NSC ADC08D1520QML

ADC08D1520QML
Low Power, 8-Bit, Dual 1.5 GSPS or Single 3.0 GSPS A/D
Converter
General Description
Features
The ADC08D1520QML is an 8–Bit, dual channel, low power,
high performance CMOS analog-to-digital converter that
builds upon the ADC08D1000 platform. The ADC08D1520QML digitizes signals to 8 bits of resolution at
sample rates up to 1.7 GSPS. It has expanded features compared to the ADC08D1000, which include a test pattern output
for system debug, clock phase adjust, and selectable output
demultiplexer modes. Consuming a typical 2.0W in Demultiplex Mode at 1.5 GSPS from a single 1.9 Volt supply, this
device is guaranteed to have no missing codes over the full
operating temperature range. The unique folding and interpolating architecture, the fully differential comparator design,
the innovative design of the internal sample-and-hold amplifier and the self-calibration scheme enable a very flat response of all dynamic parameters beyond Nyquist, producing
a high 7.2 Effective Number of Bits (ENOB) with a 748 MHz
input signal and a 1.5 GHz sample rate while providing a
10-18 Code Error Rate (C.E.R.) Output formatting is offset binary and the Low Voltage Differential Signaling (LVDS) digital
outputs are compatible with IEEE 1596.3-1996, with the exception of an adjustable common mode voltage between 0.8V
and 1.2V.
Each converter has a selectable output demultiplexer which
feeds two LVDS buses. If the 1:2 Demultiplexed Mode is selected, the output data rate is reduced to half the input sample
rate on each bus. When Non-Demultiplexed Mode is selected, that output data rate on channels DI and DQ are at the
same rate as the input sample clock. The two converters can
be interleaved and used as a single 3 GSPS ADC.
The converter typically consumes less than 2.9 mW in the
Power Down Mode and is available in a 128-pin, thermally
enhanced, multi-layer ceramic quad package and operates
over the Military (-55°C ≤ TA ≤ +125°C) temperature range.
■
■
■
■
■
■
■
■
■
■
Single +1.9V ±0.1V Operation
Interleave Mode for 2x Sample Rate
Multiple ADC Synchronization Capability
Adjustment of Input Full-Scale Range, Offset and Clock
Phase Adjustment
Choice of SDR or DDR output clocking
1:1 or 1:2 Selectable Output Demux
Second DCLK output
Duty Cycle Corrected Sample Clock
Test pattern
Serial Interface for Extended Control
Key Specifications
■
■
■
■
■
■
■
■
Resolution
Max Conversion Rate
Code Error Rate
ENOB @ 748 MHz Input
DNL
Total Ionizing Dose
Single Event Latch-up
Power Consumption
— Operating in 1:2 Demux Output
— Power Down Mode
8 Bits
1.5 GSPS (min)
10-18 (typ)
7.2 Bits (typ)
±0.15 LSB (typ)
300 krad(Si)
120 MeV-cm2/mg
2.0 W (typ)
2.9 mW (typ)
Applications
■
■
■
■
Direct RF Down Conversion
Digital Oscilloscopes
Communications Systems
Test Instrumentation
Ordering Information
NS Part Number
SMD Part Number
NS Package Number
ADC08D1520WG-QV
5962–0721401VZC
EM128A
128L, CERQUAD
GULLWING
ADC08D1520WGFQV
5962F0721401VZC
300 krad(Si)
EM128A
128L, CERQUAD
GULLWING
© 2008 National Semiconductor Corporation
300247
Package Discription
www.national.com
ADC08D1520QML Low Power, 8-Bit, Dual 1.5 GSPS or Single 3.0 GSPS A/D Converter
March 6, 2008
ADC08D1520QML
Block Diagram
30024753
www.national.com
2
ADC08D1520QML
Pin Configuration
30024701
Note: The exposed pad on the bottom of the package must be soldered to a ground plane to ensure rated performance.
3
www.national.com
ADC08D1520QML
Pin Descriptions and Equivalent Circuits
Pin Functions
Pin No.
Symbol
Equivalent Circuit
Description
3
OutV / SCLK
Output Voltage Amplitude and Serial Interface Clock. Tie this
pin high for normal differential DCLK and data amplitude.
Ground this pin for a reduced differential output amplitude
and reduced power consumption. See 1.1.6 The LVDS
Outputs. When the extended control mode is enabled, this
pin functions as the SCLK input which clocks in the serial
data. See 1.2 NON-EXTENDED CONTROL/EXTENDED
CONTROL for details on the extended control mode. See
1.3 THE SERIAL INTERFACE for description of the serial
interface.
29
PDQ
A logic high on the PDQ pin puts only the Q-Channel ADC
into the Power Down mode.
OutEdge / DDR /
SDATA
DCLK Edge Select, Double Data Rate Enable and Serial
Data Input. This input sets the output edge of DCLK+ at
which the output data transitions. See 1.1.5.2 OutEdge and
Demultiplex Control Setting When this pin is connected to
1/2 the supply voltage,VA/2, DDR clocking is enabled. When
the Extended Control Mode is enabled, this pin functions as
the SDATA input. See 1.2 NON-EXTENDED CONTROL/
EXTENDED CONTROL for details on the Extended Control
Mode. See 1.3 THE SERIAL INTERFACE for description of
the serial interface.
15
DCLK_RST/
DCLK_RST+
DCLK Reset. When single-ended DCLK_RST is selected by
setting pin 52 logic high or to VA/2, a positive pulse on this
pin is used to reset and synchronize the DCLK outputs of
multiple converters. See 1.5 MULTIPLE ADC
SYNCHRONIZATION for detailed description. When
differential DCLK_RST is selected by setting pin 52 logic low,
this pin receives the positive polarity of a differential pulse
signal used to reset and synchronize the DCLK outputs of
multiple converters.
26
PD
Power Down Pins. A logic high on the PD pin puts the entire
device into the Power Down Mode.
CAL
Calibration Cycle Initiate. A minimum tCAL_L input clock
cycles logic low followed by a minimum of tCAL_H input clock
cycles high on this pin initiates the calibration sequence. See
2.5.2 Calibration for an overview of calibration and 2.5.2.1
Initiating Calibration for a description of calibration.
FSR/DCLK_RST-
Full Scale Range Select, Alternate Extended Control Enable
and DCLK_RST-. This pin has two functions. It can
conditionally control the ADC full-scale voltage, or become
the negative polarity signal of a differential pair in differential
DCLK_RST Mode. If pin 52 and pin 41 are connected at logic
high, this pin can be used to set the full-scale-range. When
used as the FSR pin, a logic low on this pin sets the full-scale
differential input range to a reduced VIN input level. A logic
high on this pin sets the full-scale differential input range to
Higher VIN input level. See Converter Electrical
Characteristics. When pin 52 is held at logic low, this pin acts
as the DCLK_RST- pin. When in differential DCLK_RST
Mode, there is no pin-controlled FSR and the full-scale-range
is defaulted to the higher VIN input level.
4
30
14
www.national.com
4
Pin No.
Symbol
Equivalent Circuit
Description
DES / SCS
Dual Edge Sampling and Serial Interface Chip Select. With
pin 41 logic low, the device is in Extended Control Mode and
this pin is the enable pin for the Serial Interface . When in
Non-Extended Control Mode and this pin is connected to
VA/2, DES Mode is selected where the I- Channel input is
sampled at twice the input clock rate and the Q- Channel
input is ignored. See 1.1.5.1 Dual-Edge Sampling. When in
Non-Extended Controll Mode and DES is not desired, this
pin should be tied to VA.
CLK+
CLK-
LVDS Clock input pins for the ADC. The differential clock
signal must be a.c. coupled to these pins. The input signal is
sampled on the falling edge of CLK+. See 1.1.2 Acquiring
the Input for a description of acquiring the input and 2.4 THE
CLOCK INPUTS for an overview of the clock inputs.
10
11
.22
23
VINI−
VINI+
VINQ+
VINQ−
Analog signal inputs to the ADC. These differential input
signals must be a.c. coupled to these pins. The differential
full-scale input range is programmable using the FSR pin 14
in Non-Extended Control Mode and the Input Full-Scale
Voltage Adjust register in the Extended Control Mode. Refer
to the VIN specification in the Converter Electrical
Characteristics for the full-scale input range in the NonExtended Control Mode. Refer to 1.4 REGISTER
DESCRIPTION for the full-scale input range in the Extended
Control Mode.
31
VBG
Bandgap output voltage. This pin is capable of sourcing or
sinking 100 μA and can drive a load up to 80 pF.
126
CalRun
32
REXT
127
18
19
Calibration Running indication. This pin is at a logic high
when calibration is running.
External bias resistor connection. Nominal value is 3.3 kΩ
(±0.1%) to ground. See 1.1.1 Calibration.
5
www.national.com
ADC08D1520QML
Pin Functions
ADC08D1520QML
Pin Functions
Pin No.
Symbol
34
35
Tdiode_P
Tdiode_N
41
52
www.national.com
Equivalent Circuit
Description
Temperature Diode Positive (Anode and Negative
(Cathode). This pin is used for die temperature
measurements. See 2.7.2 Thermal Management.
ECE
Extended Control Enable. This pin always enables or
disables Extended Control Mode. When this pin is set logic
high, the Extended Control Mode is inactive and all control
of the device must be through control pins only . When it is
set logic low, the Extended Control Mode is active. This pin
overrides the Extended Control Enable signal set using pin
14.
DRST_SEL
DCLK_RST select. This pin selects whether the DCLK is
reset using a single-ended or differential signal. When this
pin is connected at logic high, the DCLK_RST operation is
single-ended and pin 14 functions as FSR/ALT_ECE. When
this pin is logic low, the DCLK_RST operation becomes
differential with functionality on pin 15 (DCLK_RST+) and pin
14 (DCLK_RST-). When in differential DCLK_RST Mode,
there is no pin-controlled FSR and the full-scale-range is
defaulted to 870mV. When pin 41 is set logic low, the
Extended Control Mode is active and the Full-Scale Voltage
Adjust registers can be programmed.
6
ADC08D1520QML
Pin Functions
Pin No.
Symbol
83 / 78
84 / 77
85 / 76
86 / 75
89 / 72
90 / 71
91 / 70
92 / 69
93 / 68
94 / 67
95 / 66
96 / 65
100 / 61
101 / 60
102 / 59
103 / 58
DI7− / DQ7−
DI7+ / DQ7+
DI6− / DQ6−
DI6+ / DQ6+
DI5− / DQ5−
DI5+ / DQ5+
DI4− / DQ4−
DI4+ / DQ4+
DI3− / DQ3−
DI3+ / DQ3+
DI2− / DQ2−
DI2+ / DQ2+
DI1− / DQ1−
DI1+ / DQ1+
DI0− / DQ0−
DI0+ / DQ0+
I- and Q- channel LVDS Data Outputs that are not delayed
in the output demultiplexer. Compared with the DId and DQd
outputs, these outputs represent the later time samples.
These outputs should always be terminated with a 100Ω
differential resistor.
104 / 57
105 / 56
106 / 55
107 / 54
111 / 50
112 / 49
113 / 48
114 / 47
115 / 46
116 / 45
117 / 44
118 / 43
122 / 39
123 / 38
124 / 37
125 / 36
DId7− / DQd7−
DId7+ / DQd7+
DId6− / DQd6−
DId6+ / DQd6+
DId5− / DQd5−
DId5+ / DQd5+
DId4− / DQd4−
DId4+ / DQd4+
DId3− / DQd3−
DId3+ / DQd3+
DId2− / DQd2−
DId2+ / DQd2+
DId1− / DQd1−
DId1+ / DQd1+
DId0− / DQd0−
DId0+ / DQd0+
I- and Q- channel LVDS Data Outputs that are delayed by
one CLK cycle in the output demultiplexer. Compared with
the DI and DQ outputs, these outputs represent the earlier
time sample. These outputs should always be terminated
with a 100Ω differential resistor. In Non Demux Mode, these
outputs are disabled and are high impedance. When
disabled, these outputs must be left floating.
OR+/DCLK2+
OR-/DCLK2-
Out Of Range output. A differential high at these pins
indicates that the differential input is out of range ±VIN/2 as
programmed by the FSR pin in Non-Extended Control Mode
or the Input Full-Scale Voltage Adjust register setting in the
Extended Control Mode). DCLK2 is the exact mirror of DCLK
and should output the same signal at the same rate.
DCLKDCLK+
Data Clock. Differential Clock outputs used to latch the
output data. Delayed and non-delayed data outputs are
supplied synchronous to this signal. In 1:2 Demultiplexed
Mode, this signal is at 1/2 the input clock rate in SDR Mode
and at 1/4 the input clock rate in the DDR Mode. By default,
the DCLK outputs are not active during the termination
resistor trim section of the calibration cycle. If a system
requires DCLK to run continuously during a calibration cycle,
the termination resistor trim portion of the cycle can be
disabled by setting the Resistor Trim Disable (RTD) bit to
logic high in the Extended Configuration Register (address
9h). This disables all subsequent termination resistor trims
after the initial trim which occurs during the power on
calibration. Therefore, this output is not recommended as a
system clock unless the resistor trim is disabled. When the
device is in the Non-Demultiplexed Mode, DCLK can only be
in DDR Mode and the signal is at 1/2 the input clock rate.
79
80
81
82
Equivalent Circuit
Description
7
www.national.com
ADC08D1520QML
Pin Functions
Pin No.
Symbol
2, 5, 8, 13,
16, 17, 20,
25, 28, 33,
128
VA
Analog power supply pins. Bypass these pins to ground.
40, 51, 62,
73, 88, 99,
110, 121
VDR
Output Driver power supply pins. Bypass these pins to DR
GND.
1, 6, 7, 9,
12, 21, 24,
27
GND
Ground return for VA.
42, 53, 64,
74, 87, 97,
108, 119
DR GND
Ground return for VDR.
63, 98, 109,
120
NC
www.national.com
Equivalent Circuit
Description
No Connection. Make no connection to these pins.
8
(Notes 1, 2)
Ambient Temperature Range
(Notes 1, 2)
Supply Voltage (VA, VDR)
Supply Difference
VDR - VA
Voltage on Any Input Pin
Ground Difference
|GND - DR GND|
Input Current at Any Pin (Note 3)
Package Input Current (Note 3)
Junction Temperature
VA/2 Tolerance for supply 1.9V
2.2V
Supply Voltage (VA)
Driver Supply Voltage (VDR)
VIN+, VIN- Voltage Range
(Maintaining Common Mode)
Ground Difference
(|GND - DR GND|)
CLK Pins Voltage Range
Differential CLK Amplitude
0V to 100 mV
−0.15V to (VA +0.15V)
ESD Susceptibility (Note 4)
Human Body Model
Storage Temperature
0V to 100 mV
±25 mA
±50 mA
≤ 175°C
Class 3A (6000V)
−65°C to +175°C
−55°C ≤ TA ≤ +125°C
650mV ≥ VA/2 ≤ 1.2V
+1.8V to +2.0V
+1.8V to VA
200mV to VA
0V
0V to VA
0.4VP-P to 2.0VP-P
Package Thermal Resistance
Package
θJA
θJC
Top of
Package
θJC
Thermal
Pad
128L Cer Quad
Gullwing
11.5°C/ W
3.8°C/ W
2.0°C/ W
Soldering
process
must
comply
with
National
Semiconductor’s Reflow Temperature Profile specifications.
Refer to www.national.com/packaging.
Quality Conformance Inspection
MIL-STD-883, Method 5005 - Group A
Subgroup
Description
1
Static tests at
Temp ( C)
+25
2
Static tests at
+125
3
Static tests at
-55
4
Dynamic tests at
+25
5
Dynamic tests at
+125
6
Dynamic tests at
-55
7
Functional tests at
+25
8A
Functional tests at
+125
8B
Functional tests at
-55
9
Switching tests at
+25
10
Switching tests at
+125
11
Switching tests at
-55
12
Setting time at
+25
13
Setting time at
+125
14
Setting time at
-55
9
www.national.com
ADC08D1520QML
Operating Ratings
Absolute Maximum Ratings
ADC08D1520QML
ADC08D1520QML Converter Electrical Characteristics
DC Parameters
(Note 14)
The following specifications apply after calibration for VA = VDR = +1.9V; OutV = 1.9V; VIN FSR (a.c. coupled) = differential 870
mVP-P; CL = 10 pF; Differential, a.c. coupled Sine Wave Input Clock, fCLK = 1.5 GHz at 0.5 VP-P with 50% duty cycle; VBG = Floating;
Non-Extended Control Mode; SDR Mode; REXT = 3300 Ω ±0.1%; Analog Signal Source Impedance = 100 Ω Differential; 1:2 Output
Demultiplex, duty cycle stabilizer on. Boldface limits apply for TA = TMIN to TMAX. All other limits TA = 25°C, unless otherwise
noted. (Notes 5, 6)
Symbol
Parameter
Conditions
Notes
Typical
(Note 7)
Min
Max
Units
Subgroups
STATIC CONVERTER CHARACTERISTICS
INL
Integral Non-Linearity
(Best fit)
DC Coupled, 1 MHz Sine Wave
Overanged
±0.3
±.9
LSB
1, 2, 3
DNL
Differential Non-Linearity
DC Coupled, 1 MHz Sine Wave
Overanged
±0.15
±.6
LSB
1, 2, 3
8
Bits
1, 2, 3
1.5
LSB
1, 2, 3
Resolution with No Missing
Codes
VOFF
Offset Error
PFSE
Positive Full-Scale Error
(Note
8)
−0.6
±25
mV
1, 2, 3
NFSE
Negative Full-Scale Error
(Note
8)
−1.31
±25
mV
1, 2, 3
mVP-P
1, 2, 3
mVP-P
1, 2, 3
mVP-P
1, 2, 3
mVP-P
1, 2, 3
Ω
1, 2, 3
Ω
1, 2, 3
V
1, 2, 3
V
1, 2, 3
VP-P
1, 2, 3
VP-P
1, 2, 3
VP-P
1, 2, 3
VP-P
1, 2, 3
V
1, 2, 3
V
1, 2, 3
mVP-P
1, 2, 3
mVP-P
1, 2, 3
mVP-P
1, 2, 3
mVP-P
1, 2, 3
−0.55
−1.5
ANALOG INPUT AND REFERENCE CHARACTERISTICS
FSR pin 14 Low
VIN
600
Full Scale Analog
Differential Input Range
FSR pin 14 High
RIN
900
Differential Input
Resistance
100
530
650
840
960
94
106
ANALOG OUTPUT CHARACTERISTICS
VBG
Bandgap Reference
Output Voltage
IBG = ±100 µA
1.26
Sine Wave Clock
0.6
1.20
1.33
CLOCK INPUT CHARACTERISTICS
VID
Differential Clock Input
Level
Square Wave Clock
0.6
.5
2.0
.5
2.0
DIGITAL CONTROL PIN CHARACTERISTICS
VIH
Logic High Input Voltage
VIL
Logic Low Input Voltage
0.85 x
VA
0.15 x
VA
DIGITAL OUTPUT CHARACTERISTICS
VOD
LVDS Differential Output
Voltage
www.national.com
Measured differentially,
OutV = VA, VBG = Floating
(Note
13)
Measured differentially,
OutV = GND, VBG = Floating
10
(Note
13)
780
590
580
920
380
720
Parameter
Conditions
Notes
Typical
(Note 7)
Min
Max
Units
Subgroups
POWER SUPPLY CHARACTERISTICS
820
565
1.5
875
615
mA (max)
mA (max)
mA
1, 2, 3
1, 2, 3
IA
Analog Supply Current
1:2 Demux Output
PD = PDQ = Low
PD = Low, PDQ = High
PD = PDQ = High
IDR
Output Driver Supply
Current
1:2 Demux Output
PD = PDQ = Low
PD = Low, PDQ = High
PD = PDQ = High
230
125
0.018
290
170
mA (max)
mA (max)
mA
1, 2, 3
1, 2, 3
Power Consumption
1:2 Demux Output
PD = PDQ = Low
PD = Low, PDQ = High
PD = PDQ = High
2
1.3
2.9
2.2
1.49
W (max)
W (max)
mW
1, 2, 3
1, 2, 3
Max
Units
Subgroups
Bits (min)
4, 5, 6
Bits (min)
4, 5, 6
dB (min)
4, 5, 6
dB (min)
4, 5, 6
dB (min)
4, 5, 6
dB (min)
4, 5, 6
dB (max)
4, 5, 6
dB (max)
4, 5, 6
dB (min)
4, 5, 6
dB (min)
4, 5, 6
PD
AC Parameters
Symbol
(Note 14)
Parameter
Conditions
Notes
Typical
(Note 7)
Min
Non-DES MODE DYNAMIC CONVERTER CHARACTERISTICS, 1:2 DEMUX MODE
ENOB
Effective Number of Bits
SINAD
Signal-to-Noise Plus
Distortion Ratio
SNR
Signal-to-Noise Ratio
THD
Total Harmonic Distortion
SFDR
Spurious-Free dynamic
Range
Out of Range Output Code
fIN = 373 MHz, VIN = FSR − 0.5 dB
7.4
fIN = 748 MHz, VIN = FSR − 0.5 dB
7.2
fIN = 373 MHz, VIN = FSR − 0.5 dB
46.3
fIN = 748 MHz, VIN = FSR − 0.5 dB
45.4
fIN = 373 MHz, VIN = FSR − 0.5 dB
47
fIN = 748 MHz, VIN = FSR − 0.5 dB
45
fIN = 373 MHz, VIN = FSR − 0.5 dB
−53.4
fIN = 748 MHz, VIN = FSR − 0.5 dB
−53
fIN = 373 MHz, VIN = FSR − 0.5 dB
55.5
fIN = 748 MHz, VIN = FSR − 0.5 dB
53
7
43.9
43.9
−47.5
47.5
(VIN+) − (VIN−) > + Full Scale
255
4, 5, 6
(VIN+) − (VIN−) < − Full Scale
0
4, 5, 6
INTERLEAVE MODE (DES Pin 127=VA/2) - DYNAMIC CONVERTER CHARACTERISTICS, 1:4 DEMUX MODE
ENOB
Effective Number of Bits
fIN = 373 MHz, VIN = FSR − 0.5 dB
7.0
6.6
Bits
4, 5, 6
SINAD
Signal to Noise Plus
Distortion Ratio
fIN = 373 MHz, VIN = FSR − 0.5 dB
44
41.5
dB
4, 5, 6
SNR
Signal to Noise Ratio
fIN = 373 MHz, VIN = FSR − 0.5 dB
44
41.5
dB
4, 5, 6
THD
Total Harmonic Distortion
fIN = 373 MHz, VIN = FSR − 0.5 dB
−55
dB
4, 5, 6
SFDR
Spurious Free Dynamic
Range
fIN = 373 MHz, VIN = FSR − 0.5 dB
50
dB
4, 5, 6
11
−45.2
44.1
www.national.com
ADC08D1520QML
Symbol
ADC08D1520QML
AC Timing Parameters
Symbol
Parameter
(Note 14)
Conditions
Notes
Typical
(Note 7)
Min
Max
Units
Subgroups
1.5
GHz
9, 10, 11
1.0
GHz
9, 10, 11
AC TIMING CHARACTERISTICS
fCLK(max)
Maximum Input Clock
Frequency
Non-DES Mode or DES Mode in
1:2 Output Demux
Non-DES Mode or DES Mode in
Non-demux Output
DCLK Duty Cycle
tPWR
Pulse Width DCLK_RST±
tCAL_L
CAL Pin Low Time
tCAL_H
CAL Pin High Time
www.national.com
1.7
50
See Figure 10
See Figure 10
12
%
9, 10, 11
55
%
9, 10, 11
4
CLK±
Cycles
(min)
9, 10, 11
1280
CLK±
Cycles
9, 10, 11
1280
CLK±
Cycles
9, 10, 11
45
DC Parameters
The following specifications apply after calibration for VA = VDR = +1.9V; OutV = 1.9V; VIN FSR (a.c. coupled) = differential 870
mVP-P; CL = 10 pF; Differential, a.c. coupled Sine Wave Input Clock, fCLK = 1.5 GHz at 0.5 VP-P with 50% duty cycle; VBG = Floating;
Non-Extended Control Mode; SDR Mode; REXT = 3300 Ω ±0.1%; Analog Signal Source Impedance = 100 Ω Differential; 1:2 Output
Demultiplex, duty cycle stabilizer on. Boldface limits apply for TA = TMIN to TMAX. All other limits TA = 25°C, unless otherwise
noted. (Notes 5, 6)
Symbol
Parameter
Conditions
Notes
Typical
(Note 7)
Units
STATIC CONVERTER CHARACTERISTICS
VOFF_ADJ
Input Offset Adjustment Range
Extended Control Mode
±45
mV
FS_ADJ
Full-Scale Adjustment Range
Extended Control Mode
±20
%FS
0.02
pF
1.6
pF
0.08
pF
2.2
pF
61
ppm/°C
80
pF
192 µA vs 12 µA, TJ = 25°C
71.23
mV
192 µA vs 12 µA, TJ = 125°C
94.8
mV
ANALOG INPUT AND REFERENCE CHARACTERISTICS
CIN
Analog Input Capacitance,
Normal operation
Differential
Analog Input Capacitance,
DES Mode
Differential
Each input pin to ground
Each input pin to ground
(Note 9)
(Note 9)
ANALOG OUTPUT CHARACTERISTICS
TC VBG
Bandgap Reference Voltage
Temperature Coefficient
CLOAD VBG
Maximum Bandgap Reference
load Capacitance
TA = −55°C to +125°C,
IBG = ±100 µA
TEMPERATURE DIODE CHARACTERISTICS
ΔVBE
Temperature Diode Voltage
CHANNEL-TO-CHANNEL CHARACTERISTICS
Offset Match
1
LSB
Positive Full-Scale Match
Zero offset selected in Control Register
1
LSB
Negative Full-Scale Match
Zero offset selected in Control Register
1
LSB
Phase Matching (I,Q)
fIN = 1.0 GHz
<1
Degree
X-TALK
Crosstalk from I- Channel
(Aggressor) to Q- Channel
(Victim)
Aggressor = 1160 MHz F.S.
Victim = 100 MHz F.S.
−66
dB
X-TALK
Crosstalk from Q- Channel
(Aggressor) to I- Channel
(Victim)
Aggressor = 1160 MHz F.S.
Victim = 100 MHz F.S.
−66
dB
VIN = 0 or VIN = VA
±1
µA
0.02
pF
1.5
pF
1.2
pF
±1
mV
CLOCK INPUT CHARACTERISTICS
II
Input Current
CIN
Input Capacitance
Differential
Each input to ground
(Note 9)
DIGITAL CONTROL PIN CHARACTERISTICS
CIN
Input Capacitance
Each input to ground
(Note 11)
DIGITAL OUTPUT CHARACTERISTICS
ΔVO DIFF
Change in LVDS Output Swing
Between Logic Levels
VOS
Output Offset Voltage
VBG = Floating (See Figure 1)
VOS
Output Offset Voltage
VBG = VA (See Figure 1)
ΔVOS
Output Offset Voltage Change
Between Logic Levels
IOS
Output Short Circuit Current
Output+ & Output− connected to 0.8V,
VBG = Floating, OutV = VA
13
(Note 13)
800
mV
1100
mV
±1
mV
±4
mA
www.national.com
ADC08D1520QML
Typical Electrical Characteristics
ADC08D1520QML
Symbol
Parameter
Conditions
Notes
Typical
(Note 7)
Units
ZO
Differential Output Impedance
100
Ω
VOH
CalRun H level output
IOH = −400 µA
(Note 10)
1.72
V
VOL
CalRun L level output
IOH = 400 µA
(Note 10)
0.17
V
POWER SUPPLY CHARACTERISTICS
PSRR1
D.C. Power Supply Rejection
Ratio
Change in Full Scale Error with change in
VA from 1.8V to 2.0V
30
dB
PSRR2
A.C. Power Supply Rejection
Ratio
248 MHz, 50 mVP-P injected on VA
51
dB
Typical
(Note 7)
Units
2.0
GHz
10−18
Error/Sample
d.c. to 498 MHz
±0.5
dBFS
d.c. to 1 GHz
±1.0
dBFS
fIN = 373 MHz, VIN = FSR − 0.5 dB
−60
dB
fIN = 748 MHz, VIN = FSR − 0.5 dB
−55
dB
fIN = 373 MHz, VIN = FSR − 0.5 dB
−62
dB
fIN = 748 MHz, VIN = FSR − 0.5 dB
−58
dB
fIN1 = 365 MHz, VIN = FSR − 7 dB
fIN2 = 375 MHz, VIN = FSR − 7 dB
−50
dB
AC Parameters
Symbol
Parameter
Conditions
Notes
Non-DES MODE DYNAMIC CONVERTER CHARACTERISTICS, 1:2 DEMUX MODE
FPBW
C.E.R.
Full Power Bandwidth
Code Error Rate
Gain Flatness
2nd Harm
Second Harmonic Distortion
3rd Harm
Third Harmonic Distortion
IMD
Non-DES Mode
Intermodulation Distortion
INTERLEAVE MODE (DES Pin 127=VA/2) - DYNAMIC CONVERTER CHARACTERISTICS, 1:4 DEMUX MODE
FPBW
Full Power Bandwidth
Dual Edge Sampling Mode
1.7
GHz
2nd Harm
Second Harmonic Distortion
fIN = 373 MHz, VIN = FSR − 0.5 dB
−60
dB
3rd Harm
Third Harmonic Distortion
fIN = 373 MHz, VIN = FSR − 0.5 dB
−65
dB
www.national.com
14
Symbol
Typical
(Note 7)
Units
Non-DES Mode
200
MHz
DES Mode
500
Parameter
Conditions
Notes
AC TIMING CHARACTERISTICS
fCLK(min)
Minimum Input Clock Frequency
Input Clock Duty Cycle
200 MHz ≤ fCLK ≤ 1.5 GHz
(Non-DES Mode)
(Note 10)
50
500 MHz ≤ fCLK ≤ 1.5 GHz
(DES Mode)
(Note 10)
50
MHz
% (min)
% (max)
% (min)
% (max)
tCL
Input Clock Low Time
(Note 10)
333
ps (min)
tCH
Input Clock High Time
(Note 10)
333
ps (min)
tSR
Setup Time DCLK_RST±
90
ps
tHR
Hold Time DCLK_RST±
30
ps
tSD
Synchronizing Edge to DCLK
Output Delay
tLHT
Differential Low-to-High
Transition Time
10% to 90%, CL = 2.5 pF
150
ps
tHLT
Differential High-to-Low
Transition Time
10% to 90%, CL = 2.5 pF
150
ps
tOSK
DCLK-to-Data Output Skew
50% of DCLK transition to 50% of Data
transition, SDR Mode
and DDR Mode, 0° DCLK
±50
ps (max)
tSU
Data-to-DCLK Set-Up Time
DDR Mode, 90° DCLK
400
ps
tH
DCLK-to-Data Hold Time
DDR Mode, 90° DCLK
560
ps
tAD
Sampling (Aperture) Delay
Input CLK+ Fall to Acquisition of Data
1.6
ns
tAJ
Aperture Jitter
0.4
ps rms
tOD
Input Clock-to Data Output Delay 50% of Input Clock transition to 50% of Data
(in addition to Pipeline Delay)
transition
4
ns
DI Outputs
13
tOD + tOSK
DId Outputs
Pipeline Delay (Latency)
1:2 Demux Mode
14
Non-DES Mode
DQ Outputs
DES Mode
Non-DES Mode
DQd Outputs
Pipeline Delay (Latency)
1:1 Demux Mode
13
13.5
DES Mode
14.5
13
DId Outputs
13
Non-DES Mode
DQ Outputs
DES Mode
Non-DES Mode
DES Mode
Differential VIN step from ±1.2V to 0V to get
accurate conversion
Non-DES Mode
CLK± Cycles
14
DI Outputs
DQd Outputs
Over Range Recovery Time
(Notes
10, 12)
(Notes
10, 12)
13
13.5
CLK± Cycles
13
13.5
1
CLK± Cycle
500
ns
tWU
PD low to Rated Accuracy
Conversion (Wake-Up Time)
1
µs
fSCLK
Serial Clock Frequency
15
MHz
tSSU
Data to Serial Clock Setup Time
2.5
ns (min)
tSH
Data to Serial Clock Hold Time
1
ns (min)
Serial Clock Low Time
26
ns
DES Mode
15
www.national.com
ADC08D1520QML
AC Timing Parameters
ADC08D1520QML
Symbol
tCAL
Parameter
Conditions
Notes
Typical
(Note 7)
Units
Serial Clock High Time
26
ns
Calibration Cycle Time
1.4 x 106
CLK± Cycles
Note 1: Absolute Maximum Ratings indicate limits beyond which damage to the device may occur. There is no guarantee of operation at the Absolute Maximum
Ratings. Operating Ratings indicate conditions for which the device is functional, but do not guarantee specific performance limits. For guaranteed specifications
and test conditions, see the Electrical Characteristics. The guaranteed specifications apply only for the test conditions listed. Some performance characteristics
may degrade when the device is not operated under the listed test conditions.
Note 2: All voltages are measured with respect to GND = DR GND = 0V, unless otherwise specified.
Note 3: When the input voltage at any pin exceeds the power supply limits (that is, less than GND or greater than VA), the current at that pin should be limited to
25 mA. The 50 mA maximum package input current rating limits the number of pins that can safely exceed the power supplies with an input current of 25 mA to
two. This limit is not placed upon the power, ground and digital output pins.
Note 4: Human body model is 100 pF capacitor discharged through a 1.5 kΩ resistor.
Note 5: The analog inputs are protected as shown below. Input voltage magnitudes beyond the Absolute Maximum Ratings may damage this device.
30024704
Note 6: To guarantee accuracy, it is required that VA and VDR be well bypassed. Each supply pin must be decoupled with separate bypass capacitors. Additionally,
achieving rated performance requires that the backside exposed pad be well grounded.
Note 7: Typical figures are at TA = 25°C, and represent most likely parametric norms. Test limits are guaranteed to MIL-PRF-38535.
Note 8: Calculation of Full-Scale Error for this device assumes that the actual reference voltage is exactly its nominal value. Full-Scale Error for this device,
therefore, is a combination of Full-Scale Error and Reference Voltage Error. See Figure 2. For relationship between Gain Error and Full-Scale Error, see
Specification Definitions for Gain Error.
Note 9: The analog and clock input capacitances are die capacitances only. Additional package capacitances of 0.65 pF differential and 0.95 pF each pin to
ground are isolated from the die capacitances by lead and bond wire inductances.
Note 10: This parameter is guaranteed by design and/or characterization and is not tested in production.
Note 11: The digital control pin capacitances are die capacitances only. Additional package capacitance of 1.6 pF each pin to ground are isolated from the die
capacitances by lead and bond wire inductances.
Note 12: Each of the two converters of the ADC08D1520QMLQML has two LVDS output buses, which each clock data out at one half the sample rate. The data
at each bus is clocked out at one half the sample rate. The second bus (D0 through D7) has a pipeline latency that is one Input Clock cycle less than the latency
of the first bus (Dd0 through Dd7). 1:2 Demux Mode.
Note 13: Tying VBG to the supply rail will increase the output offset voltage (VOS) by 300mv (typical), as shown in the VOS specification above. Tying VBG to the
supply rail will also affect the differential LVDS output voltage (VOD), causing it to increase by 30mV (typical).
Note 14: Pre and post irradiation limits are identical to those listed under AC and DC electrical characteristics except as listed in the Post Radiation Limits Table.
These parts may be dose rate sensitive in a space environment and demonstrate enhanced low dose rate effect. Radiation end point limits for the noted parameters
are guaranteed only for the conditions as specified in MIL-STD-883, Method 1019
DIFFERENTIAL NON-LINEARITY (DNL) is the measure of
the maximum deviation from the ideal step size of 1 LSB.
Measured at sample rate = 500 MSPS with a 1MHz input
sinewave.
EFFECTIVE NUMBER OF BITS (ENOB, or EFFECTIVE
BITS) is another method of specifying Signal-to-Noise and
Distortion Ratio, or SINAD. ENOB is defined as (SINAD −
1.76) / 6.02 and says that the converter is equivalent to a perfect ADC of this (ENOB) number of bits.
FULL POWER BANDWIDTH (FPBW) is a measure of the
frequency at which the reconstructed output fundamental
drops 3 dB below its low frequency value for a full-scale input.
GAIN ERROR is the deviation from the ideal slope of the
transfer function. It can be calculated from Offset and FullScale Errors:
Specification Definitions
APERTURE (SAMPLING) DELAYis the amount of delay,
measured from the sampling edge of the Clock input, after
which the signal present at the input pin is sampled inside the
device.
APERTURE JITTER (tAJ) is the variation in aperture delay
from sample to sample. Aperture jitter shows up as input
noise.
CODE ERROR RATE (C.E.R.) is the probability of error and
is defined as the probable number of word errors on the ADC
output per unit of time divided by the number of words seen
in that amount of time.. A C.E.R. of 10-18 corresponds to a
statistical error in one word about every four (4) years.
CLOCK DUTY CYCLE is the ratio of the time that the clock
waveform is at a logic high to the total time of one clock period.
www.national.com
16
VFS / 2N
where VFS is the differential full-scale amplitude VIN as set by
the FSR input and "N" is the ADC resolution in bits, which is
8, for the ADC08D1520QML.
LOW VOLTAGE DIFFERENTIAL SIGNALING (LVDS)
DIFFERENTIAL OUTPUT VOLTAGE (VOD) is the absolute
value of the difference between the VD+ and VD- outputs; each
measured with respect to Ground.
30024746
FIGURE 1.
LVDS OUTPUT OFFSET VOLTAGE (VOS) is the midpoint
between the D+ and D- pins output voltage with respect to
ground; ie., [(VD+) +( VD-)]/2.
MISSING CODES are those output codes that are skipped
and will never appear at the ADC outputs. These codes cannot be reached with any input value.
MSB (MOST SIGNIFICANT BIT) is the bit that has the largest
value or weight. Its value is one half of full scale.
NEGATIVE FULL-SCALE ERROR (NFSE) is a measure of
how far the first code transition is from the ideal 1/2 LSB above
a differential −VIN/2 with the FSR pin low. For the ADC08D1520QML the reference voltage is assumed to be ideal,
so this error is a combination of full-scale error and reference
voltage error.
OFFSET ERROR (VOFF) is a measure of how far the midscale point is from the ideal zero voltage differential input.
Offset Error = Actual Input causing average of 8k samples to
result in an average code of 127.5.
where Af1 is the RMS power of the fundamental (output) frequency and Af2 through Af10 are the RMS power of the first 9
harmonic frequencies in the output spectrum.
– Second Harmonic Distortion (2nd Harm) is the difference, expressed in dB, between the RMS power in the input
frequency seen at the output and the power in its 2nd harmonic level at the output.
– Third Harmonic Distortion (3rd Harm) is the difference
expressed in dB between the RMS power in the input frequency seen at the output and the power in its 3rd harmonic
level at the output.
17
www.national.com
ADC08D1520QML
OUTPUT DELAY (tOD) is the time delay (in addition to
Pipeline Delay) after the falling edge of CLK+ before the data
update is present at the output pins.
OVER-RANGE RECOVERY TIME is the time required after
the differential input voltages goes from ±1.2V to 0V for the
converter to recover and make a conversion with its rated accuracy.
PIPELINE DELAY (LATENCY) is the number of input clock
cycles between initiation of conversion and when that data is
presented to the output driver stage. New data is available at
every clock cycle, but the data lags the conversion by the
Pipeline Delay plus the tOD.
POSITIVE FULL-SCALE ERROR (PFSE) is a measure of
how far the last code transition is from the ideal 1-1/2 LSB
below a differential +VIN/2. For the ADC08D1520QML the
reference voltage is assumed to be ideal, so this error is a
combination of full-scale error and reference voltage error.
POWER SUPPLY REJECTION RATIO (PSRR) can be one
of two specifications. PSRR1 (DC PSRR) is the ratio of the
change in full-scale error that results from a power supply
voltage change from 1.8V to 2.0V. PSRR2 (AC PSRR) is a
measure of how well an a.c. signal riding upon the power
supply is rejected from the output and is measured with a 248
MHz, 50 mVP-P signal riding upon the power supply. It is the
ratio of the output amplitude of that signal at the output to its
amplitude on the power supply pin. PSRR is expressed in dB.
SIGNAL TO NOISE RATIO (SNR) is the ratio, expressed in
dB, of the rms value of the input signal at the output to the rms
value of the sum of all other spectral components below onehalf the sampling frequency, not including harmonics or d.c.
SIGNAL TO NOISE PLUS DISTORTION (S/(N+D) or
SINAD) is the ratio, expressed in dB, of the rms value of the
input signal at the output to the rms value of all of the other
spectral components below half the input clock frequency, including harmonics but excluding d.c.
SPURIOUS-FREE DYNAMIC RANGE (SFDR) is the difference, expressed in dB, between the rms values of the input
signal at the output and the peak spurious signal, where a
spurious signal is any signal present in the output spectrum
that is not present at the input, excluding d.c.
TOTAL HARMONIC DISTORTION (THD) is the ratio expressed in dB, of the rms total of the first nine harmonic levels
at the output to the level of the fundamental at the output. THD
is calculated as
Positive Gain Error = Offset Error − Positive Full-Scale
Error
Negative Gain Error = −(Offset Error − Negative FullScale Error)
Gain Error = Negative Full-Scale Error − Positive FullScale Error = Positive Gain Error + Negative Gain Error
INTEGRAL NON-LINEARITY (INL) is a measure of worst
case deviation of the ADC transfer function from an ideal
straight line drawn through the ADC transfer function. The
deviation of any given code from this straight line is measured
from the center of that code value. The best fit method is used.
INTERMODULATION DISTORTION (IMD) is the creation of
additional spectral components as a result of two sinusoidal
frequencies being applied to the ADC input at the same time.
It is defined as the ratio of the power in the second and third
order intermodulation products to the power in one of the
original frequencies. IMD is usually expressed in dBFS.
LSB (LEAST SIGNIFICANT BIT) is the bit that has the smallest value or weight of all bits. This value is
ADC08D1520QML
Transfer Characteristic
30024722
FIGURE 2. Input / Output Transfer Characteristic
www.national.com
18
ADC08D1520QML
Timing Diagrams
30024714
FIGURE 3. SDR Clocking in 1:2 Demultiplexed Mode
30024759
FIGURE 4. DDR Clocking in 1:2 Demultiplexed and Non-DES Mode
19
www.national.com
ADC08D1520QML
30024760
FIGURE 5. DDR Clocking in Non-Demultiplexed and Non-DES Mode
www.national.com
20
ADC08D1520QML
30024719
FIGURE 6. Serial Interface Timing
30024720
FIGURE 7. Clock Reset Timing in DDR Mode
30024723
FIGURE 8. Clock Reset Timing in SDR Mode with OUTEDGE Low
21
www.national.com
ADC08D1520QML
30024724
FIGURE 9. Clock Reset Timing in SDR Mode with OUTEDGE High
30024725
FIGURE 10. On-Command Calibration Timing
www.national.com
22
VA = VDR = 1.9V, fCLK = 1500 MHz, TA= 25°C, 1:2 Demux mode,
INL vs CODE
INL vs TEMPERATURE
30024764
30024765
DNL vs. CODE
DNL vs. TEMPERATURE
30024766
30024767
POWER DISSIPATION vs. SAMPLE RATE
ENOB vs. TEMPERATURE
30024781
30024776
23
www.national.com
ADC08D1520QML
Typical Performance Characteristics
Non-DES Mode unless otherwise stated.
ADC08D1520QML
ENOB vs. SUPPLY VOLTAGE
ENOB vs. SAMPLE RATE
30024777
30024778
ENOB vs. INPUT FREQUENCY
SNR vs. TEMPERATURE
30024779
30024768
SNR vs. SUPPLY VOLTAGE
SNR vs. SAMPLE RATE
30024769
www.national.com
30024770
24
ADC08D1520QML
SNR vs. INPUT FREQUENCY
THD vs. TEMPERATURE
30024771
30024772
THD vs. SUPPLY VOLTAGE
THD vs. SAMPLE RATE
30024773
30024774
THD vs. INPUT FREQUENCY
SFDR vs. TEMPERATURE
30024775
30024785
25
www.national.com
ADC08D1520QML
SFDR vs. SUPPLY VOLTAGE
SFDR vs. SAMPLE RATE
30024784
30024782
SFDR vs. INPUT FREQUENCY
Spectral Response at FIN = 373 MHz
30024783
30024787
Spectral Response at FIN = 745 MHz
CROSSTALK vs SOURCE FREQUENCY
30024788
www.national.com
30024763
26
ADC08D1520QML
FULL POWER BANDWIDTH
30024786
27
www.national.com
ADC08D1520QML
register in the Extended Control Mode. Disabling the input
termination resistor is not recommended for the initial calibration after power-up. The second portion of the calibration
cycle is the ADC calibration in which internal bias currents are
set. The ADC calibration is performed regardless of the RTD
bit setting. Running the calibration is an important part of this
chip’s functionality and is required in order to obtain specified
performance. In addition to the requirement that a calibration
be run at power-up, a calibration must be run whenever the
FSR pin is changed. For best performance, we recommend
that a calibration be run after application of power once the
power supplies have settled and the part temperature has
stabilized. Further calibrations should be run whenever the
operating temperature changes significantly relative to the
specific system performance requirements. See 2.5.2.1 Initiating Calibration for more information. Calibration can not be
initiated or run while the device is in the Power-Down Mode.
See1.1.7 Power Down for information on the interaction between Power down and calibration.
In normal operation, calibration should be performed just after
application of power and whenever a valid calibration command is given. A calibration command can be issued using
two methods. The first method is to hold the CAL pin low for
at least tCAL_L input clock cycles, then hold it high for at least
another tCAL_H input clock cycles as defined in the Converter
Electrical Characteristics. The second method is to program
the CAL bit in the Calibration register. The functionality of the
CAL bit is exactly the same as using the CAL pin. The CAL
bit must be programmed to 0b for tCAL_L input clock cycles and
then programmed to 1b for at least tCAL_H input clock cycles
to initiate a calibration cycle. The time taken by the calibration
procedure is specified as tCAL in the Converter Electrical
Characteristics.
The CAL bit does not reset itself to zero automatically, but
must be manually reset before another calibration event is
desired, the CAL bit may be left high indefinitely, with no negative consequences.
The RTD bit setting is critical for running a calibration event
with the Clock Phase Adjust enabled. If initiating a calibration
event while the Clock Phase Adjust is enabled, the RTD bit
must be set to high, or no calibration will occur. If initiating a
calibration event while the Clock Phase Adjust is not enabled,
a normal calibration will occur, regardless of the setting of the
RTD bit.
1.0 Functional Description
The ADC08D1520QML is a versatile A/D Converter with an
innovative architecture permitting very high speed operation.
The controls available ease the application of the device to
circuit solutions. Optimum performance requires adherence
to the provisions discussed here and in the Applications Information Section.
While it is not recommended in radiation environments to allow an active pin to float, pins 4, 14, 52 and 127 of the
ADC08D1520QML are designed to be left floating without
jeopardy in non radiation environments. In all discussions
throughout this data sheet, whenever a function is called by
allowing a control pin to float, connecting that pin to a potential
of one half the VA supply voltage is recommended for radiation environments.
1.1 OVERVIEW
The ADC08D1520QML uses a calibrated folding and interpolating architecture that achieves over 7.25 effective bits.
The use of folding amplifiers greatly reduces the number of
comparators and power consumption. Interpolation reduces
the number of front-end amplifiers required, minimizing the
load on the input signal and further reducing power requirements. In addition to other things, on-chip calibration reduces
the INL bow often seen with folding architectures. The result
is an extremely fast, high performance, low power converter.
The analog input signal that is within the converter's input
voltage range is digitized to eight bits at speeds of 200 MSPS
to 1.7 GSPS, typical. Differential input voltages below negative full-scale will cause the output word to consist of all
zeroes. Differential input voltages above positive full-scale
will cause the output word to consist of all ones. Either of
these conditions at either the I- or Q- Channel input will cause
the OR (Out of Range) output to be activated. This single OR
output indicates when the output code from one or both of the
channels is below negative full scale or above positive full
scale.
Each converter has a selectable output demultiplexer which
feeds two LVDS buses. If the 1:2 Demultiplexed Mode is selected, the output data rate is reduced to half the input sample
rate on each bus. When Non-Demultiplexed Mode is selected, that output data rate on channels DI and DQ are at the
same rate as the input sample clock.
The output levels may be selected to be normal or reduced.
Using reduced levels saves power but could result in erroneous data capture of some or all of the bits, especially at
higher sample rates and in marginally designed systems.
Calibration Operation Notes:
• During the calibration cycle, the OR output may be active
as a result of the calibration algorithm. All data on the
output pins and the OR output are invalid during the
calibration cycle.
• During the calibration, all clocks are halted on chip,
including internal clocks and DCLK, while the input
termination resistor is trimmed to a value that is equal to
REXT / 33. This is to reduce noise during the input resistor
calibration portion of the calibration cycle. See for
information on maintaining DCLK operation during oncommand calibration.
This external resistor is located between pin 32 and
ground. REXT must be 3300 Ω ±0.1%. With this value, the
input termination resistor is trimmed to be 100 Ω. Because
REXT is also used to set the proper current for the Track
and Hold amplifier, for the preamplifiers and for the
comparators, other values of REXT should not be used.
• The CalRun output is high whenever the calibration
procedure is running. This is true whether the calibration
is done at power-up or on-command.
1.1.1 Calibration
The ADC08D1520QML has a calibration feature which must
be invoked by the user. If the device is powered-up in the
Extended Control Mode, the registers will be in an unknown
state and no calibration is performed. For the initial calibration
after power-up, we recommend that the registers first be programmed to a known state before performing a calibration or
the part be calibrated in the pin control mode. All subsequent
calibrations can be run in either the Non-Extended Control
Mode or the Extended Control Mode.
The calibration algorithm consists of two portions. The first
portion is calibrating the analog input. This calibration trims
the 100 Ω analog input differential termination resistor and
minimizes full-scale error, offset error, DNL and INL, resulting
in maximizing SNR, THD, SINAD (SNDR) and ENOB. This
portion of the calibration can be disabled by programming the
Resistor Trim Disable (RTD) bit in the Extended Configuration
www.national.com
28
•
It is important that no digital activity take place on any of
the digital input lines during the calibration process, except
that there must be a stable, constant frequency CLK signal
present and that SCLK may be active if the Enhanced
Mode is selected. Actions that are not allowed include but
are not limited to:
Changing OUTV
Changing OutEdge or SDATA sense
Changing between SDR and DDR
Changing FSE or ECE
Changing DCLK_RST
Changing SCS
Raising PD high
Raising CAL high
Doing any of these actions can cause faulty calibration.
1.1.5 Clocking
The ADC08D1520QML must be driven with an a.c. coupled,
differential clock signal. 2.4 THE CLOCK INPUTS describes
the use of the clock input pins. A differential LVDS output
clock is available for use in latching the ADC output data into
whatever device is used to receive the data.
The ADC08D1520QML offers input and output clocking options. These options include a choice of Dual Edge Sampling
(DES) or "interleaved mode" where the ADC08D1520QML
performs as a single device converting at twice the input clock
rate, a choice of which DCLK edge the output data transitions
on, and a choice of Single Data Rate (SDR) or Double Data
Rate (DDR) outputs.
The ADC08D1520QML also has the option to use a duty cycle
corrected clock receiver as part of the input clock circuit. This
feature is enabled by default and provides improved ADC
clocking especially in the Dual-Edge Sampling Mode
(DES). This circuitry allows the ADC to be clocked with a
signal source having a duty cycle ratio of 20%/80% (worst
case) for both the Non-DES and the Dual Edge Sampling
Modes.
1.1.2 Acquiring the Input
Data is acquired at the falling edge of CLK+ (pin 18) and the
digital equivalent of that data is available at the digital outputs
13 input clock cycles later for the DI and DQ output buses and
14 input clock cycles later for the DId and DQd output buses.
There is an additional internal delay called tOD before the data
is available at the outputs. See the Timing Diagram. The ADC08D1520QML will convert as long as the input clock signal
is present. The fully differential comparator design and the
innovative design of the sample-and-hold amplifier, together
with self calibration, enables a very flat SINAD/ENOB response beyond 1.5 GHz. The ADC08D1520QML output data
signaling is LVDS and the output format is offset binary.
1.1.5.1 Dual-Edge Sampling
The DES Mode allows one of the ADC08D1520QML's inputs
(I- or Q- Channel) to be sampled by both ADCs. One ADC
samples the input on the positive edge of the input clock and
the other ADC samples the same input on the other edge of
the input clock. A single input is thus sampled twice per input
clock cycle, resulting in an overall sample rate of twice the
input clock frequency, or 3 GSPS with a 1.5 GHz input clock.
In this mode, the outputs must be carefully interleaved to reconstruct the sampled signal. If the device is programmed into
the 1:2 Demultiplex Mode while in DES Mode, the data is effectively Demultiplexed 1:4. If the input clock is 1.5 GHz, the
effective sampling rate is doubled to 3 GSPS and each of the
4 output buses have a 750 MHz output rate. All data is available in parallel. To properly reconstruct the sampled waveform, the four bytes of parallel data that are output with each
clock are in the following sampling order from the earliest to
the latest and must be interleaved as such: DQd, DId, DQ, DI.
Table 1 indicates what the outputs represent for the various
sampling possibilities. If the device is programmed into the
Non-Demultiplex Mode, two bytes of parallel data are output
with each edge of the clock in the following sampling order,
from the earliest to the latest: DQ, DI. See Table 2.
In the Non-Extended Control Mode of operation only the Ichannel input can be sampled in the DES Mode. In the
Extended Control Mode of operation, the user can select
which input is sampled.
The ADC08D1520QML also includes an automatic clock
phase background calibration feature which can be used in
DES Mode to automatically and continuously adjust the clock
phase of the I- and Q- channel. This feature removes the need
to adjust the clock phase setting manually and provides optimal Dual-Edge Sampling ENOB performance.
IMPORTANT NOTE: The background calibration feature in
DES Mode does not replace the requirement for calibration if
a large swing in ambient temperature is experienced by the
device.
1.1.3 Control Modes
Much of the user control can be accomplished with several
control pins that are provided. Examples include initiation of
the calibration cycle, Power Down Mode and full scale range
setting. However, the ADC08D1520QML also provides an
Extended Control mode whereby a serial interface is used to
access register-based control of several advanced features.
The Extended Control mode is not intended to be enabled and
disabled dynamically. Rather, the user is expected to employ
either the Non-Extended Control Mode or the Extended Control Mode at all times. When the device is in the Extended
Control Mode, pin-based control of several features is replaced with register-based control and those pin-based controls are disabled. These pins are OutV (pin 3), OutEdge/DDR
(pin 4), FSR (pin 14) and DES (pin 127). See 1.2 NON-EXTENDED CONTROL/EXTENDED CONTROL for details on
the Extended Control Mode.
1.1.4 The Analog Inputs
The ADC08D1520QML must be driven with a differential input
signal. Operation with a single-ended signal is not recommended. It is important that the input signals are a.c. coupled
to the inputs.
Two full-scale range settings are provided with pin 14 (FSR).
A high on pin 14 causes an input full-scale range setting of a
higher VIN input level, while grounding pin 14 causes an input
full-scale range setting of a reduced VIN input level. The fullscale range setting operates equally on both ADCs.
In the Extended Control Mode, the Input Full-Scale Voltage
Adjust register allows the input full-scale range to be adjusted
as described in 1.4 REGISTER DESCRIPTION and 2.3 THE
ANALOG INPUT.
29
www.national.com
ADC08D1520QML
•
ADC08D1520QML
TABLE 1. Input Channel Samples Produced at Data Outputs in 1:2 Demultiplexed Mode**
Data Outputs
(Always sourced with
respect to fall of DCLK+)
Dual-Edge Sampling Mode (DES)
Non DES Sampling Mode
I- Channel Selected
Q- Channel Selected *
DI
I- Channel Input Sampled
I- Channel Input Sampled with
with Fall of CLK 13 cycles
Fall of CLK 13 cycles earlier.
earlier.
Q- Channel Input Sampled with
Fall of CLK 13 cycles earlier.
DId
I- Channel Input Sampled
I- Channel Input Sampled with
with Fall of CLK 14 cycles
Fall of CLK 14 cycles earlier.
earlier.
Q- Channel Input Sampled with
Fall of CLK 14 cycles earlier.
DQ
I- Channel Input Sampled
Q- Channel Input Sampled with
Q- Channel Input Sampled with
with Rise of CLK 13.5 cycles
Fall of CLK 13 cycles earlier.
Rise of CLK 13.5 cycles earlier.
earlier.
DQd
I- Channel Input Sampled
Q- Channel Input Sampled with
Q- Channel Input Sampled with
with Rise of CLK 14.5 cycles
Fall of CLK 14 cycles earlier.
Rise of CLK 14.5 cycles earlier.
earlier.
* Note that, in DES + Non-DES Mode, only the I- Channel is sampled. In DES + Extended Control Mode, I- Channel or QChannel can be sampled.
** Note that, in the Non-Demultiplexed Mode, the DId and DQd outputs are disabled and are high impedance.
TABLE 2. Input Channel Samples Produced at Data Outputs in 1:1 Demultiplexed Mode
Data Outputs
(Sourced with respect to fall of DCLK+)
Non-DES Mode
DI
I- Channel Input Sampled with Fall of CLK I- Channel Input Sampled with Fall of CLK
13 cycles earlier.
13 cycles earlier.
Dld
No output.
No output.
DQ
Q- Channel Input Sampled with Fall of
CLK 13 cycles earlier.
Q- Channel Input Sampled with Fall of
CLK 13.5 cycles earlier.
DQd
No output.
No output.
the data rate and data is sent to the outputs on both edges of
DCLK. DDR clocking is enabled in Non-Extended Control
Mode by tying pin 4 to VA/2.
1.1.5.2 OutEdge and Demultiplex Control Setting
To help ease data capture in the SDR Mode, the output data
may be caused to transition on either the positive or the negative edge of the output data clock (DCLK). In the NonExtended Control Mode, this is chosen with the OutEdge input
(pin 4). A high on the OutEdge input pin causes the output
data to transition on the rising edge of DCLK+, while grounding this input causes the output to transition on the falling edge
of DCLK. See 2.5.3 Output Edge Synchronization. When in
the Extended Control Mode, the OutEdge is selected using
the OED bit in the Configuration Register. This bit has two
functions. In the single data rate (SDR) Mode, the bit functions
as OutEdge and selects the DCLK edge with which the data
transitions. In the Double Data Rate (DDR) Mode, this bit selects whether the device is in Non-Demultiplex or 1:2 Demultiplex Mode. In the DDR case, the DCLK has a 0° phase
relationship with the output data independent of the demultiplexer selection.
For 1:2 Demux DDR 0 deg Mode, there are five, as opposed
to four cycles of CLK delay from the deassertion of
DCLK_RST to the Synchronizing Edge. See 1.5 MULTIPLE
ADC SYNCHRONIZATION
1.1.5.4 Clocking Summary
The chip may be in one of four modes, depending on the DualEdge Sampling (DES) selection and the demultiplex selection. For the DES selection, there are two possibilities: NonDES Mode and DES Mode. In Non-DES Mode, each of the
channels (I-channel and Q-channel) functions independently,
i.e. the chip is a dual 1.5 GSPS A/D converter. In DES Mode,
the I- and Q-channels are interleaved and function together
as one 3.0 GSPS A/D converter. For the demultiplex selection, there are also two possibilities: Demux Mode and NonDemux Mode. The I-channel has two 8-bit output busses
associated with it: DI and DId. The Q-channel also has two 8bit output busses associated with it: DQ and DQd. In Demux
Mode, the channel is demultiplexed by 1:2. In Non-Demux
Mode, the channel is not demultiplexed. Note that Non-Demux Mode is also sometimes referred to as 1:1 Demux Mode.
For example, if the I-channel was in Non-Demux Mode, the
corresponding digital output data would be available on only
the DI bus. If the I-channel was in Demux Mode, the corresponding digital output data would be available on both the
DI and DId busses, but at half the rate of Non-Demux Mode.
Given that there are two DES Mode selections (DES Mode
and Non-DES Mode) and two demultiplex selections (Demux
Mode and Non-Demux Mode), this yields a total of four pos-
1.1.5.3 Double Data Rate
A choice of single data rate (SDR) or double data rate (DDR)
output is offered. With single data rate the output clock
(DCLK) frequency is the same as the data rate of the two output buses. With double data rate the DCLK frequency is half
www.national.com
DES Mode
30
power consumption. If the LVDS lines are long and/or the
system in which the ADC08D1520QML is used is noisy, it may
be necessary to tie the OutV pin high.
The LVDS data output have a typical common mode voltage
of 800 mV when the VBG pin is left floating. This common
mode voltage can be increased to 1.1V by tying the VBG pin
to VA if a higher common mode is required.
IMPORTANT NOTE: Tying the VBG pin to VA will also increase the differential LVDS output voltage by up to 40mV.
1.1.7 Power Down
The ADC08D1520QML is in the active state when the Power
Down pin (PD) is low. When the PD pin is high, the device is
in the Power Down Mode. In this Power Down Mode the data
output pins (positive and negative) are put into a high
impedance state and the devices power consumption is reduced to a minimal level. The DCLK+/- and OR +/- are not tristated, they are weakly pulled down to ground internally.
Therefore when both I- Channel and Q- Channel are powered
down the DCLK +/- and OR +/- should not be terminated to a
DC voltage.
A high on the PDQ pin will power down the Q- Channel and
leave the I- channel active. There is no provision to power
down the I- Channel independently of the Q- Channel. Upon
return to normal operation, the pipeline will contain meaningless information.
If the PD input is brought high while a calibration is running,
the device will not go into power down until the calibration
sequence is complete. However, if power is applied and PD
is simultaneously ramped, the device will not calibrate until
the PD input goes low. If a calibration is requested while the
device is powered down, the calibration request will be completely ignored. Calibration will function with the Q- Channel
powered down, but that channel will not be calibrated if PDQ
is high. If the Q- Channel is subsequently to be used, it is
necessary to perform a calibration after PDQ is brought low.
1.2 NON-EXTENDED CONTROL/EXTENDED CONTROL
The ADC08D1520QML may be operated in one of two
modes. In the simpler Non-Extended Control Mode, the user
affects available configuration and control of the device
through several control pins. The "Extended Control Mode"
provides additional configuration and control options through
a serial interface and a set of 9 registers. Extended Control
Mode is selected by setting pin 41 to logic low. The choice of
control modes is required to be a fixed selection and is not
intended to be switched dynamically while the device is operational.
Table 3 shows how several of the device features are affected
by the control mode chosen.
1.1.6 The LVDS Outputs
The data outputs, the Out Of Range (OR) and DCLK, are
LVDS. Output current sources provide 3 mA of output current
to a differential 100 Ohm load when the OutV input (pin 14) is
high or 2.2 mA when the OutV input is low. For short LVDS
lines and low noise systems, satisfactory performance may
be realized with the OutV input low, which results in lower
31
www.national.com
ADC08D1520QML
sible modes: (1) Non-Demux Mode, (2) Non-Demux DES
Mode, (3) 1:2 Demux Non-DES Mode, and (4) 1:4 Demux
DES Mode. The following is a brief explanation of the terms
and modes:
1. Non-Demux Mode: This mode is when the chip is in NonDemux Mode and Non-DES Mode, but it is shortened to
simply "Non-Demux Mode." The I- and Q- channels
function independently of one another. The digital output
data is available for the I- channel on DI, and for the Qchannel on DQ.
2. Non-Demux DES Mode: This mode is when the chip is
in Non-Demux Mode and DES Mode. The I- and Qchannels are interleaved and function together as one
channel. The digital output data is available on the DI and
DQ busses because although the chip is in Non-Demux
Mode, both I- and Q- channels are functioning and
passing data.
3. 1:2 Demux Non-DES Mode: This mode is when the chip
is in Demux Mode and Non-DES Mode. The I- and Qchannels function independently of one another. The
digital output data is available for the I- channel on DI and
DId, and for the Q- channel on DQ and DQd. This is
because each channel (I- channel and Q- channel) is
providing digital data in a demultiplexed manner.
4. 1:4 Demux DES Mode: This mode is when the chip is in
Demux Mode and DES Mode. The I- and Q- channels
are interleaved and function together as one channel.
The digital output data is available on the DI, DId, DQ and
DQd busses because although the chip is in Demux
Mode, both I- and Q- channels are functioning and
passing data. To avoid confusion, this mode is labeled
1:4 because the analog input signal is provided on one
channel and the digital output data is provided on four
busses.
The choice of Dual Data Rate (DDR) and Single Data Rate
(SDR) will only affect the speed of the output Data Clock
(DCLK). Once the DES Modes and Demux Modes have been
chosen, the data output rate is also fixed. In the case of SDR,
the DCLK runs at the same rate as the output data; output
data may transition with either the rising or falling edge of
DCLK. In the case of DDR, the DCLK runs at half the rate of
the output data; the output data transitions on both rising and
falling edges of the DCLK.
ADC08D1520QML
TABLE 3. Features and Modes
Feature
Non-Extended Control Mode
Extended Control Mode
SDR or DDR Clocking Selected with pin 4
Selected with bit 10 nDE in the Configuration Register
(Addr-1h; bit-10)
DDR Clock Phase
Not Selectable (0° Phase Only)
Selected with DCP in the Configuration
Register (Addr-1h; bit-11)
SDR Data transitions
with rising or falling
DCLK edge
SDR Data transitions with rising
edge of DCLK+ when pin 4 is high Selected with OED in the Configuration Register (Addr-1h; bit-8)
and on falling edge when low.
LVDS output level
Normal differential data and DCLK
amplitude selected when pin 3 is
Selected with OV in the Configuration Register (Addr-1h; bit-9)
high and reduced amplitude
selected when low.
Full-Scale Range
Normal input full-scale range
selected when pin 14 is high and
reduced range when low.
Selected range applies to both
channels.
Up to 512 step adjustments over a nominal range specified in 1.4
REGISTER DESCRIPTION. Separate range selected for IChannel and Q- Channels. Selected using Full Range
Registers (Addr-3h and Bh; bit-7 thru 15)
Input Offset Adjust
Not possible
512 steps of adjustment using the input Offset register specified
in 1.4 REGISTER DESCRIPTION for each channel using Input
Offset registers (Addr-2h and Ah; bit-7 thru 15)
Dual Edge Sampling
Selection
Enabled with pin 127 set to VA/2
Enabled by programming DEN in the Extended Configuration
Register (Addr-9h; bit-13 )
Dual Edge Sampling
Input Channel
Selection
Only I-Channel Input can be used
Either I- Channel or Q- Channel input may be sampled by both
ADCs.
Test Pattern
Not possible
A test pattern can be made present at the data outputs by setting
TPO to 1b in Extented Configuration Register (Addr-9h; bit-15)
Resistor Trim Disable Not possible
The DCLK outputs will continuously be present when RTD is set
to 1b in Extented Configuration Register (Addr-9h; bit-14)
Selectable Output
Demultiplexer
Not possible
If the device is set in DDR, the output can be programmed to be
non-demultiplex. When OED in Configuration Register is set 1b
(Addr-1h; 8-bit), this selects non-demultiplex. If OED is set 0b,
this selects 1:2 demultiplex.
Second DCLK Output Not possible
The OR outputs can be programmed to become a second DCLK
output when nSD is set 0b in Configuration Register
(Addr-1h; bit-13).
Sampling Clock Phase
Not possible
Adjust
The sampling clock phase can be manually adjusted through the
Coarse and Intermediate Register (Addr-Fh; bit-14 to 7) and Fine
register (Addr-Dh; bit-15 to 8)
When the device is powered up in the Extended Control
Mode, the Registers are loaded with invalid data and the
Registers come up on an unknown state. Before initiating a
calibration the registers must be written to and programmed
www.national.com
into a known state. If the device is powered up in the NonExtended Control Mode and the user switches to the Extended Control Mode after the part has stabilized, the registers will
load with the register default states described in Table 4.
32
ADC08D1520QML
TABLE 4. Extended Control Mode Operation
(Pin 41 Logic Low)
Feature
Extended Control Mode Default State
SDR or DDR Clocking
DDR Clocking
DDR Clock Phase
Data changes with DCLK edge (0° phase)
LVDS Output Amplitude
Normal amplitude
(VOD)
Full-Scale Range
700 mV nominal for both channels
Input Offset Adjust
No adjustment for either channel
Dual Edge Sampling (DES)
Not enabled
Test Pattern
Not present at output
Resistor Trim Disable
Trim enabled, DCLK not continuously present at output
Selectable Output Demultiplexer
1:2 demultiplex
Second DCLK Output
Not present, pin 79 and 80 function as OR+ and OR-.
Sampling Clock Phase Adjust
No adjustment for fine, intermediate or coarse
1.3 THE SERIAL INTERFACE
The 3-pin serial interface is enabled only when the device is
in the Extended Control mode. The pins of this interface are
Serial Clock (SCLK), Serial Data (SDATA) and Serial Interface Chip Select (SCS). Nine write only registers are accessible through this serial interface.
SCS: This signal should be asserted low while accessing a
register through the serial interface. Setup and hold times with
respect to the SCLK must be observed.
SCLK: Serial data input is accepted at the rising edge of this
signal.
SDATA: Each register access requires a specific 32-bit pattern at this input. This pattern consists of a header, register
address and register value. The data is shifted in MSB first.
Setup and hold times with respect to the SCLK must be observed. See the Timing Diagram.
Each Register access consists of 32 bits, as shown in Figure
6 of the Timing Diagrams. The fixed header pattern is 0000
0000 0001 (eleven zeros followed by a 1). The loading sequence is such that a "0" is loaded first. These 12 bits form
the header. The next 4 bits are the address of the register that
is to be written to and the last 16 bits are the data written to
the addressed register. The addresses of the various registers are indicated in Table 5.
Refer to the Register Description (1.4 REGISTER DESCRIPTION) for information on the data to be written to the registers.
Subsequent register accesses may be performed immediately, starting with the 33rd SCLK. This means that the SCS input
does not have to be de-asserted and asserted again between
register addresses. It is possible, although not recommended,
to keep the SCS input permanently enabled (at a logic low)
when using extended control.
IMPORTANT NOTE: Do not write to the Serial Interface when
calibrating the ADC. Doing so will impair the performance of
the device until it is re-calibrated correctly. Programming the
serial registers will also reduce dynamic performance of the
ADC for the duration of the register access time.
TABLE 5. Register Addresses
4-Bit Address
Loading Sequence:
A3 loaded after H0, A0 loaded last
33
A3
A2
A1
A0
Hex
0
0
0
0
0h
Register Addressed
Calibration
0
0
0
1
1h
Configuration
0
0
1
0
2h
I- Ch Offset
0
0
1
1
3h
I- Ch Full-Scale
Voltage Adjust
0
1
0
0
4h
Reserved
0
1
0
1
5h
Reserved
0
1
1
0
6h
Reserved
0
1
1
1
7h
Reserved
1
0
0
0
8h
Reserved
1
0
0
1
9h
Extended
Configuration
1
0
1
0
Ah
Q- Ch Offset
1
0
1
1
Bh
Q- Ch Full-Scale
Voltage Adjust
1
1
0
0
Ch
Reserved
1
1
0
1
Dh
Reserved
1
1
1
0
Eh
Sampling Clock Phase
Fine Adjust
1
1
1
1
Fh
Sample Clock Phase
Intermediate and
Coarse Adjust
www.national.com
ADC08D1520QML
1.4 REGISTER DESCRIPTION
Nine write-only registers provide several control and configuration options in the Extended Control Mode. These registers have no effect when the device is in the Non-Extended
Control Mode. Each register description below also shows the
Register Default State.
Bit 10
Calibration Register
Addr: 0h (0000b)
Write only (0x7FFF)
D15
D14
D13
D12
D11
D10
D9
D8
CAL
1
1
1
1
1
1
1
D7
D6
D5
D4
D3
D2
D1
D0
1
1
1
1
1
1
1
1
Bit 15
Bit 9
CAL: Calibration Enable. When this bit is set
1b, a command calibration cycle is initiated.
This function is exactly the same as issuing
a calibration using the CAL pin. See section
2.5.2.1, Initiating Calibration for details for
usage.
Default State: 0b
Must be set to 1b
Bits 14:0
Bit 8
Configuration Register
Addr: 1h (0001b)
Write only (0xB2FF)
D15
D14
D13
D12
D11
D10
D9
D8
1
0
nSD
DCS DCP
nDE
OV
OED
D7
D6
D5
D4
D3
D2
D1
D0
1
1
1
1
1
1
1
1
Bit 15
Bit 14
Bit 13
Bit 12
Bit 11
Must be set to 1b
Must be set to 0b
nSD: Second DCLK Output. When this bit is
1b, the device only has one DCLK output and
one OR output. When this output is 0b, the
device has two identical DCLK outputs and no
OR output.
Default State: 1b
DCS: Duty Cycle Stabilizer. When this bit is set
to 1b, a duty cycle stabilization circuit is
applied to the clock input. When this bit is set
to 0b the stabilization circuit is disabled.
Default State: 1b
DCP: DDR Clock Phase. This bit only has an
effect in the DDR Mode. When this bit is set to
0b, the DCLK edges are time-aligned with the
data bus edges ("0° Phase"). When this bit is
set to 1b, the DCLK edges are placed in the
middle of the data bit-cells ("90° Phase"),
using the one-half speed DCLK shown in
Figure 4 as the phase reference.
Default State: 0b
Bits 7:0
I-Channel Offset
Addr: 2h (0010b)
D15
D14
D13
(MSB)
D12
D11
D10
D9
Offset Value
D8
(LSB)
D7
D6
D5
D4
D3
D2
D1
D0
1
1
1
1
1
1
1
Bits 15:8
Bit 6:0
34
Write only (0x007F)
Sign
Bit 7
www.national.com
nDE: DDR Enable. When this bit is set to 0b,
data bus clocking follows the DDR (Double
Data Rate) Mode whereby a data word is
output with each rising and falling edge of
DCLK. When this bit is set to a 1b, data bus
clocking follows the SDR (single data rate)
Mode whereby each data word is output with
either the rising or falling edge of DCLK , as
determined by the OutEdge bit.
Default State: 0b
OV: Output Voltage. This bit determines the
LVDS outputs' voltage amplitude and has the
same function as the OutV pin that is used in
the Non-Extended Control Mode. When this bit
is set to 1b, the standard output amplitude of
780 mVP-P is used. When this bit is set to 0b,
the reduced output amplitude of 590 mVP-P is
used.
Default State: 1b
OED: Output Edge and Demultiplex Control.
This bit has two functions. When the device is
in SDR Mode, this bit selects the DCLK edge
with which the data words transition in the
SDR Mode and has the same effect as the
OutEdge pin in the Non-Extended Control
Mode. When this bit is set to 1b, the data
outputs change with the rising edge of DCLK
+. When this bit is set to 0b, the data output
changes with the falling edge of DCLK+. When
the device is in DDR Mode, this bit selects the
Non-Demultiplexed Mode when set to 1b.
When the bit set to 0b, the device is
programmed into the 1:2 Demultiplexed Mode.
The 1:2 Demultiplexed Mode is the default
mode. In DDR Mode, DCLK has a 0° phase
relationship with the data.
Default State: 0b
Must be set to 1b
Offset Value. The input offset of the I-Channel
ADC is adjusted linearly and monotonically by
the value in this field. 00h provides a nominal
zero offset, while FFh provides a nominal 45
mV of offset. Thus, each code step provides
0.176 mV of offset.
Default State: 0000 0000 b
Sign bit. 0b gives positive offset, 1b gives
negative offset.
Default State: 0b
Must be set to 1b
D15
D14
Bit 13
D13
D12
(MSB)
D11
D10
D9
D8
Adjust Value
D7
D6
D5
D4
D3
D2
D1
D0
(LSB)
1
1
1
1
1
1
1
Bit 15:7
Bits 6:0
Default State: 0b
Full Scale Voltage Adjust Value. The input fullscale voltage or gain of the I-Channel ADC is
adjusted linearly and monotonically with a 9 bit
data value. The adjustment range is ±20% of
the nominal 700 mVP-P differential value.
Bit 12
0000 0000 0
560mVP-P
Bit 11
Must be set to 0b
1000 0000 0
Default Value
1111 1111 1
700mVP-P
Bit 10
DLF: DES Low Frequency. When this bit is
set 1b, the dynamic performance of the
device is improved when the input clock is
less than 900 MHz.
840mVP-P
For best performance, it is recommended that
the value in this field be limited to the range of
0110 0000 0b to 1110 0000 0b. i.e., limit the
amount of adjustment to ±15%. The remaining
±5% headroom allows for the ADC's own full
scale variation. A gain adjustment does not
require ADC re-calibration.
Default State: 1000 0000 0b (no adjustment)
Must be set to 1b
Addr: 9h (1001b)
Default State: 0b
Bits 9:0
Must be set to 1b
Q- Channel Offset
Addr: Ah (1010b)
D15
D14
D13
(MSB)
Write only (0x007F)
D12
D11
D10
D9
Offset Value
D8
(LSB)
D7
D6
D5
D4
D3
D2
D1
D0
Sign
1
1
1
1
1
1
1
Write only (0x03FF)
D15
D14
D13
D12
D11
D10
D9
D8
TPO
RTD
DEN
IS
0
DLF
1
1
D7
D6
D5
D4
D3
D2
D1
D0
1
1
1
1
1
1
1
1
Bit 14
IS: Input Select. When this bit is set to 0b the
I- Channel input is operated upon by both
ADCs. When this bit is set to 1b the QChannel input is operated on by both ADCs.
Default State: 0b
Extended Configuration Register
Bit 15
DES: DES Enable. Setting this bit to 1b
enables the Dual Edge Sampling Mode. In
this mode the ADCs in this device are used
to sample and convert the same analog input
in a time-interleaved manner, accomplishing
a sample rate of twice the input clock rate.
When this bit is set to 0b, the device operates
in the Non-DES Mode.
Write only (0x807F)
Bit 15:8
TPO: Test Pattern Output. When this bit is set
1b, the ADC is disengaged and a test pattern
generator is connected to the outputs
including OR. This test pattern will work with
the device in the SDR, DDR and the NonDemultiplex output modes.
Bit 7
Default State: 0b
Bit 6:0
Offset Value. The input offset of the QChannel ADC is adjusted linearly and
monotonically by the value in this field. 00h
provides a nominal zero offset, while FFh
provides a nominal 45 mV of offset. Thus,
each code step provides about 0.176 mV of
offset.
Default State: 0000 0000 b
Sign bit. 0b gives positive offset, 1b gives
negative offset.
Default State: 0b
Must be set to 1b
RTD: Resistor Trim Disable. When this bit is
set to 1b, the input termination resistor is not
trimmed during the calibration cycle and the
DCLK output remains enabled. Note that the
ADC is calibrated regardless of this setting.
Default State: 0b
35
www.national.com
ADC08D1520QML
I-Channel Full-Scale Voltage Adjust
Addr: 3h (0011b)
ADC08D1520QML
Q- Channel Full-Scale Voltage Adjust
Addr: Bh (1011b)
D15
D14
Bit 15
Write only (0x807F)
D13
D12
(MSB)
D11
D10
D9
D8
Adjust Value
D7
D6
D5
D4
D3
D2
D1
D0
(LSB)
1
1
1
1
1
1
1
Bit 15:7
Bits 14:10 Coarse Phase Adjust. Each code value in
this field delays the sample clock by
approximately 65 ps. A value of 00000b in
this field causes zero adjustment.
Default State: 00000b
Full Scale Voltage Adjust Value. The input fullscale voltage or gain of the I- Channel ADC is
adjusted linearly and monotonically with a 9 bit
data value. The adjustment range is ±20% of
the nominal 700 mVP-P differential value.
0000 0000 0
560 mVP-P
1000 0000 0
700 mVP-P
1111 1111 1
840 mVP-P
For best performance, it is recommended that
the value in this field be limited to the range of
0110 0000 0b to 1110 0000 0b. i.e., limit the
amount of adjustment to ±15%. The remaining
±5% headroom allows for the ADC's own full
scale variation. A gain adjustment does not
require ADC re-calibration.
Default State: 1000 0000 0b (no adjustment)
Must be set to 1b
Bits 6:0
D15
D14
Write only (0x00FF)
D13
(MSB)
D12
D11
D10
D9
Fine Phase Adjust
D8
(LSB)
D7
D6
D5
D4
D3
D2
D1
D0
1
1
1
1
1
1
1
1
Bits 15:8
Bits 9:7
Intermediate Phase Adjust. Each code value
in this field delays the sample clock by
approximately 11 ps. A value of 000b in this
field causes zero adjustment. Maximum
combined adjustment using Coarse Phase
Adjust and Intermediate Phase adjust is
approximately 2.1ns.
Default State: 000b
Bits 6:0
Must be set to 1b
Note Regarding Extended Mode Offset Correction
When using the I- Channel or Q- Channel Offset Adjust registers, the following information should be noted.
For offset values of +0000 0000 and -0000 0000, the actual
offset is not the same. By changing only the sign bit in this
case, an offset step in the digital output code of about 1/10th
of an LSB is experienced. This is shown more clearly in the
Figure below.
Note Regarding Clock Phase Adjust
This is a feature intended to help the system designer remove
small imbalances in clock distribution traces at the board level
when multiple ADCs are used. Please note, however, that
enabling this feature will reduce the dynamic performance
(ENOB, SNR SFDR) some finite amount. The amount of
degradation increases with the amount of adjustment applied.
The user is strongly advised to (a) use the minimal amount of
adjustment: and (b) verify the net benefit of this feature in his
system before relying on it.
Sample Clock Phase Fine Adjust
Addr: Eh (1110b)
Polarity Select. When this bit is selected, the
polarity of the ADC sampling clock is
inverted.
Default State: 0b
Fine Phase Adjust. The phase of the ADC
sampling clock is adjusted linearly and
monotonically by the value in this field. 00h
provides a nominal zero phase adjustment,
while FFh provides a nominal 50 ps of delay.
Thus, each code step provides about 0.2 ps
of delay.
Default State: 0000 0000b
Bits 7:0
Must be set to 1b
Sample Clock Phase Intermediate/Coarse Adjust
Addr: Fh (1111b)
D15
D14
POL
D13
Write only (0x007F)
D12
D11
D10
D9
(MSB) Coarse Phase Adjust
D8
IPA
30024730
D7
D6
D5
D4
D3
D2
D1
D0
(LSB)
1
1
1
1
1
1
1
www.national.com
FIGURE 11. Extended Mode Offset Behavior
36
TABLE 6. Test Pattern by Output Port
in 1:2 Demultiplex Mode
Time
Qd
Id
Q
I
OR
T0
01h
02h
03h
04h
0
T1
FEh
FDh
FCh
FBh
1
T2
01h
02h
03h
04h
0
T3
FEh
FDh
FCh
FBh
1
T4
01h
02h
03h
04h
0
T5
01h
02h
03h
04h
0
T6
FEh
FDh
FCh
FBh
1
T7
01h
02h
03h
04h
0
T8
FEh
FDh
FCh
FBh
1
0
T9
01h
02h
03h
04h
T10
01h
02h
03h
04h
0
T11
...
...
...
...
...
Comments
Pattern
Sequence
n
Pattern
Sequence
n+1
Pattern
Sequence n+2
With the part programmed into the Non-Demultiplex Mode,
the test pattern’s order will be as described in Table 7.
TABLE 7. Test Pattern by Output Port in
Non-Demultiplex Mode
Time
1.6 ADC TEST PATTERN
To aid in system debug, the ADC08D1520QML has the capability of providing a test pattern at the four output ports
completely independent of the input signal. The ADC is disengaged and a test pattern generator is connected to the
outputs including OR. The test pattern output is the same in
DES Mode and Non-DES Mode. Each port is given a unique
Q
I
OR
T0
01h
02h
0
T1
FEh
FDh
1
T2
01h
02h
0
T3
01h
02h
0
T4
FEh
FDh
1
T5
FEh
FDh
1
T6
01h
02h
0
T7
01h
02h
0
T8
FEh
FDh
1
T9
01h
02h
0
T10
01h
02h
0
T11
FEh
FDh
1
T12
01h
02h
0
0
T13
01h
02h
T14
FEh
FDh
1
T15
...
...
...
Comments
Pattern
Sequence
n
Pattern
Sequence
n+1
To ensure that the test pattern starts synchronously in each
port, set DCLK_RST while writing the Test Pattern Output bit
in the Extended Configuration Register. The pattern appears
at the data output ports when DCLK_RST is cleared low. The
test pattern will work at speed and will work with the device in
the SDR, DDR and the Non-Demultiplex output modes.
37
www.national.com
ADC08D1520QML
8-bit word, alternating between 1's and 0's as described in the
Table 6.
1.5 MULTIPLE ADC SYNCHRONIZATION
The ADC08D1520QML has the capability to precisely reset
its sampling clock input to DCLK output relationship as determined by the user-supplied DCLK_RST pulse. This allows
multiple ADCs in a system to have their DCLK (and data) outputs transition at the same time with respect to the shared
CLK input that they all the ADCs use for sampling.
The DCLK_RST signal must observe some timing requirements that are shown in Figure 7, Figure 8 and Figure 9 of the
Timing Diagrams. The DCLK_RST pulse must be of a minimum width and its deassertion edge must observe setup and
hold times with respect to the CLK input rising edge. These
timing specifications are listed as tRH, tRS, and tRPW in the
Converter Electrical Characteristics.
The DCLK_RST signal can be asserted asynchronous to the
input clock. If DCLK_RST is asserted, the DCLK output is held
in a designated state. The state in which DCLK is held during
the reset period is determined by the mode of operation (SDR/
DDR) and the setting of the Output Edge configuration pin or
bit. (Refer to Figure 7, Figure 8 and Figure 9 for the DCLK
reset state conditions). Therefore, depending upon when the
DCLK_RST signal is asserted, there may be a narrow pulse
on the DCLK line during this reset event. When the
DCLK_RST signal is de-asserted in synchronization with the
CLK rising edge, the 4th or 5th CLK falling edge synchronizes
the DCLK output with those of other ADC08D1520QMLs in
the system. The DCLK output is enabled again after a constant delay (relative to the input clock frequency) which is
equal to the CLK input to DCLK output delay (tSD). The device
always exhibits this delay characteristic in normal operation.
As shown in Figure 7, Figure 8 and Figure 9of the Timing Diagrams, there is a delay from the deassertion of DCLK_RST
to the reappearance of DCLK, which is equal to several CLK
cycles of delay plus tSD. Note that the deassertion of
DCLK_RST is not latched in until the next falling edge of CLK.
For 1:2 Demux DDR 0 deg Mode, there are five CLK cycles
of delay; for all other modes, there are four CLK cycles of
delay.
If the device is not programmed to allow DCLK to run continuously, DCLK will become inactive during a calibration cycle.
Therefore, it is strongly recommended that DCLK only be
used as a data capture clock and not as a system clock.
The DCLK_RST pin should NOT be brought high while the
calibration process is running (while CalRun is high). Doing
so could cause a digital glitch in the digital circuitry, resulting
in corruption and invalidation of the calibration. (See Application Information Section 2.4.3)
ADC08D1520QML
FSR pin grounded, the millivolt values in Table 8 are reduced
to 75% of the values indicated. In the Extended Control Mode,
these values will be determined by the full scale range and
offset settings in the Control Registers.
2.0 Applications Information
2.1 APPLICATIONS IN RADIATION ENVIRONMENTS
Applying the ADC08D1520QML in a radiation environment
should be done with careful consideration to that environment. The QMLV version of this part has been rated to tolerate
a high total dose of ionizing radiation by test method 1019 of
MIL-STD-883. The part is also immune to SEE (Single Event
Effects) hard errors such as Single Event Latch-up and Functional Interrupts. However, there are still some recommendations and cautions.
Floating pins. There are four tri-level pins which activate the
following modes when left floating: FSR/DCLK_RST-, OutEdge/DDR/SDATA, DRST_SEL and DES/SCS. If modes requiring a floating pin are needed to be used, then it is strongly
recommended that the floating method of establishing Va/2
on these pins not be employed. Due to the potential of increased leakage of the input protection diodes after large
ionizing doses, the midpoint voltage (Va/2 or 0.95V) should
be voltage forced or formed with a resistor divider from the
analog supply to ground with two 2K ohm resistors. The tolerance for this mid point voltage is 650mV ≥ VA/2 ≤ 1.2V. The
internal voltage divider resistors provide too little current to
set the midpoint voltage reliably in radiation environments.
TABLE 8. Differential Input To Output Relationship
(Non-Extended Control Mode, FSR High)
VIN−
Output Code
VCM + 225 mV
0000 0000
VCM − 113 mV
VCM + 113 mV
0100 0000
VCM
VCM
0111 1111 /
1000 0000
VCM + 109 mV
VCM −109 mV
1100 0000
VCM + 217.5 mV
VCM − 217.5 mV
1111 1111
The buffered analog inputs simplify the task of driving these
inputs and the RC pole that is generally used at sampling ADC
inputs is not required. If it is desired to use an amplifier circuit
before the ADC, use care in choosing an amplifier with adequate noise and distortion performance and adequate gain at
the frequencies used for the application.
IMPORTANT NOTE: An Analog input channel that is not used
(e.g. in DES Mode) should be left floating when the inputs are
a.c. coupled. Do not connect an unused analog input to
ground.
2.2 THE REFERENCE VOLTAGE
The voltage reference for the ADC08D1520QML is derived
from a 1.254V bandgap reference, a buffered version of which
is made available at pin 31, VBG, for user convenience. This
output has an output current capability of ±100 μA and should
be buffered if more current is required.
The internal bandgap-derived reference voltage has a nominal value VIN, as determined by the FSR pin and described in
1.1.4 The Analog Inputs.
There is no provision for the use of an external reference voltage, but the full-scale input voltage can be adjusted through
a Full Scale Register in the Extended Control Mode, as explained in 1.2 NON-EXTENDED CONTROL/EXTENDED
CONTROL.
Differential input signals up to the chosen full-scale level will
be digitized to 8 bits. Signal excursions beyond the full-scale
range will be clipped at the output. These large signal excursions will also activate the OR output for the time that the
signal is out of range. See 2.3.2 Out Of Range (OR) Indication.
One extra feature of the VBG pin is that it can be used to raise
the common mode voltage level of the LVDS outputs. The
output offset voltage (V OS) is typically 800 mV when the VBG
pin is used as an output or left unconnected. To raise the
LVDS offset voltage to a typical value of 1100 mV the VBG pin
can be connected directly to the supply rails.
30024744
FIGURE 12. Differential Input Drive
2.3.1 Handling Single-Ended Input Signals
There is no provision for the ADC08D1520QML to adequately
process single-ended input signals. The best way to handle
single-ended signals is to convert them to differential signals
before presenting them to the ADC. The easiest way to accomplish single-ended to differential signal conversion is with
an appropriate balun-connected transformer, as shown in
Figure 13.
2.3.1.1. a.c. Coupled Input
The easiest way to accomplish single-ended a.c. Input to differential a.c. signal is with an appropriate balun, as shown in
Figure 13.
2.3 THE ANALOG INPUT
The analog input is a differential one to which the signal
source must be a.c. coupled as shown in Figure 12. In the
Non-Extended Control Mode, the full-scale input range is selected with the FSR pin as specified in the Converter Electrical
Characteristics. In the Extended Control Mode, the full-scale
input range is selected by programming the Full-Scale Voltage Adjust register through the Serial Interface. For best
performance, when adjusting the input full-scale range in the
Extended Control, refer to 1.4 REGISTER DESCRIPTION for
guidelines on limiting the amount of adjustment.
Table 8 gives the input to output relationship with the FSR pin
high when the normal (Non-Extended) Mode is used. With the
www.national.com
VIN+
VCM − 225 mV
38
30024743
FIGURE 13. Single-Ended to Differential Signal
Conversion using a Balun
Figure 13 is a generic depiction of a single-ended to differential signal conversion using a balun. The circuitry specific to
the balun will depend upon the type of balun selected and the
overall board layout. It is recommended that the system designer contact the manufacturer of the balun they have selected to aid in designing the best performing single-ended to
differential conversion circuit using that particular balun.
When selecting a balun, it is important to understand the input
architecture of the ADC. There are specific balun parameters
of which the system designer should be mindful. A designer
should match the impedance of the analog source to the
ADC08D1520QML’s on-chip 100Ω differential input termination resistor. The range of this input termination resistor is
described in the Converter Electrical Characteristics as the
specification RIN.
Also, the phase and amplitude balance are important. The
lowest possible phase and amplitude imbalance is desired
when selecting a balun. The phase imbalance should be no
more than ±2.5° and the amplitude imbalance should be limited to less than 1dB at the desired input frequency range.
Finally, when selecting a balun, the VSWR (Voltage Standing
Wave Ratio), bandwidth and insertion loss of the balun should
also be considered. The VSWR aids in determining the overall
transmission line termination capability of the balun when interfacing to the ADC input. The insertion loss should be
considered so that the signal at the balun output is within the
specified input range of the ADC as described in the Converter Electrical Characteristics as the specification VIN.
30024747
FIGURE 14. Differential (LVDS) Input Clock Connection
The differential input clock line pair should have a characteristic impedance of 100Ω and (when using a balun), be terminated at the clock source in that (100 Ω) characteristic
impedance. The input clock line should be as short and as
direct as possible. The ADC08D1520QML clock input is internally terminated with an untrimmed 100Ω resistor.
Insufficient input clock levels will result in poor dynamic performance. Excessively high input clock levels could cause a
change in the analog input offset voltage. To avoid these
problems, keep the input clock level within the range specified
in the Converter Electrical Characteristics.
The low and high times of the input clock signal can affect the
performance of any A/D Converter. The ADC08D1520QML
features a duty cycle clock correction circuit which can maintain performance over temperature even in DES Mode. The
ADC will meet its performance specification if the input
clock high and low times are maintained within the range
(20/80% ratio).
High speed, high performance ADCs such as the ADC08D1520QML require a very stable input clock signal with
minimum phase noise or jitter. ADC jitter requirements are
defined by the ADC resolution (number of bits), maximum
ADC input frequency and the input signal amplitude relative
to the ADC input full scale range. The maximum jitter (the sum
of the jitter from all sources) allowed to prevent a jitter-induced
reduction in SNR is found to be
2.3.2 Out Of Range (OR) Indication
When the conversion result is clipped the Out of Range output
is activated such that OR+ goes high and OR- goes low. This
output is active as long as accurate data on either or both of
the buses would be outside the range of 00h to FFh. Note that
when the device is programmed to provide a second DCLK
output, the OR signals become DCLK2. Refer to 1.4 REGISTER DESCRIPTION.
2.3.3 Full-Scale Input Range
As with all A/D Converters, the input range is determined by
the value of the ADC's reference voltage. The reference voltage of the ADC08D1520QML is derived from an internal
band-gap reference. The FSR pin controls the effective reference voltage of the ADC08D1520QML such that the differential full-scale input range at the analog inputs is a normal
amplitude with the FSR pin high, or a reduced amplitude with
FSR pin low as defined by the specification VIN in the Converter Electrical Characteristics. Best SNR is obtained with
FSR high.
tJ(MAX) = (VIN(P-P)/VINFSR) x (1/(2(N+1) x π x fIN))
where tJ(MAX) is the rms total of all jitter sources in seconds,
VIN(P-P) is the peak-to-peak analog input signal, VINFSR is the
full-scale range of the ADC, "N" is the ADC resolution in bits
and fIN is the maximum input frequency, in Hertz, at the ADC
analog input.
Note that the maximum jitter described above is the RSS sum
of the jitter from all sources, including that in the ADC input
clock, that added by the system to the ADC input clock and
2.4 THE CLOCK INPUTS
The ADC08D1520QML has differential LVDS clock inputs,
CLK+ and CLK-, which must be driven with an a.c. coupled,
differential clock signal. Although the ADC08D1520QML is
39
www.national.com
ADC08D1520QML
tested and its performance is guaranteed with a differential
1.5 GHz clock, it typically will function well with input clock
frequencies indicated in the Converter Electrical Characteristics. The clock inputs are internally terminated and biased.
The input clock signal must be capacitive coupled to the clock
pins as indicated in Figure 14.
Operation up to the sample rates indicated in the Converter
Electrical Characteristics is typically possible if the maximum
ambient temperatures indicated are not exceeded. Operating
at higher sample rates than indicated for the given ambient
temperature may result in reduced device reliability and product lifetime. This is because of the higher power consumption
and die temperatures at high sample rates. Important also for
reliability is proper thermal management. See 2.7.2 Thermal
Management.
ADC08D1520QML
input signals and that added by the ADC itself. Since the effective jitter added by the ADC is beyond user control, the best
the user can do is to keep the sum of the externally added
input clock jitter and the jitter added by the analog circuitry to
the analog signal to a minimum.
Input clock amplitudes above those specified in the Converter
Electrical Characteristics may result in increased input offset
voltage. This would cause the converter to produce an output
code other than the expected 127/128 when both input pins
are at the same potential.
and then programmed to 1b for a minimum of tCAL_H input
clock cycles to initiate a calibration cycle. The CalRun signal
should be monitored to determine when the calibration cycle
has completed. The CalRun pin will become a logic high indicating an active calibration cycle regardless of which
method was used to initiate the calibration cycle. Note that the
DCLK outputs are not active during a calibration cycle; therefore, it is not recommended for use as a system clock.
The minimum number of tCAL_L and tCAL_H input clock cycle
sequences are required to ensure that random noise does not
cause a calibration to begin when it is not desired. As mentioned in 1.1 OVERVIEW, for best performance, a calibration
should be performed 20 seconds or more after power up and
repeated when the operating temperature changes significantly relative to the specific system design performance
requirements. Dynamic performance changes slightly with increasing junction temperature and can be easily corrected by
performing a calibration.
2.5 CONTROL PINS
Six control pins (without the use of the serial interface) provide
a wide range of possibilities in the operation of the ADC08D1520QML and facilitate its use. These control pins provide Full-Scale Input Range setting, Self Calibration, Output
Edge Synchronization choice, LVDS Output Level choice and
a Power Down feature.
2.5.1 Full-Scale Input Range Setting
The input full-scale range can be selected with the FSR control input (pin 14) in the Normal Mode of operation. The input
full-scale range is specified as VIN in the Converter Electrical
Characteristics. In the Extended Control Mode, the input fullscale range may be programmed using the Full-Scale Adjust
Voltage register. See 2.3 THE ANALOG INPUT for more information.
2.5.3 Output Edge Synchronization
DCLK signals are available to help latch the converter output
data into external circuitry. The output data can be synchronized with either edge of these DCLK signals. That is, the
output data transition can be set to occur with either the rising
edge or the falling edge of the DCLK signal, so that either
edge of that DCLK signal can be used to latch the output data
into the receiving circuit.
When OutEdge (pin 4) is high, the output data is synchronized
with (changes with) the rising edge of the DCLK+ (pin 82).
When OutEdge is low, the output data is synchronized with
the falling edge of DCLK+.
At the very high speeds of which the ADC08D1520QML is
capable, slight differences in the lengths of the DCLK and
data lines can mean the difference between successful and
erroneous data capture. The OutEdge pin is used to capture
data on the DCLK edge that best suits the application circuit
and layout.
2.5.2 Calibration
The ADC08D1520QML calibration must be run to achieve
specified performance. The calibration must be initiated by
the user. The calibration procedure is exactly the same
whether there is an input clock present upon power up or if
the clock begins some time after application of power. The
CalRun output indicator is high while a calibration is in
progress. Note that the DCLK outputs are not active during a
calibration cycle by default, therefore it is not recommended
for use as a system clock. The DCLK outputs are continuously
present at the output only when the Resistor Trim Disable is
activated.
2.5.4 LVDS Output Level Control
The output level can be set to one of two levels with OutV
(pin3). The strength of the output drivers is greater with OutV
high. With OutV low there is less power consumption in the
output drivers, but the lower output level means decreased
noise immunity.
For short LVDS lines and low noise systems, satisfactory performance may be realized with the OutV input low. If the LVDS
lines are long and/or the system in which the ADC08D1520QML is used is noisy, it may be necessary to tie
the OutV pin high.
2.5.2.1 Initiating Calibration
A calibration may be run at any time in both the Non-DES and
DES Modes. After power-up, we recommend that the part be
calibrated with the Resistor Trim Disable inactive once the
power supplies have stabilized and the temperature of the
chip has stabilized. When a calibration is run with the Resistor
Trim Disable inactive, both the ADC and the input termination
resistor are calibrated. However, since the input termination
resistance changes only marginally with temperature, the user has the option to disable the input termination resistor
calibration for subsequent calibrations, which will guarantee
that the DCLK is continuously present at the output. The Resistor Trim Disable can be programmed in the Extended
Configuration register (Addr: 9h) when in the Extended Control Mode. Refer to Extended Configuration Register for register programming information.
As dynamic performance changes slightly with junction temperature, a calibration may be executed to bring the performance of the ADC in line. Two methods can be used initiate
a calibration. The first method is to hold the CAL pin low for
at least tCAL_L input clock cycles, then hold it high for at least
another tCAL_H input clock cycles. The second method is to
program the CAL bit in the Calibration register while in Extended Control Mode. The functionality of the CAL bit is
exactly the same as using the CAL pin. The CAL bit must be
programmed to 0b for a minimum of tCAL_Linput clock cycles
www.national.com
2.5.5 Dual Edge Sampling
The Dual Edge Sampling (DES) feature causes one of the two
input pairs to be routed to both ADCs. The other input pair is
deactivated. One of the ADCs samples the input signal on the
rising input clock edge (duty cycle corrected), the other samples the input signal on the falling input clock edge (duty cycle
corrected). If the device is in the 1:4 Demux DES Mode, the
result is an output data rate 1/4 that of the interleaved sample
rate which is twice the input clock frequency. Data is presented in parallel on all four output buses in the following order:
DQd, DId, DQ, DI. If the device is the Non-Demultiplex output
mode, the result is an output data rate 1/2 that of the interleaved sample rate. Data is presented in parallel on two
output buses in the following order: DQ, DI.
40
centimeter. Leadless chip capacitors are preferred because
they have low lead inductance.
The VA and VDR supply pins should be isolated from each
other to prevent any digital noise from being coupled into the
analog portions of the ADC. A ferrite choke, such as the JW
Miller FB20009-3B, is recommended between these supply
lines when a common source is used for them.
As is the case with all high speed converters, the ADC08D1520QML should be assumed to have little power supply noise rejection. Any power supply used for digital circuitry
in a system where a lot of digital power is being consumed
should not be used to supply power to the ADC08D1520QML.
The ADC supplies should be the same supply used for other
analog circuitry, if not a dedicated supply.
2.5.6 Power Down Feature
The Power Down pins (PD and PDQ) allow the ADC08D1520QML to be entirely powered down (PD) or the QChannel channel to be powered down and the I- Channel to
remain active. See 1.1.7 Power Down for details on the power
down feature.
The digital data (+/-) output pins are put into a high impedance
state when the PD pin for the respective channel is high. Upon
return to normal operation, the pipeline will contain meaningless information and must be flushed.
If the PD input is brought high while a calibration is running,
the device will not go into power down until the calibration
sequence is complete. However, if power is applied and PD
is simultaneously ramped, the device will not calibrate until
the PD input goes low. When PD is high and a calibration is
initiated, the request for calibration is completely ignored. Refer to 1.1.7 Power Down
2.7.1 Supply Voltage
The ADC08D1520QML is specified to operate with a supply
voltage of 1.9V ±0.1V. It is very important to note that, while
this device will function with slightly higher supply voltages,
these higher supply voltages may reduce product lifetime.
No pin should ever have a voltage on it that is in excess of the
supply voltage or below ground by more than 150 mV, not
even on a transient basis. This can be a problem upon application of power and power shut-down. Be sure that the supplies to circuits driving any of the input pins, analog or digital,
do not come up any faster than does the voltage at the ADC08D1520QML power pins.
The Absolute Maximum Ratings should be strictly observed,
even during power up and power down. A power supply that
produces a voltage spike at turn-on and/or turn-off of power
can destroy the ADC08D1520QML. The circuit of Figure 15
will provide supply overshoot protection.
Many linear regulators will produce output spiking at poweron unless there is a minimum load provided. Active devices
draw very little current until their supply voltages reach a few
hundred millivolts. The result can be a turn-on spike that can
destroy the ADC08D1520QML, unless a minimum load is
provided for the supply. The 100Ω resistor at the regulator
output provides a minimum output current during power-up to
ensure there is no turn-on spiking.
In the circuit of Figure 15, an LM317 linear regulator is satisfactory if its input supply voltage is 4V to 5V . If a 3.3V supply
is used, an LM1086 linear regulator is recommended.
2.6 THE DIGITAL OUTPUTS
The ADC08D1520QML demultiplexes the output data of each
of the two ADCs on the die onto two LVDS output buses (total
of four buses, two for each ADC). For each of the two converters, the results of successive conversions started on the
odd falling edges of the CLK+ pin are available on one of the
two LVDS buses, while the results of conversions started on
the even falling edges of the CLK+ pin are available on the
other LVDS bus. This means that, the word rate at each LVDS
bus is 1/2 the ADC08D1520QML input clock rate and the two
buses must be multiplexed to obtain the entire 1.5 GSPS
conversion result.
Since the minimum recommended input clock rate for this
device is 200 MSPS (Non DES Mode), the effective rate can
be reduced to as low as 100 MSPS by using the results available on just one of the two LVDS buses and a 200 MHz input
clock, decimating the 200 MSPS data by two.
There is one LVDS output clock pair (DCLK+/-) available for
use to latch the LVDS outputs on all buses. Whether the data
is sent at the rising or falling edge of DCLK is determined by
the sense of the OutEdge pin, as described in 2.5.3 Output
Edge Synchronization.
DDR (Double Data Rate) clocking can also be used. In this
mode a word of data is presented with each edge of DCLK,
reducing the DCLK frequency to 1/4 the input clock frequency.
See the Timing Diagram section for details.
The OutV pin is used to set the LVDS differential output levels.
See 2.5.4 LVDS Output Level Control.
The output format is Offset Binary. Accordingly, a full-scale
input level with VIN+ positive with respect to VIN− will produce
an output code of all ones, a full-scale input level with VIN−
positive with respect to VIN+ will produce an output code of all
zeros and when VIN+ and VIN− are equal, the output code will
vary between codes 127 and 128.
30024754
FIGURE 15. Non-Spiking Power Supply
The output drivers should have a supply voltage, VDR, that is
within the range specified in the Operating Ratings table. This
voltage should not exceed the VA supply voltage.
If the power is applied to the device without an input clock
signal present, the current drawn by the device might be below 200 mA. This is because the ADC08D1520QML gets
reset through clocked logic and its initial state is unknown. If
the reset logic comes up in the "on" state, it will cause most
of the analog circuitry to be powered down, resulting in less
2.7 POWER CONSIDERATIONS
A/D converters draw sufficient transient current to corrupt
their own power supplies if not adequately bypassed. A 33 µF
capacitor should be placed within an inch (2.5 cm) of the A/D
converter power pins. A 0.1 µF capacitor should be placed as
close as possible to each VA pin, preferably within one-half
41
www.national.com
ADC08D1520QML
To use this feature in the Non-Extended Control Mode, tie pin
127 to VA/2 and the signal at the I- channel input will be sampled by both converters.
In the Extended Control Mode, either input may be used for
dual edge sampling. See 1.1.5.1 Dual-Edge Sampling.
ADC08D1520QML
top and bottom copper areas. These thermal vias act as "heat
pipes" to carry the thermal energy from the device side of the
board to the opposite side of the board where it can be more
effectively dissipated. The use of approximately 100 thermal
vias is recommended. Use of a higher weight copper on the
internal ground plane is recommended, (i.e. 2OZ instead of
1OZ, for thermal considerations only.
The thermal vias should be placed on a 61mil grid spacing
and have a diameter of 15 mil typically. These vias should be
barrel plated to avoid solder wicking into the vias during the
soldering process as this wicking could cause voids in the
solder between the package exposed pad and the thermal
land on the PCB. Such voids could increase the thermal resistance between the device and the thermal land on the
board, which would cause the device to run hotter.
If it is desired to monitor die temperature, a temperature sensor may be mounted on the heat sink area of the board near
the thermal vias. Allow for a thermal gradient between the
temperature sensor and the ADC08D1520QML die of θJ-PAD
times typical power consumption.
than 100 mA of current draw. This current is greater than the
power down current because not all of the ADC is powered
down. The device current will be normal after the input clock
is established.
2.7.2 Thermal Management
The ADC08D1520QML is capable of impressive speeds and
performance at very low power levels for its speed. However,
the power consumption is still high enough to require attention
to thermal management. For reliability reasons, the die temperature should be kept to a maximum of 150°C. That is, TA
(ambient temperature) plus ADC power consumption times
θJA (junction to ambient thermal resistance) should not exceed 150°C.
Please note that the following are recommendations for
mounting this device onto a PCB. This should be considered
the starting point in PCB and assembly process development.
It is recommended that the process be developed based upon
past experience in package mounting.
The bottom of the package of the ADC08D1520QML provides
the primary heat removal path as well as excellent electrical
grounding to the printed circuit board. The land pattern design
for lead attachment to the PCB should be the same as for a
conventional LQFP, but the bottom of the package must be
attached to the board to remove the maximum amount of heat
from the package, as well as to ensure best product parametric performance.
To maximize the removal of heat from the package, a thermal
land pattern must be incorporated on the PC board within the
footprint of the package. The bottom of the device must be
soldered down to ensure adequate heat conduction out of the
package. The land pattern for this exposed pad should be as
large as the 600 x 600 mil bottom of the package and be located such that the bottom of the device is entirely over that
thermal land pattern. This thermal land pattern should be
electrically connected to ground.
2.8 LAYOUT AND GROUNDING
Proper grounding and proper routing of all signals are essential to ensure accurate conversion. A single ground plane
should be used, instead of splitting the ground plane into analog and digital areas.
Since digital switching transients are composed largely of
high frequency components, the skin effect tells us that total
ground plane copper weight will have little effect upon the
logic-generated noise. Total surface area is more important
than is total ground plane volume. Coupling between the typically noisy digital circuitry and the sensitive analog circuitry
can lead to poor performance that may seem impossible to
isolate and remedy. The solution is to keep the analog circuitry well separated from the digital circuitry.
High power digital components should not be located on or
near any linear component or power supply trace or plane that
services analog or mixed signal components as the resulting
common return current path could cause fluctuation in the
analog input “ground” return of the ADC, causing excessive
noise in the conversion result.
Generally, we assume that analog and digital lines should
cross each other at 90° to avoid getting digital noise into the
analog path. In high frequency systems, however, avoid
crossing analog and digital lines altogether. The input clock
lines should be isolated from ALL other lines, analog AND
digital. The generally accepted 90° crossing should be avoided as even a little coupling can cause problems at high
frequencies. Best performance at high frequencies is obtained with a straight signal path.
The analog input should be isolated from noisy signal traces
to avoid coupling of spurious signals into the input. This is
especially important with the low level drive required of the
ADC08D1520QML. Any external component (e.g., a filter capacitor) connected between the converter's input and ground
should be connected to a very clean point in the analog
ground plane. All analog circuitry (input amplifiers, filters, etc.)
should be separated from any digital components.
30024721
FIGURE 16. Recommended Package Land Pattern
Since a large aperture opening may result in poor release, the
aperture opening should be subdivided into an array of smaller openings, similar to the land pattern of Figure 16.
To minimize junction temperature, it is recommended that a
simple heat sink be built into the PCB. This is done by including a copper area of about 2.25 square inches (14.52 square
cm) on the opposite side of the PCB. This copper area may
be plated or solder coated to prevent corrosion, but should
not have a conformal coating, which could provide some thermal insulation. Thermal vias should be used to connect these
www.national.com
2.9 DYNAMIC PERFORMANCE
The ADC08D1520QML is a.c. tested and its dynamic performance is guaranteed. To meet the published specifications
and avoid jitter-induced noise, the clock source driving the
CLK input must exhibit low rms jitter. The allowable jitter is a
function of the input frequency and the input signal level, as
described in 2.4 THE CLOCK INPUTS.
42
TABLE 10. Extended Control Mode Operation (Pin 41
Logic Low and Pin 52 VA/2 or Logic High)
Pin
2.10 USING THE SERIAL INTERFACE
The ADC08D1520QML may be operated in the Non-Extended Control (non-Serial Interface) Mode or in the extended
control mode. Table 9 and Table 10 describe the functions of
pins 3, 4, 14 and 127 in the Non-Extended Control Mode and
the Extended Control Mode, respectively.
TABLE 9. Non-Extended Control Mode Operation
(Pin 41 VA/2 and Pin 52 VA/2 or Logic High)
Low
High
VA/2
3
Reduced VOD
Normal VOD
n/a
4
OutEdge = Neg
OutEdge = Pos
DDR
127
N/A
N/A
DES
Normal VIN
Extended
Control
Mode
14
Reduced VIN
3
SCLK (Serial Clock)
4
SDATA (Serial Data)
127
SCS (Serial Interface Chip Select)
2.11 COMMON APPLICATION PITFALLS
Driving the inputs (analog or digital) beyond the power
supply rails. For device reliability, no input should go more
than 150 mV below the ground pins or 150 mV above the
supply pins. Exceeding these limits on even a transient basis
may not only cause faulty or erratic operation, but may impair
device reliability. It is not uncommon for high speed digital
circuits to exhibit undershoot that goes more than a volt below
ground. Controlling the impedance of high speed lines and
terminating these lines in their characteristic impedance
should control overshoot.
Care should be taken not to overdrive the inputs of the ADC08D1520QML. Such practice may lead to conversion inaccuracies and even to device damage.
Driving the VBG pin to change the reference voltage. As
mentioned in 2.2 THE REFERENCE VOLTAGE, the reference voltage is intended to be fixed to provide one of two
different full-scale values (650 mVP-P and 870 mVP-P). Over
driving this pin will not change the full scale value, but can be
used to change the LVDS common mode voltage from 0.8V
to 1.2V by tying the VBG pin to VA.
Driving the clock input with an excessively high level
signal. The ADC input clock level should not exceed the level
described in the Operating Ratings Table or the input offset
could change.
Inadequate input clock levels. As described in 2.4 THE
CLOCK INPUTS, insufficient input clock levels can result in
poor performance. Excessive input clock levels could result
in the introduction of an input offset.
Using a clock source with excessive jitter, using an excessively long input clock signal trace, or having other
signals coupled to the input clock signal trace. This will
cause the sampling interval to vary, causing excessive output
noise and a reduction in SNR performance.
Failure to provide adequate heat removal. As described in
2.7.2 Thermal Management, it is important to provide adequate heat removal to ensure device reliability. This can be
done either with adequate air flow or the use of a simple heat
sink built into the board. The backside pad should be grounded for best performance.
2.10.1 Non-Extended Control Mode Operation
Non-extended Control Mode operation means that the Serial
Interface is not active and all controllable functions are controlled with various pin settings. Pin 41 is the primary control
of the extended control enable function. When pin 41 is logic
high, the device is in the Non-Extended Control Mode. If pin
41 is tied to VA/2 and pin 52 connected to VA/2 or logic high,
the extended control enable function is controlled by pin 14.
The device has functions which are pin programmable when
in the Non-Extended Control Mode. An example is the fullscale range is controlled in the Non-Extended Control Mode
by setting pin 14 high or low. Table 9 indicates the pin functions of the ADC08D1520QML in the Non-Extended Control
Mode.
Pin
Function
Pin 3 can be either high or low in the Non-Extended Control
Mode. See 1.2 NON-EXTENDED CONTROL/EXTENDED
CONTROL for more information.
Pin 4 can be high or low in the Non-Extended Control Mode.
In the Non-Extended Control Mode, pin 4 high or low defines
the edge at which the output data transitions. See 2.5.3 Output Edge Synchronization for more information. If this pin is
tied to VA/2, the output clock (DCLK) is a DDR (Double Data
Rate) clock (see 1.1.5.3 Double Data Rate) and the output
edge synchronization is irrelevant since data is clocked out
on both DCLK edges.
43
www.national.com
ADC08D1520QML
When in Normal Mode, Pin 127 must be tied high. If pin 127
is tied to VA/2, the converter performs dual edge sampling
(DES).
It is good practice to keep the ADC input clock line as short
as possible, to keep it well away from any other signals and
to treat it as a transmission line. Other signals can introduce
jitter into the input clock signal. The clock signal can also introduce noise into the analog path if not isolated from that
path.
Best dynamic performance is obtained when the exposed pad
at the back of the package has a good connection to ground.
This is because this path from the die to ground is a lower
impedance than offered by the package pins.
ADC08D1520QML
Revision History
Date
Released
Revision
01/09/08
03/05/08
www.national.com
Section
Originator
Changes
A
Initial Release, New Product
R. Eddy/R.
Rennie
New Product Data Sheet, Released at Edit
16
B
Whole data sheet
R. Rennie
Edit clarification Updates, Revision A will be
Archived.
44
ADC08D1520QML
Physical Dimensions inches (millimeters) unless otherwise noted
NOTES: UNLESS OTHERWISE SPECIFIED
REFERENCE JEDEC REGISTRATION MS-026, VARIATION BFB.
128-Lead Ceramic Quad
(Gold Lead Finish)
NS Package Number EL128A
45
www.national.com
ADC08D1520QML Low Power, 8-Bit, Dual 1.5 GSPS or Single 3.0 GSPS A/D Converter
Notes
For more National Semiconductor product information and proven design tools, visit the following Web sites at:
Products
Design Support
Amplifiers
www.national.com/amplifiers
WEBENCH
www.national.com/webench
Audio
www.national.com/audio
Analog University
www.national.com/AU
Clock Conditioners
www.national.com/timing
App Notes
www.national.com/appnotes
Data Converters
www.national.com/adc
Distributors
www.national.com/contacts
Displays
www.national.com/displays
Green Compliance
www.national.com/quality/green
Ethernet
www.national.com/ethernet
Packaging
www.national.com/packaging
Interface
www.national.com/interface
Quality and Reliability
www.national.com/quality
LVDS
www.national.com/lvds
Reference Designs
www.national.com/refdesigns
Power Management
www.national.com/power
Feedback
www.national.com/feedback
Switching Regulators
www.national.com/switchers
LDOs
www.national.com/ldo
LED Lighting
www.national.com/led
PowerWise
www.national.com/powerwise
Serial Digital Interface (SDI)
www.national.com/sdi
Temperature Sensors
www.national.com/tempsensors
Wireless (PLL/VCO)
www.national.com/wireless
THE CONTENTS OF THIS DOCUMENT ARE PROVIDED IN CONNECTION WITH NATIONAL SEMICONDUCTOR CORPORATION
(“NATIONAL”) PRODUCTS. NATIONAL MAKES NO REPRESENTATIONS OR WARRANTIES WITH RESPECT TO THE ACCURACY
OR COMPLETENESS OF THE CONTENTS OF THIS PUBLICATION AND RESERVES THE RIGHT TO MAKE CHANGES TO
SPECIFICATIONS AND PRODUCT DESCRIPTIONS AT ANY TIME WITHOUT NOTICE. NO LICENSE, WHETHER EXPRESS,
IMPLIED, ARISING BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS
DOCUMENT.
TESTING AND OTHER QUALITY CONTROLS ARE USED TO THE EXTENT NATIONAL DEEMS NECESSARY TO SUPPORT
NATIONAL’S PRODUCT WARRANTY. EXCEPT WHERE MANDATED BY GOVERNMENT REQUIREMENTS, TESTING OF ALL
PARAMETERS OF EACH PRODUCT IS NOT NECESSARILY PERFORMED. NATIONAL ASSUMES NO LIABILITY FOR
APPLICATIONS ASSISTANCE OR BUYER PRODUCT DESIGN. BUYERS ARE RESPONSIBLE FOR THEIR PRODUCTS AND
APPLICATIONS USING NATIONAL COMPONENTS. PRIOR TO USING OR DISTRIBUTING ANY PRODUCTS THAT INCLUDE
NATIONAL COMPONENTS, BUYERS SHOULD PROVIDE ADEQUATE DESIGN, TESTING AND OPERATING SAFEGUARDS.
EXCEPT AS PROVIDED IN NATIONAL’S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, NATIONAL ASSUMES NO
LIABILITY WHATSOEVER, AND NATIONAL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY RELATING TO THE SALE
AND/OR USE OF NATIONAL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR
PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY
RIGHT.
LIFE SUPPORT POLICY
NATIONAL’S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT DEVICES OR
SYSTEMS WITHOUT THE EXPRESS PRIOR WRITTEN APPROVAL OF THE CHIEF EXECUTIVE OFFICER AND GENERAL
COUNSEL OF NATIONAL SEMICONDUCTOR CORPORATION. As used herein:
Life support devices or systems are devices which (a) are intended for surgical implant into the body, or (b) support or sustain life and
whose failure to perform when properly used in accordance with instructions for use provided in the labeling can be reasonably expected
to result in a significant injury to the user. A critical component is any component in a life support device or system whose failure to perform
can be reasonably expected to cause the failure of the life support device or system or to affect its safety or effectiveness.
National Semiconductor and the National Semiconductor logo are registered trademarks of National Semiconductor Corporation. All other
brand or product names may be trademarks or registered trademarks of their respective holders.
Copyright© 2008 National Semiconductor Corporation
For the most current product information visit us at www.national.com
National Semiconductor
Americas Technical
Support Center
Email:
[email protected]
Tel: 1-800-272-9959
www.national.com
National Semiconductor Europe
Technical Support Center
Email: [email protected]
German Tel: +49 (0) 180 5010 771
English Tel: +44 (0) 870 850 4288
National Semiconductor Asia
Pacific Technical Support Center
Email: [email protected]
National Semiconductor Japan
Technical Support Center
Email: [email protected]