FEATURES
140 MSPS Guaranteed Conversion Rate
100 MSPS Low Cost Version Available
330 MHz Analog Bandwidth
1 V p-p Analog Input Range
Internal 2.5 V Reference
Differential or Single-Ended Clock Input
3.3 V/5.0 V Three-State CMOS Outputs
Single or Demultiplexed Output Ports
Data Clock Output Provided
Low Power: 1.0 W Typical
5 V Converter Power Supply
APPLICATIONS
RGB Graphics Processing
High Resolution Video
LCD Monitors and Projectors
Micromirror Projectors
Plasma Display Panels
Scan Converters
R AIN
R AIN
G AIN
G AIN
B AIN
B AIN
ENCODE
ENCODE
DS
DS
A/D Converter
AD9483
FUNCTIONAL BLOCK DIAGRAM
AD9483
T/H
T/H
T/H
TIMING
2.5V
VREF
RVREFINGVREFINBVREFINV
OUT
QUANTIZER
QUANTIZER
QUANTIZER
8
8
8
CONTROL
CC
GND
V
DD
DRA
7-0
DRB
7-0
DGA
7-0
DGB
7-0
DBA
7-0
B
D
7-0
B
CLKOUT
CLKOUT
OMS
I/P
PD
GENERAL DESCRIPTION
The AD9483 is a triple 8-bit monolithic analog-to-digital
converter optimized for digitizing RGB graphics signals from
personal computers and workstations. Its 140 MSPS encode
rate capability and full-power analog bandwidth of 330 MHz
supports display resolutions of up to 1280 × 1024 at 75 Hz with
sufficient input bandwidth to accurately acquire and digitize
each pixel.
To minimize system cost and power dissipation, the AD9483
includes an internal 2.5 V reference and track-and-hold circuit.
The user provides only a 5 V power supply and an encode clock.
No external reference or driver components are required for
many applications. The digital outputs are three-state CMOS
outputs. Separate output power supply pins support interfacing
with 3.3 V or 5 V logic.
The AD9483’s encode input interfaces directly to TTL, CMOS,
or positive ECL logic and will operate with single-ended or
differential inputs. The user may select dual channel or single
channel digital outputs. The Dual Channel (demultiplexed)
mode interleaves ADC data through two 8-bit channels at onehalf the clock rate. Operation in Dual Channel mode reduces
the speed and cost of external digital interfaces while allowing
the ADCs to be clocked to the full 140 MSPS conversion rate.
In the Single Channel mode, all data is piped at the full clock
rate to the Channel A outputs and the ADCs conversion rate is
limited to 100 MSPS. A data clock output is provided at the
Channel A output data rate for both Dual Channel or Single
Channel output modes.
Fabricated in an advanced BiCMOS process, the AD9483 is
provided in a space-saving 100-lead MQFP surface-mount
plastic package (S-100) and is specified over the 0°C to 85°C
temperature range.
REV. C
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 owners.
FullVI5050µA
Analog Bandwidth, Full Power25°CV 330330MHz
REFERENCE OUTPUT
Output VoltageFullVI+2.4 +2.5+2.6+2.4+2.5+2.6V
Temperature CoefficientFullV110110ppm/°C
SWITCHING PERFORMANCE
Maximum Conversion RateFullVI140100MSPS
Minimum Conversion RateFullIV1010MSPS
Encode Pulse Width High (t
Encode Pulse Width Low (t
Aperture Delay (t
)25°CV 1.51.5ns
A
)25°CIV2.8504.050ns
EH
)25°CIV2.8504.050ns
EL
Aperture Delay Matching25°CV 100100ps
Aperture Uncertainty (Jitter)25°CV 2.32.3ps rms
Data Sync Setup Time (t
Data Sync Hold Time (t
Data Sync Pulsewidth (t
Output Valid Time (t
Output Propagation Delay (t
Clock Valid Time (t
CV
Clock Propagation Delay (t
Data to Clock Skew (t
Data to Clock Skew (tPD–t
)25°CIV00ns
SDS
)25°CIV0.50.5ns
HDS
)25°CIV2.02.0ns
PWDS
2
)
V
PD
3
)
CPD
)FullVI–1.00+1.0–1.00+1.0ns
V–tCV
)FullVI–2.00+2.0–2.00+2.0ns
CPD
FullVI4.06.34.06.3ns
2
)
FullVI8.0108.010ns
FullVI3.86.23.86.2ns
3
)
FullVI8.0108.010ns
DIGITAL INPUTS
Input Capacitance25°CV 33pF
DIFFERENTIAL INPUTS
Differential Signal Amplitude (VID)FullIV400400mV
HIGH Input Voltage (V
LOW Input Voltage (V
IHD
ILD
Common-Mode Input (V
HIGH Level Current (I
IH
)FullIV0.4V
CC
)FullIV00V
)FullIV1.51.5V
ICM
)FullVI1.21.2mA
0.4V
CC
V
LOW Level Current (IIL)FullVI1.21.2mA
VREF IN
Input Resistance25°CV 2.52.5kΩ
–2–
REV. C
Page 3
AD9483
Test AD9483KS-140 AD9483KS-100
ParameterTemperatureLevelMinTypMaxMinTypMaxUnit
SINGLE-ENDED INPUTS
HIGH Input Voltage (VIH)FullIV2.0V
LOW Input Voltage (V
HIGH Level Current (I
Maximum Case Temperature . . . . . . . . . . . . . . . . . . . . 150°C
*Stresses above those listed under Absolute Maximum Ratings may cause perma-
nent 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
sections of this specification is not implied. Exposure to absolute maximum ratings
for extended periods may effect device reliability.
EXPLANATION OF TEST LEVELS
Test Level
I–100% production tested.
II – 100% production tested at 25°C and sample tested at
specified temperatures.
III – Periodically sample tested.
IV – Parameter is guaranteed by design and characterization
testing.
V–Parameter is a typical value only.
VI – 100% production tested at 25°C; guaranteed by design
AD9483KS-1000°C to 85°CMetric Quad Flat Package S-100B
AD9483KS-1400°C to 85°CMetric Quad Flat Package S-100B
AD9483/PCBEvaluation Board
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 AD9483 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.
2ENCODEEncode Clock for ADC (ADC Samples on Rising Edge of ENCODE)
3ENCODEEncode Clock Complement (ADC Samples on Falling Edge of ENCODE)
4DSData Sync Aligns Output Channels in Dual-Channel Mode
5DSData Sync Complement
8DCOData Clock Output. Clock Output at Channel A Data Rate
9DCOData Clock Output Complement
11, 21, 31, 41, 51, 61, 71V
79, 82, 83, 93, 94, 98, 99V
12–19D
22–29D
32–39D
42–49D
52–59D
62–69D
DD
CC
BB7–DBB0
BA7–DBA0
GB7–DGB0
GA7–DGA0
RB7–DRB0
RA7–DRA0
72NCNo Connect
74OMSSelects Single Channel or Dual Channel Output Mode, (HIGH = Single,
75I/PSelects Interleaved or Parallel Output Mode, (HIGH = Interleaved, LOW = Parallel)
76PDPower-Down and Three-State Select (HIGH = Power-Down)
84R AINAnalog Input Complement for Converter “R”
85R AINAnalog Input True for Converter “R”
86R REF INReference Input for Converter “R” (2.5 V Typical, ±10%)
87G AINAnalog Input Complement for Converter “G”
88G AINAnalog Input True for Converter “G”
89G REF INReference Input for Converter “G” (2.5 V Typical, ±10%)
90B AINAnalog Input Complement for Converter “B”
91B AINAnalog Input True for Converter “B”
92B REF INReference Input for Converter “B” (2.5 V Typical, ± 10%)
97REF OUTInternal Reference Output (2.5 V Typical); Bypass with 0.01 µF to Ground
Output Power Supply. Nominally 3.3 V
Converter Power Supply. Nominally 5.0 V
Digital Outputs of Converter “B,” Channel B. DBB7 is the MSB
Digital Outputs of Converter “B,” Channel A. DBA7 is the MSB
Digital Outputs of Converter “G,” Channel B. DGB7 is the MSB
Digital Outputs of Converter “G,” Channel A. DGA7 is the MSB
Digital Outputs of Converter “R,” Channel B. DRB7 is the MSB
Digital Outputs of Converter “R,” Channel A. DRA7 is the MSB
LOW = Demuxed)
REV. C
–5–
Page 6
AD9483
GND
1
GND
GND
DCO
DCO
GND
V
DBB
DBB
DBB
DBB
DBB
DBB
DBB
DBB
GND
V
DBA
DBA
DBA
DBA
DBA
DBA
DBA
DBA
GND
DS
DS
2
3
4
5
6
7
8
9
10
11
DD
12
7
13
6
14
5
15
4
16
3
17
2
18
1
19
0
20
21
DD
22
7
23
6
24
5
25
4
26
3
27
2
28
1
29
0
30
ENCODE
ENCODE
NC = NO CONNECT
PIN CONFIGURATION
Metric Quad Flat Package (S-100B)
CC
CC
V
V
GND
100
PIN 1
IDENTIFIER
GND
REF OUT
99989796959493
GND
CC
CC
B AIN
V
V
B REF IN
929190
B AIN
G REF IN
G AIN
89
88
G AIN
8786858483
AD9483
TOP VIEW
(PINS DOWN)
31
33
32
7
6
B
B
DD
G
G
V
D
D
37
35
34
5
4
B
B
G
G
D
D
39
40
38
36
3
2
B
B
G
G
D
D
41
43
44
1
B
G
D
42
0
B
G
GND
D
5
7
6
A
A
A
DD
G
G
G
V
D
D
D
R AIN
R REF IN
46
45
4
3
A
A
G
G
D
D
R AIN
47
2
A
G
D
CC
CC
GND
V
V
81
82
80
GND
79
V
CC
78
GND
77
GND
76
PD
75
I/P
74
OMS
73
GND
72
NC
71
V
DD
70
GND
69
DRA
0
68
DRA
1
67
DRA
2
66
DRA
3
65
DRA
4
64
DRA
5
63
DRA
6
62
DRA
7
61
V
DD
60
GND
59
DRB
0
58
DRB
1
57
DRB
2
56
DRB
3
55
DRB
4
54
DRB
5
DRB
53
6
52
DRB
7
51
V
DD
50
49
48
1
0
A
A
G
G
GND
D
D
–6–
REV. C
Page 7
TIMING
AD9483
AIN
ENCODE
ENCODE
DS
AIN
ENCODE
ENCODE
D7–D 0
CLOCK OUT
CLOCK OUT
SAMPLE N–2
SAMPLE N–1
SAMPLE N
t
EH
SAMPLE N+3
A
SAMPLE N+1
t
EL
1/f
t
SAMPLE N+2
S
SAMPLE N+4
t
DATA N–5DATA N–4DATA N–3DATA N–2DATA N–1DATA N
t
CPD
Figure 1. Timing—Single Channel Mode
SAMPLE N–1
t
EH
t
HDS
t
EL
t
SDS
SAMPLE N
t
A
1/f
SAMPLE N+1
S
SAMPLE N+2
SAMPLE N+3
SAMPLE N+4
SAMPLE N+5
PD
t
V
t
CV
SAMPLE N+6
DS
PORT A
D7–D0
PORT B
D7–D0
PORT A
D7–D0
PORT B
D7–D0
CLKOUT
CLKOUT
DATA N–7
OR N–8
DATA N–8
OR N–7
DATA N–9
OR N–8
DATA N–8
OR N–7
DATA N–7
DATA N–6
OR N–7
DATA N–7
OR N–8
DATA N–6
OR N–7
INTERLEAVED DATA OUT
OR N–6
INVALID IF OUT OF SYNC
DATA N–5 IF IN SYNC
INVALID IF OUT OF SYNC
DATA N–5 IF IN SYNC
INVALID IF OUT OF SYNC
DATA N–4 IF IN SYNC
PARALLEL DATA OUT
DATA N–7
OR N–6
DATA N–3
INVALID IF OUT OF SYNC
DATA N–4 IF IN SYNC
DATA N–3
Figure 2. Timing—Dual Channel Mode
t
PD
DATA N–2
t
V
DATA N
DATA N–1
DATA N–2
DATA N–1
t
t
CV
CPD
DATA N+1
DATA N
DATA N+1
REV. C
–7–
Page 8
AD9483
V
CC
AD9483
DIGITAL
INPUTS
EQUIVALENT CIRCUITS
V
AIN
AIN
AD9483
Figure 3. Equivalent Analog Input Circuit
V
CC
VREF IN
500⍀
2k⍀
AD9483
Figure 4. Equivalent Reference Input Circuit
17.5k⍀
ENCODE
DS
300⍀
AD9483
300⍀
7.5k⍀
CC
V
CC
ENCODE
DS
V
DD
AD9483
DIGITAL
OUTPUTS
Figure 7. Equivalent Digital Output Circuit
V
CC
VREF
OUT
AD9483
Figure 8. Equivalent Reference Output Circuit
Figure 5. Equivalent Encode and Data Select Input Circuit
TPC 21. 3rd Harmonic vs. Temperature, fS = 140 MSPS
–45
–40
–25
0406080100
TEMPERATURE – ⴗC
TPC 23. 2nd Harmonic vs. Temperature, fS = 140 MSPS
0
–10
–20
–30
–40
–50
dB
–60
–70
–80
–90
–100
0
102030405060708090 100
MHz
F1 = 55.0MHz
F2 = 56.0MHz
F1 = F2 = –7.0dBFS
TPC 24. Two Tone Intermodulation Distortion
–12–
REV. C
Page 13
AD9483
APPLICATION NOTES
Theory of Operation
The AD9483 combines Analog Devices’ patented MagAmp bitper-stage architecture with flash converter technology to create a
high performance, low power ADC. For ease of use the part
includes an on board reference and input logic that accepts TTL,
CMOS, or PECL levels.
Each of the three analog input signals is buffered by a high speed
differential amplifier and applied to a track-and-hold (T/H)
circuit. This T/H captures the value of the input at the sampling
instant and maintains it for the duration of the conversion. The
sampling and conversion process is initiated by a rising edge on
the ENCODE input. Once the signal is captured by the T/H,
the four Most Significant Bits (MSBs) are sequentially encoded
by the MagAmp string. The residue signal is then encoded by a
flash comparator string to generate the four Least Significant
Bits (LSBs). The comparator outputs are decoded and combined into the 8-bit result.
If the user has selected Single Channel mode (OMS = HIGH)
the 8-bit data word is directed to an A output bank. Data are
strobed to the output on the rising edge of the ENCODE input
with four pipeline delays. If the user has selected Dual Channel
mode (OMS = LOW) the data are alternately directed between
the A and B output banks and the data has five pipeline delays.
At power-up, the N sample data can appear at either the A or B
Port. To align the data in a known state, the user must strobe
DATA SYNC (DS, DS) per the conditions described in the
Timing section.
Graphics Applications
The high bandwidth and low power of the AD9483 makes it very
attractive for applications that require the digitization of presampled waveforms, wherein the input signal rapidly slews from
one level to another, then is relatively stable for a period of time.
Examples of these include digitizing the output of computer
graphic display systems, and very high speed solid state imagers.
These applications require the converter to process inputs with
frequency components well in excess of the sampling rate (often
with subnanosecond rise times), after which the A/D must settle
and sample the input in well under one pixel time. The architecture of the AD9483 is vastly superior to older flash architectures,
which not only exhibit excessive input capacitance (which is very
hard to drive), but can make major errors when fed a very rapidly slewing signal. The AD9483’s extremely wide bandwidth
Track/Hold circuit processes these signals without difficulty.
Using the AD9483
Good high speed design practices must be followed when using the
AD9483. Decoupling capacitors should be physically as close as
possible to the chip to obtain maximum benefit. We recommend
placing a 0.1 µF capacitor at each power ground pin pair (14 total)
for high frequency decoupling and including one 10 µF capacitor for
local low frequency decoupling. Each of the three VREF IN pins
should also be decoupled by a 0.1 µF capacitor.
The part should be located on a solid ground plane and output trace
lengths should be short (<1 inch) to minimize transmission line effects.
This will avoid the need for termination resistors on the output bus
and reduces the load capacitance that needs to be driven, which in
turn minimizes on-chip noise due to heavy current flow in the
outputs. We have obtained optimum performance on our evaluation
board by tying all V
pins to a quiet analog power supply system
CC
and tying all GND pins to a quiet analog system ground.
Minimum Encode Rate
The minimum sampling rate for the AD9483 is 10 MHz for the
140 MSPS and 100 MSPS versions. To achieve this sampling
rate, the Track/Hold circuit employs a very small hold capacitor.
When operated below the minimum guaranteed sampling rate,
the T/H droop becomes excessive. This is first observed as an
increase in offset voltage, followed by degraded linearity at even
lower frequencies.
Lower effective sampling rates may be easily supported by operating the converter in Dual Port output mode and using only
one output channel. A majority of the power dissipated by the
AD9483 is static (not related to conversion rate), so the penalty
for clocking at twice the desired rate is not high.
Digital Inputs
SNR performance is directly related to the sampling clock stability in A/D converters, particularly for high input frequencies
and wide bandwidths.
ENCODE and Data Select (DS) can be driven differentially or
single-ended. For single-ended operation, the complement
inputs (ENCODE, DS) are internally biased to V
/3 (~1.5 V)
DD
by a high impedance on-chip resistor divider (Figure 5), but
they may be externally driven to establish an alternate threshold
if desired. A 0.1 µF decoupling capacitor to ground is sufficient
to maintain a threshold appropriate for TTL or CMOS logic.
When driven differentially, ENCODE and DS will accommodate differential signals centered between 1.5 V and 4.5 V with a
total differential swing ≥800 mV (V
≥400 mV).
ID
Note the 6-diode clock input protection circuitry in Figure 5.
This limits the differential input voltage to ±2.1 V. When the
diodes turn on, current is limited by the 300 Ω series resistor.
Exceeding 2.1 V across the differential inputs will have no
impact on the performance of the converter, but be aware of the
clock signal distortion that may be produced by the nonlinear
impedance at the converter.
DRIVING DIFFERENTIAL INPUTS DIFFERENTIALLY
CLOCK
CLOCK
CLOCK
0.1F
ENC
ENC
ENC
ENC
V
IH D
V
IC M
V
IL D
DRIVING DIFFERENTIAL INPUTS SINGLE-ENDEDLY
V
IN D
V
IC M
V
IL D
V
V
ID
ID
Figure 10. Input Signal Level Definitions
ADC Gain Control
Each of the three ADC channels has independent limited gain
control. The full-scale signal amplitude for a given ADC is set by
the dc voltage on its VREF In pin. The equation relating the full
scale amplitude to VREF In is as follows: FS = (0.4) × (VREF
IN). The three ADCs are optimized for a full-scale signal amplitude of 1 V, but will accommodate up to ±10% variation.
REV. C
–13–
Page 14
AD9483
ADC Offset Control
The offset for each of the three ADCs can be independently
controlled. For a single-ended analog input where the analog
input is connected to a reference, offset can be adjusted simply
by adjusting the dc voltage of the reference. For differential
analog inputs, the user must provide the offset in their signal.
Offset can be adjusted up or down as far as the common-mode
input range will allow.
Power Dissipation
Power dissipation for the AD9483 has two components, V
CC
and VDD. Power dissipation from VCC is relatively constant for a
given supply voltage, whereas power dissipation from VDD can
vary greatly. V
supplies power to the analog circuity. V
CC
DD
supplies power to the digital outputs and can be approximated
by the following equation:
P (V
) = 1/2 C × V2 ×F×N
DD
C =Output Load Capacitance
V =V
Supply Voltage
DD
F =Encode Frequency
N =Number of Outputs Switching
Nominally, C = 10 pF, V = 3.3 V, F = 140 MSPS, and N = 26.
N comes from the 24 output bits plus two clock outputs,
P(V
) = 197 mW.
DD
Power-Down
The power-down function allows users to reduce power dissipation when output data is not required. A TTL/CMOS HIGH
signal on pin 76, (PD), shuts down most of the chip and brings
the total power dissipation to less than 100 mW. The internal
bandgap voltage reference remains active during power-down
mode to minimize reactivation time. If the power-down function
is not desired, the PD pin should be tied to ground or held to a
TTL/CMOS LOW level.
Bandgap Voltage Reference
The AD9483 internal reference, VREF OUT (Pin 97), provides
a simple, cost effective reference for many applications. It exhibits reasonable accuracy and excellent stability over power supply
and temperature variations. The reference output can be used to
set the three ADCs’ gain and offset. The reference is capable of
providing up to 1 mA of additional current beyond the requirements of the AD9483.
As the ADC gain and offset are set by the reference inputs,
some applications may require a reference with greater accuracy
or temperature performance. In these cases, an external reference may be connected directly to the VREF IN pins. VREF
OUT, if unused, should be left floating. Note, each of the three
VREF IN pins will require up to 1 mA of current.
Modes of Operation
The AD9483 has three modes of operation, Single Channel
output mode, and a Dual Channel output mode with two possible data formats, interleaved or parallel. Two pins control which
mode of operation the chip is in, Pin 74 Output Mode Select
(OMS) and Pin 75 Interleaved/Parallel Select (I/P). Table II
shows the configuration required for each mode.
Table II. Output Mode Selection
MODEOMSI/P
Dual Channel—ParallelLOWLOW
Dual Channel—InterleavedLOWHIGH
Single ChannelHIGHDON’T CARE
Demuxed Output Mode
In demuxed mode, (Pin 74 OMS = LOW), the ADC output
data are alternated between the two output ports (Port A and
Port B). This limits the data output rate to 1/2 the rate of
ENCODE, and facilitates conversion rates up to 140 MSPS.
Demuxed output mode is recommended for guaranteed operation above 100 MSPS, but may be enabled at any specified
conversion rate.
Two data formats are possible in Dual Channel output mode,
parallel data out and interleaved data out. Pin 75 I/P should be
LOW for parallel format and HIGH for interleaved format.
Figures 1 and 2 show the timing requirements for each format.
Note that the Data Sync input, (DS), is required in Dual Channel output mode for both formats. The section on Data Sync
describes the requirements of the Data Sync input.
As shown in Figures 1 and 2, when using the interleaved data
format, a sample is taken on an ENCODE rising edge N. The
resulting data is produced on an output port following the fifth
rising edge of ENCODE after the sample was taken, (five pipeline delays). The following sample, (N+1), will be produced on
the opposite port, also five pipeline delays after it was taken.
The state of CLKOUT when the sample was taken will determine out of which port the data will come. If CLKOUT was
LOW, the data will come out Port A. If CLKOUT was HIGH,
the data will come out Port B.
In order to achieve parallel data format on the two output data
ports, the data is internally aligned. This is accomplished by adding
an extra pipeline delay to just the A Data Port. Thus, data coming out Port A will have six pipeline delays and data coming out
Port B will have five pipeline delays. As with the interleaved
format, the state of Data Sync when a sample is taken will
determine out of which port the data will come. If CLKOUT
was LOW, the data will come out Port A. If CLKOUT was
HIGH, the data will come out Port B.
–14–
REV. C
Page 15
AD9483
Data Sync
The Data Sync input, DS, is required to be driven for most
applications to guarantee at which output port a given sample
will appear. When DS is held high, the ADC data outputs and clockoutputs do not switch—they are held static. Synchronization is
accomplished by the assertion (falling edge) of DS, within the
timing constraints T
SDS
and T
edge. (On initial synchronization T
falls T
before a given encode rising edge N, the analog value
SDS
relative to an encode rising
HDS
is not relevant.) If DS
HDS
at that point in time will be digitized and available at Port A five
cycles later (interleaved mode). The very next sample, N+1, will
be sampled by the next rising encode edge and available at Port
B five cycles after that encode edge (interleaved mode). In dual
parallel mode the A port has a six cycle latency, the B port has a
five cycle latency as described in Demuxed Outputs Mode section.
DS can be asserted once per video line if desired by using the
horizontal sync signal (HSYNC). The start of HSYNC should
occur after the end of active video by at least the chip latency.
The HSYNC front porch is usually much greater than this in a
typical SXGA system. If this is true in a given system then DS
can be reset high by the HSYNC leading edge (the samples at
that point should not be required in a typical system). DS can
then be reasserted (brought low), by triggering from HSYNC
trailing edge—observing T
of the next rising encode edge.
SDS
The first pixel data (on A Port) would be available five cycles
after the first rising encode after HSYNC goes high.
It is possible to use the phase of the data clock outputs and
software programming to accommodate situations where DS is
not driven. The data clock outputs (CLKOUT and CLKOUT)
can be used to determine when data is valid on the output ports.
In these cases DS should be grounded and DS left floating or
connected to V
. If CLKOUT was low when a given sample
CC
was taken, the digitized value will be available on Port A, five
cycles later. Data Sync has no effect when Single Channel
Mode is selected, it should be grounded
Figure 2 shows how to use DS properly. The DS rising edge
does not have any special timing requirements except that no
data will come out of either port while it is held HIGH. The
falling edge of DS must, however, meet a minimum setup-andhold time with respect to the rising edge of ENCODE.
Single Channel Outputs Mode
In Single Channel mode, (Pin 74 OMS = HIGH), the timing
of the AD9483 is similar to any high speed ADC (Figure 1). A
sample is taken on every rising edge of ENCODE, and the resulting data is produced on the output pins following the fourth
rising edge of ENCODE after the sample was taken, (four pipeline delays). The output data are valid t
of ENCODE, and remain valid until at least t
after the rising edge
PD
after the next
V
rising edge of ENCODE.
The maximum conversion rate in the mode should be limited to
100 MSPS. This is recommended because the guaranteed output data valid time minus the propagation delay is only 4 ns at
100 MSPS. This is about as fast as standard logic is able to capture
the data with reasonable design margins. The AD9483 will
operate faster in this mode if the user is able to capture the data.
When operating in single channel mode, all data comes out the
A Ports while the B Ports are held static in a random state.
Data Clock Outputs
The data clock outputs will switch at two potential frequencies.
In Single Channel mode, where all data comes out of Port A
at the full ENCODE rate, the data clock outputs switch at the
same frequency as the ENCODE. In Dual Channel mode,
where the data alternates between the two ports, each of which
operate at 1/2 the full ENCODE rate, the data clock outputs
also switch at 1/2 the full ENCODE rate.
The data clock outputs have two potential purposes. The first is
to act as a latch signal for capturing output data. In order to do
this, simply drive the data latches with the appropriate data
clock output. The second use is in Dual Channel data mode to
help determine out of which data port data will come out. Refer
to Figure 2 for a complete timing diagram, but in this mode, a
rising edge on data clock will correspond to data switching on
data Port B.
LAYOUT AND BYPASSING CONSIDERATIONS
Proper high speed layout and bypassing techniques should be
used with the AD9483. Each V
and VDD power pin should be
CC
bypassed as close to the pin as possible with a 0.01 µF to 0.1 µF
capacitor Also, one 10 µF capacitor to ground should be used
per supply per board. The VREF OUT pin and each of the
three VREF IN pins should also be bypassed with a 0.01 µF to
0.1 µF capacitor to ground.
A single, substantial, low impedance ground plane should be
place under and around the AD9483. Try to maximize the
distance between the sensitive analog signals, (AIN, VREF),
and the digital signals. Capacitive loading on the digital outputs
should be kept to a minimum. This can be facilitated by keeping
the traces short and in the case of the clock outputs by driving
as few other devices as possible. Socketing the AD9483 should
also be avoided. Try to match trace lengths of similar signals to
avoid mismatches in propagation delays, (the encode inputs,
analog inputs, digital outputs).
POWER SUPPLIES
At power up, VCC must come up before VDD. VCC is considered
the converter supply, nominally 5.0 V (±5.0%) V
is consider
DD
output power supply, nominally 3.3 V (±10%) or 5.0 V (±5%).
At power off, V
must turn off first. Failure to observe the
DD
correct power supply sequencing many damage this device.
REV. C
–15–
Page 16
AD9483
EVALUATION BOARD
The AD9483 evaluation board offers an easy way to test the
AD9483. It provides ac or dc biasing for the analog input, it
generates the output latch clocks for Single Mode, Dual
Parallel Mode and Dual Interleaved Mode. Each of the three
channels has a reconstruction DAC (A Port only). The board
has several different modes of operation, and is shipped in
the following configuration:
• Single-ended ac coupled analog input (1 V p-p centered
at ground)
• Differential clock inputs (PECL) (See ENCODE section
for TTL drive)
• Internal voltage references connected to externally buffered on-chip reference (VREF OUT)
• Preset for Dual Mode Interleaved
Analog Input
The evaluation board accepts a 1 V p-p input signal centered
at ground for ac coupled input mode (Set Jumpers W4, W5,
W12, W13, W18, W17 to jump Pin 1 to Pin 2). This signal
biased up to 2.5 V by the on-chip reference. Note: input
signal should be bandlimited (filtered) prior to sampling to
avoid aliasing. The analog inputs are terminated to ground
by a 75 Ω resistor on the board. The analog inputs are ac
coupled through 0.1 µF caps C2, C4, C6 on top of the
board. These can be increased to accommodate lower frequency inputs if desired using test points PR1–PR6 on bottom of board. In dc-coupled input mode (Set Jumpers W4,
W5, W12, W13, W18, W17 to jump Pin 3 to Pin 2 ) the
board accepts typical video level signal levels (0 mV to 700 mV)
the signal is level shifted and amplified to 1 V p-p by the
AD8055 preamp. Variable Resistors R98–R100 are used to
adjust dc black level to 2 V at ADC inputs.
Encode
The AD9483 ENCODE input can be driven two ways.
1. Differential PECL (V
= 3, VHI = 4 nominal). It is
LO
shipped in this mode.
2. Single ended TTL or CMOS. (At Encode Bar–Remove
50 Ω termination resistor R10, add 0.1 µF capacitor C7)
Voltage Reference
The AD9483 has an internal 2.5 V voltage reference (VREF
OUT). This is buffered externally on board to support additional level shifting circuitry (the AD9483 VREF OUT pin can
drive the three VREF IN pins in applications where level shifting
is not required with no additional buffering). An external reference may be employed instead to drive each VREF IN pin
independently (requires moving Jumpers W14, W15, and W16).
Single Channel Mode
Single Channel mode sets the AD9483 to produce data on
every clock cycle on output port A only. The maximum speed
in Single Channel mode is 100 MSPS.
Dual Channel Modes (Outputs Clocked at 1/2 Encode Clock)
Dual Channel Interleaved
Sets the ADC to produce data alternately on Port A and Port B.
The maximum speed in this mode is 140 MSPS.
Dual Channel Parallel
Sets the ADC to produce data concurrently on Port A and
Port B. Maximum speed in this mode is 140 MSPS.
DAC Out
The DAC output is a representation of the data on output Port
A only. The DAC is terminated on the board into 75 Ω. Fullscale voltage swing at DAC output is nominally 0 mV to 800 mV
when terminated into external 75 Ω (doubly terminated).
Output Port B is not reconstructed. The DAC outputs are NOT
filtered and will exhibit sampling noise. The DACs can be powered down at W1, W2, and W3 (jumper not installed).
Data Ready
An output clock for latching the ADC outputs is available at
Pin 1 at the 25-pin connector. Its complement is located at
Pin 14. The clocks are terminated on the board by a 75 Ω
Thevenin termination to V
/2. The timing on these clock out-
D
puts can be inverted at W9, W10 (jumper not installed).
Schematics
The schematics for the evaluation board follow. (Note bypass
capacitors for ADC are shown in Figure 15.)
DESIGN NOTES
Maximum frequency for PARALLEL is 140 MHz.
Maximum frequency for INTERLEAVED is 140 MHz.
Maximum frequency for SINGLE is 100 MHz.
DS is tied to ground through a 50 Ω resistor.DS is left floating.
Figure 14. Digital Outputs Connectors and Terminations Section
–20–
P16
P15
R38
100⍀
BLUE_A4BL_A4
P14
R39
100⍀
BLUE_A5BL_A5
P13
R40
BLUE_A6BL_A6
P12
P10
10
11
P11
P10
10
10
P10
100⍀
BLUE_A7BL_A7
ST1
ST4
R61
100⍀
BLUE_B0BL_B0
U13
12345
12345
R60
R62
100⍀
BLUE_B1BL_B1
P1P2P3P4P5
P1P2P3P4P5
R63
R64
100⍀
100⍀
100⍀
BLUE_B2BL_B2
BLUE_B3BL_B3
ST7
ST8
R59
BLUE_B4BL_B4
GND
R58
100⍀
BLUE_B5BL_B5
REV. C
100⍀
BLUE_B6BL_B6
R57
100⍀
BLUE_B7BL_B7
Page 21
VA
C55
C26
C25
C24
C22
C21
C23
C57
C19
C50
C20
10F
0.1F
0.1F
0.1F
0.1F
0.1F
0.1F
0.1F
0.1F
0.1F
0.1F
–VA
C65
0.1F
C62
0.1F
C61
0.1F
C60
0.1F
C63
10F
VD
C34
0.1F
C33
0.1F
C32
0.1F
C31
0.1F
C30
0.1F
BYPASS CAPS
C29
0.1F
C28
0.1F
VD
C56
C45
C44
C43
C42
C40
C39
C38
AD9483
16
REF
B
R96
1.3k⍀
R100
A_LAT
5
2
B
TRIM
500⍀
R97
4
3
1.5k⍀
REF
C
DATA_LOCK_OUT
W11
LATCH CLK SOURCE SELECT
10F
0.1F
0.1F
0.1F
0.1F
0.1F
0.1F
0.1F
REF
A
R92
A
TRIM
1.3k⍀
R93
1.5k⍀
R98
500⍀
B_LAT
DATA_LOCK_OUT
C
TRIM
REV. C
C37
C18
0.1F
C36
0.1F
0.1F
7
6
AD9483
4
3
OUT
REF
REF SOURCE SELECT
C46
0.1F
3
1
2
W14
C41
–VA
0.1F
C51
10F
EXT REF A
W15
1
2
C35
0.1F
C49
10F
VA
2
C47
3
EXT REF B
0.1F
C53
R95
1.3k⍀
R94
1.5k⍀
R99
500⍀
C48
0.1F
3
1
2
W16
10F
C54
10F
EXT REF C
C27
C52
AD9483 EXTERNAL
REFERENCES
REF A
REF B
REF C
POWER/DC INPUTS
–VA AD9483 SUPPORT LOGIC – SUPPLY
VA AD9483 ANALOG SUPPLY
–VA AD9483 DIGITAL SUPPLY
–VA AD9483 SUPPORT LOGIC + SUPPLY
EXT
EXT
EXT
1234567
TB1
8
0.1F
10F
GND
Figure 15. Power Connector, Decoupling Capacitors, DC Adjust Variable Resistors Section
–21–
Page 22
AD9483
PCB LAYOUT
The PCB is designed on a four layer (1 oz. Cu) board. Components and routing are on the top layer with a ground flood for
additional isolation. Test and ground points were judiciously
placed to facilitate high speed probing. Each channel has a
separate 25-pin connector for it’s digital outputs. A common
ground plane exists on the second layer.
The third layer has the 3 split power planes:
1. 5 V analog for the ADC and preamps,
2. 3.3 V (or 5 V) ADC output supply, and
3. A separate 3.3 V supply for support logic. The fourth layer
contains the –5 V plane for the preamps and additional
components and routing. There is additional space for two
extra components on top of the board to allow for modification.
NOTES
All resistors are surface mount (size 1206) and have a 1% tolerance.
Jumpers are Samtec parts TSW-110-08-G-D and TSW-110-08-G-S.
Jumpers W1, W2, W3, W9, W8, W10 are omitted.
REV. C
–25–
Page 26
AD9483
OUTLINE DIMENSIONS
100-Lead Metric Quad Flat Package [MQFP]
(S-100B)
Dimensions shown in millimeters
23.20 BSC
20.00 BSC
18.85 REF
80
81
51
50
51
50
80
81
14.00
TOP VIEW
(PINS DOWN)
PIN 1
100
1
0.65 BSC
3.40
MAX
1.03
0.88
0.73
NOTE: THE AD9483KS PACKAGE USES A COPPER INSERT TO HELP DISSIPATE HEAT AND ENSURE RELIABL
OPERATION OVER THE FULL 0ⴗ C TO +85ⴗ C TEMPERATURE RANGE. THIS COPPER INSERT IS EXPOSED ON
THE UNDERSIDE OF THE DEVICE. IT IS RECOMMENDED THAT DURING THE DESIGN OF THE PC BOARD NO
THROUGH HOLES OR SIGNAL TRACES BE PLACED UNDER THE AD9483 THAT COULD COME IN CONTACT WITH
THE COPPER INSERT. COMMONLY ACCEPTED BOARD LAYOUT PRACTICES FOR HIGH SPEED CONVERTERS
SPECIFY THAT ONLY GROUND PLANES SHALL BE LOCATED UNDER THESE DEVICES TO MINIMIZE NOISE OR
DISTORTION OF VIDEO SIGNALS.
SEATING
PLANE
COMPLIANT TO JEDEC STANDARDS MS-022-GC-1, WITH THE ADDITION OF THE HEATSINK