NSC ADC08D1000CIYB High performance, low power, dual 8-bit, 1 gsps a/d converter Datasheet

ADC08D1000
High Performance, Low Power, Dual 8-Bit, 1 GSPS A/D
Converter
General Description
Features
The ADC08D1000 is a dual, low power, high performance
CMOS analog-to-digital converter that digitizes signals to 8
bits resolution at sampling rates up to 1.3 GSPS. Consuming
a typical 1.6 Watts at 1 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 7.4 ENOB with a 500 MHz input signal and a 1 GHz
sample rate while providing a 10-18 B.E.R. Output formatting
is offset binary and the 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 1:2 demultiplexer that feeds two LVDS
buses and reduces the output data rate on each bus to half
the sampling rate. The two converters can be interleaved and
used as a single 2 GSPS ADC.
The converter typically consumes less than 3.5 mW in the
Power Down Mode and is available in a 128-lead, thermally
enhanced exposed pad LQFP and operates over the Industrial (-40°C ≤ TA ≤ +85°C) temperature range.
■
■
■
■
■
■
■
■
■
Internal Sample-and-Hold
Single +1.9V ±0.1V Operation
Choice of SDR or DDR output clocking
Interleave Mode for 2x Sampling Rate
Multiple ADC Synchronization Capability
Guaranteed No Missing Codes
Serial Interface for Extended Control
Fine Adjustment of Input Full-Scale Range and Offset
Duty Cycle Corrected Sample Clock
Key Specifications
■
■
■
■
■
■
■
Resolution
Max Conversion Rate
Bit Error Rate
ENOB @ 500 MHz Input
DNL
Power Consumption
— Operating
— Power Down Mode
8 Bits
1 GSPS (min)
10-18 (typ)
7.4 Bits (typ)
±0.15 LSB (typ)
1.6 W (typ)
3.5 mW (typ)
Applications
■
■
■
■
■
Direct RF Down Conversion
Digital Oscilloscopes
Satellite Set-top boxes
Communications Systems
Test Instrumentation
Block Diagram
20097453
© 2009 National Semiconductor Corporation
200974
www.national.com
ADC08D1000 High Performance, Low Power, Dual 8-Bit, 1 GSPS A/D Converter
April 16, 2009
ADC08D1000
Ordering Information
Industrial Temperature Range (-40°C < TA < +85°C)
NS Package
ADC08D1000CIYB
128-Pin Exposed Pad LQFP
ADC08D1000DEV
Development Board
Pin Configuration
20097401
* Exposed pad on back of package must be soldered to ground plane to ensure rated performance.
www.national.com
2
ADC08D1000
Pin Descriptions and Equivalent Circuits
Pin Functions
Pin No.
Symbol
Equivalent Circuit
Description
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. See1.1.6 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 NORMAL/EXTENDED
CONTROL for details on the extended control mode. See 1.3 THE
SERIAL INTERFACE for description of the serial interface.
4
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 Setting). When this pin is
floating or connected to 1/2 the supply voltage, DDR clocking is
enabled. When the extended control mode is enabled, this pin
functions as the SDATA input. See 1.2 NORMAL/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-
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 self calibration sequence. See 2.4.2 Self
Calibration for an overview of self-calibration and 2.4.2.2 OnCommand Calibration for a description of on-command
calibration.
PDQ
A logic high on the PDQ pin puts only the "Q" ADC into the Power
Down mode.
FSR/ECE
Full Scale Range Select and Extended Control Enable. In nonextended control mode, 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 a higher
VIN input level. See Converter Electrical Characteristics. To enable
the extended control mode, whereby the serial interface and
control registers are employed, allow this pin to float or connect it
to a voltage equal to VA/2. See 1.2 NORMAL/EXTENDED
CONTROL for information on the extended control mode.
3
30
29
14
DCLK Reset. A positive pulse on this pin is used to reset and
synchronize the DCLK outs of multiple converters. See 1.5
MULTIPLE ADC SYNCHRONIZATION for detailed description.
3
www.national.com
ADC08D1000
Pin Functions
Pin No.
Symbol
Equivalent Circuit
Description
CalDly / DES /
SCS
Calibration Delay, Dual Edge Sampling and Serial Interface Chip
Select. With a logic high or low on pin 14, this pin functions as
Calibration Delay and sets the number of input clock cycles after
power up before calibration begins (See 1.1.1 Self-Calibration).
With pin 14 floating, this pin acts as the enable pin for the serial
interface input and the CalDly value becomes "0" (short delay with
no provision for a long power-up calibration delay). When this pin
is floating or connected to a voltage equal to VA/2, DES (Dual Edge
Sampling) mode is selected where the "I" input is sampled at twice
the input clock rate and the "Q" input is ignored. See 1.1.5.1 DualEdge Sampling.
18
19
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 Section 2.3 for an overview
of the clock inputs.
11
10
VINI+
VINI−
22
23
VINQ+
VINQ−
127
7
VCMO
31
VBG
126
CalRun
www.national.com
Analog signal inputs to the ADC. The differential full-scale input
range of this input is programmable using the FSR pin 14 in normal
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 normal mode. Refer to 1.4 REGISTER DESCRIPTION for
the full-scale input range in the extended control mode.
Common Mode Voltage. The voltage output at this pin is required
to be the common mode input voltage at VIN+ and VIN− when d.c.
coupling is used. This pin should be grounded when a.c. coupling
is used at the analog inputs. This pin is capable of sourcing or
sinking 100μA. See 2.2 THE ANALOG INPUT.
Bandgap output voltage capable of 100 μA source/sink.
Calibration Running indication. This pin is at a logic high when
calibration is running.
4
ADC08D1000
Pin Functions
Pin No.
Symbol
32
REXT
External bias resistor connection. Nominal value is 3.3k-Ohms
(±0.1%) to ground. See Section 1.1.1.
Tdiode_P
Tdiode_N
Temperature Diode Positive (Anode) and Negative (Cathode).
These pins may be used for die temperature measurements,
however no specified accuracy is implied or guaranteed. Noise
coupling from adjacent output data signals has been shown to
affect temperature measurements using this feature. See 2.6.2
Thermal Management.
34
35
Equivalent Circuit
Description
5
www.national.com
ADC08D1000
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+
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/DQ
outputs, these outputs represent the earlier time sample. These
outputs should always be terminated with a 100Ω differential
resistor.
OR+
OR-
Out Of Range output. A differential high at these pins indicates that
the differential input is out of range (outside the 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).
82
81
DCLK+
DCLK-
Differential Clock outputs used to latch the output data. Delayed
and non-delayed data outputs are supplied synchronous to this
signal. 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. The DCLK outputs
are not active during a calibration cycle, therefore this is not
recommended as a system clock.
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, 9, 12,
21, 24, 27,
41
GND
Ground return for VA.
79
80
www.national.com
Equivalent Circuit
Description
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.
6
ADC08D1000
Pin Functions
Pin No.
Symbol
42, 53, 64,
74, 87, 97,
108, 119
DR GND
52, 63, 98,
109, 120
NC
Equivalent Circuit
Description
Ground return for VDR.
No Connection. Make no connection to these pins.
7
www.national.com
ADC08D1000
Operating Ratings
Absolute Maximum Ratings
(Notes 1, 2)
−40°C ≤ TA ≤ +85°C
(Notes 1, 2)
Ambient Temperature Range
If Military/Aerospace specified devices are required,
please contact the National Semiconductor Sales Office/
Distributors for availability and specifications.
Supply Voltage (VA)
Driver Supply Voltage (VDR)
Analog Input Common Mode Voltage
VIN+, VIN- Voltage Range
(Maintaining Common Mode)
Supply Voltage (VA, VDR)
Supply Difference
VDR - VA
Voltage on Any Input Pin
(Except VIN+, VIN- )
Voltage on VIN+, VIN(Maintaining Common Mode)
Ground Difference
|GND - DR GND|
Input Current at Any Pin (Note 3)
Package Input Current (Note 3)
2.2V
+1.8V to +2.0V
+1.8V to VA
VCMO ±50mV
0V to 2.15V
(100% duty cycle)
0V to 2.5V
(10% duty cycle)
0V to 100 mV
Ground Difference
(|GND - DR GND|)
CLK Pins Voltage Range
Differential CLK Amplitude
−0.15V to (VA +0.15V)
-0.15V to 2.5V
0V to 100 mV
±25 mA
±50 mA
Power Dissipation at TA ≤ 85°C
ESD Susceptibility (Note 4)
Human Body Model
Machine Model
Soldering Temperature, Infrared,
10 seconds, (Note 5), (Applies
to standard plated package only)
Storage Temperature
0V
0V to VA
0.4VP-P to 2.0VP-P
Package Thermal Resistance
Package
2.0 W
128-Lead
Exposed Pad
LQFP
2500V
250V
θJA
25°C / W
θJC(Top of
θJ-PAD
Package)
(Thermal Pad)
10°C / W
2.8°C / W
Soldering
process
must
comply
with
National
Semiconductor’s Reflow Temperature Profile specifications.
Refer to www.national.com/packaging.
235°C
−65°C to +150°C
Converter Electrical Characteristics
The following specifications apply after calibration for VA = VDR = +1.9VDC, OutV = 1.9V, VIN FSR (a.c. coupled) = differential
870mVP-P, CL = 10 pF, Differential, a.c. coupled Sinewave Input Clock, fCLK = 1 GHz at 0.5VP-P with 50% duty cycle, VBG = Floating,
Non-Extended Control Mode, SDR Mode, REXT = 3300Ω ±0.1%, Analog Signal Source Impedance = 100Ω Differential. Boldface
limits apply for TA = TMIN to TMAX. All other limits TA = 25°C, unless otherwise noted. (Notes 6, 7)
Symbol
Parameter
Conditions
Typical
(Note 8)
Limits
(Note 8)
Units
(Limits)
STATIC CONVERTER CHARACTERISTICS
INL
Integral Non-Linearity (Best fit)
DC Coupled, 1MHz Sine Wave Over
ranged
±0.3
±0.9
LSB (max)
DNL
Differential Non-Linearity
DC Coupled, 1MHz Sine Wave Over
ranged
±0.15
±0.6
LSB (max)
8
Bits
−1.5
0.5
LSB (min)
LSB (max)
Resolution with No Missing
Codes
VOFF
Offset Error
VOFF_ADJ
Input Offset Adjustment Range
PFSE
Positive Full-Scale Error (Note 9)
−0.6
±25
mV (max)
NFSE
Negative Full-Scale Error (Note
9)
−1.31
±25
mV (max)
FS_ADJ
Full-Scale Adjustment Range
±20
±15
%FS
-0.45
Extended Control Mode
Extended Control Mode
±45
mV
NORMAL MODE (Non DES) DYNAMIC CONVERTER CHARACTERISTICS
FPBW
B.E.R.
Full Power Bandwidth
1.7
GHz
10-18
Error/Sample
d.c. to 500 MHz
±0.5
dBFS
d.c. to 1 GHz
±1.0
dBFS
fIN = 100 MHz, VIN = FSR − 0.5 dB
7.5
fIN = 248 MHz, VIN = FSR − 0.5 dB
7.4
7.0
Bits (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
7.4
7.0
Bits (min)
Bit Error Rate
Gain Flatness
ENOB
Normal Mode (non DES)
Effective Number of Bits
www.national.com
8
Bits
SINAD
SNR
THD
2nd Harm
3rd Harm
SFDR
IMD
Parameter
Signal-to-Noise Plus Distortion
Ratio
Signal-to-Noise Ratio
Total Harmonic Distortion
Second Harmonic Distortion
Third Harmonic Distortion
Spurious-Free dynamic Range
Intermodulation Distortion
Out of Range Output Code
(In addition to OR Output high)
Conditions
Typical
(Note 8)
Limits
(Note 8)
Units
(Limits)
fIN = 100 MHz, VIN = FSR − 0.5 dB
47
fIN = 248 MHz, VIN = FSR − 0.5 dB
46.3
43.9
dB (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
46.3
43.9
dB (min)
dB
fIN = 100 MHz, VIN = FSR − 0.5 dB
48
fIN = 248 MHz, VIN = FSR − 0.5 dB
47.1
44.0
dB (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
47.1
44.0
dB (min)
fIN = 100 MHz, VIN = FSR − 0.5 dB
-55
fIN = 248 MHz, VIN = FSR − 0.5 dB
-55
-47.5
dB (max)
fIN = 498 MHz, VIN = FSR − 0.5 dB
-55
-47.5
dB (max)
fIN = 100 MHz, VIN = FSR − 0.5 dB
−60
dB
fIN = 248 MHz, VIN = FSR − 0.5 dB
−60
dB
fIN = 498 MHz, VIN = FSR − 0.5 dB
−60
dB
fIN = 100 MHz, VIN = FSR − 0.5 dB
−65
dB
fIN = 248 MHz, VIN = FSR − 0.5 dB
−65
dB
fIN = 498 MHz, VIN = FSR − 0.5 dB
−65
dB
fIN = 100 MHz, VIN = FSR − 0.5 dB
55
fIN = 248 MHz, VIN = FSR − 0.5 dB
55
47.5
dB (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
55
47.5
dB (min)
fIN1 = 321 MHz, VIN = FSR − 7 dB
fIN2 = 326 MHz, VIN = FSR − 7 dB
-50
dB
dB
dB
dB
(VIN+) − (VIN−) > + Full Scale
255
(VIN+) − (VIN−) < − Full Scale
0
INTERLEAVE MODE (DES Pin 127=Float) - DYNAMIC CONVERTER CHARACTERISTICS
FPBW
(DES)
Full Power Bandwidth
ENOB
Effective Number of Bits
SINAD
Signal to Noise Plus Distortion
Ratio
SNR
Signal to Noise Ratio
THD
2nd Harm
Total Harmonic Distortion
Second Harmonic Distortion
3rd Harm
Third Harmonic Distortion
SFDR
Spurious Free Dynamic Range
Dual Edge Sampling Mode
900
MHz
fIN = 248 MHz, VIN = FSR − 0.5 dB
7.3
6.8
Bits (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
7.3
6.8
Bits (min)
fIN = 248 MHz, VIN = FSR − 0.5 dB
46
42.5
dB (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
46
42.5
dB (min)
fIN = 248 MHz, VIN = FSR − 0.5 dB
46.4
43
dB (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
46.4
43
dB (min)
fIN = 248 MHz, VIN = FSR − 0.5 dB
-58
-49
dB (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
-58
-49
dB (min)
fIN = 248 MHz, VIN = FSR − 0.5 dB
-64
dB
fIN = 498 MHz, VIN = FSR − 0.5 dB
-64
dB
fIN = 248 MHz, VIN = FSR − 0.5 dB
-69
dB
fIN = 498 MHz, VIN = FSR − 0.5 dB
-69
dB
fIN = 248 MHz, VIN = FSR − 0.5 dB
57
47
dB (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
57
47
dB (min)
ANALOG INPUT AND REFERENCE CHARACTERISTICS
VIN
VCMI
FSR pin 14 Low
650
FSR pin 14 High
870
Full Scale Analog Differential
Input Range
Analog Input Common Mode
Voltage
VCMO
9
570
mVP-P (min)
730
mVP-P (max)
790
mVP-P (min)
950
mVP-P (max)
VCMO − 50
VCMO + 50
mV (min)
mV (max)
www.national.com
ADC08D1000
Symbol
ADC08D1000
Symbol
CIN
RIN
Parameter
Conditions
Differential
Analog Input Capacitance,
Normal operation (Notes 10, 11) Each input pin to ground
Analog Input Capacitance, DES
Mode (Notes 10, 11)
Typical
(Note 8)
Limits
(Note 8)
Units
(Limits)
0.02
pF
1.6
pF
Differential
0.08
pF
Each input pin to ground
2.2
pF
Differential Input Resistance
100
94
Ω (min)
106
Ω (max)
0.95
1.45
V (min)
V (max)
ANALOG OUTPUT CHARACTERISTICS
VCMO
Common Mode Output Voltage
ICMO = ±100 µA
1.26
VCMO_LVL
VCMO input threshold to set DC
Coupling mode
VA = 1.8V
0.60
V
VA = 2.0V
0.66
V
TC VCMO
Common Mode Output Voltage
Temperature Coefficient
TA = −40°C to +85°C
118
ppm/°C
CLOAD VCMO
Maximum VCMO load
Capacitance
VBG
Bandgap Reference Output
Voltage
IBG = ±100 µA
TC VBG
Bandgap Reference Voltage
Temperature Coefficient
TA = −40°C to +85°C,
IBG = ±100 µA
CLOAD VBG
Maximum Bandgap Reference
load Capacitance
1.26
80
pF
1.20
1.33
V (min)
V (max)
28
ppm/°C
80
pF
TEMPERATURE DIODE CHARACTERISTICS
ΔVBE
Temperature Diode Voltage
192 µA vs. 12 µA,
TJ = 25°C
71.23
mV
192 µA vs. 12 µA,
TJ = 85°C
85.54
mV
1
LSB
1
LSB
CHANNEL-TO-CHANNEL CHARACTERISTICS
Offset Match
Positive Full-Scale Match
Zero offset selected in Control Register
Negative Full-Scale Match
Zero offset selected in Control Register
Phase Matching (I, Q)
FIN = 1.0 GHz
1
LSB
<1
Degree
X-TALK
Crosstalk from I (Aggressor) to Q Aggressor = 867 MHz F.S.
(Victim) Channel
Victim = 100 MHz F.S.
-71
dB
X-TALK
Crosstalk from Q (Aggressor) to I Aggressor = 867 MHz F.S.
(Victim) Channel
Victim = 100 MHz F.S.
-71
dB
CLOCK INPUT CHARACTERISTICS
VID
Sine Wave Clock
0.6
0.4
2.0
VP-P (min)
VP-P (max)
Square Wave Clock
0.6
0.4
2.0
VP-P (min)
VP-P (max)
VIN = 0 or VIN = VA
±1
µA
Differential
0.02
pF
Each input to ground
1.5
pF
Differential Clock Input Level
II
Input Current
CIN
Input Capacitance (Notes 10, 11)
DIGITAL CONTROL PIN CHARACTERISTICS
VIH
Logic High Input Voltage
(Note 12)
VIL
Logic Low Input Voltage
(Note 12)
CIN
Input Capacitance (Notes 11, 13) Each input to ground
DIGITAL OUTPUT CHARACTERISTICS
www.national.com
10
1.2
0.85 x VA
V (min)
0.15 x VA
V (max)
pF
VOD
Conditions
Typical
(Note 8)
Measured differentially, OutV = VA, VBG
= Floating (Note 15)
710
Measured differentially, OutV = GND,
VBG = Floating (Note 15)
510
Parameter
LVDS Differential Output Voltage
Limits
(Note 8)
Units
(Limits)
400
mVP-P (min)
920
mVP-P (max)
280
mVP-P (min)
720
mVP-P (max)
Δ VO DIFF
Change in LVDS Output Swing
Between Logic Levels
±1
mV
VOS
Output Offset Voltage, see Figure
VBG = Floating
1
800
mV
VOS
Output Offset Voltage, see Figure
VBG = VA (Note 15)
1
1200
mV
Δ VOS
Output Offset Voltage Change
Between Logic Levels
±1
mV
IOS
Output Short Circuit Current
±4
mA
ZO
Differential Output Impedance
VOH
CalRun High level output
IOH = -400uA (Note 12)
1.65
1.5
V
VOL
CalRun Low level output
IOH = 400uA (Note 12)
0.15
0.3
V
Output+ & Output- connected to 0.8V
100
Ohms
POWER SUPPLY CHARACTERISTICS
IA
Analog Supply Current
PD = PDQ = Low
PD = Low, PDQ = High
PD = PDQ = High
660
430
1.8
765
508
mA (max)
mA (max)
mA
IDR
Output Driver Supply Current
PD = PDQ = Low
PD = Low, PDQ = High
PD = PDQ = High
200
112
0.012
275
157
mA (max)
mA (max)
mA
PD
Power Consumption
PD = PDQ = Low
PD = Low, PDQ = High
PD = PDQ = High
1.6
1.0
3.5
1.97
1.27
W (max)
W (max)
mW
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, 50mVP-P riding on VA
51
dB
AC ELECTRICAL CHARACTERISTICS
fCLK1
Maximum Input Clock Frequency Normal Mode (non DES) or DES Mode
1.3
fCLK2
Minimum Input Clock Frequency Normal Mode (non DES)
200
MHz
fCLK2
Minimum Input Clock Frequency DES Mode
500
MHz
1.0
GHz (min)
Input Clock Duty Cycle
200 MHz ≤ Input clock frequency ≤ 1
GHz (Normal Mode) (Note 12)
50
20
80
% (min)
% (max)
Input Clock Duty Cycle
500MHz ≤ Input clock frequency ≤ 1
GHz (DES Mode) (Note 12)
50
20
80
% (min)
% (max)
tCL
Input Clock Low Time
(Note 11)
500
200
ps (min)
tCH
Input Clock High Time
(Note 11)
500
200
ps (min)
45
55
% (min)
% (max)
DCLK Duty Cycle
(Note 11)
50
tRS
Reset Setup Time
(Note 11)
150
ps
tRH
Reset Hold Time
(Note 11)
250
ps
tSD
Synchronizing Edge to DCLK
Output Delay
tRPW
Reset Pulse Width
tLHT
Differential Low to High Transition
10% to 90%, CL = 2.5 pF
Time
tOD + tOSK
(Note 11)
4
11
250
Clock Cycles
(min)
ps
www.national.com
ADC08D1000
Symbol
ADC08D1000
Symbol
Parameter
Conditions
tHLT
Differential High to Low Transition
10% to 90%, CL = 2.5 pF
Time
tOSK
DCLK to Data Output Skew
tSU
tH
Typical
(Note 8)
Limits
(Note 8)
Units
(Limits)
250
ps
50% of DCLK transition to 50% of Data
transition, SDR Mode
and DDR Mode, 0° DCLK (Note 11)
±50
ps (max)
Data to DCLK Set-Up Time
DDR Mode, 90° DCLK (Note 11)
750
ns
DCLK to Data Hold Time
DDR Mode, 90° DCLK (Note 11)
890
ns
tAD
Sampling (Aperture) Delay
Input CLK+ Fall to Acquisition of Data
1.3
ns
tAJ
Aperture Jitter
0.4
ps rms
tOD
Input Clock to Data Output Delay 50% of Input Clock transition to 50% of
(in addition to Pipeline Delay)
Data transition
3.1
ns
Pipeline Delay (Latency)
(Notes 11, 14)
DI Outputs
13
DId Outputs
14
Normal Mode
DQ Outputs
tWU
13.5
Normal Mode
DQd Outputs
Over Range Recovery Time
13
DES Mode
14
DES Mode
Differential VIN step from ±1.2V to 0V to
get accurate conversion
PD low to Rated Accuracy
Conversion (Wake-Up Time)
Input Clock
Cycles
14.5
1
Input Clock
Cycle
500
ns
DCS
(Note 11)
1
μs
fSCLK
Serial Clock Frequency
(Note 11)
100
MHz
tSSU
Data to Serial Clock Setup Time (Note 11)
2.5
ns (min)
tSH
Data to Serial Clock Hold Time
(Note 11)
1
Serial Clock Low Time
ns (min)
ns (min)
4
Serial Clock High Time
ns (min)
4
tCAL
Calibration Cycle Time
tCAL_L
CAL Pin Low Time
See Figure 9 (Note 11)
80
Clock Cycles
(min)
tCAL_H
CAL Pin High Time
See Figure 9 (Note 11)
80
Clock Cycles
(min)
CalDly = Low
See 1.1.2 Acquiring the Input, Figure 9,
(Note 11)
225
Clock Cycles
(min)
CalDly = High
See1.1.1 Self-Calibration, Figure 9,
(Note 11)
231
Clock Cycles
(max)
tCalDly
Calibration delay determined by
pin 127
1.4 x 105
Clock 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. Machine model is 220 pF discharged through ZERO Ohms.
Note 5: See AN-450, “Surface Mounting Methods and Their Effect on Product Reliability”.
Note 6: The analog inputs are protected as shown below. Input voltage magnitudes beyond the Absolute Maximum Ratings may damage this device.
www.national.com
12
ADC08D1000
20097404
Note 7: 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 8: Typical figures are at TA = 25°C, and represent most likely parametric norms. Test limits are guaranteed to National's AOQL (Average Outgoing Quality
Level).
Note 9: 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 10: 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 11: This parameter is guaranteed by design and is not tested in production.
Note 12: This parameter is guaranteed by design and/or characterization and is not tested in production.
Note 13: 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 14: Each of the two converters of the ADC08D1000 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).
Note 15: Tying VBG to the supply rail will increase the output offset voltage (VOS) by 400mv (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 40mV (typical).
13
www.national.com
ADC08D1000
Specification Definitions
APERTURE (SAMPLING) DELAY is that time required after
the fall of the clock input for the sampling switch to open. The
Sample/Hold circuit effectively stops capturing the input signal and goes into the “hold” mode the aperture delay time
(tAD) after the input clock goes low.
APERTURE JITTER (tAJ) is the variation in aperture delay
from sample to sample. Aperture jitter shows up as input
noise.
Bit Error Rate (B.E.R.) is the probability of error and is defined as the probable number of errors per unit of time divided
by the number of bits seen in that amount of time. A B.E.R. of
10-18 corresponds to a statistical error in one bit about every
four (4) years.
CLOCK DUTY CYCLE is the ratio of the time that the clock
wave form is at a logic high to the total time of one clock period.
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:
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 step. 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
20097446
FIGURE 1.
LVDS OUTPUT OFFSET VOLTAGE (VOS) is the midpoint
between the D+ and D- pins output voltage; i.e., [(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 last code transition is from the ideal 1/2 LSB above
a differential - VIN/2. For the ADC08D1000 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.
OUTPUT DELAY (tOD) is the time delay (in addition to
Pipeline Delay) after the falling edge of DCLK 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 ADC08D1000 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.
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 ADC08D1000.
LVDS DIFFERENTIAL OUTPUT VOLTAGE (VOD) is the absolute value of the difference between the VD+ & VD- outputs;
each measured with respect to Ground.
www.national.com
14
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.
Transfer Characteristic
20097422
FIGURE 2. Input / Output Transfer Characteristic
15
www.national.com
ADC08D1000
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
ADC08D1000
Timing Diagrams
20097414
FIGURE 3. ADC08D1000 Timing — SDR Clocking
20097459
FIGURE 4. ADC08D1000 Timing — DDR Clocking
www.national.com
16
ADC08D1000
20097419
FIGURE 5. Serial Interface Timing
20097420
FIGURE 6. Clock Reset Timing in DDR Mode
20097423
FIGURE 7. Clock Reset Timing in SDR Mode with OUTEDGE Low
17
www.national.com
ADC08D1000
20097424
FIGURE 8. Clock Reset Timing in SDR Mode with OUTEDGE High
20097425
FIGURE 9. Self Calibration and On-Command Calibration Timing
www.national.com
18
VA=VDR=1.9V, FCLK=1000MHz, TA=25°C unless otherwise stated.
INL vs. CODE
INL vs. TEMPERATURE
20097464
20097465
DNL vs. CODE
DNL vs. TEMPERATURE
20097466
20097467
POWER DISSIPATION vs. SAMPLE RATE
ENOB vs. CLOCK DUTY CYCLE
20097481
20097480
19
www.national.com
ADC08D1000
Typical Performance Characteristics
ADC08D1000
ENOB vs. TEMPERATURE
ENOB vs. SUPPLY VOLTAGE
20097476
20097477
ENOB vs. SAMPLE RATE
ENOB vs. INPUT FREQUENCY
20097478
20097479
SNR vs. TEMPERATURE
SNR vs. SUPPLY VOLTAGE
20097468
www.national.com
20097469
20
ADC08D1000
SNR vs. SAMPLE RATE
SNR vs. INPUT FREQUENCY
20097470
20097471
THD vs. TEMPERATURE
THD vs. SUPPLY VOLTAGE
20097472
20097473
THD vs. SAMPLE RATE
THD vs. INPUT FREQUENCY
20097474
20097475
21
www.national.com
ADC08D1000
SFDR vs. TEMPERATURE
SFDR vs. SUPPLY VOLTAGE
20097485
20097484
SFDR vs. SAMPLE RATE
SFDR vs. INPUT FREQUENCY
20097482
20097483
Spectral Response at FIN = 248 MHz
Spectral Response at FIN = 498 MHz
20097487
www.national.com
20097488
22
ADC08D1000
CROSSTALK vs. SOURCE FREQUENCY
FULL POWER BANDWIDTH
20097463
20097486
STEP RESPONSE
STEP RESPONSE DETAIL VIEW
20097461
20097462
23
www.national.com
ADC08D1000
During the calibration process, the input termination resistor
is trimmed to a value that is equal to REXT / 33. 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.
In normal operation, calibration is performed just after application of power and whenever a valid calibration command is
given, which is holding the CAL pin low for at least tCAL_L clock
cycles, then hold it high for at least another tCAL_H clock cycles
as defined in the Converter Electrical Characteristics. The
time taken by the calibration procedure is specified as tCALin
Converter Electrical Characteristics. Holding the CAL pin high
upon power up will prevent the calibration process from running until the CAL pin experiences the above-mentioned
tCAL_L clock cycles followed by tCAL_H clock cycles.
CalDly (pin 127) is used to select one of two delay times after
the application of power to the start of calibration. This calibration delay time is depedent on the setting of the CalDly pin
and is specified as tCalDly in the Converter Electrical Characteristics. These delay values allow the power supply to come
up and stabilize before calibration takes place. If the PD pin
is high upon power-up, the calibration delay counter will be
disabled until the PD pin is brought low. Therefore, holding
the PD pin high during power up will further delay the start of
the power-up calibration cycle. The best setting of the CalDly
pin depends upon the power-on settling time of the power
supply.
1.0 Functional Description
The ADC08D1000 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 generally poor practice to allow an active pin to float,
pins 4, 14 and 127 of the ADC08D1000 are designed to be
left floating without jeopardy. 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 will have the same effect as allowing it to
float.
1.1 OVERVIEW
The ADC08D1000 uses a calibrated folding and interpolating
architecture that achieves 7.5 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 1.3 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" 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 of the two converters has a 1:2 demultiplexer that feeds
two LVDS output buses. The data on these buses provide an
output word rate on each bus at half the ADC sampling rate
and must be interleaved by the user to provide output words
at the full conversion rate.
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 power-up calibration and during the oncommand 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 2.4.2 Self
Calibration for information on maintaining DCLK operation
during on-command 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 Self-Calibration
A self-calibration is performed upon power-up and can also
be invoked by the user upon command. Calibration trims the
100Ω analog input differential termination resistor and minimizes full-scale error, offset error, DNL and INL, resulting in
maximizing SNR, THD, SINAD (SNDR) and ENOB. Internal
bias currents are also set with the calibration process. All of
this is true whether the calibration is performed upon power
up or is performed upon command. Running the self calibration is an important part of this chip's functionality and is
required in order to obtain adequate performance. In addition
to the requirement to be run at power-up, self calibration must
be re-run whenever the sense of the FSR pin is changed. For
best performance, we recommend that self calibration be run
20 seconds or more after application of power and whenever
the operating temperature changes significantly, according to
the system design performance specifications. See 2.4.2.2
On-Command Calibration for more information. Calibration
can not be initiated or run while the device is in the powerdown mode. See 1.1.7 Power Down for information on the
interaction between Power Down and Calibration.
www.national.com
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 ADC08D1000 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 1.0 GHz. The ADC08D1000 output data signaling is LVDS and the output format is offset binary.
24
1.1.5.1 Dual-Edge Sampling
The DES mode allows one of the ADC08D1000'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 2 GSPS with a 1 GHz input clock.
In this mode the outputs are interleaved such that the data is
effectively demultiplexed 1:4. Since the sample rate is doubled, each of the 4 output buses have a 500 MSPS output rate
with a 1 GHz input clock. All data is available in parallel. The
four bytes of parallel data that are output with each clock is in
the following sampling order, from the earliest to the latest:
DQd, DId, DQ, DI. Table 1 indicates what the outputs represent for the various sampling possibilities.
In the non-extended mode of operation only the "I" input can
be sampled in the DES mode. In the extended mode of operation the user can select which input is sampled.
The ADC08D1000 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 On-Command Calibration which should be run before entering DES
mode, or if a large swing in ambient temperature is experienced by the device.
1.1.4 The Analog Inputs
The ADC08D1000 must be driven with a differential input signal. Operation with a single-ended signal is not recommended. It is important that the inputs either be a.c. coupled to the
inputs with the VCMO pin grounded, or d.c. coupled with the
VCMO pin left floating. An input common mode voltage equal
to the VCMO output must be provided when d.c. coupling is
used.
Two full-scale range settings are provided with pin 14 (FSR).
The input full-scale range is programmable in the normal
mode by setting a level on pin 14 (FSR) as defined in by the
specification VIN in the Converter Electrical Characteristics.
The full-scale range setting operates equally on both ADCs.
In the Extended Control mode, programming the Input FullScale Voltage Adjust register allows the input full-scale range
to be adjusted as described in ??1.4 and ??2.2.
1.1.5 Clocking
The ADC08D1000 must be driven with an a.c. coupled, differential clock signal. Section 2.3 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 ADC08D1000 offers two options for input and output
clocking. These options include a choice of Dual Edge Sampling (DES) or "interleaved mode" where the ADC08D1000
TABLE 1. Input Channel Samples Produced at Data Outputs
Data Outputs (Always
sourced with respect to fall
of DCLK)
Dual-Edge Sampling Mode (DES)
Normal Sampling Mode
I-Channel Selected
Q-Channel Selected *
DI
"I" Input Sampled with Fall of "I" Input Sampled with Fall of "Q" Input Sampled with Fall of
CLK 13 cycles earlier.
CLK 13 cycles earlier.
CLK 13 cycles earlier.
DId
"I" Input Sampled with Fall of "I" Input Sampled with Fall of "Q" Input Sampled with Fall of
CLK 14 cycles earlier.
CLK 14 cycles earlier.
CLK 14 cycles earlier.
DQ
"Q" Input Sampled with Fall of "I" Input Sampled with Rise of "Q" Input Sampled with Rise
CLK 13 cycles earlier.
CLK 13.5 cycles earlier.
of CLK 13.5 cycles earlier.
DQd
"Q" Input Sampled with Fall of
"I" Input Sampled with Rise of "Q" Input Sampled with Rise
CLK 14 cycles after being
CLK 14.5 cycles earlier.
of CLK 14.5 cycles earlier.
sampled.
* Note: In DES + normal mode, only the I Channel is sampled. In DES + extended control mode, I or Q channel can be
sampled.
25
www.national.com
ADC08D1000
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 ADC08D1000 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 80 / 20 % (worst case) for
both the normal and the Dual Edge Sampling modes.
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 ADC08D1000 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 normal 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 registerbased control and those pin-based controls are disabled.
These pins are OutV (pin 3), OutEdge/DDR (pin 4), FSR (pin
14) and CalDly/DES (pin 127). See 1.2 NORMAL/EXTENDED CONTROL for details on the Extended Control mode.
ADC08D1000
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 tri-stated, they
are weakly pulled down to ground internally. Therefore when
both I and Q 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 already high, the device will not begin the calibration sequence until the PD input goes low. If a manual calibration is
requested while the device is powered down, the calibration
will not begin at all. That is, the manual calibration input is
completely ignored in the power down state. 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.
1.1.5.2 OutEdge 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). 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 2.4.3 Output Edge
Synchronization.
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
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 allowing pin 4 to float.
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
power consumption. If the LVDS lines are long and/or the
system in which the ADC08D1000 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 800mV when the VBG pin is unconnected and floating. This
common mode voltage can be increased to 1.2V 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.
1.2 NORMAL/EXTENDED CONTROL
The ADC08D1000 may be operated in one of two modes. In
the simpler standard 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. The two control modes are selected with
pin 14 (FSR/ECE: Extended Control Enable). 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 2 shows how several of the device features are affected
by the control mode chosen.
1.1.7 Power Down
The ADC08D1000 is in the active state when the Power Down
pin (PD) is low. When the PD pin is high, the device is in the
www.national.com
26
Feature
Normal Control Mode
Extended Control Mode
SDR or DDR Clocking
Selected with nDE in the Configuration
DDR Clocking selected with pin 4
Register (1h; bit-10). When the device is
floating. SDR clocking selected when pin
in DDR mode, address 1h, bit-8 must be
4 not floating.
set to 0b.
DDR Clock Phase
Not Selectable (0° Phase Only)
Selected with DCP bit in the
Configuration Register (1h; bit-11).
SDR Data transitions with rising edge of
SDR Data transitions with rising or
Selected with OE in the Configuration
DCLK+ when pin 4 is high and on falling
falling DCLK edge
Register (1h; bit-8).
edge when low.
LVDS output level
Normal differential data and DCLK
amplitude selected when pin 3 is high
and reduced amplitude selected when
low.
Power-On Calibration Delay
Short delay selected when pin 127 is low
Short delay only.
and longer delay selected when high
Full-Scale Range
Options (650 mVP-P or 870 mVP-P)
selected with pin 14. Selected range
applies to both channels.
Up to 512 step adjustments over a
nominal range specified in ??1.4.
Selected using the Input Full-Scale
Adjust register (3h; bits-7 thru 15).
Input Offset Adjust
Not possible
512 steps of adjustment using the Input
Offset register (2h; bits-7 thru 15) as
specified in ??1.4
Dual Edge Sampling Selection
Enabled with pin 127
Enabled through DES Enable Register.
Dual Edge Sampling Input
Channel Selection
Only I-Channel Input can be used
Either I- or Q-Channel input may be
sampled by both ADCs.
The Clock Phase is adjusted
automatically
Automatic Clock Phase control can be
selected by setting bit 14 in the DES
Enable register (Dh). The clock phase
can also be adjusted manually through
the Coarse & Fine registers (Eh and Fh).
DES Sampling Clock Adjustment
The default state of the Extended Control Mode is set upon
power-on reset (internally performed by the device) and is
shown in Table 3.
1.3 THE SERIAL INTERFACE
IMPORTANT NOTE: During the initial write using the serial
interface, all 8 user registers must be written with desired or
default values. In addition, the first write to the DES Enable
register (Dh) must load the default value (0x3FFFh). Once all
registers have been written once, other desired settings, including enabling DES can be loaded.
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) Eight 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 with the rising edge of
this signal. There is no minimum frequency requirement for
SCLK.
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
5 of the Timing Diagrams. The fixed header pattern is 0000
0000 0001 (eleven zeros followed by a 1). The loading se-
TABLE 3. Extended Control Mode Operation
(Pin 14 Floating)
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
(710 mVP-P)
Calibration Delay
Short Delay
Full-Scale Range
700 mV nominal for both
channels
Input Offset Adjust
No adjustment for either
channel
Dual Edge Sampling (DES)
Not enabled
Selected with the OV in the
Configuration Register (1h; bit-9).
27
www.national.com
ADC08D1000
TABLE 2. Features and modes
ADC08D1000
quence 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 4.
Refer to the Register Description (Section 1.4) 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: The Serial Interface should not be used
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.
IMPORTANT: The Configuration Register should not be
written if the DES Enable bit = 1. The DES Enable bit
should first be changed to 0, then the Configuration
Register can be written. Failure to follow this procedure
can cause the internal DES clock generation circuitry to
stop.
Bit 15
Must be set to 1b
Bit 14
Must be set to 0b
Bit 13
Must be set to 1b
Bit 12
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.
POR State: 1b
Bit 11
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.
POR State: 0b
Bit 10
nDE: DDR Enable. When this bit is set to 0b,
data bus clocking follows the DDR (Dual 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.
POR State: 0b
Bit 9
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 normal control mode. When this bit is set
to 1b, the standard output amplitude of 710
mVP-P is used. When this bit is set to 0b, the
reduced output amplitude of 510 mVP-P is
used.
POR State: 1b
Bit 8
OE: Output Edge. 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 normal control mode.
When this bit is 1, the data outputs change with
the rising edge of DCLK+. When this bit is 0,
the data output change with the falling edge of
DCLK+.
POR State: 0b
Bits 7:0
Must be set to 1b.
TABLE 4. Register Addresses
4-Bit Address
Loading Sequence:
A3 loaded after Fixed Header Pattern, A0 loaded last
A3
A2
A1
A0
Hex
Register Addressed
0
0
0
0
0h
Reserved
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
Reserved
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
DES Enable
1
1
1
0
Eh
DES Coarse Adjust
1
1
1
1
Fh
DES Fine Adjust
1.4 REGISTER DESCRIPTION
Eight 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 Normal Control
Mode. Each register description below also shows the PowerOn Reset (POR) state of each control bit.
Configuration Register
Addr: 1h (0001b)
D15
D14
D13
1
0
1
W only (0xB2FF)
D11
D10
D9
D8
DCS DCP
D12
nDE
OV
OE
I-Channel Offset
Addr: 2h (0010b)
D7
D6
D5
D4
D3
D2
D1
D0
1
1
1
1
1
1
1
1
D15
(MSB)
www.national.com
28
D14
D13
W only (0x007F)
D12
D11
Offset Value
D10
D9
D8
(LSB)
D6
D5
D4
D3
D2
D1
D0
Sign
1
1
1
1
1
1
1
Bits 15:8
Bit 15:8
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.
POR State: 0000 0000 b
Sign bit. 0b gives positive offset, 1b gives
negative offset.
POR State: 0b
Must be set to 1b
Bit 7
Bit 6:0
Bit 7
Bit 6:0
Q-Channel Full-Scale Voltage Adjust
Addr: Bh (1011b)
I-Channel Full-Scale Voltage Adjust
Addr: 3h (0011b)
D15
D14
D15
W only (0x807F)
D13
D12
(MSB)
D11
D10
D9
D7
D6
D5
D4
D3
D2
D1
D0
(LSB)
1
1
1
1
1
1
1
Bits 6:0
560mVP-P
1000 0000 0
Default Value
1111 1111 1
700mVP-P
D15
D14
D13
D12
D11
D10
D9
D10
D9
D8
Adjust Value
D5
D4
D3
D2
D1
D0
1
1
1
1
1
1
1
Bits 6:0
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
560mVP-P
1000 0000 0
700mVP-P
1111 1111 1
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.
POR State: 1000 0000 0b (no adjustment)
Must be set to 1b
DES Enable
Addr: Dh (1101b)
W only (0x3FFF)
D15
D14
D13
D12
D11
D10
D9
D8
DEN
ACP
1
1
1
1
1
1
D8
D7
D6
D5
D4
D3
D2
D1
D0
(LSB)
1
1
1
1
1
1
1
1
W only (0x007F)
Offset Value
D11
D6
840mVP-P
Q-Channel Offset
(MSB)
D12
D7
Bit 15:7
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.
POR State: 1000 0000 0b (no adjustment)
Must be set to 1b
Addr: Ah (1010b)
D13
(LSB)
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
D14
W only (0x807F)
(MSB)
D8
Adjust Value
Bit 15:7
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.
POR State: 0000 0000 b
Sign bit. 0b gives positive offset, 1b gives
negative offset.
POR State: 0b
Must be set to 1b
D7
D6
D5
D4
D3
D2
D1
D0
Sign
1
1
1
1
1
1
1
29
www.national.com
ADC08D1000
D7
ADC08D1000
Bit 15
Bit 14
Bits 13:0
DES Coarse Adjust
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 timeinterleaved manner, accomplishing a
sampling rate of twice the input clock rate.
When this bit is set to 0b, the device operates
in the normal dual channel mode.
POR State: 0b
Automatic Clock Phase Control. (ACP) Setting
this bit to 1b enables the Automatic Clock
Phase Control. In this mode the DES Coarse
and Fine manual controls are disabled. A
phase detection circuit continually adjusts the
I and Q sampling edges to be 180 degrees out
of phase. When this bit is set to 0b, the sample
(input) clock delay between the I and Q
channels is set manually using the DES
Coarse and Fine Adjust registers. (See 2.4.5
Dual Edge Sampling for important application
information) Using the ACP Control option
is recommended over the manual DES
settings.
POR State: 0b
Must be set to 1b
Addr: Eh (1110b)
W only (0x07FF)
D15
D14
IS
ADS
D13
D12
D7
D6
D5
D4
1
1
1
1
D11
D10
D9
D8
1
1
1
D3
D2
D1
D0
1
1
1
1
CAM
Bit 15
Input Select. When this bit is set to 0b the "I"
input is operated upon by both ADCs. When
this bit is set to 1b the "Q" input is operated on
by both ADCs.
POR State: 0b
Bit 14
Adjust Direction Select. When this bit is set to
0b, the programmed delays are applied to the
"I" channel sample clock while the "Q" channel
sample clock remains fixed. When this bit is
set to 1b, the programmed delays are applied
to the "Q" channel sample clock while the "I"
channel sample clock remains fixed.
POR State: 0b
Bits 13:11 Coarse Adjust Magnitude. Each code value in
this field delays either the "I" channel or the "Q"
channel sample clock (as determined by the
ADS bit) by approximately 20 picoseconds. A
value of 000b in this field causes zero
adjustment.
POR State: 000b
Bits 10:0 Must be set to 1b
DES Fine Adjust
Addr: Fh (1111b)
D15
D14
D13
W only (0x007F)
D12
D11
(MSB)
D10
D9
D8
FAM
D7
D6
D5
D4
D3
D2
D1
D0
(LSB)
1
1
1
1
1
1
1
Bits 15:7
Bit 6:0
Fine Adjust Magnitude. Each code value in this
field delays either the "I" channel or the "Q"
channel sample clock (as determined by the
ADS bit of the DES Coarse Adjust Register) by
approximately 0.1 ps. A value of 0000 0000 0b
in this field causes zero adjustment. Note that
the amount of adjustment achieved with each
code will vary with the device conditions as
well as with the Coarse Adjustment value
chosen.
POR State: 0000 0000 0b
Must be set to 1b
1.4.1 Note Regarding Extended Mode Offset Correction
When using the I 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.
www.national.com
30
20097430
FIGURE 10. Extended Mode Offset Behavior
2.2 THE ANALOG INPUT
The analog input is a differential one to which the signal
source may be a.c. coupled or d.c. coupled. In the normal
mode, the full-scale input range is selected using 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. for guidelines on limiting the amount of adjustment.
Table 5 gives the input to output relationship with the FSR pin
high and the normal (non-extended) mode is used. With the
FSR pin grounded, the millivolt values in Table 5 are reduced
to 75% of the values indicated. In the Enhanced Control
Mode, these values will be determined by the full scale range
and offset settings in the Control Registers.
1.5 MULTIPLE ADC SYNCHRONIZATION
The ADC08D1000 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 all the ADCs use for sampling.
The DCLK_RST signal must observe some timing requirements that are shown in Figure 6, Figure 7 and Figure 8 of the
Timing Diagrams. The DCLK_RST pulse must be of a minimum width and its desertion edge must observe setup and
hold times with respect to the CLK input rising edge. These
times are specified in the AC 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 6, Figure 7 and Figure 8 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 next CLK falling edge synchronizes the
DCLK output with those of other ADC08D1000s 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.
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.
TABLE 5. DIFFERENTIAL INPUT TO OUTPUT
RELATIONSHIP
(Non-Extended Control Mode, FSR High)
VIN+
VIN−
Output Code
VCM − 217.5mV
VCM + 217.5mV
0000 0000
VCM − 109 mV
VCM + 109 mV
0100 0000
VCM
VCM
0111 1111 /
1000 0000
VCM + 109 mV
VCM −109 mV
1100 30000
VCM + 217.5mV
VCM − 217.5mV
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.
Note that a precise d.c. common mode voltage must be
present at the ADC inputs. This common mode voltage,
VCMO, is provided on-chip when a.c. input coupling is used
and the input signal is a.c. coupled to the ADC.
When the inputs are a.c. coupled, the VCMO output must be
grounded, as shown in Figure 11. This causes the on-chip
VCMO voltage to be connected to the inputs through on-chip
50k-Ohm resistors.
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.
2.0 Applications Information
2.1 THE REFERENCE VOLTAGE
The voltage reference for the ADC08D1000 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 than this is required.
The internal bandgap-derived reference voltage has a nominal value VIN, as determined by the FSR pin and described in
Section 1.1.4.
There is no provision for the use of an external reference voltage, but the full-scale input voltage can be adjusted through
31
www.national.com
ADC08D1000
a Configuration Register in the Extended Control mode, as
explained in Section 1.2.
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 2.2.2 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 (VOS) is typically 800mV when the VBG
pin is used as an output or left unconnected. To raise the
LVDS offset voltage to a typical value of 1200mV the VBG pin
can be connected directly to the supply rails.
ADC08D1000
The Input impedance in the d.c. coupled mode (VCMO pin not
grounded) consists of a precision 100Ω resistor between VIN
+ and VIN− and a capacitance from each of these inputs to
ground. In the a.c. coupled mode the input appears the same
except there is also a resistor of 50K between each analog
input pin and the VCMO potential.
Driving the inputs beyond full scale will result in a saturation
or clipping of the reconstructed output.
2.2.1 Handling Single-Ended Input Signals
There is no provision for the ADC08D1000 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.
20097444
FIGURE 11. Differential Input Drive
When the d.c. coupled mode is used, a common mode voltage must be provided at the differential inputs. This common
mode voltage should track the VCMO output pin. Note that the
VCMO output potential will change with temperature. The common mode output of the driving device should track this
change.
IMPORTANT NOTE: An analog input channel that is not used
(e.g. in DES Mode) should be tied to the VCMO voltage when
the inputs are d.c. coupled. Do not connect unused analog
inputs to ground.
Full-scale distortion performance falls off rapidly as the
input common mode voltage deviates from VCMO. This is
a direct result of using a very low supply voltage to minimize power. Keep the input common voltage within 50
mV of VCMO.
Performance is as good in the d.c. coupled mode as it is
in the a.c. coupled mode, provided the input common
mode voltage at both analog inputs remain within 50 mV
of VCMO.
If d.c. coupling is used, it is best to servo the input common
mode voltage, using the VCMO pin, to maintain optimum performance. An example of this type of circuit is shown in Figure
12.
2.2.1.1 a.c. Coupled Input
The easiest way to accomplish single-ended a.c. Input to differential a.c. signal is with an appropriate balun-connected
transformer, as shown in ?? F13.
20097443
FIGURE 13. Single-Ended to Differential Signal
Conversion Using a Balun
The 100 Ohm external resistor placed across the output terminals of the balun in parallel with the ADC08D1000's on-chip
100 Ohm resistor makes a 50 Ohms differential impedance
at the balun output. Or, 25 Ohms to virtual ground at each of
the balun output terminals.
Looking into the balun, the source sees the impedance of the
first coil in series with the impedance at the output of that coil.
Since the transformer has a 1:1 turns ratio, the impedance
across the first coil is exactly the same as that at the output
of the second coil, namely 25 Ohms to virtual ground. So, the
25 Ohms across the first coil in series with the 25 Ohms at its
output gives 50 Ohms total impedance to match the source.
2.2.2 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.
20097455
FIGURE 12. Example of Servoing the Analog Input with
VCMO
2.2.3 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 ADC08D1000 is derived from an internal band-gap
reference. The FSR pin controls the effective reference voltage of the ADC08D1500 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, but better distortion and SFDR are obtained with the FSR pin low.
One such circuit should be used in front of the VIN+ input and
another in front of the VIN− input. In that figure, RD1, RD2 and
RD3 are used to divide the VCMO potential so that, after being
gained up by the amplifier, the input common mode voltage
is equal to VCMO from the ADC. RD1 and RD2 are split to allow
the bypass capacitor to isolate the input signal from VCMO.
RIN, RD2 and RD3 will divide the input signal, if necessary. Capacitor "C" in Figure 12 should be chosen to keep any component of the input signal from affecting VCMO.
Be sure that the current drawn from the VCMO output does not
exceed 100 μA.
www.national.com
32
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, to the ADC
analog input.
Note that the maximum jitter described above is the arithmetic
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 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.
2.3 THE CLOCK INPUTS
The ADC08D1000 has differential LVDS clock inputs, CLK+
and CLK-, which must be driven with an a.c. coupled, differential clock signal. Although the ADC08D1000 is tested and
its performance is guaranteed with a differential 1.0 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 capacitively 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 2.6.2 Thermal
Management.
2.4 CONTROL PINS
Six control pins (without the use of the serial interface) provide
a wide range of possibilities in the operation of the ADC08D1000 and facilitate its use. These control pins provide
Full-Scale Input Range setting, Self Calibration, Calibration
Delay, Output Edge Synchronization choice, LVDS Output
Level choice and a Power Down feature.
2.4.1 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 is
specified as VIN in the Converter Electrical Characteristics. In
the extended control mode, the input full-scale range may be
programmed using the Full-Scale Adjust Voltage register.
See 2.2 THE ANALOG INPUT for more information.
20097447
2.4.2 Self Calibration
The ADC08D1000 self-calibration must be run to achieve
specified performance. The calibration procedure is run upon
power-up and can be run any time on command. 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 DCLK outputs are not active during a calibration cycle, therefore it is not
recommended as a system clock.
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 (100Ω) characteristic
impedance. The input clock line should be as short and as
direct as possible. The ADC08D1000 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 ADC08D1000 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)
as specified in the Converter Electrical Characteristics.
High speed, high performance ADCs such as the ADC08D1000 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
2.4.2.1 Power-On Calibration
Power-on calibration begins after a time delay following the
application of power. This time delay is determined by the
setting of CalDly, as described in the Calibration Delay Section, below.
The calibration process will be not be performed if the CAL
pin is high at power up. In this case, the calibration cycle will
not begin until the on-command calibration conditions are
met. The ADC08D1000 will function with the CAL pin held
high at power up, but no calibration will be done and performance will be impaired. A manual calibration, however, may
be performed after powering up with the CAL pin high. See
2.4.2.2 On-Command Calibration.
The internal power-on calibration circuitry comes up in an unknown logic state. If the input clock is not running at power up
and the power on calibration circuitry is active, it will hold the
analog circuitry in power down and the power consumption
will typically be less than 200 mW. The power consumption
will be normal after the clock starts.
tJ(MAX) = (VINFSR / VIN(P-P)) x (1/(2(N+1) x π x fIN))
33
www.national.com
ADC08D1000
The LMH6555 of Figure 12 is suitable for any Full Scale
Range.
ADC08D1000
2.4.2.2 On-Command Calibration
On-command calibration may be run at any time in NORMAL
(non-DES) mode only. Do not run a calibration while operating the ADC in Auto DES Mode.
If the ADC is operating in Auto DES mode and a calibration
cycle is required then the controlling application should bring
the ADC into normal (non DES) mode before an On Command calibration is initiated. Once calibration has completed,
the ADC can be put back into Auto DES mode.
To initiate an on-command calibration, bring the CAL pin high
for a minimum of tCAL_H input clock cycles after it has been
low for a minimum of tCAL_L input clock cycles. Holding the
CAL pin high upon power up will prevent execution of poweron calibration until the CAL pin is low for a minimum of
tCAL_L input clock cycles, then brought high for a minimum of
another tCAL_H input clock cycles. The calibration cycle will
begin tCAL_H input clock cycles after the CAL pin is thus
brought high. The CalRun signal should be monitored to determine when the calibration cycle has completed.
The minimum tCAL_H and tCAL_L 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
Section 1.1.1 for best performance, a self calibration should
be performed 20 seconds or more after power up and repeated when the operating temperature changes significantly,
according to the particular system performance requirements.
ENOB drops slightly with increasing junction temperature,
and a self calibration eliminates the change. In the first example, (see Figure 15) a sample clock of 1GSPS is used to
capture a full-scale 749MHz signal at the I-channel input as
the junction temperature (TJ) is increased from 65°C to 125°
C with no intermediate calibration cycles. The vertical line at
125°C is the result of an on-command calibration cycle that
essentially eliminates the drop in ENOB. Of course, calibration cycles can be run more often, at smaller intervals of
temperature change, if system design specifications require
it. In the second example, (see Figure 16) the test method is
the same and the I-channel input signal is 249MHz. The variation in ENOB vs. TJ has a smaller range then the previous
example, and is again removed by an on-command calibration cycle at the maximum test temperature.
20097458
FIGURE 16. ENOB vs. Junction Temperature, 249 MHz
Input
2.4.2.3 Calibration Delay
The CalDly input (pin 127) is used to select one of two delay
times after the application of power to the start of calibration,
as described in Section 1.1.1. The calibration delay values
allow the power supply to come up and stabilize before calibration takes place. With no delay or insufficient delay, calibration would begin before the power supply is stabilized at
its operating value and result in non-optimal calibration coefficients. If the PD pin is high upon power-up, the calibration
delay counter will be disabled until the PD pin is brought low.
Therefore, holding the PD pin high during power up will further
delay the start of the power-up calibration cycle. The best
setting of the CalDly pin depends upon the power-on settling
time of the power supply.
Note that the calibration delay selection is not possible in the
Extended Control mode and the short delay time is used.
2.4.3 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 ADC08D1000 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.
2.4.4 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
20097457
FIGURE 15. ENOB vs. Junction Temperature, 749 MHz
Input
www.national.com
34
2.5 THE DIGITAL OUTPUTS
The ADC08D1000 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 ADC08D1000 input clock rate and the two buses must be multiplexed to obtain the entire 1 GSPS conversion result.
Since the minimum recommended input clock rate for this
device is 200 MSPS (normal 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 Section 2.4.3.
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 2.4.4 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.
2.4.5 Dual Edge Sampling
IMPORTANT NOTE: When using the ADC in Extended Control Mode, the Configuration Register must only be written
when the DES Enable bit = 0. Writing to the Configuration
Register when the DES Enable bit = 1 can cause the internal
DES clock generation circuitry to stop.
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 one
input clock edge (duty cycle corrected), the other samples the
input signal on the other input clock edge (duty cycle corrected). The result is a 1:4 demultiplexed output with a sample
rate that is twice the input clock frequency.
To use this feature in the non-enhanced control mode, allow
pin 127 to float and the signal at the "I" channel input will be
sampled by both converters. The Calibration Delay will then
only be a short delay.
In the enhanced control mode, either input may be used for
dual edge sampling. See 1.1.5.1 Dual-Edge Sampling.
IMPORTANT NOTES :
1) For the Extended Control Mode - When using the Automatic Clock Phase Control feature in dual edge sampling
mode, it is important that the automatic phase control is disabled (set bit 14 of DES Enable register Dh to 0) before the
ADC is powered up. Not doing so may cause the device not
to wake up from the power down state.
2) For the Non-Extended Control Mode - When the ADC08D1000 is powered up and DES mode is required, ensure
that pin 127 (CalDly/DES/SCS) is initially pulled low during or
after the power up sequence. The pin can then be allowed to
float or be tied to V A / 2 to enter the DES mode. This will ensure that the part enters the DES mode correctly.
3) The automatic phase control should also be disabled if the
input clock is interrupted or stopped for any reason. This is
also the case if a large abrupt change in the clock frequency
occurs.
4) If a calibration of the ADC is required in Auto DES mode,
the device must be returned to the Normal Mode of operation
before performing a calibration cycle. Once the Calibration
has been completed, the device can be returned to the Auto
DES mode and operation can resume.
2.6 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
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 ADC08D1000 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 ADC08D1000. The
ADC supplies should be the same supply used for other analog circuitry, if not a dedicated supply.
2.4.6 Power Down Feature
The Power Down pins (PD and PDQ) allow the ADC08D1000
to be entirely powered down (PD) or the "Q" 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 already high, the device will not begin the calibration sequence until the PD input goes low. If a manual calibration is
requested while the device is powered down, the calibration
will not begin at all. That is, the manual calibration input is
completely ignored in the power down state.
2.6.1 Supply Voltage
The ADC08D1000 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,
35
www.national.com
ADC08D1000
lines are long and/or the system in which the ADC08D1000
is used is noisy, it may be necessary to tie the OutV pin high.
ADC08D1000
cess development. It is recommended that the process be
developed based upon past experience in package mounting.
The package of the ADC08D1000 has an exposed pad on its
back that 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 exposed pad
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 exposed pad 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
at least as large as the 5 x 5 mm of the exposed pad of the
package and be located such that the exposed pad of the
device is entirely over that thermal land pattern. This thermal
land pattern should be electrically connected to ground. A
clearance of at least 0.5 mm should separate this land pattern
from the mounting pads for the package pins.
do not come up any faster than does the voltage at the ADC08D1000 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 ADC08D1000. The circuit of Figure 17 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 ADC08D1000, 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 17, 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.
20097454
FIGURE 17. 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 ADC08D1000 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 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.
20097421
FIGURE 18. 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 18.
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 square inches (6.5 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 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 9 to 16 thermal vias is recommended.
The thermal vias should be placed on a 1.2 mm grid spacing
and have a diameter of 0.30 to 0.33 mm. 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
2.6.2 Thermal Management
The ADC08D1000 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 130°C. That is, TA
(ambient temperature) plus ADC power consumption times
θJA (junction to ambient thermal resistance) should not exceed 130°C. This is not a problem if the ambient temperature
is kept to a maximum of +85°C as specified in the Operating
Ratings section.
As a convenience to the user, the ADC08D100 incorporates
a thermal diode to aid in temperature measurement. However, this diode has not been characterized and National Semiconductor has no information to provide regarding its
characteristics. Hence, no information is available as to the
temperature accuracy attainable when using this diode.
Please note that the following are general recommendations
for mounting exposed pad devices onto a PCB. This should
be considered the starting point in PCB and assembly pro-
www.national.com
36
2.9 USING THE SERIAL INTERFACE
The ADC08D1000 may be operated in the non-extended control (non-Serial Interface) mode or in the extended control
mode. Table 6 and Table 7 describe the functions of pins 3,
4, 14 and 127 in the non-extended control mode and the extended control mode, respectively.
2.9.1 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. That is, the full-scale range,
single-ended or differential input, the power on calibration delay, the output voltage and the input coupling (a.c. or d.c.).
The non-extended control mode is used by setting pin 14 high
or low, as opposed to letting it float. Table 6 indicates the pin
functions of the ADC08D1000 in the non-extended control
mode.
2.7 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
ADC08D1000. 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.
TABLE 6. Non-Extended Control Mode Operation
(Pin 14 High or Low)
Pin
Low
High
Floating
3
Reduced
VOD
Normal VOD
n/a
4
OutEdge =
Neg
OutEdge =
Pos
DDR
127
14
CalDly Short CalDly Long
Reduced VIN
Normal VIN
n/a
Extended
Control Mode
Pin 3 can be either high or low in the non-extended control
mode. Pin 14 must not be left floating to select this mode. See
1.2 NORMAL/EXTENDED CONTROL for more information.
Pin 4 can be high or low or can be left floating in the nonextended control mode. In the non-extended control mode,
pin 4 high or low defines the edge at which the output data
transitions. See 2.4.3 Output Edge Synchronization for more
information. If this pin is floating, 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.
Pin 127 in the non-extended control mode sets the calibration
delay. Pin 127 is not designed to remain floating.
TABLE 7. Extended Control Mode Operation
(Pin 14 Floating)
Pin
2.8 DYNAMIC PERFORMANCE
The ADC08D1000 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
Section 2.3.
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.
Function
3
SCLK (Serial Clock)
4
SDATA (Serial Data)
127
SCS (Serial Interface Chip Select)
2.10 COMMON APPLICATION PITFALLS
Failure to write all register locations when using extended control mode. When using the serial interface, all 8 user
registers must be written at least once with the default or desired values before calibration and subsequent use of the
ADC. In addition, the first write to the DES Enable register
(Dh) must load the default value (0x3FFFh). Once all registers
have been written once, other desired settings, including enabling DES can be loaded.
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
37
www.national.com
ADC08D1000
temperature sensor and the ADC08D1000 die of θJ-PAD times
typical power consumption = 2.8 x 1.6 = 4.5°C. Allowing for a
5.5°C (including an extra 1°C) temperature drop from the die
to the temperature sensor, then, would mean that maintaining
a maximum pad temperature reading of 124.5°C will ensure
that the die temperature does not exceed 130°C, assuming
that the exposed pad of the ADC08D1000 is properly soldered down and the thermal vias are adequate. (The inaccuracy of the temperature sensor is additional to the above
calculation).
ADC08D1000
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 ADC08D1000. Such practice may lead to conversion inaccuracies and even to device damage.
Incorrect analog input common mode voltage in the d.c.
coupled mode. As discussed in sections 1.1.4 and 2.2, the
Input common mode voltage must remain within 50 mV of the
VCMO output , which has a variability with temperature that
must also be tracked. Distortion performance will be degraded if the input common mode voltage is more than 50 mV from
VCMO .
Using an inadequate amplifier to drive the analog input.
Use care when choosing a high frequency amplifier to drive
the ADC08D1000 as many high speed amplifiers will have
higher distortion than will the ADC08D1000, resulting in overall system performance degradation.
Driving the VBG pin to change the reference voltage. As
mentioned in Section 2.1, the reference voltage is intended to
be fixed to provide one of two different full-scale values (650
www.national.com
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 Section 2.3,
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
Section 2.6.2, 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.
38
ADC08D1000
Physical Dimensions inches (millimeters) unless otherwise noted
NOTES: UNLESS OTHERWISE SPECIFIED
REFERENCE JEDEC REGISTRATION MS-026, VARIATION BFB.
128-Lead Exposed Pad LQFP
Order Number ADC08D1000CIYB
NS Package Number VNX128A
39
www.national.com
ADC08D1000 High Performance, Low Power, Dual 8-Bit, 1 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® Tools
www.national.com/webench
Audio
www.national.com/audio
App Notes
www.national.com/appnotes
Clock and Timing
www.national.com/timing
Reference Designs
www.national.com/refdesigns
Data Converters
www.national.com/adc
Samples
www.national.com/samples
Interface
www.national.com/interface
Eval Boards
www.national.com/evalboards
LVDS
www.national.com/lvds
Packaging
www.national.com/packaging
Power Management
www.national.com/power
Green Compliance
www.national.com/quality/green
Switching Regulators
www.national.com/switchers
Distributors
www.national.com/contacts
LDOs
www.national.com/ldo
Quality and Reliability
www.national.com/quality
LED Lighting
www.national.com/led
Feedback/Support
www.national.com/feedback
Voltage Reference
www.national.com/vref
Design Made Easy
www.national.com/easy
PowerWise® Solutions
www.national.com/powerwise
Solutions
www.national.com/solutions
Serial Digital Interface (SDI)
www.national.com/sdi
Mil/Aero
www.national.com/milaero
Temperature Sensors
www.national.com/tempsensors
SolarMagic™
www.national.com/solarmagic
Wireless (PLL/VCO)
www.national.com/wireless
Analog University®
www.national.com/AU
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© 2009 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]
National Semiconductor Asia
Pacific Technical Support Center
Email: [email protected]
National Semiconductor Japan
Technical Support Center
Email: [email protected]