should be used
with this type of CCD (see Figure 17c). The color steering
performs the proper multiplexing of the R, G, and B gain values
(loaded into the PxGA gain registers), and is synchronized by
the user with vertical (VD) and horizontal (HD) sync pulses.
For timing information, see Figure 18c.

CCD: 3-FIELD READOUT

FIRST FIELD

RR

Gr

Gb GbBB

RR

Gr

Gb GbBB

SECOND FIELD

Gb GbBB

RR

Gr

Gb GbBB

RR

Gr

THIRD FIELD

RR

Gr

Gb GbBB

RR

Gr

Gb GbBB

Gr

Gr

Gr

Gr

Gr

Gr

COLOR STEERING MODE:
THREE FIELD

LINE0 GAIN0, GAIN1, GAIN0, GAIN1, ...

LINE1

LINE2

LINE0 GAIN2, GAIN3, GAIN2, GAIN3, ...

LINE1

LINE2

LINE0 GAIN0, GAIN1, GAIN0, GAIN1, ...

LINE1

LINE2

GAIN2, GAIN3, GAIN2, GAIN3, ...

GAIN0, GAIN1, GAIN0, GAIN1, ...

GAIN0, GAIN1, GAIN0, GAIN1, ...

GAIN2, GAIN3, GAIN2, GAIN3, ...

GAIN2, GAIN3, GAIN2, GAIN3, ...

GAIN0, GAIN1, GAIN0, GAIN1, ...

Gb GbBB

Figure 17b. CCD Color Filter Example—Interlaced Readout

Figure 17c. CCD Color Filter Example—Three-Field Readout

REV. 0

–21–

AD9948

FIELDVAL

VD

HD

PxGA GAIN

REGISTER

NOTES

1. VD FALLING EDGE WILL RESET THE PxGA GAIN REGISTER STEERING TO 0101 LINE.

2. HD FALLING EDGES WILL ALTERNATE THE PxGA GAIN REGISTER STEERING BETWEEN 0101 AND 2323 LINES.

3. FIELDVAL IS ALWAYS RESET TO 0 ON VD FALLING EDGES.

FIELDVAL = 0

0

110XX

22033 11

022033 11

FIELDVAL = 0

0

110

Figure 18a. PxGA Color Steering—Progressive Mode

FIELDVAL

VD

HD

PxGA GAIN

REGISTER

NOTES

1. FIELDVAL = 0 (START OF FIRST FIELD) WILL RESET THE PxGA GAIN REGISTER STEERING TO 0101 LINE.

2. FIELDVAL = 1 (START OF SECOND FIELD) WILL RESET THE PxGA GAIN REGISTER STEERING TO 2323 LINE.

3. HD FALLING EDGES WILL RESET THE PxGA GAIN REGISTER STEERING TO EITHER 0 (FIELDVAL = 0) OR 2 (FIELDVAL = 1).

4. FIELDVAL WILL TOGGLE BETWEEN 0 AND 1 ON EACH VD FALLING EDGE.

FIELDVAL = 0

110XX

0

00 311 22

FIELDVAL = 1 FIELDVAL = 0

31100 00

2

332

Figure 18b. PxGA Color Steering—Interlaced Mode

0

1

0

1

FIELDVAL

VD

HD

PxGA GAIN

REGISTER

NOTES

1. FIELDVAL = 0 (START OF FIRST FIELD) WILL RESET THE PxGA GAIN REGISTER STEERING TO 0101 LINE.

2. FIELDVAL = 1 (START OF SECOND FIELD) WILL RESET THE PxGA GAIN REGISTER STEERING TO 2323 LINE.

2. FIELDVAL = 2 (START OF THIRD FIELD) WILL RESET THE PxGA GAIN REGISTER STEERING TO 0101 LINE.

3. HD FALLING EDGES WILL ALTERNATE THE PxGA GAIN REGISTER STEERING BETWEEN 0101 AND 2323 LINES.

4. FIELDVAL WILL INCREMENT AT EACH VD FALLING EDGE, REPEATING THE 0...1...2...0...1...2 PATTERN.

FIELDVAL = 0

110XX

0

22 333 22

FIELDVAL = 1 FIELDVAL = 2

31100 22

1

001

Figure 18c. PxGA Color Steering—Three-Field Mode

3

3

REV. 0–22–

AD9948

The PxGA gain for each of the four channels is variable from 0 dB
to 18 dB in 512 steps, specified using the PxGA GAIN01 and
PxGA GAIN23 registers. The PxGA gain curve is shown in
Figure 19. The PxGA GAIN01 registers contains nine bits each
for PxGA Gain0 and Gain1, and the PxGA GAIN23 registers
contains nine bits each for PxGA Gain2 and Gain3.

18

15

12

9

PxGA GAIN (dB)

6

3

0

64 128 192 256 320 384 448 511

0

PxGA GAIN REGISTER CODE

Figure 19. PxGA Gain Curve

Variable Gain Amplifier

The VGA stage provides a gain range of 6 dB to 42 dB, program­mable with 10-bit resolution through the serial digital interface.
The minimum gain of 6 dB is needed to match a 1 V input signal
with the ADC full-scale range of 2 V. When compared to 1 V
full-scale systems, the equivalent gain range is 0 dB to 36 dB.

The VGA gain curve follows a linear-in-dB characteristic. The
exact VGA gain can be calculated for any gain register value by
using the equation

Gain dB Code dB()(. )=×+0 0351 6

where the code range is 0 to 1023.

There is a restriction on the maximum amount of gain that can
be applied to the signal. The PxGA can add as much as 18 dB,
and the VGA is capable of providing up to 42 dB. However, the
maximum total gain from the PxGA and VGA is restricted to
42 dB. If the registers are programmed to specify a total gain
higher than 42 dB, the total gain will be clipped at 42 dB.

A/D Converter

The AD9948 uses a high performance ADC architecture, opti­mized for high speed and low power. Differential nonlinearity
(DNL) performance is typically better than 0.5 LSB. The ADC
uses a 2 V input range. See TPC 1 and TPC 2 for typical linearity
and noise performance plots for the AD9948.

Optical Black Clamp

The optical black clamp loop is used to remove residual offsets
in the signal chain and to track low frequency variations in the
CCD’s black level. During the optical black (shielded) pixel
interval on each line, the ADC output is compared with a fixed
black level reference, selected by the user in the clamp level register.
The value can be programmed between 0 LSB and 63.75 LSB
in 256 steps. The resulting error signal is filtered to reduce noise,
and the correction value is applied to the ADC input through a
D/A converter. 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 the postprocessing, the AD9948 optical black
clamping may be disabled using Bit D2 in the OPRMODE register.
When the loop is disabled, the clamp level register may still be
used to provide programmable offset adjustment.

The CLPOB pulse should be placed during the CCD’s optical
black pixels. It is recommended that the CLPOB pulse duration
be at least 20 pixels wide to minimize clamp noise. Shorter
pulsewidths may be used, but clamp noise may increase and the
ability to track low frequency variations in the black level will be
reduced. See the Horizontal Clamping and Blanking and the
Applications Information sections for timing examples.

Digital Data Outputs

The AD9948 digital output data is latched using the DOUT
phase register value, as shown in Figure 15. Output data timing is
shown in Figure 7. It is also possible to leave the output latches
transparent, so that the data outputs are valid immediately from
the A/D converter. Programming the AFE control register Bit D4
to a 1 will set the output latches transparent. The data outputs
can also be disabled (three-stated) by setting the AFE control
register Bit D3 to a 1.

The data output coding is normally straight binary, but the
coding my be changed to gray coding by setting the AFE
control register Bit D5 to a 1.

REV. 0

42

36

30

24

VGA GAIN (dB)

18

12

6

0

127 255 383 511 639 767 895 1023

VGA GAIN REGISTER CODE

Figure 20. VGA Gain Curve (PxGA Not Included)

–23–

AD9948

APPLICATIONS INFORMATION
Circuit Configuration

The AD9948 recommended circuit configuration is shown in
Figure 21. Achieving good image quality from the AD9948
requires careful attention to PCB layout. All signals should be
routed to maintain low noise performance. The CCD output
signal should be directly routed to Pin 27 through a 0.1 µF
capacitor. The master clock CLI should be carefully routed to
Pin 25 to minimize interference with the CCDIN, REFT, and
REFB signals.

The digital outputs and clock inputs are located on Pins 2 to 13
and Pins 31 to 39, and should be connected to the digital ASIC
away from the analog and CCD clock signals. Placing series
resistors close to the digital output pins may help to reduce
digital code transition noise. If the digital outputs must drive a
load larger than 20 pF, buffering is recommended to minimize
additional noise. If the digital ASIC can accept gray code, the
AD9948’s outputs can be selected to output data in gray code
format using the control register Bit D5. Gray coding will
help reduce potential digital transition noise compared with
binary coding.

The H1–H4 and RG traces should have low inductance to
avoid excessive distortion of the signals. Heavier traces are
recommended because of the large transient current demand
on H1–H4 from the capacitive load of the CCD. If possible,
physically locating the AD9948 closer to the CCD will reduce
the inductance on these lines. As always, the routing path
should be as direct as possible from the AD9948 to the CCD.

Grounding and Decoupling Recommendations

As shown in Figure 21, a single ground plane is recommended
for the AD9948. This ground plane should be as continuous as
possible, particularly around Pins 23 to 30. This will ensure that
all analog decoupling capacitors provide the lowest possible
impedance path between the power and bypass pins and their
respective ground pins. All high frequency decoupling capacitors
should be located as close as possible to the package pins. It is
recommended that the exposed paddle on the bottom of the
package be soldered to a large pad, with multiple vias connect­ing the pad to the ground plane.

All the supply pins must be decoupled to ground with good
quality, high frequency chip capacitors. There should also be a

4.7 µF or larger bypass capacitor for each main supply—AVDD,
RGVDD, HVDD, and DRVDD—although this is not necessary
for each individual pin. In most applications, it is easier to share
the supply for RGVDD and HVDD, which may be done as long
as the individual supply pins are separately bypassed. A separate
3 V supply may be used for DRVDD, but this supply pin should
still be decoupled to the same ground plane as the rest of the
chip. A separate ground for DRVSS is not recommended.

The reference bypass pins (REFT, REFB) should be decoupled
to ground as close as possible to their respective pins. The analog
input (CCDIN) capacitor should also be located close to the pin.

DRIVER

SUPPLY

VD/HD/HBLK INPUTS

3V

+

4.7F 0.1F

OUTPUTS

CLP/BLK OUTPUT

10

DATA

ANALOG

4

NC 1

(LSB) D0 2

D1 3

D2 4

DRVSS 5

DRVDD 6

D3 7

D4 8

D5 9

D6 10

3V

SUPPLY

40 NC

38 HBLK

39 CLP/PBLK

PIN 1
IDENTIFIER

D7 11

D8 12

(MSB) D9 13

0.1F

37 DVDD

36 DVSS

35 HD

AD9948

TOP VIEW

H1 14

H2 15

HVSS 16

34 VD

32 SDI

33 SCK

H4 19

H3 18

HVDD 17

31 SL

30 REFB

29 REFT

28 AVSS
27 CCDIN

26 AVDD

25 CLI

24 TCVDD

23 TCVSS

22 RGVDD

21 RG

RGVSS 20

0.1F

SERIAL

+

4.7F

INTERFACE

1F

1F

0.1F

0.1F

0.1F

+

4.7F

0.1F

+

4.7F

CCD SIGNAL

MASTER CLOCK INPUT

3V
ANALOG
SUPPLY

RG OUTPUT

RG DRIVER
SUPPLY

H DRIVER
SUPPLY

3

4

Figure 21. Recommended Circuit Configuration

H1–H4

REV. 0–24–

AD9948

CLI

25

MASTER CLOCK

AD9948

ASIC

Driving the CLI Input

The AD9948’s master clock input (CLI) may be used in two
different configurations, depending on the application. Figure 23a
shows a typical dc-coupled input from the master clock source.
When the dc-coupled technique is used, the master clock signal
should be at standard 3 V CMOS logic levels. As shown in

CCDIN

SIGNAL

OUT

27

AD9948

18

19 14 15 21

H2 RGH3 H4 H1

H1

H2

CCD IMAGER

RG

Figure 22a. CCD Connections (2 H-Clock)

Figure 23b, a 1000 pF ac coupling capacitor may be used
between the clock source and the CLI input. In this configura­tion, the CLI input will self-bias to the proper dc voltage level
of approximately 1.4 V. When the ac-coupled technique is
used, the master clock signal can be as low as ±500 mV in
amplitude.

Figure 23a. CLI Connection, DC-Coupled

AD9948

CCDIN

SIGNAL

OUT

H2 H1

27

AD9948

14 15 21

H1 H2

18 19

H1

H2

CCD IMAGER

RGH3 H4

RG

Figure 22b. CCD Connections (4 H-Clock)

CLI

25

LPF

1nF

ASIC

MASTER CLOCK

Figure 23b. CLI Connection, AC-Coupled

REV. 0

–25–

AD9948

HORIZONTAL TIMING SEQUENCE EXAMPLE

Figure 24 shows an example CCD layout. The horizontal regis­ter contains 28 dummy pixels, which will occur on each line
clocked from the CCD. In the vertical direction, there are 10
optical black (OB) lines at the front of the readout and two at
the back of the readout. The horizontal direction has four OB
pixels in the front and 48 in the back.

To configure the AD9948 horizontal signals for this CCD, three
sequences can be used. Figure 25 shows the first sequence, to be
used during vertical blanking. During this time, there are no

V

4 OB PIXELS

EFFECTIVE IMAGE AREA

H

48 OB PIXELS

HORIZONTAL CCD REGISTER

valid OB pixels from the sensor, so the CLPOB signal is not
used. PBLK may be enabled during this time, because no valid
data is available.

Figure 26 shows the recommended sequence for the vertical OB
interval. The clamp signals are used across the whole lines in
order to stabilize the clamp loop of the AD9948.

Figure 27 shows the recommended sequence for the effective
pixel readout. The 48 OB pixels at the end of each line are used
for the CLPOB signal.

SEQUENCE 2 (OPTIONAL)

2 VERTICAL OB LINES

USE SEQUENCE 3

10 VERTICAL OB LINES

USE SEQUENCE 2

SEQUENCE 1: VERTICAL BLANKING

CCDIN

SHP

SHD

H1/H3

H2/H4

HBLK

PBLK

CLPOB

28 DUMMY PIXELS

VERTICAL SHIFT

Figure 25. Horizontal Sequence during Vertical Blanking

Figure 24. Example CCD Configuration

DUMMY INVALID PIXELSINVALID PIX

VERT SHIFT

REV. 0–26–

SEQUENCE 2: VERTICAL OPTICAL BLACK LINES

AD9948

CCDIN

OPTICAL BLACK

SHP

SHD

H1/H3

H2/H4

HBLK

PBLK

CLPOB

SEQUENCE 3: EFFECTIVE PIXEL LINES

OPTICAL BLACK

CCDIN

SHP

SHD

VERTICAL SHIFT

DUMMY

OPTICAL BLACK

Figure 26. Horizontal Sequences during Vertical Optical Black Pixels

OB

VERTICAL SHIFT

DUMMY

EFFECTIVE PIXELS

OPTICAL BLACK

VERT SHIFT

VERT SHIFT

H1/H3

H2/H4

HBLK

PBLK

CLPOB

Figure 27. Horizontal Sequences during Effective Pixels

REV. 0

–27–

AD9948

OUTLINE DIMENSIONS

40-Lead Lead Frame Chip Scale Package [LFCSP]

6 mm  6 mm Body

(CP-40)

Dimensions shown in millimeters

PIN 1

INDICATOR

1.00

0.90

0.80

12 MAX

SEATING
PLANE

6.00

BSC SQ

TOP

VIEW

0.80 MAX

0.65 NOM

0.30

0.23

0.18

COMPLIANT TO JEDEC STANDARDS MO-220-VJJD-2

5.75

BSC SQ

0.20 REF

0.05 MAX

0.02 NOM

COPLANARITY

0.60 MAX

0.50

BSC

0.50

0.40

0.30

0.08

0.60 MAX

31

30

21

20

BOTTOM

VIEW

4.50
REF

PIN 1

40

11

INDICATOR

1

4.25

3.70 SQ

1.75

10

C03752–0–5/03(0)

–28–

REV. 0