The AD7621 is a 16-bit, 3 MSPS, charge redistribution SAR,
fully differential analog-to-digital converter (ADC) that
operates from a single 2.5 V power supply. It contains a high
speed 16-bit sampling ADC, an internal conversion clock, an
internal reference (and buffer), error correction circuits, and
both serial and parallel system interface ports. It features two
very high sampling rate modes (wideband warp and warp), a
fast mode (normal) for asynchronous rate applications, and a
reduced power mode (impulse) for low power applications
where the power is scaled with the throughput. Operation is
specified from −40°C to +85°C.
Rev. 0
Information furnished by Analog Devices is believed to be accurate and reliable.
However, no responsibility is assumed by Analog Devices for its use, nor for any
infringements of patents or other rights of third parties that may result from its use.
Specifications subject to change without notice. No license is granted by implication
or otherwise under any patent or patent rights of Analog Devices. Trademarks and
registered trademarks are the property of their respective owners.
PRODUCT HIGHLIGHTS
1. Fast Throughput.
The AD7621 is a 3 MSPS, charge redistribution,
16-bit SAR ADC.
2. Superior Linearity.
The AD7621 has no missing 16-bit code.
3. Internal Reference.
The AD7621 has a 2.048 V internal reference with a
typical drift of ±7 ppm/°C.
4. Single-Supply Operation.
The AD7621 operates from a 2.5 V single supply and
typically dissipates 65 mW. In impulse mod
e
dissipation decreases with th
Se
rial or Parallel Interface.
5.
Versatile parallel (16- or 8-bit bus) or 2-wire serial interf
arrangement compatible with 2.5 V, 3.3 V, or 5 V logic.
VIN+ − VIN−
Operating Input Voltage VIN+, VIN− to AGND −0.1 AVDD
Analog Input CMRR fIN = 100 kHz 55 dB
Input Current 3 MSPS throughput 25 μA
Input Impedance
CONVERSION AND RESET (Refer to Figure 31 and Figure 32)
Convert Pulse Width t
Time Between Conversions (Warp2 Mode/Normal Mode/Impulse Mode)3t
CNVST Low to BUSY High Delay
BUSY High All Modes (Except Master Serial Read After Convert) t
Aperture Delay t
End of Conversion to BUSY Low Delay t
Conversion Time (Warp Mode/Normal Mode/Impulse Mode) t
Acquisition Time (Warp Mode/Normal Mode/Impulse Mode) t
RESET Pulse Width t
RESET Low to BUSY High Delay
BUSY High Time from RESET Low
4
4
PARALLEL INTERFACE MODES (Refer to Figure 33 and Figure 35)
CNVST Low to DATA Valid Delay
(Warp Mode/Normal Mode/Impulse Mode)
DATA Valid to BUSY Low Delay t
Bus Access Request to DATA Valid t
Bus Relinquish Time t
MASTER SERIAL INTERFACE MODES5 (Refer to Figure 37 and Figure 38)
CS Low to SYNC Valid Delay
CS Low to Internal SCLK Valid Delay
5
CS Low to SDOUT Delay
CNVST Low to SYNC Delay
(Warp Mode/Normal Mode/Impulse Mode) t
SYNC Asserted to SCLK First Edge Delay t
Internal SCLK Period
Internal SCLK High
Internal SCLK Low
SDOUT Valid Setup Time
SDOUT Valid Hold Time
SCLK Last Edge to SYNC Delay
6
6
6
6
6
6
CS High to SYNC HI-Z
CS High to Internal SCLK HI-Z
CS High to SDOUT HI-Z
BUSY High in Master Serial Read after Convert
CNVST Low to SYNC Asserted Delay (All Modes)
SYNC Deasserted to BUSY Low Delay t
SLAVE SERIAL INTERFACE MODES5 (Refer to Figure 40 and Figure 41)
External SCLK Setup Time t
External SCLK Active Edge to SDOUT Delay t
SDIN Setup Time t
SDIN Hold Time t
External SCLK Period t
External SCLK High t
External SCLK Low t
1
See the Conversion Control section.
2
All timings for wideband warp mode are the same as warp mode.
3
In warp mode only, the time between conversions is 1 ms; otherwise, there is no required maximum time.
4
See the Digital Interface, and RESET sections.
5
In serial interface modes, the SYNC, SCLK, and SDOUT timings are defined with a maximum load CL of 10 pF; otherwise, the load is 60 pF maximum.
6
In serial master read during convert mode. See Table 4 for serial master read after convert mode timing specifications.
31
32
33
34
35
36
37
SERIAL CLOCK TIMING SPECIFICATIONS
Table 4. Serial Clock Timings in Master Read After Convert Mode
DIVSCLK[1] 0 0 1 1
DIVSCLK[0] Symbol 0 1 0 1 Unit
SYNC to SCLK First Edge Delay Minimum t
Internal SCLK Period Minimum t
Internal SCLK Period Maximum t
Internal SCLK High Minimum t
Internal SCLK Low Minimum t
SDOUT Valid Setup Time Minimum t
SDOUT Valid Hold Time Minimum t
SCLK Last Edge to SYNC Delay Minimum t
BUSY High Width Maximum (Wideband and Warp Modes) t
BUSY High Width Maximum (Normal Mode) t
BUSY High Width Maximum (Impulse Mode) t
NOTE
IN SERIAL INTERFACE MODES, THE SYNC, SCLK AND
SDOUT ARE DEFINED WITH A MAXIMUM LOAD.
C
OF 10pF; OTHERWISE, THE LOAD IS 60pF MAXIMUM.
L
Figure 2. Load Circuit for Digital Interface Timing,
, SYNC, and SCLK Outputs, C
SDOUT
OL
1.4V
OH
04565-002
= 10 pF
L
0.8V
t
DELAY
2V
0.8V
Figure 3. Voltage Reference Levels for Timing
Rev. 0 | Page 6 of 32
2V
t
DELAY
2V
0.8V
05665-003
Page 7
AD7621
www.BDTIC.com/ADI
ABSOLUTE MAXIMUM RATINGS
Table 5.
Parameter Rating
Analog Inputs/Outputs
IN+1, IN−, REF, REFBUFIN, TEMP,
INGND, REFGND to AGND
Ground Voltage Differences
AGND, DGND, OGND ±0.3 V
Supply Voltages
AVDD, DVDD –0.3 V to +2.7 V
OVDD –0.3 V to +3.8 V
AVDD to DVDD ±2.8 V
AVDD to OVDD +2.8 V to −3.8 V
OVDD to DVDD
Digital Inputs −0.3 V to +5.5 V
PDREF, PDBUF
Internal Power Dissipation
Internal Power Dissipation
Junction Temperature 125°C
Storage Temperature Range –65°C to +125°C
1
See the Analog Inputs section.
2
See the Power Supply section.
3
See the Voltage Reference Input section.
4
Specification is for the device in free air: 48-Lead LQFP; θJA = 91°C/W,
θJC = 30°C/W.
5
Specification is for the device in free air: 48-Lead LFCSP; θJA = 26°C/W.
2
3
4
5
AVDD + 0.3 V to
AGND − 0.3 V
≤ +0.3 V if DVDD < 2.3 V
±20 mA
700 mW
2.5 W
Stresses above those listed under Absolute Maximum Ratings
may cause permanent damage to the device. This is a stress
rating only; functional operation of the device at these or any
other conditions above those indicated in the operational
section of this specification is not implied. Exposure to absolute
maximum rating conditions for extended periods may affect
device reliability.
ESD CAUTION
ESD (electrostatic discharge) sensitive device.
Charged devices and circuit boards can discharge
without detection. Although this product features
patented or proprietary protection circuitry, damage may
occur on devices subjected to high energy ESD.
Therefore, proper ESD precautions should be taken to
performance degradation or loss of functionality.
avoid
Rev. 0 | Page 7 of 32
Page 8
AD7621
www.BDTIC.com/ADI
PIN CONFIGURATION AND FUNCTION DESCRIPTIONS
PDBUF
PDREF
REFBUFIN
TEMP
AVDD
IN+
AGND
AGNDNCIN–
DGND
D8/SDOUT
D9/SCLK
REFGND
D10/SYNC
48 47 46 45 4439 38 3743 42 41 40
1
AGND
AVDD
NC
BYTESWAP
OB/2C
WARP
IMPULSE
SER/PAR
D0
D1
D2/DIVSCLK[0]
D3/DIVSCLK[1]
NC = NO CONNECT
PIN 1
IDENTIFIER
2
3
4
5
6
7
8
9
10
11
12
13 14 15 16 17 18 19 20 21 22 23 24
D4/EXT/INT
D6/INVSCLK
D5/INVSYNC
AD7621
TOP VIEW
(Not to Scale)
DVDD
OVDD
OGND
D7/RDC/SDIN
Figure 4. Pin Configuration
Table 6. Pin Function Descriptions
Pin No. Mnemonic Type1 Description
1, 41, 42 AGND P Analog Power Ground Pin.
2, 44 AVDD P Input Analog Power Pins. Nominally 2.5 V.
3, 40 NC
4 BYTESWAP DI
No Connect.
Parallel Mode Selection (8-Bit/16-Bit). When high, the LSB is output on D[15:8] and the MSB is output
on D[7:0]; when low, the LSB is output on D[7:0] and the MSB is output on D[15:8].
5
OB/2C
DI
Straight Binary/Binary Twos Complement Output. When high, the digital output is straight binary;
when low, the MSB is inverted resulting in a twos complement output from its internal shift register.
6 WARP DI
Conversion Mode Selection. When WARP = high and IMPULSE = high, this selects wideband mode
with slightly improved linearity and THD. When WARP = high and IMPULSE = low, this selects warp
mode. In either mode, these are the fastest modes; maximum throughput is achievable, and a
minimum conversion rate must be applied in order to guarantee full specified accuracy.
When WARP = low and IMPULSE = low, this input selects normal mode where full accuracy is
maintained independent of the minimum conversion rate.
7 IMPULSE DI
Conversion Mode Selection. When IMPULSE = high and WARP = low, this input selects impulse mode,
a reduced power mode. In this mode, the power dissipation is approximately proportional to the
sampling rate.
8
SER/PAR
DI
Serial/Parallel Selection Input. When high, the serial interface is selected and some bits of the data bus
are used as a serial port; the remaining data bits are high impedance outputs. When SER/PAR
the parallel port is selected.
9, 10 D[0:1] DO Bit 0 and Bit 1 of the Parallel Port Data Output Bus.
11, 12 D[2:3] DI/O
or
DIVSCLK[0:1]
When SER/PAR = low, these outputs are used as Bit 2 and Bit 3 of the parallel port data output bus.
When SER/PAR
mode (EXT/INT
= high, serial clock division selection. When using serial master read after convert
= low, RDC/SDIN = low) these inputs can be used to slow down the internally
generated serial clock that clocks the data output. In other serial modes, these pins are high
impedance outputs.
13 D4 DI/O
or EXT/INT
When SER/PAR
When SER/PAR
= low, this output is used as Bit 4 of the parallel port data output bus.
= high, serial clock source select. This input is used to select the internally generated
(master ) or external (slave) serial data clock.
When EXT/INT
When EXT/INT
= low: master mode. The internal serial clock is selected on SCLK output.
= high: slave mode. The output data is synchronized to an external clock signal, gated
15 D6 DI/O
or INVSCLK Invert SCLK Select. In all serial modes, this input is used to invert the SCLK signal.
16 D7 DI/O Bit 7 of the Parallel Port Data Output Bus.
or RDC
or SDIN
17 OGND P Input/Output Interface Digital Power Ground.
18 OVDD P
19 DVDD P Digital Power. Nominally at 2.5 V.
20 DGND P Digital Power Ground.
21 D8 DO
or SDOUT
22 D9 DI/O
or SCLK
23 D10 DO
or SYNC
24 D11 DO
or RDERROR
25 to 28 D[12:15] DO Bit 12 to Bit 15 of the Parallel Port Data Output Bus.
29 BUSY DO
30 DGND P Digital Power Ground.
31
RD
DI
When SER/PAR
When SER/PAR
select the active state of the SYNC signal.
When INVSYNC = low, SYNC is active high.
When INVSYNC = high, SYNC is active low.
When SER/
When SER/PAR
used to select the read mode.
When RDC = high, the previous conversion result is read during
SCLK changes (see the Master Serial Interface section).
When RDC = low (read after convert), the current result is read after conversi
Serial Data In. When using serial slave mode, (EXT/INT
daisy-chain the conversion results from two or more ADCs onto a single SDOUT line. The digital data
level on SDIN is output on SDOUT with a delay of 16 SCLK periods after the initiation of the read
sequence.
Input/Output Interface Digital Power. Nomin
(2.5 V or 3 V).
When SER/PAR
When SER/PAR
synchronized to SCLK. Conversion results are stored in an on-chip register. The AD7621 provides the
conversion result, MSB first, from its internal shift register. The data format is determined by the logic
level of OB/2C.
In master mode (EXT/INT
In slave mode (EXT/INT
When INVSCLK = low, SDOUT is updated on SCLK rising edge and valid on the next falling edge.
When INVSCLK = high, SDOUT is updated on SCLK falling edge and valid on the next rising edge.
Parallel Port Data Output Bus Bit 9. When SER/PAR
data output bus.
Serial Clock. When SER/PAR
clock input or output, dependent upon the logic state of the EXT/INT
data SDOUT is updated depends upon the logic state of the INVSCLK pin.
When SER/PAR
When SER/PAR
used as a digital output frame synchronization for use with the internal data clock.
When a read sequence is initiated and INVSYNC = low, SYNC is driven high and remains high while
SDOUT output i
When a read sequence is initiated and INVSYNC = high, SYNC is driven low and remains low while
SDOUT output is valid.
Parallel Port Data Output Bus Bit 11. When SER/PAR
port data output bus.
Read Error. When SER/PAR
as an incomplete read error flag. If a data read is started and not completed when the current
conversion is complete, the current data is lost and RDERROR is pulsed high.
Busy Output. Transitions high when a conversion is started, and remains hig
complete and the data is latched into the on-chip shift register. The falling edge of BUSY can be used
as a data ready clock signal.
Read Data. When CS
= low this output is used as Bit 5 of the parallel port data output bus.
= high, invert sync select. In serial master mode (EXT/INT = low), this input is used to
PAR
= low this output is used as Bit 6 of the parallel port data output bus.
= high, read during convert. When using Serial Master mode (EXT/INT = low), RDC is
current conversion and the period of
on.
= high), SDIN could be used as a data input to
ally at the same supply as the supply of the host interface
= low this output is used as Bit 8 of the parallel port data output bus.
= high, serial data output. In serial mode, this pin is used as the serial data output
= low). SDOUT is valid on both edges of SCLK.
= high):
= low, this output is used as Bit 9 of the parallel port
= high, serial clock. In all serial modes, this pin is used as the serial data
pin. The active edge where the
= low, this output is used as Bit 10 of the parallel port data output bus.
= high, frame synchronization. In serial master mode (EXT/INT= low), this output is
s valid.
= low, this output is used as Bit 11 of the parallel
= high, read error. In serial slave mode (EXT/INT = high), this output is used
h until the conversion is
and RD are both low, the interface parallel or serial output bus is enabled.
Rev. 0 | Page 9 of 32
Page 10
AD7621
www.BDTIC.com/ADI
Pin No. Mnemonic Type1 Description
32
33 RESET DI
34 PD DI
35
36 AGND P Analog Power Ground Pin.
37 REF AI/O
38 REFGND AI Reference Input Analog Ground.
39 IN− AI Differential Negative Analog Input.
43 IN+ AI Differential Positive Analog Input.
45 TEMP AO Temperature Sensor Analog Output.
46 REFBUFIN AI/O
47 PDREF DI
48 PDBUF DI
1
AI = analog input; AI/O = bidirectional analog; AO = analog output; DI = digital input; DI/O = bidirectional digital; DO = digital output; P = power.
CS
CNVST
DI
DI
Chip Select. When CS and RD are both low, the interface parallel or serial output bus is enabled. CS is
also used to gate the external clock in slave serial mode.
Reset Input. When high, reset the AD7621. Current conversi
enables the calibration mode indicated by pulsing BUSY high. Refer to the Digital Interface section. If
not used, this pin can be tied to DGND.
Power-Down Input. When high, power down the ADC. Power consumption is reduced and
sions are inhibited after the current one is completed.
conver
Conversion Start. A falling edge on CNVST
initiates a conversion.
Reference Output/Input.
PDREF/PDBUF = low, the internal reference and buffer are enabled producing 2.048 V on this pin.
When
When PDREF/PDBUF = high, the internal reference and buffer are disabled allowing an externally
supplied voltage reference up to AVDD volts. Decoupling is required with or without the internal
reference and buffer. Refer to the Voltage Reference Input section.
Internal Reference Output/Reference Buffer Input.
When PDREF/PDBUF = low, the
(typical) bandgap output on this pin, which needs external decoupling. The internal fixed gain
reference buffer uses this to produce 2.048V on the REF pin.
When using an external reference with the internal reference buffer (PDBUF = low, PDREF = high),
applying 1.2 V on this pin produces 2.048 V on the REF pin. Refer to the Voltage Reference Input section.
Internal Reference Power-Down Input.
When low, the i
When high, the internal reference is powered down and an external reference must been used.
Internal Reference Buffer Power-Down Input.
When low, the buffer is enabled (must be low when using inter
When high, the buffer is powered-down.
nternal reference is enabled.
puts the internal sample-and-hold into the hold state and
internal reference and buffer are enabled producing the 1.2 V
on if any is aborted. Falling edge of RESET
nal reference).
Rev. 0 | Page 10 of 32
Page 11
AD7621
www.BDTIC.com/ADI
TERMINOLOGY
Integral Nonlinearity Error (INL)
Linearity error refers to the deviation of each individual code
from a line drawn from negative full-scale through positive fullscale. The point used as negative full-scale occurs ½ LSB before
the first code transition. Positive full-scale is defined as a level
1½ LSBs beyond the last code transition. The deviation is
measured from the middle of each code to the true straight line.
Differential Nonlinearity Error (DNL)
n an ideal ADC, code transitions are 1 LSB apart. Differential
I
nonlinearity is the maximum deviation from this ideal value. It
is often specified in terms of resolution for which no missing
codes are guaranteed.
Gain Error
irst transition (from 000…00 to 000…01) should occur for
The f
an analog voltage ½ LSB above the nominal negative full-scale
(−2.0479688 V for the ±2.048 V range). The last transition
(from 111…10 to 111…11) should occur for an analog voltage
1½ LSBs below the nominal full-scale (2.0479531 V for the
±2.048 V range). The gain error is the deviation of the
difference between the actual level of the last transition and the
actual level of the first transition from the difference between
the ideal levels.
Zero Error
The zer
o error is the difference between the ideal midscale
input voltage (0 V) and the actual voltage producing the
midscale output code.
Dynamic Range
D
ynamic range is the ratio of the rms value of the full-scale to
the rms noise measured with the inputs shorted together. The
value for dynamic range is expressed in decibels.
Signal-to-Noise Ratio (SNR)
S
NR is the ratio of the rms value of the actual input signal to the
rms sum of all other spectral components below the Nyquist
frequency, excluding harmonics and dc. The value for SNR is
expressed in decibels.
Signal-to-(Noise + Distortion) Ratio (SINAD)
INAD is the ratio of the rms value of the actual input signal to
S
the rms sum of all other spectral components below the Nyquist
frequency, including harmonics but excluding dc. The value for
SINAD is expressed in decibels.
Spurious-Free Dynamic Range (SFDR)
ference, in decibels (dB), between the rms amplitude of
The dif
the input signal and the peak spurious signal.
Effective Number of Bits (ENOB)
OB is a measurement of the resolution with a sine wave
EN
input. It is related to SINAD and is expressed in bits by
− 1.76)/6.02]
ENOB = [(SI
NAD
dB
Aperture Delay
A
perture delay is a measure of the acquisition performance
measured from the falling edge of the
CNVST
input to when
the input signal is held for a conversion.
Transi ent Res p onse
The t
ime required for the AD7621 to achieve its rated accuracy
after a full-scale step function is applied to its input.
Reference Voltage Temperature Coefficient
ence voltage temperature coefficient is derived from the
Refer
typical shift of output voltage at 25°C on a sample of parts at the
maximum and minimum reference output voltage (V
measured at T
MIN
REF
, T(25°C), and T
)(TCV×
Cppm/
. It is expressed in ppm/°C as
MAX
REFREF
×°=°C
)(V–)(V
MinMax
)T–T()25(V
MINMAXREF
)
REF
6
10
where:
V
(Max) = maximum V
REF
V
(Min) = minimum V
REF
(25°C) = V
V
REF
T
MAX
T
MIN
= +85°C
= –40°C
REF
at 25°C
REF
REF
at T
at T
MIN
MIN
, T (25°C), or T
, T (25°C), or T
MAX
MAX
Total Harmonic Distortion (THD)
HD is the ratio of the rms sum of the first five harmonic
T
components to the rms value of a full-scale input signal and is
expressed in decibels.
Rev. 0 | Page 11 of 32
Page 12
AD7621
www.BDTIC.com/ADI
TYPICAL PERFORMANCE CHARACTERISTICS
2.0
1.5
1.0
1.5
1.0
0.5
0
INL (LSB)
–0.5
–1.0
–1.5
–2.0
0163843276849152
CODE
Figure 5. Integral Nonlinearity vs. Code
160000
140000
120000
100000
80000
COUNTS
60000
40000
20000
013
0
7FFC 7FFD 7FFE 7FFF 8000 8001 8002 8003 8004
2080
149969
51847
CODE IN HEX
54791
2415
Figure 6. Histogram of 261,120 Conversions of a DC Input
at
the Code Center (External Reference)
2.0530
2.0525
2.0520
2.0515
2.0510
(V)
2.0505
REF
V
2.0500
2.0495
2.0490
2.0485
2.0480
–55–35 –15525456585105125
Figure 7. Typical Reference Voltage Out
TEMPERATURE (°C)
put vs. Temperature (3 Units)
65536
σ
= 0.69
50
04565-005
04565-006
04565-007
0.5
DNL (LSB)
0
–0.5
–1.0
0163843276849152
CODE
65536
Figure 8. Differential Nonlinearity vs. Code
160000
140000
120000
100000
80000
COUNTS
60000
40000
20000
0
0
7FFC 7FFD 7FFE 7FFF 8000 8001 8002 8003 8004
176
6644
127317
60842
CODE IN HEX
59637
6376
σ
128
= 0.82
0
Figure 9. Histogram of 261,120 Conversions of a DC Input
Center (Internal Reference)
–FS
ZERO ERROR
TEMPERATURE (°C)
10
8
6
4
2
0
–2
–4
–6
–8
ZERO SCALE, FULL-SCALE ERROR (LSB)
–10
–55 –35–15125105856545255
at the Code
+FS
Figure 10. Zero Error, Positive and Negative Full Scale vs. Temperature
Figure 17. SNR and SINAD vs. Input Level (Referred to Full Scale)
16
14
12
A)
μ
10
8
6
DVDD, OVDD (
4
2
0
–55 –35 –15525456585105125
AVDD
OVDD, 3.3V
OVDD, 2.5V
DVDD
TEMPERATURE (°C)
Figure 18. Power-Down Operating Currents vs. Temperature
280
270
260
250
240
230
220
210
200
0
04565-017
A)
μ
AVDD (
04565-019
1k
AVDD, IMPULSE
100
OPERATING CURRENTS (μA)
0.1
OVDD = 3.3V, ALL MODES
10
DVDD, ALL MODES
1
101001k10k100M1M10M
SAMPLING RATE (SPS)
OVDD, 2.5V, ALL MODES
PDREF = PDBUF = HIGH
Figure 19. Operating Current vs. Sample Rate
20
18
16
14
12
DELAY (ns)
12
10
t
8
6
4
050100150200
CL (pF)
OVDD = 2.5V @ 85°C
OVDD = 2.5V @ 25°C
OVDD = 3.3V @ 85°C
OVDD = 3.3V @ 25°C
Figure 20. Typical Delay vs. Load Capacitance C
04565-020
04565-022
L
Rev. 0 | Page 14 of 32
Page 15
AD7621
www.BDTIC.com/ADI
THEORY OF OPERATION
IN+
MSB
32,768C 16,384C4C2CCC
REF
REFGND
32,768C 16,384C4C2CCC
MSB
IN–
Figure 21. ADC Simplified Schematic
CIRCUIT INFORMATION
The AD7621 is a very fast, low power, single-supply, precise,
16-bit analog-to-digital converter (ADC) using successive
approximation architecture. The AD7621 features different
modes to optimize performances according to the applications.
In warp mode, the AD7621 is capable of converting 3,000,000
samples per second (3 MSPS).
The AD7621 provides the user with an on-chip track-and-hold,
s
uccessive approximation ADC that does not exhibit any
pipeline or latency, making it ideal for multiple multiplexed
channel applications.
The AD7621 can be operated from a single 2.5 V supply and
be
interfaced to either 5 V, 3.3 V, or 2.5 V digital logic. It
is housed in 48-lead LQFP or tiny LFCSP packages that
combine space savings with flexibility, allowing the AD7621
to be configured as either a serial or parallel interface. The
AD7621 is pin-to-pin-compatible with, and a speed upgrade
of, the AD7677.
CONVERTER OPERATION
The AD7621 is a successive approximation analog-to-digital
converter (ADC) based on a charge redistribution DAC. Figure 21
s
hows the simplified schematic of the ADC. The capacitive
DAC consists of two identical arrays of 16 binary weighted
capacitors which are connected to the two comparator inputs.
During the acquisition phase, terminals of the array tied to the
mparator’s input are connected to AGND via SW+ and SW−.
co
All independent switches are connected to the analog inputs.
Thus, the capacitor arrays are used as sampling capacitors and
acquire the analog signal on IN+ and IN− inputs. A conversion
phase is initiated once the acquisition phase is complete and the
CNVST
SW+ and SW− are opened first. The two capacitor arrays are
then disconnected from the inputs and connected to the
REFGND input. Therefore, the differential voltage between the
inputs (IN+ and IN−) captured at the end of the acquisition
input goes low. When the conversion phase begins,
AGND
SWITCHES
COMP
CONTROL
CONTROL
LOGIC
CNVST
BUSY
OUTPUT
CODE
04565-024
LSB
LSB
SW+
SW–
AGND
phase is applied to the comparator inputs, causing the
comparator to become unbalanced. By switching each element
of the capacitor array between REFGND and REF, the
comparator input varies by binary weighted voltage steps
(V
/2, V
REF
/4 through V
REF
/65536). The control logic toggles
REF
these switches, starting with the MSB first, in order to bring the
comparator back into a balanced condition. After the
completion of this process, the control logic generates the ADC
output code and brings BUSY output low.
MODES OF OPERATION
The AD7621 features four modes of operation: wideband warp,
warp, normal, and impulse. Each of these modes is more
suitable to specific applications.
Wideband warp (WARP = high, IMPULSE = high) and warp
ARP = high, IMPULSE = low) modes allow the fastest
(W
conversion rate up to 3 MSPS. However, in these modes, the full
specified accuracy is guaranteed only when the time between
conversions does not exceed 1 ms. If the time between two
consecutive conversions is longer than 1 ms (after power up),
the first conversion result should be ignored. These modes
make the AD7621 ideal for applications where both high
accuracy and fast sample rate are required. Wideband warp
mode offers slightly improved linearity and THD over warp
mode.
Normal mode (WARP = low, IMPULSE = low) is the fastest
ode (2 MSPS) without any limitation on time between
m
conversions. This mode makes the AD7621 ideal for
asynchronous applications such as data acquisition systems,
where both high accuracy and fast sample rate are required.
Impulse mode (WARP = low, IMPULSE = high), the lowest
ower dissipation mode, allows power saving between
p
conversions. The maximum throughput in this mode is
1.25 MSPS. In this mode, the ADC powers down circuits after
conversion making the AD7621 ideal for battery-powered
applications.
Rev. 0 | Page 15 of 32
Page 16
AD7621
www.BDTIC.com/ADI
TRANSFER FUNCTIONS
Using the OB/2C digital input, the AD7621 offers two output
codings: straight binary and twos complement. The LSB size
with V
Figure 22 and Table 7 for the ideal transfer characteristic.
This is also the code for overrange analog input (V
V
− V
Analog Input
REF
2
This is also the code for underrange analog input (V
−V
).
REFGND
+ V
REFGND
).
(2.5V OR 3.3V)
10μF
REF
Straight
Binary
DIGITAL
INTERFACE
SUPPLY
− V
IN+
− V
IN+
Twos
Complement
1
0x7FFF
2
0x8000
above
IN−
below
IN−
AVDD
AGND DGNDDVDD OVDDOGND
REF
C
REF
10μF
100nF
NOTE 4
NOTE 2
ANALOG
INPUT +
ANALOG
INPUT –
1. SEE ANALOG INPUT SECTION.
2. THE AD8021 IS RECOMMENDED. SEE DRIVER AMPLIFIER CHOICE SECTION.
3. THE CONFIGURATION SHOWN IS USING THE INTERNAL REFERENCE. SEE VOLTAGE REFERENCE INPUT SECTION.
4. A 10μF CERAMIC CAPACITOR (X5R, 1206 SIZE) IS RECOMMENDED (e.g., PANASONIC ECJ3YB0J106M).
SEE VOLTAGE REFERENCE INPUT SECTION.
5. OPTION, SEE POWER SUPPLY SECTION.
6. OPTION, SEE POWER UP SECTION.
7. OPTIONAL LOW JITTER CNVST, SEE CONVERSION CONTROL SECTION.
U1
C
NOTE 2
U2
C
10Ω
C
C
1nF
NOTE 1
10Ω
1nF
NOTE 1
REFBUFIN
REFGND
IN+
IN–
NOTE 3
AD7621
PD
PDREF
NOTE 3
PDBUF
SCLK
SDOUT
BUSY
CNVST
OB/2C
SER/PAR
WARP
IMPULSE
CS
RD
RESET
Figure 23. Typical Connection Diagram
50Ω
50pF
50pF
10kΩ
NOTE 6
SERIAL
PORT
NOTE 7
OVDD
D
CLOCK
MICROCONVERTER/
MICROPROCESSOR/
DSP
04565-026
Rev. 0 | Page 16 of 32
Page 17
AD7621
www.BDTIC.com/ADI
TYPICAL CONNECTION DIAGRAM
Figure 23 shows a typical connection diagram for the AD7621.
Different circuitry from that shown in this diagram are optional
and are discussed below.
ANALOG INPUTS
Figure 24 shows an equivalent circuit of the input structure of
the AD7621.
The two diodes, D
analog inputs, IN+ and IN−. Care must be taken to ensure that
the analog input signal never exceeds the supply rails by more
than 0.3 V as this causes the diodes to become forward-biased
and start conducting current. These diodes can handle a
forward-biased current of 100 mA maximum. For instance,
these conditions could eventually occur when the input buffer’s
U1 or U2 supplies are different from AVDD. In such a case, an
input buffer with a short-circuit current limitation can be used
to protect the part.
IN+ OR IN–
AGND
The analog input of AD7621 is a true differential structure. By
using this differential input, small signals common to both
inputs are rejected, as shown in Figure 25, representing the
t
ypical CMRR over frequency with internal and external
references.
75
70
65
60
CMRR (dB)
55
50
45
1101001k10k
During the acquisition phase for ac signals, the impedance of
the analog inputs, IN+ and IN−, can be modeled as a parallel
combination of Capacitor C
series connection of R
capacitance. R
and D2, provide ESD protection for the
1
AVDD
D
1
C
PIN
Figure 24. AD7621 Simplified Analog Input.
Figure 25. Analog Input C MRR vs. Frequency
is typically 350 Ω and is a lumped component
IN
D
EXT REF
FREQUENCY (kHz)
PIN
and CIN. C
IN
2
INT REF
and the network formed by the
is primarily the pin
PIN
C
R
IN
IN
04565-027
04565-099
comprised of some serial resistors and the on resistance of the
switches. C
is typically 12 pF and is mainly the ADC sampling
IN
capacitor. During the conversion phase, when the switches are
opened, the input impedance is limited to C
. RIN and CIN
PIN
make a one-pole, low-pass filter that has a typical −3 dB cutoff
frequency of 50 MHz, thereby reducing an undesirable aliasing
effect while limiting noise from the inputs.
Since the input impedance of the AD7621 is very high, the
AD7621 can
be directly driven by a low impedance source
without gain error. To further improve the noise filtering
achieved by the AD7621 analog input circuit, an external, onepole RC filter between the amplifier’s outputs and the ADC
analog inputs can be used, as shown in
rge source impedances significantly affect the ac performance,
la
Figure 23. However,
especially total harmonic distortion (THD). The maximum
source impedance depends on the amount of THD that can be
tolerated. The THD degrades as a function of the source
impedance and the maximum input frequency, as shown in
Figure 26.
–60
PDBUF = PDREF = LOW
–65
–70
–75
–80
–85
THD (dB)
–90
–95
–100
–105
1101001k
Figure 26. THD vs. Analog Input Frequency and Source Resistance
INPUT FREQUENCY (kHz)
R
= 50Ω
S
R
S
RS = 500Ω
= 100Ω
= 10Ω
R
S
04565-029
DRIVER AMPLIFIER CHOICE
Although the AD7621 is easy to drive, the driver amplifier
needs to meet the following requirements:
ogether, the driver amplifier and the AD7621 analog
•T
input circuit must be able to settle for a full-scale step of
the capacitor array at a 16-bit level (0.0015%). In the
amplifier data sheet, settling at 0.1% to 0.01% is more
commonly specified. This could differ significantly from
the settling time at a 16-bit level and should be verified
prior to driver selection. The AD8021 op amp, which
combines ultralow noise and high gain bandwidth, meets
this settling time requirement even when used with gains
up to 13.
•The
noise generated by the driver amplifier needs to be
kept as low as possible in order to preserve the SNR and
transition noise performance of the AD7621. The noise
Rev. 0 | Page 17 of 32
Page 18
AD7621
www.BDTIC.com/ADI
coming from the driver is filtered by the AD7621 analog
input circuit one-pole, low-pass filter made by R
or by the external filter, if one is used. The SNR
degradation due to the amplifier is
and CIN
IN
ANALOG INPUT
(UNIPOLAR 0V TO 2.048V)
U1
AD8021
10pF
SNR
LOSS
⎛
⎜
=
log20
⎜
⎜
2809
⎝
53
+
π
−23
⎞
⎟
⎟
⎟
()
Nef
NdB
⎠
where:
f
is the input bandwidth of the AD7621 (50 MHz) or
–3dB
the cutoff frequency of the input filter (16 MHz), if
one is used.
N is t
he noise factor of the amplifier (+1 in buffer
configuration).
e
is the equivalent input voltage noise density of the op
N
amp, in nV/√Hz.
For instance, a driver with an equivalent input noise
den
sity of 2.1 nV/√Hz, like the AD8021 with a noise gain
of +1 when configured as a buffer, degrades the SNR by
only 0.33 dB when using the RC filter in Figure 23, and by
1 dB wi
thout.
•The dr
iver needs to have a THD performance suitable to
that of the AD7621. Figure 13 gives the THD vs. frequency
t
hat the driver should exceed.
The AD8021
meets these requirements and is appropriate for
almost all applications. The AD8021 needs a 10 pF external
compensation capacitor that should have good linearity as an
NPO ceramic or mica type. Moreover, the use of a noninverting
+1 gain arrangement is recommended and helps to obtain the
best signal-to-noise ratio.
The AD8022
can also be used when a dual version is needed
and a gain of 1 is present. The AD829 is an alternative in
applications where high frequency (above 100 kHz) performance
is not required. In applications with a gain of 1, an 82 pF
compensation capacitor is required. The AD8610 is an option
when low bias current is needed in low frequency applications.
Single-to-Differential Driver
For applications using unipolar analog signals, a single-endedto-differential driver, as shown in Figure 27, allows for a
ferential input into the part. This configuration, when
dif
provided an input signal of 0 to V
±V
with midscale at V
REF
/2. The one-pole filter using R = 10 Ω
REF
, will produce a differential
REF
and C = 1 nF provides a corner frequency of 16 MHz.
If the application can tolerate more noise, the AD8139
ferential driver can be used.
dif
590Ω
590Ω
1kΩ
1kΩ
100nF
Figure 27. Single-Ended-to-Dif
U2
AD8021
10pF
(Internal Reference Buffer Used)
10Ω
1nF
10Ω
1nF
ferential Driver Circuit
IN+
IN–
AD7621
REF
10μF
VOLTAGE REFERENCE INPUT
The AD7621 allows the choice of either a very low temperature
drift internal voltage reference or an external reference.
Unlike many ADCs with internal references, the internal
r
eference of the AD7621 provides excellent performance and
can be used in almost all applications.
Internal Reference
(PDBUF = Low, PDREF = Low)
To use the internal reference, the PDREF and PDBUF inputs
must be low. This produces a 1.2 V band gap output on
REFBUFIN which, amplified by the internal buffer, results in a
2.048 V reference on the REF pin.
The internal reference is temperature-compensated to 2.048 V ±
10 mV
. The reference is trimmed to provide a typical drift of
7 ppm/°C. This typical drift characteristic is shown in
The output resistance of the REFBUFIN is 6.33 kΩ (minimum)
hen the internal reference is enabled. It is necessary to
w
decouple this with a ceramic capacitor greater than 100 nF.
Thus, the capacitor provides an RC filter for noise reduction.
Since the output impedance of REFBUFIN is typically 6.33 kΩ,
r
elative humidity (among other industrial contaminates) can
directly affect the drift characteristics of the reference. Typically,
a guard ring is used to reduce the effects of drift under such
circumstances. However, since the AD7621 has a fine lead pitch,
guarding this node is not practical. Therefore, in these
industrial and other types of applications, it is recommended to
use a conformal coating such as Dow Corning 1-2577 or
Humiseal 1B73.
External 1.2 V Reference and Internal Buffer
(PDREF = High, PBBUF = Low)
To use an external reference with the internal buffer, PDREF
should be high and PDBUF should be low. This powers down
the internal reference and allows the 1.2 V reference to be
applied to REFBUFIN.
Figure 7.
04565-030
Rev. 0 | Page 18 of 32
Page 19
AD7621
www.BDTIC.com/ADI
External Reference (PDBUF = High, PRBUF = High)
To use an external reference directly on the REF pin, PDREF
and PDBUF should both be high.
For improved drift performance, an external reference, such as
e AD780 or ADR431, can be used. The advantages of directly
th
using the external voltage reference are:
•S
NR and dynamic range improvement (about 1.7 dB)
resulting from the use of a reference voltage very close to
the supply (2.5 V) instead of a typical 2.048 V reference
when the internal reference is used. This is calculated by
048.2
⎞
=
⎛
log20SNR
⎜
⎝
ower savings when the internal reference is powered
• P
⎟
50.2
⎠
down (PBREF = PDBUF = high).
PDREF and PDBUF power down the internal reference and the
nternal reference buffer, respectively.
i
Reference Decoupling
Whether using an internal or external reference, the AD7621
voltage reference input (REF) has a dynamic input impedance;
therefore, it should be driven by a low impedance source with
efficient decoupling between the REF and REFGND inputs.
This decoupling depends on the choice of the voltage reference,
but usually consists of a low ESR capacitor connected to REF
and REFGND with minimum parasitic inductance. A 10 μF
(X5R, 1206 size) ceramic chip capacitor (or 47 μF tantalum
capacitor) is appropriate when using either the internal
reference or one of these recommended reference voltages:
• The lo
• The lo
• The lo
w noise, low temperature drift ADR431 and AD780
w power ADR291
w cost AD1582
The placement of the reference decoupling is also important to
th
e performance of the AD7621. The decoupling capacitor
should be mounted on the same side as the ADC right at the
REF pin with a thick PCB trace. The REFGND should also
connect to the reference decoupling capacitor with the shortest
distance.
Temperature Sensor
The TEMP pin measures the temperature of the AD7621. To
improve the calibration accuracy over the temperature range,
the output of the TEMP pin is applied to one of the inputs of the
analog switch (such as, ADG779), and the ADC itself is used to
measure its own temperature. This configuration is shown in
Figure 28.
ANALOG INPUT
(UNIPOLAR)
ADG779
AD8021
Figure 28. Use of the Temperature Sensor
C
C
IN+
TEMP
AD7621
TEMPERATURE
SENSOR
04565-031
POWER SUPPLY
The AD7621 uses three sets of power supply pins: an analog
2.5 V supply AVDD, a digital 2.5 V core supply DVDD, and a
digital input/output interface supply OVDD. The OVDD supply
allows direct interface with any logic working between 2.3 V
and 5.25 V. To reduce the number of supplies needed, the digital
core (DVDD) can be supplied through a simple RC filter from
the analog supply as shown in
Power Sequencing
The AD7621 is independent of power supply sequencing once
OVDD does not exceed DVDD by more than 0.3 V until
DVDD = 2.3 V during any time; for instance, at power-up or
power-down (see the
dditionally, it is very insensitive to power supply variations
A
Absolute Maximum Ratings section).
over a wide frequency range as shown in
75
70
65
60
PSRR (dB)
55
50
EXT REF
Figure 23.
Figure 29.
INT REF
For applications that use multiple AD7621 devices, it is more
ef
fective to use the internal reference buffer in order to buffer
45
1101001k10k
the reference voltage.
The voltage reference temperature coefficient (TC) directly
pacts full scale; therefore, in applications where full-scale
im
Figure 29. PSRR v s. Frequency
accuracy matters, care must be taken with the TC. For instance,
a ±15 ppm/°C TC of the reference changes full-scale by ±1 LSB/°C.
Rev. 0 | Page 19 of 32
FREQUENCY (kHz)
04565-098
Page 20
AD7621
www.BDTIC.com/ADI
Power-Up
At power-up, or returning to operational mode from the powerdown mode (PD = high), the AD7621 engages an initialization
process. During this time, the first 128 conversions should be
ignored or the RESET input could be pulsed to engage a faster
initialization process. Refer to the
RES
ET and timing details.
A simple power-on reset circuit, as shown in Figure 23, can be
ed to minimize the digital interface. As OVDD powers up, the
us
capacitor is shorted and brings RESET high; it is then charged
returning RESET to low. However, this circuit only works when
powering up the AD7621 because the power down mode (PD =
high) does not power down any of the supplies. As a result,
RESET is low.
POWER DISSIPATION VS. THROUGHPUT
In impulse mode, the AD7621 automatically reduces its power
consumption at the end of each conversion phase. During the
acquisition phase, the operating currents are very low which
allows a significant power saving when the conversion rate is
reduced (see
f
or very low power, battery-operated applications.
It should be noted that the digital interface remains active even
d
uring the acquisition phase. To reduce the operating digital
supply currents even further, drive the digital inputs close to the
power rails (that is, OVDD and OGND).
100k
10k
Figure 30). This feature makes the AD7621 ideal
WARP MODE POWER
Digital Interface section for
CONVERSION CONTROL
t
CNVST
CNVST
input. A falling edge
2
t
6
t
8
CNVST
should not
low time, t1, or until the end
Figure 23.
The AD7621 is controlled by the
CNVST
on
is all that is necessary to initiate a conversion.
Detailed timing diagrams of the conversion process are shown
in Figure 31. Once initiated, it cannot be restarted or aborted,
ven by the power-down input, PD, until the conversion is
e
complete. The
RD
signals.
CNVST
BUSY
MODE
CNVST
signal operates independently of CS and
t
1
t
1
t
t
3
t
5
4
CONVERTACQUIREACQUIRECONVERT
t
7
Figure 31. Basic Conversion Timing
For optimal performance, the rising edge of
occur after the maximum
of conversion.
CNVST
Although
is a digital signal, it should be designed with
special care with fast, clean edges, and levels with minimum
overshoot, undershoot, or ringing.
CNVST
The
trace should be shielded with ground and a low
value (such as 50 Ω) serial resistor termination should be added
close to the output of the component that drives this line. Also,
a 60 pF capacitor is recommended to further reduce the effects
of overshoot and undershoot as shown in
04565-034
IMPULSE MODE POWER
1k
POWER DISSIPATION (μW)
For applications where SNR is critical, the
have very low jitter. This can be achieved by using a dedicated
oscillator for
CNVST
generation, or by clocking
high frequency, low jitter clock, as shown in Figure 23.
100
1001k10k100k1M10M
SAMPLING RATE (SPS)
Figure 30. Power Dissipation vs. Sample Rate
PDREF = PDBUF = HIGH
04565-032
Rev. 0 | Page 20 of 32
CNVST
signal should
CNVST
with a
Page 21
AD7621
A
www.BDTIC.com/ADI
INTERFACES
DIGITAL INTERFACE
The AD7621 has a versatile digital interface that can be set up
as either a serial or parallel interface with the host system. The
serial interface is multiplexed on the parallel data bus. The
AD7621 digital interface also accommodates 2.5 V, 3.3 V, or 5 V
logic with either OVDD at 2.5 V or 3.3 V. OVDD defines the
logic high output voltage. In most applications, the OVDD
supply pin of the AD7621 is connected to the host system
interface 2.5 V or 3.3 V digital supply. Finally, by using the
2C
OB/
input pin, both twos complement or straight binary
coding can be used.
CS
The two signals,
one of these signals is high, the interface outputs are in high
impedance. Usually,
multicircuit applications and is held low in a single AD7621
RD
design.
is generally used to enable the conversion result on
the data bus.
RESET
The RESET input is used to reset the AD7621 and generate a
fast initialization. A rising edge on RESET aborts the current
conversion (if any) and tristates the data bus. The falling edge of
RESET clears the data bus and engages the initialization process
indicated by pulsing BUSY high. Conversions can take place
after the falling edge of BUSY. Refer to
ti
ming details.
RESET
and RD, control the interface. When at least
CS
allows the selection of each AD7621 in
Figure 32 for the RESET
t
9
PARALLEL INTERFACE
The AD7621 is configured to use the parallel interface when
PA R
SER/
Master Parallel Interface
Data can be continuously read by tying CS and RD low thus
requiring minimal microprocessor connections. However, in
this mode the data bus is always driven and cannot be used in
shared bus applications (unless the device is held in RESET).
Figure 33 details the timing for this mode.
CS = RD = 0
CNVST
BUSY
DATA
Slave Parallel Interface
In slave parallel reading mode, the data can be read either after
each conversion, which is during the next acquisition phase, or
during the following conversion, as shown in Figure 34 and
Figure 35, respectively. When the data is read during the
co
first half of the conversion phase. This avoids any potential
feedthrough between voltage transients on the digital interface
and the most critical analog conversion circuitry.
is held low.
t
1
t
10
t
t
3
BUS
Figure 33. Master Parallel Data Timing for Reading (Continuous Read)
PREVIOUS CONVERSION DATANEW DATA
4
t
11
nversion, it is recommended that it is read-only during the
04565-036
CNVST
DATA
BUSY
t
38
t
39
Figure 32. RESET Timing
CS
RD
t
8
04565-035
Rev. 0 | Page 21 of 32
BUSY
DAT
BUS
t
12
Figure 34. Slave Parallel Data Timing for Reading (Read After Convert)
CURRENT
CONVERSION
t
13
04565-037
Page 22
AD7621
www.BDTIC.com/ADI
CS = 0
CNVST,
RD
BUSY
t
DATA
BUS
t
12
Figure 35. Slave Parallel Data Timing for Reading (Read During Convert)
8-Bit Interface (Master or Slave)
The BYTESWAP pin allows a glueless interface to an 8-bit bus.
As shown in Figure 36, when BYTESWAP is low, the LSB byte is
output
on D[7:0] and the MSB is output on D[15:8]. When
BYTESWAP is high, the LSB and MSB bytes are swapped, and
the LSB is output on D[15:8] and the MSB is output on D[7:0].
By connecting BYTESWAP to an address line, the 16-bit data
can be read in two bytes on either D[15:8] or D[7:0]. This
interface can be used in both master and slave parallel reading
modes.
CS
RD
BYTESWAP
PINS D[15:8]
PINS D[7:0]
HI-Z
HI-Z
Figure 36. 8-Bit and 16-Bit Parallel Interface
t
1
3
PREVIOUS
CONVERSION
t
13
HIGH BYTELOW BYTE
t
12
LOW BYTEHIGH BYTE
t
4
HI-Z
t
t
12
13
HI-Z
04565-038
04565-039
SERIAL INTERFACE
The AD7621 is configured to use the serial interface when
PA R
SER/
is held high. The AD7621 outputs 16 bits of data,
MSB first, on the SDOUT pin. This data is synchronized with
the 16 clock pulses provided on the SCLK pin. The output data
is valid on both the rising and falling edge of the data clock.
MASTER SERIAL INTERFACE
Internal Clock
The AD7621 is configured to generate and provide the serial
INT
data clock SCLK when the EXT/
AD7621 also generates a SYNC signal to indicate to the host
when the serial data is valid. The serial clock SCLK and the
SYNC signal can be inverted, if desired. Depending on the read
during convert input, RDC/SDIN, the data can be read after
each conversion or during the following conversion.
a
nd Figure 38 show detailed timing diagrams of these two
mo
des.
Usually, because the AD7621 is used with a fast throughput, the
ter read during conversion mode is the most recommended
mas
serial mode. In this mode, the serial clock and data toggle at
appropriate instants, minimizing potential feedthrough between
digital activity and critical conversion decisions. In this mode,
the SCLK period changes since the LSBs require more time to
settle and the SCLK is derived from the SAR conversion cycle.
In read after conversion mode, unlike other modes, the BUSY
nal returns low after the 16 data bits are pulsed out and not at
sig
the end of the conversion phase resulting in a longer BUSY
width. As a result, the maximum throughput cannot be
achieved in this mode.
pin is held low. The
Figure 37
Rev. 0 | Page 22 of 32
Page 23
AD7621
S
www.BDTIC.com/ADI
CS, RD
CNVST
BUSY
SYNC
SCLK
DOUT
EXT/INT = 0
t
3
t
29
t
14
t
20
t
15
X
t
16
t
22
RDC/SDIN = 0INVSCLK = INVSYNC = 0
t
28
t
30
t
18
t
19
t
21
123141516
D15D14D2D1D0
t
23
t
24
t
25
t
26
t
27
04565-040
Figure 37. Master Serial Data Timing for Reading (Read After Convert)
RDC/SDIN = 1INVSCLK = INVSYNC = 0
23
t
25
t
24
t
26
t
27
04565-041
CS, RD
CNVST
BUSY
SYNC
SCLK
SDOUT
EXT/INT = 0
t
1
t
3
t
17
t
14
t
15
t
18
t
16
t
22
t
19
t20t
21
123141516
D15D14D2D1D0X
t
Figure 38. Master Serial Data Timing for Reading (Read Previous Conversion During Convert)
Rev. 0 | Page 23 of 32
Page 24
AD7621
www.BDTIC.com/ADI
SLAVE SERIAL INTERFACE
External Clock
The AD7621 is configured to accept an externally supplied
INT
serial data clock on the SCLK pin when the EXT/
held high. In this mode, several methods can be used to read
the data. The external serial clock is gated by
RD
are both low, the data can be read after each conversion or
CS
during the following conversion. The external clock can be
either a continuous or a discontinuous clock. A discontinuous
clock can be either normally high or normally low when
inactive.
d
Figure 40 and Figure 41 show the detailed timing
iagrams of these methods.
While the AD7621 is performing a bit decision, it is important
at voltage transients be avoided on digital input/output pins
th
or degradation of the conversion result could occur. This is
particularly important during the second half of the conversion
phase because the AD7621 provides error correction circuitry
that can correct for an improper bit decision made during the
first half of the conversion phase. For this reason, it is
recommended that when an external clock is being provided, it
is a discontinuous clock that is toggling only when BUSY is low
or, more importantly, that it does not transition during the latter
half of BUSY high.
External Discontinuous Clock Data Read After
Conversion
Though the maximum throughput cannot be achieved using
this mode, it is the most recommended of the serial slave
modes. Figure 40 shows the detailed timing diagrams of this
m
ethod. After a conversion is complete, indicated by BUSY
returning low, the conversion result can be read while both
are low. Data is shifted out MSB first with 16 clock
and
RD
pulses and is valid on the rising and falling edges of the clock.
Among the advantages of this method is the fact that
nversion performance is not degraded because there are no
co
voltage transients on the digital interface during the conversion
process. Another advantage is the ability to read the data at any
speed up to 80 MHz, which accommodates both the slow digital
host interface and the fastest serial reading.
Finally, in this mode only, the AD7621 provides a daisy-chain
f
eature using the RDC/SDIN pin for cascading multiple
converters together. This feature is useful for reducing
component count and wiring connections when desired, as, for
instance, in isolated multiconverter applications.
An example of the concatenation of two devices is shown in
Figure 39. Simultaneous sampling is possible by using a
pin is
. When CS and
CS
CNVST
mmon
co
signal. It should be noted that the RDC/SDIN
input is latched on the edge of SCLK opposite to the one used to
shift out the data on SDOUT. Hence, the MSB of the upstream
converter just follows the LSB of the downstream converter on
the next SCLK cycle.
BUSY
OUT
BUSYBUSY
AD7621
#2
(UPSTREAM)
RDC/SDINSDOUT
CNVST
CS
SCLK
SCLK IN
CS IN
CNVST IN
Figure 39. Two AD7621 Devices in a Daisy-Chain Configuration
AD7621
#1
(DOWNSTREAM)
SDOUTRDC/SDIN
CNVST
SCLK
CS
DATA
OUT
04565-042
External Clock Data Read During Previous Conversion
Figure 41 shows the detailed timing diagrams of this method.
During a conversion, while both
CS
and RD are low, the result
of the previous conversion can be read. The data is shifted out,
MSB first, with 16 clock pulses and is valid on both the rising
and falling edge of the clock. The 16 bits have to be read before
the current conversion is complete, otherwise; RDERROR is
pulsed high and can be used to interrupt the host interface to
prevent incomplete data reading. There is no daisy-chain
feature in this mode and RDC/SDIN input should always be
tied either high or low.
To reduce performance degradation due to digital activity, a fast
iscontinuous clock (at least 30 MHz when impulse mode is
d
used, 60 MHz when normal mode is used, or 80 MHz when
warp mode is used) is recommended to ensure that all the bits
are read during the first half of the SAR conversion phase.
It is also possible to begin to read data after conversion and
ntinue to read the last bits after a new conversion has been
co
initiated. However, this is not recommended when using the
fastest throughput of any mode since the acquisition times are
only 70 ns, 100 ns, and 50 ns for warp, normal, and impulse
modes.
If the maximum throughput is not used, thus allowing more
acq
uisition time, then the use of a slower clock speed can be
used to read the data.
Rev. 0 | Page 24 of 32
Page 25
AD7621
S
S
www.BDTIC.com/ADI
BUSY
SCLK
DOUT
SDIN
CS
CS
EXT/INT = 1
t
35
t36t
37
1231415161718
t
31
X
D15D14D1
t
16
t
33
t
34
X15X14X13X1X0Y15Y14
Figure 40. Slave Serial Data Timing for Reading (Read After Convert)
t
32
D13
EXT/INT = 1INVSCLK = 0
INVSCLK = 0
RD = 0
RD = 0
D0
X15X14
04565-043
CNVST
BUSY
SCLK
DOUT
t
3
t
31
t
16
t
35
t36t
37
123141516
t
32
X
D15D14D13
Figure 41. Slave Serial Data Timing for Reading (Read Previous Conversion During Convert)
D1
D0
04565-044
Rev. 0 | Page 25 of 32
Page 26
AD7621
www.BDTIC.com/ADI
MICROPROCESSOR INTERFACING
The AD7621 is ideally suited for traditional dc measurement
applications supporting a microprocessor, and ac signal
processing applications interfacing to a digital signal processor.
The AD7621 is designed to interface with a parallel 8-bit or
16-bit wide interface, or with a general-purpose serial port or
I/O ports on a microcontroller. A variety of external buffers can
be used with the AD7621 to prevent digital noise from coupling
into the ADC. The following section illustrates the use of the
AD7621 with an ADSP-219x SPI-equipped DSP.
SPI Interface (ADSP-219x)
Figure 42 shows an interface diagram between the AD7621 and
an SPI-equipped DSP, ADSP-219x. To accommodate the slower
speed of the DSP, the AD7621 acts as a slave device and data
must be read after conversion. This mode also allows the daisychain feature. The convert command could be initiated in
response to an internal timer interrupt. The reading process can
be initiated in response to the end-of-conversion signal (BUSY
going low) using an interrupt line of the DSP. The serial
p
eripheral interface (SPI) on the ADSP-219x is configured for
master mode (MSTR) = 1, clock polarity bit (CPOL) = 0, clock
phase bit (CPHA) = 1, and SPI interrupt enable (TIMOD) = 00
by writing to the SPI control register (SPICLTx). It should be
noted that to meet all timing requirements, the SPI clock should
be limited to 17 Mb/s allowing it to read an ADC result in less
than 1 μs. When a higher sampling rate is desired, use one of
the parallel interface modes.
DVDD
AD7621*
SER/PAR
EXT/INT
RD
INVSCLK
*ADDITIONAL PINS OMITTED FOR CLARITY
Figure 42. Interfacing the AD7621 to SPI Interface
BUSY
CS
SDOUT
SCLK
CNVST
ADSP-219x*
PFx
SPIxSEL (PFx)
MISOx
SCKx
PFx OR TFSx
04565-045
Rev. 0 | Page 26 of 32
Page 27
AD7621
www.BDTIC.com/ADI
APPLICATION
LAYOUT
While the AD7621 has very good immunity to noise on the
power supplies, exercise care with the grounding layout. To
facilitate the use of ground planes that can be easily separated,
design the printed circuit board that houses the AD7621 so that
the analog and digital sections are separated and confined to
certain areas of the board. Digital and analog ground planes
should be joined in only one place, preferably underneath the
AD7621, or as close as possible to the AD7621. If the AD7621 is
in a system where multiple devices require analog-to-digital
ground connections, the connections should still be made at
one point only, a star ground point, established as close as
possible to the AD7621.
To prevent coupling noise onto the die, avoid radiating noise,
d to reduce feedthrough:
an
• D
o not run digital lines under the device.
• Do r
un the analog ground plane under the AD7621.
CNVST
o shield fast switching signals, like
• D
digital ground to avoid radiating noise to other sections of
the board, and never run them near analog signal paths.
• A
void crossover of digital and analog signals.
• R
un traces on different but close layers of the board, at right
angles to each other, to reduce the effect of feedthrough
through the board.
The power supply lines to the AD7621 should use as large a
race as possible to provide low impedance paths and reduce the
t
effect of glitches on the power supply lines. Good decoupling is
also important to lower the impedance of the supplies presented
to the AD7621, and to reduce the magnitude of the supply
spikes. Decoupling ceramic capacitors, typically 100 nF, should
be placed on each of the power supplies pins, AVDD, DVDD,
and OVDD. The capacitors should be placed close to, and
ideally right up against, these pins and their corresponding
ground pins. Additionally, low ESR 10 μF capacitors should be
located in the vicinity of the ADC to further reduce low
frequency ripple.
or clocks, with
The DVDD supply of the AD7621 can be either a separate
pply or come from the analog supply, AVDD, or from the
su
digital interface supply, OVDD. When the system digital supply
is noisy, or fast switching digital signals are present, and no
separate supply is available, it is recommended to connect the
DVDD digital supply to the analog supply AVDD through an
RC filter, and to connect the system supply to the interface
digital supply OVDD and the remaining digital circuitry. Refer
to
Figure 23 for an example of this configuration. When DVDD
is
powered from the system supply, it is useful to insert a bead
to further reduce high frequency spikes.
The AD7621 has four different ground pins: REFGND, AGND,
D
GND, and OGND. REFGND senses the reference voltage and,
because it carries pulsed currents, should be a low impedance
return to the reference. AGND is the ground to which most
internal ADC analog signals are referenced; it must be
connected with the least resistance to the analog ground plane.
DGND must be tied to the analog or digital ground plane
depending on the configuration. OGND is connected to the
digital system ground.
The layout of the decoupling of the reference voltage is
portant. To minimize parasitic inductances, place the
im
decoupling capacitor close to the ADC and connect it with
short, thick traces.
EVALUATING THE AD7621 PERFORMANCE
A recommended layout for the AD7621 is outlined in the
documentation of the EVAL-AD7621CB evaluation board for
the AD7621. The evaluation board package includes a fully
assembled and tested evaluation board, documentation, and
software for controlling the board from a PC via the EVAL-