NSC ADC08D1000DEV

ADC081000
High Performance, Low Power 8-Bit, 1 GSPS A/D Converter
General Description
Features
The ADC081000 is a low power, high performance CMOS
analog-to-digital converter that digitizes signals to 8 bits resolution at sampling rates up to 1.6 GSPS. Consuming a typical
1.4 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 7.5
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 compliant with IEEE
1596.3-1996, with the exception of a reduced common mode
voltage of 0.8V.
The converter has a 1:2 demultiplexer that feeds two LVDS
buses, reducing the output data rate on each bus to half the
sampling rate. The data on these buses are interleaved in
time to provide a 500 MHz output rate per bus and a combined
output rate of 1 GSPS.
The converter typically consumes less than 10 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
Adjustable Output Levels
Guaranteed No Missing Codes
Low Power Standby Mode
Key Specifications
■
■
■
■
■
■
■
Resolution
Max Conversion Rate
ENOB @ 500 MHz Input
DNL
Conversion Latency
Power Consumption
— Operating
— Power Down Mode
8 Bits
1 GSPS (min)
7.5 Bits (typ)
±0.25 LSB (typ)
7 and 8 Clock Cycles
1.45 W (typ)
9 mW (typ)
Applications
■
■
■
■
■
Direct RF Down Conversion
Digital Oscilloscopes
Satellite Set-top boxes
Communications Systems
Test Instrumentation
Block Diagram
20068153
© 2009 National Semiconductor Corporation
200681
www.national.com
ADC081000 High Performance, Low Power 8-Bit, 1 GSPS A/D Converter
April 21, 2009
ADC081000
Ordering Information
Extended Commercial Temperature Range (-40°C < TA < +85°C)
NS Package
ADC081000CIYB
128-Pin Exposed Pad LQFP
ADC08D1000DEV (use Dual product)
Development Board
Pin Configuration
20068101
* Exposed pad on back of package must be soldered to ground plane to ensure rated performance.
www.national.com
2
ADC081000
Pin Descriptions and Equivalent Circuits
Pin Functions
Pin No.
Symbol
Equivalent Circuit
Description
OutV
Output Voltage Amplitude set. Tie this pin high for normal
differential output amplitude. Ground this pin for a reduced
differential output amplitude and reduced power
consumption. See Section 1.5.
OutEdge
Output Edge Select. Sets the edge of the DCLK+ (pin 82) at
which the output data transitions. The output transitions with
the DCLK+ rising edge when this pin is high or on the falling
edge when this pin is low. See Section 5.3.
14
DC_Coup
DC Coupling select. When this pin is high, the VIN+ and VINanalog inputs are d.c. coupled and the input common mode
voltage should equal the VCMO (pin 7) output voltage. When
this pin is low, the analog input pins are internally biased and
the input signal should be a.c. coupled to the analog input
pins. See Section 3.0.
26
PD
Power Down Pin. A logic high on this pin puts the ADC into
the Power Down mode. A logic low on this pin allows normal
operation.
30
CAL
Calibration. A minimum 10 clock cycles low followed by a
minimum of 10 clock cycles high on this pin will initiate the
self calibration sequence. See Section 1.1.
35
FSR
Full scale Range Select. With a logic low on this pin, the fullscale differential input is 600 mVP-P. With a logic high on this
pin, the full-scale differential input is 800 mVP-P. See Section
1.3.
127
CalDly
Calibration Delay. This sets the number of clock cycles after
power up before calibration begins. See Section 1.1.
18
19
CLK+
CLK-
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+.
11
10
VIN+
VIN-
Analog Signal Differential Inputs to the ADC.
3
4
3
www.national.com
ADC081000
Pin Functions
Pin No.
Symbol
Equivalent Circuit
Description
Common Mode Output voltage for VIN+ and VIN- when d.c.
input coupling is used, in which case the voltage at this pin
is required to be the common mode input voltage at VIN+ and
VIN−. See Section 3.0.
7
VCMO
31
VBG
126
CalRun
Calibration Running indication. This pin is at a logic high
when calibration is running.
32
REXT
External Bias Resistor connection. The required value is
3.3k-Ohms (±0.1%) to ground. See Section 1.1.
www.national.com
Bandgap output voltage. This pin is capable of sourcing or
sinking up to 1.0 µA.
4
ADC081000
Pin Functions
Pin No.
Symbol
83
84
85
86
89
90
91
92
93
94
95
96
100
101
102
103
D7D7+
D6D6+
D5D5+
D4D4+
D3D3+
D2D2+
D1D1+
D0D0+
Equivalent Circuit
Description
LVDS data output bits sampled second in time sequence.
These outputs should always be terminated with a differential
100Ω resistance.
104
105
106
107
111
112
113
114
115
116
117
118
122
123
124
125
Dd7Dd7+
Dd6Dd6+
Dd5Dd5+
Dd4Dd4+
Dd3Dd3+
Dd2Dd2+
Dd1Dd1+
Dd0Dd0+
LVDS data output bits sampled first in time sequence. These
outputs should always be terminated with a differential
100Ω resistance.
79
80
OR+
OR-
Out of Range output. A differential high at these pins
indicates that the differential input is out of range (outside the
range of ±300 mV or ±400 mV as defined by the FSR pin).
See Section 1.6.
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.
2, 5, 8, 13, 16,
17, 20, 25, 28,
33, 128
VA
Analog power supply pins. Bypass these pins to GND.
40, 51, 62, 73,
88, 99, 110, 121
VDR
Output Driver power supply pins. Bypass these pins to DR
GND.
1, 6, 9, 12, 15,
21, 24, 27
GND
Ground return for VA
42, 53, 64, 74,
87, 97, 108, 119
DR GND
Ground return for VDR
22, 23, 29, 34,
36 - 39,
41, 43 - 50, 52,
54 - 61, 63,
65 - 72,
75 - 78, 98, 109,
120
NC
No Connection. Make no connection to these pins.
5
www.national.com
ADC081000
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)
Analog Supply Voltage (VA, VDR)
Digital Supply above Analog Supply
(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)
Power Dissipation at TA = 25°C
ESD Susceptibility (Note 4)
Human Body Model
Machine Model
Soldering Temperature, Infrared,
10 seconds (Note 5)
Storage Temperature
2.2V
300 mV
−0.15V to
(VA +0.15V)
+1.8V to +2.0V
+1.8V to VA
1.2V to 1.3V
0V to 2.15V
(100% duty cycle)
0V to 2.5V
(10% duty cycle)
Ground Difference
(|GND - DR GND|)
CLK Pins Voltage Range
Differential CLK Amplitude
-0.15V to 2.5V
0V to 100 mV
±25 mA
±50 mA
2.0 W
0V
0V to VA
0.6VP-P to 2.0VP-P
Package Thermal Resistances
2500V
250V
235°C
−65°C to +150°C
Package
θJ-C
(Top of
Package)
θJ-PAD
(Thermal Pad)
128-Lead Exposed Pad
LQFP
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.
Converter Electrical Characteristics
The following specifications apply after calibration for VA = VDR = +1.9VDC, OutV = 1.9V, VIN FSR (a.c. coupled) = differential
800mVP-P, CL = 10 pF, Differential, a.c. coupled Sinewave Clock, fCLK = 1 GHz at 0.5VP-P with 50% duty cycle, REXT = 3300Ω ±
0.1%, Analog Signal Source Impedance = 100Ω. Boldface limits apply for TA = TMIN to TMAX. All other limits TA = 25°C, unless
otherwise stated. (Notes 6, 7)
Symbol
Parameter
Conditions
Typical
(Note 8)
Limits
(Note 8)
Units
(Limits)
STATIC CONVERTER CHARACTERISTICS
INL
Integral Non-Linearity
±0.35
±0.9
LSB (max)
DNL
Differential Non-Linearity
±0.25
±0.7
LSB (max)
8
Bits
−1.5
0.5
LSB (min)
LSB (max)
−2.2
±25
mV (max)
−1.1
±25
mV (max)
Resolution with No Missing Codes
VOFF
Offset Error
TC VOFF
Offset Error Tempco
PFSE
Positive Full-Scale Error (Note 9)
NFSE
Negative Full-Scale Error (Note 9)
TC PFSE
Positive Full Scale Error Tempco
−40°C to +85°C
20
ppm/°C
TC NFSE
Negative Full Scale Error Tempco
−40°C to +85°C
13
ppm/°C
−0.45
−40°C to +85°C
−3
ppm/°C
Dynamic Converter Characteristics
FPBW
Full Power Bandwidth
B.E.R.
Bit Error Rate
Gain Flatness
ENOB
SINAD
Effective Number of Bits
Signal-to-Noise Plus Distortion Ratio
www.national.com
1.7
GHz
10-18
Error/Bit
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
Bits
fIN = 248 MHz, VIN = FSR − 0.5 dB
7.5
7.1
Bits (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
7.5
7.1
Bits (min)
fIN = 100 MHz, VIN = FSR − 0.5 dB
47
fIN = 248 MHz, VIN = FSR − 0.5 dB
47
44.8
dB (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
47
44.8
dB (min)
6
dB
SNR
THD
2nd Harm
3rd Harm
SFDR
IMD
Parameter
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
48
fIN = 248 MHz, VIN = FSR − 0.5 dB
48
45.5
dB (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
48
45.5
dB (min)
fIN = 100 MHz, VIN = FSR − 0.5 dB
-57
fIN = 248 MHz, VIN = FSR − 0.5 dB
-57
−50
dB (max)
fIN = 498 MHz, VIN = FSR − 0.5 dB
-57
−50
dB (max)
fIN = 100 MHz, VIN = FSR − 0.5 dB
−64
dB
fIN = 248 MHz, VIN = FSR − 0.5 dB
−64
dB
fIN = 498 MHz, VIN = FSR − 0.5 dB
−64
dB
fIN = 100 MHz, VIN = FSR − 0.5 dB
−64
dB
fIN = 248 MHz, VIN = FSR − 0.5 dB
−64
dB
fIN = 498 MHz, VIN = FSR − 0.5 dB
−64
dB
fIN = 100 MHz, VIN = FSR − 0.5 dB
58.5
fIN = 248 MHz, VIN = FSR − 0.5 dB
58.5
50
dB (min)
fIN = 498 MHz, VIN = FSR − 0.5 dB
58.5
50
dB (min)
fIN1 = 121 MHz, VIN = FSR − 7 dB
fIN2 = 126 MHz, VIN = FSR − 7 dB
-51
dB
dB
dB
dB
(VIN+) − (VIN−) > + Full Scale
255
(VIN+) − (VIN−) < − Full Scale
0
ANALOG INPUT AND REFERENCE CHARACTERISTICS
VIN
600
FSR pin High
800
Full Scale Analog Differential Input
Range
VCMI
Common Mode Analog Input Voltage
CIN
Analog Input Capacitance
(Note 10)
RIN
FSR pin Low
VCMO
Differential
0.02
Each input to ground
1.6
Differential Input Resistance
100
550
mVP-P (min)
650
mVP-P (max)
750
mVP-P (min)
850
mVP-P (max)
VCMO − 50
VCMO+ 50
mV (min)
mV (max)
pF
pF
94
Ω (min)
106
Ω (max)
0.95
1.45
V (min)
V (max)
ANALOG OUTPUT CHARACTERISTICS
VCMO
Common Mode Output Voltage
ICMO = ±1 µA
1.21
TC VCMO
Common Mode Output Voltage
Temperature Coefficient
TA = −40°C to +85°C
118
VBG
Bandgap Reference Output Voltage
IBG = ±100 µA
1.26
TC VBG
Bandgap Reference Voltage
Temperature Coefficient
TA = −40°C to +85°C,
IBG = ±100 µA
-28
Square Wave Clock
0.6
0.4
2.0
VP-P (min)
VP-P (max)
Sine Wave Clock
0.6
0.4
2.0
VP-P (min)
VP-P (max)
ppm/°C
1.22
1.33
V (min)
V (max)
ppm/°C
CLOCK INPUT CHARACTERISTICS
VID
II
CIN
Differential Clock Input Level
Input Current
Input Capacitance (Note 10)
VIN = 0V or VIN = VA
±1
µA
Differential
0.02
pF
Each Input to Ground
1.5
pF
7
www.national.com
ADC081000
Symbol
ADC081000
Symbol
Parameter
Conditions
Typical
(Note 8)
Limits
(Note 8)
Units
(Limits)
DIGITAL CONTROL PIN CHARACTERISTICS
VIH
Logic High Input Voltage
(Note 12)
1.4
V (min)
VIL
Logic Low Input Voltage
(Note 12)
0.5
V (max)
II
Input Current
VIN = 0 or VIN = VA
±1
µA
CIN
Logic Input Capacitance (Note 13)
Each input to ground
1.2
pF
OutV = VA, measured single-ended
300
OutV = GND, measured single-ended
225
DIGITAL OUTPUT CHARACTERISTICS
VOD
LVDS Differential Output Voltage
200
mVP-P (min)
450
mVP-P (max)
140
mVP-P (min)
340
mVP-P (max)
Δ VOD DIFF
Change in LVDS Output Swing
Between Logic Levels
±1
mV
VOS
Output Offset Voltage
800
mV
Δ VOS
Output Offset Voltage Change
Between Logic Levels
±1
mV
IOS
Output Short Circuit Current
ZO
Differential Output Impedance
Output+ & Output- connected to 0.8V
−4
mA
100
Ohms
POWER SUPPLY CHARACTERISTICS
IA
Analog Supply Current
IDR
Output Driver Supply Current
PD
PSRR1
Power Consumption
D.C. Power Supply Rejection Ratio
PD = Low
PD = High
646
4.5
792
mA (max)
mA
PD = Low
108
160
mA (max)
1.8
W (max)
PD = High
0.1
PD = Low
1.43
mA
PD = High
8.7
mW
Change in Offset Error with change in
VA from 1.8V to 2.0V
73
dB
TA = 85°C
1.1
TA ≤ 75°C
1.3
AC ELECTRICAL CHARACTERISTICS
fCLK1
Maximum Conversion Rate
fCLK2
Minimum Conversion Rate
TA ≤ 70°C
Input Clock Duty Cycle
200 MHz ≤ Input clock frequency < 1
GHz
1.0
GHz (min)
GHz
1.6
GHz
200
MHz
50
20
80
% (min)
% (max)
tCL
Input Clock Low Time (Note 12)
500
200
ps (min)
tCH
Input Clock High Time (Note 12)
500
200
ps (min)
50
45
55
% (min)
% (max)
DCLK Duty Cycle (Note 12)
tLHT
Differential Low to High Transition
Time
10% to 90%, CL = 2.5 pF
250
ps
tHLT
Differential High to Low Transition
Time
10% to 90%, CL = 2.5 pF
250
ps
tOSK
DCLK to Data Output Skew
(Note 11)
50% of DCLK transition to 50% of Data
transition
0
tAD
Sampling (Aperture) Delay
Input CLK+ Fall to Acquisition of Data
930
ps
tAJ
Aperture Jitter
0.4
ps rms
2.7
ns
tOD
Input Clock to Data Output Delay
Pipeline Delay (Latency) (Note 11)
www.national.com
50% of Input Clock transition to 50% of
Data transition
±200
"D" Outputs
7
"Dd" Outputs
8
8
ps (max)
Clock Cycles
Parameter
tWU
PD low to Rated Accuracy Conversion
(Wake-Up Time)
tCAL
Calibration Cycle Time
Typical
(Note 8)
Conditions
Limits
(Note 8)
Units
(Limits)
500
ns
46,000
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 the section entitled "Surface Mount" found in any post 1986 National Semiconductor Linear Data Book for methods of soldering surface mount
devices.
Note 6: The analog inputs are protected as shown below. Input voltage magnitudes beyond the Absolute Maximum Ratings may damage this device.
20068104
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 TJ = 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 Transfer Characteristic Figure 2. For relationship between Gain Error and FullScale 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 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: The ADC081000 has two interleaved 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 clock cycle less than the latency of the first bus (Dd0 through
Dd7).
9
www.national.com
ADC081000
Symbol
ADC081000
LVDS OUTPUT OFFSET VOLTAGE (VOS) is the midpoint
between the the D+ and D− pins' output voltages; i.e.,
[ (VD+) + (VD−) ] / 2.
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 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.
COMMON MODE VOLTAGE is the d.c. potential that is common to both pins of a differential pair. For a voltage to be a
common mode one, the signal departure from this d.c. common mode voltage at any given instant must be the same for
each of the pins, but in opposite directions from each other.
DIFFERENTIAL NON-LINEARITY (DNL) is the measure of
the maximum deviation from the ideal step size of 1 LSB.
Measured at 1 GSPS with a ramp input.
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:
– PGE = OE − PFSE
– NGE = −(OE − NFSE) = NFSE − OE
– Gain Error = NFSE − PFSE = PGE + NGE
where PGE is Positive Gain Error, NGE is Negative Gain Error, OE is Offset Error, PFSE is Positive Full-Scale Error and
NFSE is Negative Full-Scale Error.
INTEGRAL NON-LINEARITY (INL) is a measure of the deviation of each individual code from a straight line through the
input to output transfer function. The deviation of any given
code from this straight line is measured from the center of that
code value. The best fit method is used.
INTERMODULATION DISTORTION (IMD) is the creation of
additional spectral components as a result of two sinusoidal
frequencies being applied to the ADC input at the same time.
it is defined as the ratio of the power in the second and third
order intermodulation products to the power in one of the
original frequencies. IMD is usually expressed in dBFS.
LSB (LEAST SIGNIFICANT BIT) is the bit that has the smallest value or weight of all bits. This value is
20068190
FIGURE 1.
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 is a measure of how far
the last code transition is from the ideal 1/2 LSB above a differential −VIN/2. For the ADC081000 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.
OUTPUT DELAY (tOD) is the time delay after the falling edge
of the 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 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 ADC081000 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 offset 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.
LSB = 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 ADC081000.
LVDS DIFFERENTIAL OUTPUT VOLTAGE (VOD) is this absolute value of the difference between the VD+ and VD−
outputs, each measured with respect to Ground.
www.national.com
10
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.
11
www.national.com
ADC081000
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 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
ADC081000
Transfer Characteristic
20068156
FIGURE 2. Input / Output Transfer Characteristic
Timing Diagrams
20068114
FIGURE 3. ADC081000 Timing
www.national.com
12
VA = VDR = +1.9V, fCLK = 1 GHz (differential clock),
fIN = 248 MHz, Differential Inputs, unless otherwise stated. Parameters shown across temperature were measured after
recalibration at each temperature.
INL
INL vs. Temperature
20068117
20068118
INL vs. Supply Voltage
INL vs. Output Driver Voltage
20068119
20068120
INL vs. Sample Rate
DNL
20068121
20068122
13
www.national.com
ADC081000
Typical Performance Characteristics
ADC081000
DNL vs. Temperature
DNL vs. Supply Voltage
20068123
20068124
DNL vs. Output Driver Voltage
DNL vs. Sample Rate
20068125
20068126
SNR vs. Temperature
SNR vs. Supply Voltage
20068127
www.national.com
20068128
14
ADC081000
SNR vs. Output Driver Voltage
SNR vs. Sample Rate
20068129
20068130
SNR vs. Clock Duty Cycle
SNR vs. Input Frequency
20068131
20068132
Distortion vs. Temperature
Distortion vs. Supply Voltage
20068133
20068134
15
www.national.com
ADC081000
Distortion vs. Output Driver Voltage
Distortion vs. Sample Rate
20068135
20068136
Distortion vs. Clock Duty Cycle
Distortion vs. Input Frequency
20068137
20068138
Distortion vs. Input Common Mode
SINAD vs. Temperature
20068181
www.national.com
20068139
16
ADC081000
SINAD vs. Supply Voltage
SINAD vs. Output Driver Voltage
20068140
20068141
SINAD vs. Sample Rate
SINAD vs. Clock Duty Cycle
20068142
20068143
SINAD vs. Input Frequency
ENOB vs. Temperature
20068160
20068161
17
www.national.com
ADC081000
ENOB vs. Supply Voltage
ENOB vs. Output Driver Voltage
20068162
20068163
ENOB vs. Sample Rate
ENOB vs. Clock Duty Cycle
20068164
20068165
ENOB vs. Input Frequency
ENOB vs. Input Common Mode
20068166
www.national.com
20068182
18
ADC081000
SFDR vs. Temperature
SFDR vs. Supply Voltage
20068167
20068168
SFDR vs. Output Driver Voltage
SFDR vs. Sample Rate
20068169
20068170
SFDR vs. Clock Duty Cycle
SFDR vs. Input Frequency
20068171
20068172
19
www.national.com
ADC081000
Power Consumption vs. Sample Rate
Spectral Response @ fIN = 100 MHz
20068174
20068175
Spectral Response @ fIN = 248 MHz
Spectral Response @ fIN = 498 MHz
20068176
20068177
Spectral Response @ fIN = 1.5 GHz
Spectral Response @ fIN = 1.6 GHz
20068179
20068178
www.national.com
20
ADC081000
Intermodulation Distortion
20068183
21
www.national.com
ADC081000
ever the operating ambient temperature changes more than
30°C since calibration was last performed. See Section 5.1.2
for more information.
During the calibration process, the input termination resistor
is trimmed to a value that is equal to REXT / 33. This external
resistor must be placed between pin 32 and ground and 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 bias 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 10 clock
cycles, then holding it high for at least another 10 clock cycles.
There is no need to bring the CAL pin low after the 10 clock
cycles of CAL high to begin the calibration routine. Holding
the CAL pin high upon power up, however, will prevent the
calibration process from running until the CAL pin experiences the above-mentioned 10 clock cycles low followed by
10 cycles high.
The CalDly pin is used to select one of two delay times after
the application of power to the start of calibration. This calibration delay is 224 clock cycles (about 16.8 ms at 1 GSPS)
with CalDly low, or 230 clock cycles (about 1.07 seconds at 1
GSPS) with CalDly high. 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.
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.
Calibration can not be initiated or run while the device is in the
power-down mode. See Section 1.7 for information on the interaction between Power Down and Calibration.
Functional Description
The ADC081000 is a versatile, high performance, easy to use
A/D Converter with an innovative architecture permitting very
high speed operation. The controls available ease the application of the device to circuit solutions. The ADC081000 uses
a calibrated folding and interpolating architecture that
achieves over 7.5 effective bits. The use of folding amplifiers
greatly reduces the number of comparators and power consumption, while 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. Optimum performance requires adherence to the provisions discussed here and in the
Applications Information Section.
1.0 OVERVIEW
The analog input signal that is within the converter's input
voltage range is digitized to eight bits at speeds of 200 MSPS
to 1.6 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. The OR (Out
of Range) output is activated whenever the correct output
code would be outside of the 00h to FFh range.
The converter 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.
The voltage reference for the ADC081000 is derived from a
1.254V bandgap reference which is made available to the user at the VBG pin. This output is capable of sourcing or sinking
±100 μA.
The internal bandgap derived reference voltage has a nominal value of 600 mV or 800 mV, as determined by the FSR
pin and described in Section 1.3. There is no provision for the
use of an external reference voltage.
The fully differential comparator design and the innovative
design of the sample-and-hold amplifier, together with self
calibration, enables flat SINAD/ENOB response beyond 1.0
GHz. The ADC081000 output data signaling is LVDS and the
output format is offset binary.
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
7 clock cycles later for the "D" output bus and 8 clock cycles
later for the "Dd" output bus. There is an additional internal
delay called tOD before the data is available at the outputs.
See the Timing Diagram. The ADC081000 will convert as long
as the clock signal is present and the PD pin is low.
1.3 The Analog Inputs
The ADC081000 must be driven with a differential input signal. It is important that the inputs either be a.c. coupled to the
inputs with the DC_Coup pin grounded or d.c. coupled with
the DC_Coup pin high and have an input common mode voltage that is equal to and tracks the VCMO output.
Two full-scale range settings are provided with the FSR pin.
A high on that pin causes an input differential full-scale range
setting of 800 mVP-P, while grounding that pin causes an input
differential full-scale range setting of 600 mVP-P.
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), SFDR 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 important for 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 whenwww.national.com
1.4 Clocking
The ADC081000 must be driven with an a.c. coupled, differential clock signal. Section 4 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 receives that
data.
22
TABLE 1. Digital Control Pins
PIN DESCRIPTION
1.5 The LVDS Outputs
The data outputs, the Out Of Range (OR) and DCLK are
LVDS compliant outputs. Output current sources provide 3
mA of output current to a differential 100 Ohm load when the
OutV input 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 ADC081000 is used is noisy, it may
be necessary to tie the OutV pin high.
Note that the LVDS levels are not intended to meet any given
LVDS specification, but output levels are such that interfacing
with LVDS receivers is practical.
LOW
HIGH
3
OutV
440mV Outputs 600mV Outputs
4
OutEdge
Data Transition Data Transition
at DCLK Fall
at DCLK Rise
14
DC_Coup
A.C. Coupled
Inputs
D.C. Coupled
Inputs
26
PD
Normal
Operation
Power Down
30
CAL
Normal
Operation
Run Calibration
35
FSR
127
CalDly
600 mVP-P Full- 800 mVP-P FullScale In
Scale In
224 Clock
Cycles
230 Clock
Cycles
TABLE 2. Convenience Output Pins
PIN DESCRIPTION
1.6 Out Of Range (OR) Indication
The input signal is out of range whenever the correct code
would be above positive full-scale or below negative full scale.
When the input signal for any given sample is thus out of
range, the OR output is high for that word time.
1.7 Power Down
The ADC081000 is in the active state when the Power Down
pin (PD) is low. When the PD pin is high, the device is in the
power down mode, where the device power consumption is
reduced to a minimal level and the outputs are in a high
impedance state. 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.
USE / INDICATION
7
VCMO
31
VBG
1.25V Convenience Output.
79
80
OR+
OR−
Differential Out-Of-Range
Indication; active high.
126
CalRun
Common Mode Output Voltage.
Low is normal operation. High
indicates Calibration is running.
Applications Information
2.0 THE REFERENCE VOLTAGE
The voltage reference for the ADC081000 is derived from a
1.254V bandgap reference which is made available at the
VBG output for user convenience and has an output current
capability of ±100 μA. The VBG output should be buffered if
more current than this is required of it.
The internal bandgap-derived reference voltage causes the
full-scale peak-to-peak input swing to be either 600 mV or 800
mV, as determined by the FSR pin and described in Section
1.3. There is no provision for the use of an external reference
voltage.
1.8 Summary of Control Pins and Convenience Outputs
Table 1 and Table 2 are provided as a guide to the use of the
various control and convenience pins of the ADC081000.
Note that this table is only a guide and that the rest of this data
sheet should be consulted for the full meaning and use of
these pins.
3.0 THE ANALOG INPUT
The analog input is a differential one to which the signal
source may be a.c. coupled or d.c. coupled. Table 3 gives the
input to output relationship with the FSR pin high. With the
FSR pin grounded, the millivolt values in Table 3 are reduced
to 75% of the values indicated.
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.
23
www.national.com
ADC081000
To help ease data capture, 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. A high on the OutEdge input 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.
ADC081000
The ADC081000 is not designed to work with single-ended
signals, so it is NOT RECOMMENDED that this be done.
However, if the resulting drop in performance is allowable,
drive the ADC08100 with a single-ended signal by bypassing
the unused input to a.c. ground with a capacitor or connect it
directly to the VCMO pin. DO NOT connect either input pin directly to ground.
TABLE 3. DIFFERENTIAL INPUT TO OUTPUT
RELATIONSHIP (FSR High)
VIN+
VIN−
Output Code
VCM − 200 mV
VCM + 200 mV
0000 0000
VCM − 99 mV
VCM + 99 mV
0100 0000
VCM
VCM
0111 1111 /
1000 0000
VCM + 101 mV
VCM − 101 mV
1100 0000
VCM + 200mV
VCM − 200 mV
1111 1111
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 DC_Coup (pin 14) is low and
the input signal is a.c. coupled to the ADC. See Figure 4.
20068146
FIGURE 5. Single-Ended to Differential signal conversion
with a balun
When d.c. coupling to the ADC081000 analog inputs is required, single-ended to differential conversion may be easily
accomplished with the LMH6555, as shown in Figure 6. In
such applications, the LMH6555 performs the task of singleended to differential conversion while delivering low distortion
and noise, as well as output balance, that supports the operation of the ADC081000. Connecting the ADC081000 VCMO
pin to the VCM_REF pin of the LMH6555, through the appropriate buffer, will ensure that the ADC081000 common mode
input voltage is as needed for optimum performance of the
ADC081000. See Figure 6. The LMV321 was chosen as the
buffer in Figure 6 for its low voltage operation and reasonable
offset voltage. Be sure to limit output current from the
ADC081000 VCMO pin to 1.0 μA.
20068149
FIGURE 4. Differential Input Drive
When pin 14 is high, the analog inputs are d.c. coupled and
a common mode voltage must be externally provided at the
analog input pins. This common mode voltage should track
the VCMO output voltage. Note that the VCMO output potential
will change with temperature. The common mode output of
the driving device should track this change. 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 of the ADC081000 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 input pins 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.
Be sure that any current drawn from the VCMO output does not
exceed ±1 μA.
The Input impedance in the d.c. coupled mode (DC_Coup pin
high) consists of a precision 100 Ohm resistor between VIN+
and VIN- and a capacitance from each of these inputs to
ground. Driving the inputs beyond full scale will result in saturation or clipping of the reconstructed output.
20068155
FIGURE 6. Example of Servoing the Analog Input with
VCMO
3.2 Out Of Range (OR) Indication
When the conversion result is clipped the Out of Range output
is activated such that OR+ goes high and OR- goes low. This
output is active as long as accurate data on either or both of
the buses would be outside the range of 00h to FFh.
3.1 Handling Single-Ended Analog Signals
There is no provision for the ADC081000 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, as shown in Figure 5.
A balun is especially designed for very high frequencies and
has a wider bandwidth than does a transformer so is perferred
over a transformer for use with very high frequencies.
www.national.com
3.3 Full-Scale Input Range
As with all A/D Converters, the input range is determined by
the value of the ADC's reference voltage. The reference voltage of the ADC081000 is derived from an internal bandgap
reference. The FSR pin controls the effective reference voltage of the ADC081000 such that the differential full-scale
input range at the analog inputs is 800 mVP-P with the FSR
pin high, or is 600 mVP-P with FSR pin low. Best SNR is obtained with FSR high, but better distortion and SFDR are
24
and fIN is the maximum input frequency, in Hertz, to the ADC
analog input.
Note that the maximum jitter described above is the rms total
of the jitter from all sources, including that in the ADC clock,
that added by the system to the ADC 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 clock jitter
and the jitter added by the analog circuitry to the analog signal
to a minimum.
4.0 THE CLOCK INPUTS
The ADC081000 has differential LVDS clock inputs, CLK+
and CLK-, which must be driven with an a.c. coupled, differential clock signal. Although the ADC081000 is tested and its
performance is guaranteed with a differential 1.0 GHz clock,
it typically will function well with clock frequencies indicated
in the Electrical Characteristics Table. The clock inputs are
internally terminated and biased. The clock signal must be
capacitive coupled to the clock pins as indicated in Figure 7.
Operation up to the sample rates indicated in the Electrical
Characteristics Table is typically possible if the conditions of
the Operating Ratings 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 Section 7.2.
5.0 CONTROL PINS
Seven control pins provide a wide range of possibilities in the
operation of the ADC081000 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.
5.1 Self Calibration
The ADC081000 self-calibration must be run to achieve rated
performance. This procedure is performed upon power-up
and can be run any time on command. The calibration procedure is exactly the same whether there is a 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.
5.1.1 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 Section 1.1.
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 on-command calibration conditions are met.
The ADC081000 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 OnCommand Calibration Section 5.1.2.
The internal power-on calibration circuitry comes up in a random state. If the 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.
20068147
FIGURE 7. Differential (LVDS) Clock Connection
The differential Clock line pair should have a characteristic
impedance of 100Ω and be terminated at the clock source in
that (100Ω) characteristic impedance. The clock line should
be as short and as direct as possible. The ADC081000 clock
input is internally terminated with an untrimmed 100Ω resistor.
Insufficient clock levels will result in poor dynamic performance. Excessively high clock levels could cause a change
in the analog input offset voltage. To avoid these problems,
keep the clock level within the range specified in the Operating Ratings.
While it is specified and performance is guaranteed at 1.0
GSPS with a 50% clock duty cycle, ADC081000 performance
is essentially independent of clock duty cycle. However, to
ensure performance over temperature, it is recommended
that the input clock duty cycle be such that the minimum clock
high and low times are maintained within the range specified
in the Electrical Characteristics Table.
High speed, high performance ADCs such as the ADC081000
require very stable clock signals 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 total of the jitter from all
sources) allowed to prevent a jitter-induced reduction in SNR
is found to be
5.1.2 On-Command Calibration
Calibration may be run at any time by bringing the CAL pin
high for a minimum of 10 clock cycles after it has been low for
a minimum of 10 clock cycles. Holding the CAL pin high upon
power up will prevent execution of power-on calibration until
the CAL pin is low for a minimum of 10 clock clock cycles,
then brought high for a minimum of another 10 clock cycles.
The calibration cycle will begin 10 clock cycles after the CAL
pin is thus brought high.
The minimum 10 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, for best
performance, a self calibration should be performed 20 seconds or more after power up and repeated when the ambient
temperature changes more than 30°C since the last self calibration was run. SINAD drops about 1.5 dB for every 30°C
change in die temperature and ENOB drops about 0.25 bit for
every 30°C change in die temperature.
tJ(MAX) = (VINFSR / VIN(P-P)) x (1/(2(N+1) x π x fIN))
where tJ(MAX) is the rms total of all jitter sources in seconds,
VIN(P-P) is the peak-to-peak analog input signal, VINFSR is the
full-scale range of the ADC, "N" is the ADC resolution in bits
25
www.national.com
ADC081000
obtained with the FSR pin low. The LMH6555 is suitable for
both settings.
ADC081000
5.2 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. 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.
of the the two LVDS buses and a 200 MHz input clock, decimating the 200 MSPS data by two.
There is one LVDS clock pair available for use to latch the
LVDS outputs on both 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 5.3.
The OutV pin is used to set the LVDS differential output levels.
See Section 1.5.
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 127 and 128.
5.3 Output Edge Synchronization
DCLK signals are available to help latch the converter output
data into external circuitry. The output data can be synchronized with either edge of these clock 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 clock signal can be used to latch the output data
into the receiving circuit.
When the OutEdge pin is high, the output data is synchronized with (changes with) the rising edge of DCLK+. When
OutEdge is low, the output data is synchronized with the
falling edge of DCLK+.
At the very high speeds of which the ADC081000 is capable,
slight differences in the lengths of the clock 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.
7.0 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. Having power and ground
planes in adjacent layers of the PC Board will provide the best
supply bypass capacitance in terms of low ESL.
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 ADC081000
should be assumed to have little power supply noise rejection.
Any power supply used for digital circuity in a system where
a lot of digital power is being consumed should not be used
to supply power to the ADC081000. The ADC supplies should
be the same supply used for other analog circuitry, if not a
dedicated supply.
5.4 Power Down Feature
The Power Down (PD) pin, when high, puts the ADC081000
into a low power mode where power consumption is greatly
reduced.
The digital output pins retain the last conversion output code
when the clock is stopped, but are in a high impedance state
when the PD pin is high. However, upon return to normal operation (re-establishment of the clock and/or lowering of the
PD pin), 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.
7.1 Supply Voltage
The ADC081000 is specified to operate with a supply voltage
of 1.9V ±0.1V. It is very important to note that, while this device will function with slightly higher supply voltages, these
higher supply voltages may reduce product lifetime.
No pin should ever have a voltage on it that is in excess of the
supply voltage or below ground by more than 150 mV, not
even on a transient basis. This can be a problem upon application of power and power shut-down. Be sure that the supplies to circuits driving any of the input pins, analog or digital,
do not come up any faster than does the voltage at the
ADC081000 power pins.
6.0 THE DIGITAL OUTPUTS
The ADC081000 demultiplexes its output data onto two LVDS
output buses.
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 ADC081000 clock rate and the two buses must be
interleaved to obtain the entire 1 GSPS conversion result.
Since the minimum recommended clock rate for this device
is 200 MSPS, the effective sample rate can be reduced to as
low as 100 MSPS by using the results available on just one
www.national.com
20068154
FIGURE 8. Non-Spiking Power Supply
The Absolute Maximum Ratings should be strictly observed,
even during power up and power down. A power supply that
26
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 9.
20068151
FIGURE 9. Recommended Package Land Pattern
To minimize junction temperature, it is recommended that a
simple heat sink be built into the PCB. This is done by including a minimum copper pad of 2 inches by 2 inches (5.1 cm by
5.1 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.
On a board of FR-4 material and the built in heat sink described above (4 square inch pad and 9 thermal vias), the die
temperature stabilizes at about 30°C above the ambient temperature in about 20 seconds.
If it is desired to monitor die temperature, a temperature sensor may be mounted on the heat sink area of the board near
the thermal vias. Allow for a thermal gradient between the
temperature sensor and the ADC081000 die of θJC times typical power consumption = 2.8 x 1.43 = 4°C. Allowing for a 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 125°C will ensure that
the die temperature does not exceed 130°C, assuming that
the exposed pad of the ADC081000 is properly soldered
down and the thermal vias are adequate.
7.2 Thermal Management
The ADC081000 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, the device is soldered to a
PC Board and the sample rate is at or below 1 Gsps.
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 process
development. It is recommended that the process be developed based upon past experience in package mounting.
The package of the ADC081000 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.
8.0 LAYOUT AND GROUNDING
Proper grounding and proper routing of all signals are essential to ensure accurate conversion. A single ground plane
should be used, as opposed to 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
27
www.national.com
ADC081000
produces a voltage spike at turn-on and/or turn-off of power
can destroy the ADC081000. The circuit of Figure 8 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 ADC081000, unless a minimum load is provided
for the supply. The 100Ω resistor at the regulator output in
Figure 8 provides a minimum output current during power-up
to ensure there is no turn-on spiking.
In this circuit, 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. Also, be sure that
the impedance of the power distribution system is low to minimize resistive losses and minimize noise on the 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 and should
never spike to a voltage greater than (VA + 100mV).
If the power is applied to the device without a clock signal
present, the current drawn by the device might be below 100
mA. This is because the ADC081000 gets reset through
clocked logic and its initial state is random. 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 clock is established.
ADC081000
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. 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
ADC081000. 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.
justable linear regulator such as the LM317 or LM1086 set for
1.9V as discussed in Section 7.1.
Driving the inputs (analog or digital) beyond the power
supply rails. For device reliability, all inputs should not go
more than 150 mV below the ground pins or 150 mV above
the supply pins. Exceeding these limits on even a transient
basis may not only cause faulty or erratic operation, but may
impair device reliability. It is not uncommon for high speed
digital circuits to exhibit undershoot that goes more than a volt
below ground. Controlling the impedance of high speed lines
and terminating these lines in their characteristic impedance
should control overshoot.
Care should be taken not to overdrive the inputs of the
ADC081000. 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 1.3 and 3.0, the
Input common mode voltage must remain within 50 mV of the
VCMO output and track that 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 ADC081000 as many high speed amplifiers will have
higher distortion than will the ADC081000, resulting in overall
system performance degradation.
Driving the VBG pin to change the reference voltage. As
mentioned in Section 1.3, the reference voltage is intended to
be fixed to provide one of two different full-scale values (600
mVP-P and 800 mVP-P). Over driving this pin will not change
the full scale value, but can otherwise upset operation.
Driving the clock input with an excessively high level
signal. The ADC clock level should not exceed the level described in the Operating Ratings Table or the input offset error
could increase.
Inadequate clock levels. As described in Section 4.0, insufficient clock levels can result in poor performance. Excessive
clock levels could result in the introduction of an input offset.
Using an excessively long clock signal trace, or having
other signals coupled to the 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 7.2, it is important to provide an adequate heat removal to ensure device reliability. This can either be done 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.
9.0 DYNAMIC PERFORMANCE
The ADC081000 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 4.0.
It is good practice to keep the ADC 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 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 that offered by the package pins.
10.0 COMMON APPLICATION PITFALLS
Allowing loose power supply voltage tolerance. The
ADC081000 is specified for operation between 1.8 Volts to
2.0 Volts. Using a 1.8 Volt power supply then implies the need
for no negative tolerance. The best solution is to use an ad-
www.national.com
28
ADC081000
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 ADC081000CIYB
NS Package Number VNX128A
29
www.national.com
ADC081000 High Performance, Low Power 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]