BB ADS820U

®
ADS
ADS820
820
ADS
U
820
E
10-Bit, 20MHz Sampling
ANALOG-TO-DIGITAL CONVERTER
TM
FEATURES
DESCRIPTION
● NO MISSING CODES
The ADS820 is a low power, monolithic 10-bit, 20MHz
analog-to-digital converter utilizing a small geometry
CMOS process. This COMPLETE converter includes
a 10-bit quantizer with internal track/hold, reference,
and a power down feature. It operates from a single
+5V power supply and can be configured to accept
either differential or single-ended input signals.
The ADS820 employs digital error correction to provide excellent Nyquist differential linearity performance for demanding imaging applications. Its low
distortion, high SNR and high oversampling capability
give it the extra margin needed for telecommunications
and video applications.
This high performance converter is specified for AC
and DC performance at a 20MHz sampling rate. The
ADS820 is available in 28-pin SOIC and SSOP
packages.
● INTERNAL REFERENCE
● LOW DIFFERENTIAL LINEARITY ERROR:
0.2LSB
● LOW POWER: 195mW
● HIGH SNR: 60dB
● WIDEBAND TRACK/HOLD: 65MHz
● PACKAGES: 28-Pin SOIC and 28-PIN SSOP
APPLICATIONS
● SET-TOP BOXES
● CABLE MODEMS
● VIDEO DIGITIZING
● CCD IMAGING
Camcorders
Copiers
Scanners
Security Cameras
● IF AND BASEBAND DIGITIZATION
CLK
MSBI
OE
Error
Correction
Logic
3-State
Outputs
Timing
Circuitry
IN
Pipeline
A/D
T/H
IN
10-Bit
Digital
Data
+3.25V
REFT
CM
REFB
+1.25V
International Airport Industrial Park • Mailing Address: PO Box 11400, Tucson, AZ 85734 • Street Address: 6730 S. Tucson Blvd., Tucson, AZ 85706 • Tel: (520) 746-1111 • Twx: 910-952-1111
Internet: http://www.burr-brown.com/ • FAXLine: (800) 548-6133 (US/Canada Only) • Cable: BBRCORP • Telex: 066-6491 • FAX: (520) 889-1510 • Immediate Product Info: (800) 548-6132
©
1995 Burr-Brown Corporation
PDS-1288D
Printed in U.S.A. October, 1996
SPECIFICATIONS
At TA = +25°C, VS = +5V, Sampling Rate = 20MHz, with a 50% duty cycle clock having a 2ns rise/fall time, unless otherwise noted.
ADS820U, E
PARAMETER
CONDITIONS
Resolution
Specified Temperature Range
ANALOG INPUT
Differential Full Scale Input Range
Common-Mode Voltage
Analog Input Bandwidth (–3dB)
Small Signal
Full Power
TEMP
MIN
–40
+1.25
–20dBFS(1) Input
0dB Input
+25°C
+25°C
+85
Bits
°C
2.25
+3.25
V
V
120
65
MHz
MHz
1.25 || 4
MΩ || pF
±0.6
±1.0
±85
0.01
±2.1
0.02
+25°C
Full
Delta +VS = ±5%
+25°C
Full
+25°C
Delta +VS = ±5%
CONVERSION CHARACTERISTICS
Sample Rate
Data Latency
10k
±1.5
±2.5
0.1
±3.0
0.1
±0.15
±0.15
±0.2
±0.2
Guaranteed
±0.5
+25°C
Full
+25°C
Full
Full
Full
f = 10MHz
No Missing Codes
Integral Linearity Error at f = 500kHz
Spurious-Free Dynamic Range (SFDR)
f = 500kHz (–1dBFS input)
+25°C
Full
+25°C
Full
f = 10MHz (–1dBFS input)
Two-Tone Intermodulation Distortion (IMD)(3)
f = 4.4MHz and 4.5MHz (referred to –1dBFS envelope)
67
64
59
57
+25°C
Full
Signal-to-Noise Ratio (SNR)
f = 500kHz (–1dBFS input)
f = 10MHz (–1dBFS input)
Signal-to-(Noise + Distortion) (SINAD)
f = 500kHz (–1dBFS input)
f = 10MHz (–1dBFS input)
NTSC or PAL
NTSC or PAL
fIN = 3.58MHz
Sample/s
Convert Cycle
±1.0
±1.0
±1.0
±1.0
LSB
LSB
LSB
LSB
±2.0
LSB
77
74
63
62
dBFS
dBFS
dBFS
dBFS
–61
–60
dBc
dBc
+25°C
Full
+25°C
Full
58
56
58
56
60.5
60
60
60
dB
dB
dB
dB
+25°C
Full
+25°C
Full
+25°C
+25°C
58
55
56
54
60.5
60
58
57
0.5
0.1
9.5
2
7
2
dB
dB
dB
dB
%
degrees
Bits
ns
ps rms
ns
+25°C
+25°C
+25°C
1.5x Full Scale Input
%
%
ppm/°C
%FSR/%
%
%FSR/%
20M
6.5
DYNAMIC CHARACTERISTICS
Differential Linearity Error
f = 500kHz
Differential Gain Error
Differential Phase Error
Effective Bits(4)
Aperture Delay Time
Aperture Jitter
Overvoltage Recovery Time(5)
UNITS
TTL/HCT Compatible CMOS
Falling Edge
Start Conversion
ACCURACY(2)
Gain Error
Gain Tempco
Power Supply Rejection of Gain
Input Offset Error
Power Supply Rejection of Offset
MAX
10 Guaranteed
TAMBIENT
Input Impedance
DIGITAL INPUT
Logic Family
Convert Command
TYP
NOTE: (1) dBFS refers to dB below Full Scale. (2) Percentage accuracies are referred to the internal A/D Full Scale Range of 4Vp-p. (3) IMD is referred to the
larger of the two input signals. If referred to the peak envelope signal (≈0dB), the intermodulation products will be 7dB lower. (4) Based on (SINAD – 1.76)/6.02.
(5) No "rollover" of bits.
The information provided herein is believed to be reliable; however, BURR-BROWN assumes no responsibility for inaccuracies or omissions. BURR-BROWN assumes
no responsibility for the use of this information, and all use of such information shall be entirely at the user’s own risk. Prices and specifications are subject to change
without notice. No patent rights or licenses to any of the circuits described herein are implied or granted to any third party. BURR-BROWN does not authorize or warrant
any BURR-BROWN product for use in life support devices and/or systems.
®
ADS820
2
SPECIFICATIONS
(CONT)
At TA = +25°C, VS = +5V, Sampling Rate = 20MHz, with a 50% duty cycle clock having a 2ns rise/fall time, unless otherwise noted.
ADS820U, E
PARAMETER
CONDITIONS
OUTPUTS
Logic Family
Logic Coding
Logic Levels
TEMP
MIN
TYP
Full
Full
3-State Enable Time
3-State Disable Time
20
2
0.4
+VS
40
10
V
V
V
ns
ns
+5
39
40
195
200
+5.25
47
55
235
275
V
mA
mA
mW
mW
0
2.5
Full
Operating
Operating
Operating
Operating
Operating
Power Consumption
UNITS
TTL/HCT Compatible CMOS
SOB or BTC
Logic Selectable
Logic “LO”, CL = 15pF
Logic “HI”, CL = 15pF
POWER SUPPLY REQUIREMENTS
Supply Voltage: +VS
Supply Current: +IS
MAX
Full
+25°C
Full
+25°C
Full
Thermal Resistance, θJA
28-Pin SOIC
28-Pin SSOP
+4.75
°C/W
°C/W
75
50
ABSOLUTE MAXIMUM RATINGS
ELECTROSTATIC
DISCHARGE SENSITIVITY
+VS ....................................................................................................... +6V
Analog Input .............................................................. 0V to (+VS + 300mV)
Logic Input ................................................................ 0V to (+VS + 300mV)
Case Temperature ......................................................................... +100°C
Junction Temperature .................................................................... +150°C
Storage Temperature ..................................................................... +125°C
External Top Reference Voltage (REFT) .................................. +3.4V Max
External Bottom Reference Voltage (REFB) .............................. +1.1V Min
This integrated circuit can be damaged by ESD. Burr-Brown
recommends that all integrated circuits be handled with appropriate precautions. Failure to observe proper handling and
installation procedures can cause damage.
Electrostatic discharge can cause damage ranging from
performance degradation to complete device failure. BurrBrown Corporation recommends that all integrated circuits be
handled and stored using appropriate ESD protection
methods.
NOTE: Stresses above these ratings may permanently damage the device.
PACKAGE/ORDERING INFORMATION
PRODUCT
PACKAGE
PACKAGE
DRAWING
NUMBER(1)
ADS820U
ADS820E
28-Pin SOIC
28-Pin SSOP
217
324
TEMPERATURE
RANGE
–40°C to +85°C
–40°C to +85°C
NOTE: (1) For detailed drawing and dimension table, please see end of data
sheet, or Appendix C of Burr-Brown IC Data Book.
®
3
ADS820
PIN DESCRIPTIONS
PIN CONFIGURATION
TOP VIEW
SOIC/SSOP
PIN
DESIGNATOR
GND
B1
B2
B3
B4
B5
B6
B7
B8
B9
B10
DNC
DNC
GND
+VS
CLK
+VS
OE
GND
1
28
GND
Bit 1(MSB)
2
27
IN
Bit 2
3
26
IN
Bit 3
4
25
GND
Bit 4
5
24
+VS
Bit 5
6
23
REFT
Bit 6
7
22
CM
Bit 7
8
21
REFB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Bit 8
9
20
+VS
19
MSBI
Bit 9
10
19
MSBI
Bit 10 (LSB)
11
18
OE
20
21
+VS
REFB
DNC
12
17
+VS
DNC
13
16
CLK
22
CM
GND
14
15
+VS
23
REFT
24
25
26
27
28
+VS
GND
IN
IN
GND
ADS820
DNC: Do Not Connect
DESCRIPTION
Ground
Bit 1, Most Significant Bit
Bit 2
Bit 3
Bit 4
Bit 5
Bit 6
Bit 7
Bit 8
Bit 9
Bit 10, Least Significant Bit
Do not connect.
Do not connect.
Ground
+5V Power Supply
Convert Clock Input, 50% Duty Cycle
+5V Power Supply
HI: High Impedance State. LO or Floating: Normal Operation. Internal pull-down resistor.
Most Significant Bit Inversion, HI: MSB inverted
for complementary output. LO or Floating: Straight
output. Internal pull-down resistor.
+5V Power Supply
Bottom Reference Bypass. For external bypassing of internal +1.25V reference.
Common-Mode Voltage. It is derived by (REFT +
REFB)/2.
Top Reference Bypass. For external bypassing
of internal +3.25V reference.
+5V Power Supply
Ground
Input
Complementary Input
Ground
TIMING DIAGRAM
tCONV
tL
CONVERT
CLOCK
tD
tH
DATA LATENCY
(6.5 Clock Cycles)
Hold
Hold
Hold
Hold
Hold
Hold
Track "N + 1" Track "N + 2" Track "N + 3" Track "N + 4" Track "N + 5" Track "N + 6" Track
(1)
Track
INTERNAL
TRACK/HOLD
Hold
"N"
t2
OUTPUT
DATA
Data Valid
N-8
Data Valid
N-7
Data Valid
N-6
N-5
N-4
N-3
N-2
N-1
t1
Data Invalid
SYMBOL
tCONV
tL
tH
tD
t1
t2
NOTE: (1) “
DESCRIPTION
MIN
Convert Clock Period
Clock Pulse Low
Clock Pulse High
Aperture Delay
Data Hold Time, CL = 0pF
New Data Delay Time, CL = 15pF max
50
24
24
MAX
UNITS
100µs
ns
ns
ns
ns
ns
ns
25
25
2
3.9
12.5
” indicates the portion of the waveform that will stretch out at slower sample rates.
®
ADS820
TYP
4
N
TYPICAL PERFORMANCE CURVES
At TA = +25°C, VS = +5V, Sampling Rate = 20MHz, with a 50% duty cycle clock having a 2ns rise/fall time, unless otherwise noted.
SPECTRAL PERFORMANCE
SPECTRAL PERFORMANCE
0
0
fIN = 500kHz
fS = 10MHz
fIN = 4.8MHz
–20
–40
Amplitude (dB)
Amplitude (dB)
–20
–60
–80
–100
–40
–60
–80
–100
–120
–120
0
1.0
2.0
3.0
4.0
5.0
0
2.0
4.0
Frequency (MHz)
6.0
0
fIN = 9.8MHz
f1 = 4.5MHz
–20
–20
Amplitude (dB)
Amplitude (dB)
10.0
TWO-TONE INTERMODULATION
SPECTRAL PERFORMANCE
0
–40
–60
–80
–100
f2 = 4.4MHz
–40
–60
–80
–100
–120
–120
0
2.0
4.0
6.0
8.0
0.0
10.0
2.50
Frequency (MHz)
5.00
7.50
10.00
Frequency (MHz)
DIFFERENTIAL LINEARITY ERROR
DIFFERENTIAL LINEARITY ERROR
2.0
2.0
fIN = 10MHz
fIN = 500kHz
1.0
DLE (LSB)
1.0
DLE (LSB)
8.0
Frequency (MHz)
0
0
–1.0
–1.0
–2.0
–2.0
24
224
424
624
824
24
1024
224
424
624
824
1024
Code
Code
®
5
ADS820
TYPICAL PERFORMANCE CURVES
(CONT)
At TA = +25°C, VS = +5V, Sampling Rate = 20MHz, with a 50% duty cycle clock having a 2ns rise/fall time, unless otherwise noted.
SWEPT POWER SFDR
DYNAMIC PERFORMANCE vs INPUT FREQUENCY
100
85
fIN = 10MHz
80
80
75
SFDR (dBFS)
SFDR, SNR (dB)
SFDR
70
65
SNR
60
40
20
60
55
100k
0
1M
10M
–50
100M
–40
–30
–20
–10
0
10
Input Amplitude (dBm)
Frequency (Hz)
SWEPT POWER SNR
INTEGRAL LINEARITY ERROR
60
4.0
fIN = 500kHz
fIN = 10MHz
50
2.0
ILE (LSB)
SNR (dB)
40
30
0
20
–2.0
10
0
–48.125
–4.0
–40
–30
–20
–10
0
10
24
424
624
824
Code
DYNAMIC PERFORMANCE
vs SINGLE-ENDED FULL-SCALE INPUT RANGE
DYNAMIC PERFORMANCE
vs DIFFERENTIAL FULL-SCALE INPUT RANGE
80
80
SFDR (fIN = 500kHz)
1024
SFDR (fIN = 500kHz)
75
Dynamic Range (dB)
75
Dynamic Range (dB)
224
Input Amplitude (dBm)
70
SFDR (fIN = 10MHz)
65
SNR (fIN = 500kHz)
60
70
SFDR (fIN = 10MHz)
65
SNR (fIN = 500kHz)
60
SNR (fIN = 10MHz)
SNR (fIN = 10MHz)
55
55
1
2
3
4
5
1
Single-Ended Full-Scale Input Range (Vp-p)
NOTE: REFTEXT varied, REFB is fixed at the internal value of +1.25V.
3
4
5
NOTE: REFTEXT varied, REFB is fixed at internal value of +1.25V.
®
ADS820
2
Differential Full-Scale Input Range (Vp-p)
6
TYPICAL PERFORMANCE CURVES
(CONT)
At TA = +25°C, VS = +5V, Sampling Rate = 20MHz, with a 50% duty cycle clock having a 2ns rise/fall time, unless otherwise noted.
SPURIOUS FREE DYNAMIC RANGE
vs TEMPERATURE
DIFFERENTIAL LINEARITY ERROR vs TEMPERATURE
0.3
90
fIN = 500kHz
80
0.2
SFDR (dBFS)
DLE (LSB)
fIN = 10MHz
0.1
fIN = 500kHz
70
60
fIN = 10MHz
0
50
–50
–25
0
25
50
75
100
–50
–25
0
25
50
75
100
75
100
Ambient Temperature (°C)
Ambient Temperature (°C)
SIGNAL-TO-NOISE RATIO vs TEMPERATURE
SIGNAL-TO-(NOISE + DISTORTION)
vs TEMPERATURE
70
62
fIN = 500kHz
65
60
SINAD (dB)
SNR (dB)
fIN = 500kHz
60
fIN = 10MHz
55
58
fIN = 10MHz
56
50
54
–50
–25
0
25
50
75
100
–50
–25
Ambient Temperature (°C)
0
25
50
Ambient Temperature (°C)
SUPPLY CURRENT vs TEMPERATURE
POWER DISSIPATION vs TEMPERATURE
42
210
200
IQ (mA)
Power (mW)
40
38
190
180
36
–50
–25
0
25
50
75
170
100
–50
Ambient Temperature (°C)
–25
0
25
50
75
100
Ambient Temperature (°C)
®
7
ADS820
TYPICAL PERFORMANCE CURVES
(CONT)
At TA = +25°C, VS = +5V, Sampling Rate = 20MHz, with a 50% duty cycle clock having a 2ns rise/fall time, unless otherwise noted.
GAIN ERROR vs TEMPERATURE
OFFSET ERROR vs TEMPERATURE
Offset (% FSR)
–1.75
–0.55
–1.05
–1.55
–2.0
–2.25
–2.50
–50
–25
0
25
50
75
100
–50
–25
0
Ambient Temperature (°C)
0
1M
–1
0.8M
Counts
1.2M
–2
0.4M
–4
0.2M
–5
1M
10M
100M
0.0
N–2
1G
Frequency (Hz)
N–1
N
Code
®
ADS820
75
100
0.6M
–3
100k
50
OUTPUT NOISE HISTOGRAM (NO SIGNAL)
1
10k
25
Ambient Temperature (°C)
TRACK-MODE SMALL-SIGNAL INPUT BANDWIDTH
Track-Mode Input Response (dB)
Gain (% FSR)
–0.05
8
N+1
N+2
THEORY OF OPERATION
Op Amp
Bias
The ADS820 is a high speed sampling analog-to-digital
converter with pipelining. It uses a fully differential architecture and digital error correction to guarantee 10-bit resolution. The differential track/hold circuit is shown in Figure
1. The switches are controlled by an internal clock which has
a non-overlapping two phase signal, φ1 and φ2. At the
sampling time the input signal is sampled on the bottom
plates of the input capacitors. In the next clock phase, φ2, the
bottom plates of the input capacitors are connected together
and the feedback capacitors are switched to the op amp
output. At this time the charge redistributes between CI and
CH, completing one track/hold cycle. The differential output
is a held DC representation of the analog input at the sample
time. The track/hold circuit can also convert a single-ended
input signal into a fully differential signal for the quantizer.
The pipelined quantizer architecture has 9 stages with each
stage containing a two-bit quantizer and a two bit digital-toanalog converter, as shown in Figure 2. Each two-bit quantizer stage converts on the edge of the sub-clock, which is
twice the frequency of the externally applied clock. The
output of each quantizer is fed into its own delay line to
IN
IN
φ1
φ1
CH
φ2
CI
IN
IN
φ1
φ2
OUT
φ1
OUT
φ1
CI
φ2
CH
φ1
φ1
Input Clock (50%)
Op Amp
Bias
VCM
Internal Non-overlapping Clock
φ1
φ2
φ1
FIGURE 1. Input Track/Hold Configuration with Timing
Signals.
Digital Delay
Input
T/H
2-Bit
Flash
STAGE 1
VCM
2-Bit
DAC
+
Σ
–
x2
Digital Delay
STAGE 2
B1 (MSB)
2-Bit
DAC
B2
Digital Error Correction
2-Bit
Flash
+
Σ
–
x2
B3
B4
B5
B6
B7
B8
B9
Digital Delay
2-Bit
Flash
STAGE 8
B10 (LSB)
2-Bit
DAC
+
Σ
–
x2
STAGE 9
2-Bit
Flash
Digital Delay
FIGURE 2. Pipeline A/D Architecture.
®
9
ADS820
DIGITAL OUTPUT DATA
time-align it with the data created from the following quantizer stages. This aligned data is fed into a digital error
correction circuit which can adjust the output data based on
the information found on the redundant bits. This technique
gives the ADS820 excellent differential linearity and guarantees no missing codes at the 10-bit level.
The 10-bit output data is provided at CMOS logic levels.
The standard output coding is Straight Offset Binary where
a full scale input signal corresponds to all “1’s” at the output.
This condition is met with pin 19 “LO” or Floating due to an
internal pull-down resistor. By applying a high voltage to
this pin, a Binary Two’s Complement output will be provided where the most significant bit is inverted. The digital
outputs of the ADS820 can be set to a high impedance state
by driving OE (pin 18) with a logic “HI”. Normal operation
is achieved with pin 18 “LO” or Floating due to internal
pull-down resistors. This function is provided for testability
purposes and is not meant to drive digital buses directly or
be dynamically changed during the conversion process.
There is a 6.5 clock cycle data latency from the start convert
signal to the valid output data. The output data is available in
Straight Offset Binary (SOB) or Binary Two’s Complement
(BTC) format.
THE ANALOG INPUT AND INTERNAL REFERENCE
The analog input of the ADS820 can be configured in
various ways and driven with different circuits, depending
on the nature of the signal and the level of performance
desired. The ADS820 has an internal reference that sets the
full scale input range of the A/D. The differential input range
has each input centered around the common-mode of +2.25V,
with each of the two inputs having a full scale range of
+1.25V to +3.25V. Since each input is 2V peak-to-peak and
180° out of phase with the other, a 4V differential input
signal to the quantizer results. As shown in Figure 3, the
positive full scale reference (REFT) and the negative full
scale reference (REFB) are brought out for external bypassing. In addition, the common-mode voltage (CM) may be
used as a reference to provide the appropriate offset for the
driving circuitry. However, care must be taken not to appreciably load this reference node. For more information regarding external references, single-ended inputs, and
ADS820 drive circuits, refer to the applications section.
OUTPUT CODE
DIFFERENTIAL
INPUT(1)
SOB
PIN 19
FLOATING or LO
BTC
PIN 19
HI
1111111111
1111111111
1111111110
1110000000
1100000000
1010000000
1000000001
1000000000
0111111111
0110000000
0100000000
0010000000
0000000001
0000000000
0111111111
0111111111
0111111110
0110000000
0100000000
0010000000
0000000001
0000000000
1111111111
1110000000
1100000000
1010000000
1000000001
1000000000
+FS (IN = +3.25V, IN = +1.25V)
+FS –1LSB
+FS –2LSB
+3/4 Full Scale
+1/2 Full Scale
+1/4 Full Scale
+1LSB
Bipolar Zero (IN = IN = +2.25V)
–1LSB
–1/4 Full Scale
–1/2 Full Scale
–3/4 Full Scale
–FS +1LSB
–FS (IN = +1.25V, IN = +3.25V)
Note: In the single-ended input mode, +FS = +4.25V and –FS = +0.25V.
TABLE I. Coding Table for the ADS820.
ADS820
+3.25V
23
APPLICATIONS
REFT
DRIVING THE ADS820
0.1µF
2kΩ
+2.25V
22
The ADS820 has a differential input with a common-mode
of +2.25V. For AC-coupled applications, the simplest way
to create this differential input is to drive the primary
winding of a transformer with a single-ended input. A
differential output is created on the secondary if the center
tap is tied to the common-mode voltage (CM) of +2.25V per
Figure 4. This transformer-coupled input arrangement pro-
To
Internal
Comparators
CM
2kΩ
21
REFB
0.1µF
+1.25V
FIGURE 3. Internal Reference Structure.
22 CM
CLOCK REQUIREMENTS
0.1µF
The CLK pin accepts a CMOS level clock input. The rising
and falling edge of the externally applied convert command
clock control the various interstage conversions in the pipeline. Therefore, the duty cycle of the clock should be held at
50% with low jitter and fast rise/fall times of 2ns or less. This
is especially important when digitizing a high frequency
input and operating at the maximum sample rate. Deviation
from a 50% duty cycle will effectively shorten some of the
interstage settling times, thus degrading the SNR and DNL
performance.
ADS820
22pF
Mini-Circuits
TT1-6-KK81
or equivalent
27 IN
22pF
FIGURE 4. AC-Coupled Single-Ended to Differential Drive
Circuit Using a Transformer.
®
ADS820
26 IN
AC Input
Signal
10
vides good high frequency AC performance. It is important
to select a transformer that gives low distortion and does not
exhibit core saturation at full scale voltage levels. Since the
transformer does not appreciably load the ladder, there is no
need to buffer the common-mode (CM) output in this instance. In general, it is advisable to keep the current draw
from the CM output pin below 0.5µA to avoid nonlinearity
in the internal reference ladder. A FET input operational
amplifier such as the OPA130 can provide a buffered reference for driving external circuitry. The analog IN and IN
inputs should be bypassed with 22pF capacitors to minimize
track/hold glitches and to improve high input frequency
performance.
each input. The cut-off frequency of the filter is determined
by fC = 1/(2πRSER•(CSH+CADC)) where RSER is the resistor in
series with the input, CSH is the external capacitor from the
input to ground, and CADC is the internal input capacitance of
the A/D converter (typically 4pF).
Resistors R1 and R2 are used to derive the necessary common
mode voltage from the buffered top and bottom references.
The total load of the resistor string should be selected so that
the current does not exceed 1mA. Although the circuit in
Figure 5 uses two resistors of equal value so that the common
mode voltage is centered between the top and bottom reference (+2.25V), it is not necessary to do so. In all cases the
center point, VCM, should be bypassed to ground in order to
provide a low impedance AC ground.
Figure 5 illustrates another possible low cost interface circuit
which utilizes resistors and capacitors in place of a transformer. Depending on the signal bandwidth, the component
values should be carefully selected in order to maintain the
performance outlined in the data sheet. The input capacitors,
CIN, and the input resistors, RIN, create a high-pass filter with
the lower corner frequency at fC = 1/(2πRINCIN). The corner
frequency can be reduced by either increasing the value of
RIN or CIN. If the circuit operates with a 50Ω or 75Ω
impedance level, the resistors are fixed and only the value of
the capacitor can be increased. Usually AC-coupling capacitors are electrolytic or tantalum capacitors with values of
1µF or higher. It should be noted that these large capacitors
become inductive with increased input frequency, which
could lead to signal amplitude errors or oscillation. To
maintain a low AC-coupling impedance throughout the signal band, a small value (e.g. 1µF) ceramic capacitor could be
added in parallel with the polarized capacitor.
If the signal needs to be DC coupled to the input of the
ADS820, an operational amplifier input circuit is required.
In the differential input mode, any single-ended signal must
be modified to create a differential signal. This can be
accomplished by using two operational amplifiers, one in
the noninverting mode for the input and the other amplifier
in the inverting mode for the complementary input. The low
distortion circuit in Figure 6 will provide the necessary input
shifting required for signals centered around ground. It also
employs a diode for output level shifting to guarantee a low
distortion +3.25V output swing. Another DC-coupled circuit
is shown in Figure 7. Other amplifiers can be used in place
of the OPA642s if the lowest distortion is not necessary. If
output level shifting circuits are not used, care must be taken
to select operational amplifiers that give the necessary performance when swinging to +3.25V with a ±5V supply
operational amplifier. The OPA620 and OPA621, or the
lower power OPA650 and OPA651 can be used in place of
the OPA642s in Figure 6. In that configuration, the OPA650
and OPA651 will typically swing to within 100mV of
positive full scale. If the OPA621 or OPA651 is used, the
input buffer must be configured in a gain of 2.
Capacitors CSH1 and CSH2 are used to minimize current
glitches resulting from the switching in the input track and
hold stage and to improve signal-to-noise performance. These
capacitors can also be used to establish a low-pass filter and
effectively reduce the noise bandwidth. In order to create a
real pole, resistors RSER1 and RSER2 were added in series with
C1
0.1µF
CIN
0.1µF
*RSER1
49.9Ω
R1
(6kΩ)
+3.25V
Top Reference
IN
RIN1
25Ω
RIN2
25Ω
CIN
0.1µF
CSH1
22pF
R3
1kΩ
C2
0.1µF
*RSER2
49.9Ω
ADS8xx
VCM
R2
(6kΩ)
IN
CSH2
22pF
+1.25V
Bottom Reference
C3
0.1µF
NOTE: * indicates optional component.
FIGURE 5. AC-Coupled Differential Input Circuit.
®
11
ADS820
+5V
604Ω
+5V
301Ω
BAS16(1)
(3)
Optional
High Impedance
Input Amplifier
301Ω
301Ω
27 IN
OPA642
2.49kΩ
+5V(2)
22pF
0.1µF
0.1µF
–5V
DC-Coupled
Input Signal
+5V
(3)
OPA642
604Ω
604Ω
ADS820
49.9Ω
OPA130
+5V
–5V
2.49kΩ +2.25V
22 CM
+5V
301Ω
24.9Ω
Input Level
Shift Buffer
301Ω
BAS16(1)
(3)
26 IN
OPA642
22pF
0.1µF
–5V
NOTES: (1) A Philips BAS16 diode or equivalent may be used.
(2) Supply bypassing not shown. (3) OPA620 or OPA650 may be
substituted. See "Driving the ADS820" section.
604Ω
301Ω
FIGURE 6. A Low Distortion DC-Coupled, Single-Ended to Differential Input Driver Circuit.
DC-Coupled
Input Signal
2kΩ
+5V
–5V
7
4
1
3
243Ω
–5V
B
C
2
E
26 IN
+1
OTA
6
1kΩ
22pF
OPA660
8
200Ω
1nF
5
500Ω
+5V
1kΩ
ADS820
2
50Ω
OPA130
1kΩ
C1
15pF
200Ω
500Ω
8
2
E
3
B
3
22 CM
0.1µF
200Ω
5
OTA
VOUT
+1
6
C
243Ω
–5V
OPA660
27 IN
22pF
NOTE: Power supplies and bypassing not shown. The measured SNR performance with 12.5MHz input signal is 57dB with this driver circuit.
FIGURE 7. A Wideband DC-Coupled, Single-Ended to Differential Input Driver Circuit.
The ADS820 can also be configured with a single-ended
input full scale range of +0.25V to +4.25V by tying the
complementary input to the common-mode reference voltage
as shown in Figure 8. This configuration will result in
increased even-order harmonics, especially at higher input
frequencies. However, this tradeoff may be quite acceptable
for time-domain applications. The driving amplifier must
give adequate performance with a +0.25V to +4.25V output
swing in this case.
22 CM
0.1µF
ADS820
Single-Ended
Input Signal
26 IN
27 IN
22pF
Full Scale = +0.25V to +4.25V with internal references.
FIGURE 8. Single-Ended Input Connection.
®
ADS820
12
DYNAMIC PERFORMANCE TESTING
EXTERNAL REFERENCES AND ADJUSTMENT OF
FULL SCALE RANGE
The ADS820 is a high performance converter and careful
attention to test techniques is necessary to achieve accurate
results. Highly accurate phase-locked signal sources allow
high resolution FFT measurements to be made without using
data windowing functions. A low jitter signal generator such
as the HP8644A for the test signal, phase-locked with a low
jitter HP8022A pulse generator for the A/D clock, gives
excellent results. Low pass filtering (or bandpass filtering)
of test signals is absolutely necessary to test the low distortion of the ADS820. Using a signal amplitude slightly lower
than full scale will allow a small amount of “headroom” so
that noise or DC offset voltage will not overrange the A/D
and cause clipping on signal peaks.
The internal reference buffers are limited to approximately
1mA of output current. As a result, these internal +1.25V
and +3.25V references may be overridden by external references that have at least 18mA (at room temperature) of
output drive capability. In this instance, the common-mode
voltage will be set halfway between the two references. This
feature can be used to adjust the gain error, improve gain
drift, or to change the full scale input range of the ADS820.
Changing the full scale range to a lower value has the benefit
of easing the swing requirements of external input amplifiers. The external references can vary as long as the value of
the external top reference (REFTEXT) is less than or equal to
+3.4V and the value of the external bottom reference
(REFBEXT) is greater than or equal to +1.1V and the difference between the external references are greater than or
equal to 800mV.
DYNAMIC PERFORMANCE DEFINITIONS
1. Signal-to-Noise-and-Distortion Ratio (SINAD):
Sinewave Signal Power
10 log
Noise + Harmonic Power (first 15 harmonics)
For the differential configuration, the full scale input range
will be set to the external reference values that are selected.
For the single-ended mode, the input range is 2•(REFTEXT –
REFBEXT), with the common-mode being centered at
(REFTEXT + REFBEXT)/2. Refer to the typical performance
curves for expected performance vs full scale input range.
2. Signal-to-Noise Ratio (SNR):
Sinewave Signal Power
10 log
Noise Power
The circuit in Figure 9 works completely on a single +5V
supply. As a reference element, it uses the micro-power
reference REF1004-2.5, which is set to a quiescent current
of 0.1mA. Amplifier A2 is configured as a follower to buffer
the +1.25V generated from the resistor divider. To provide
the necessary current drive, a pull-down resistor, RP is
added.
Amplifier A1 is configured as an adjustable gain stage, with
a range of approximately 1 to 1.32. The pull-up resistor
again relieves the op amp from providing the full current
drive. The value of the pull-up/down resistors is not critical
and can be varied to optimize power consumption. The need
for pull-up/down resistors depends only on the drive capability of the selected drive amplifiers and thus can be
omitted.
3. Intermodulation Distortion (IMD):
Highest IMD Product Power (to 5th-order)
10 log
Sinewave Signal Power
IMD is referenced to the larger of the test signals f1 or f2.
Five “bins” either side of peak are used for calculation of
fundamental and harmonic power. The “0” frequency bin
(DC) is not included in these calculations as it is of little
importance in dynamic signal processing applications.
PC BOARD LAYOUT AND BYPASSING
A well-designed, clean PC board layout will assure proper
operation and clean spectral response. Proper grounding and
bypassing, short lead lengths, and the use of ground planes
are particularly important for high frequency circuits. Multilayer PC boards are recommended for best performance
but if carefully designed, a two-sided PC board with large,
heavy ground planes can give excellent results. It is recommended that the analog and digital ground pins of the
ADS820 be connected directly to the analog ground plane.
In our experience, this gives the most consistent results. The
A/D power supply commons should be tied together at the
analog ground plane. Power supplies should be bypassed
with 0.1µF ceramic capacitors as close to the pin as possible.
®
13
ADS820
+5V
RP
220Ω
A1
1/2
OPA2234
+5V
Top
Reference
+2.5V to +3.25V
2kΩ
10kΩ
6.2kΩ
10kΩ
REF1004
+2.5V
A2
0.1µF
+1.25V
1/2
OPA2234
10kΩ*
10kΩ
Bottom
Reference
RP
220Ω
10kΩ*
NOTE: (*) Use parts alternatively for adjustment capability.
FIGURE 9. Optional External Reference to Set the Full-Scale Range Utilizing a Dual, Single-Supply Op Amp.
IDT74FCT2245
+5V
0.1µF
+VS
CLK
Ext
Clk
0.1µF
R1
50Ω
+VS
OE
MSBI
0.1µF
+VS
REFB
CM
REFT
0.1µF
0.1µF
0.1µF
0.1µF
AC Input
Signal
+VS
GND
IN
R2
50Ω
IN
22pF
Mini-Circuits
TT1-6-KK81
or equivalent
(1)
22pF
GND
15
14
16
13
17
12
18
11
19
10
20
9
GND
24
25
26
27
28
7
14
6
15
5
16
4
17
3
18
2
DNC
1
LSB
19
9
12
8
13
7
14
6
15
5
16
4
17
3
18
2
7
6
5
4
3
2
1
MSB
GND
FIGURE 10. ADS820 Interface Schematic with AC-Coupling and External Buffers.
14
G+
11
1
19
®
Dir
IDT74FCT2245
NOTE: (1) All capacitors should be located as close to the pins as the manufacturing
process will allow. Ceramic X7R surface-mount capacitors or equivalent are recommended.
ADS820
8
13
DNC
ADS820
23
9
12
8
21
22
11
Dir
G+