EVAL-AD7606-6SDZ [ADI]

8-/6-/4-Channel DAS with 16-Bit, Bipolar Input, Simultaneous Sampling ADC;
EVAL-AD7606-6SDZ
型号: EVAL-AD7606-6SDZ
厂家: ADI    ADI
描述:

8-/6-/4-Channel DAS with 16-Bit, Bipolar Input, Simultaneous Sampling ADC

文件: 总36页 (文件大小:955K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
8-/6-/4-Channel DAS with 16-Bit, Bipolar  
Input, Simultaneous Sampling ADC  
Data Sheet  
AD7606/AD7606-6/AD7606-4  
FEATURES  
APPLICATIONS  
8/6/4 simultaneously sampled inputs  
True bipolar analog input ranges: ±10 V, ±± V  
Single ± V analog supply and 2.3 V to ± V VDRIVE  
Fully integrated data acquisition solution  
Analog input clamp protection  
Power-line monitoring and protection systems  
Multiphase motor control  
Instrumentation and control systems  
Multiaxis positioning systems  
Data acquisition systems (DAS)  
Input buffer with 1 MΩ analog input impedance  
Second-order antialiasing analog filter  
On-chip accurate reference and reference buffer  
16-bit ADC with 200 kSPS on all channels  
Oversampling capability with digital filter  
Flexible parallel/serial interface  
SPI/QSPI™/MICROWIRE™/DSP compatible  
Performance  
7 kV ESD rating on analog input channels  
9±.± dB SNR, −107 dB THD  
Table 1. High Resolution, Bipolar Input, Simultaneous  
Sampling DAS Solutions  
Single-  
Ended  
Inputs  
True  
Number of  
Differential Simultaneous  
Inputs  
Resolution  
18 Bits  
Sampling Channels  
AD7608  
AD7606  
AD7606-6  
AD7606-4  
AD7607  
AD7609  
8
8
6
4
8
16 Bits  
14 Bits  
±0.± LSB INL, ±0.± LSB DNL  
Low power: 100 mW  
Standby mode: 2± mW  
Temperature range: −40°C to +8±°C  
64-lead LQFP package  
FUNCTIONAL BLOCK DIAGRAM  
AVCC  
AVCC  
REGCAP REGCAP  
REFCAPB REFCAPA  
RFB  
1MΩ  
V1  
CLAMP  
CLAMP  
T/H  
SECOND-  
ORDER LPF  
V1GND  
2.5V  
LDO  
2.5V  
LDO  
RFB  
RFB  
1MΩ  
1MΩ  
REFIN/REFOUT  
V2  
CLAMP  
CLAMP  
T/H  
T/H  
T/H  
T/H  
T/H  
T/H  
T/H  
SECOND-  
V2GND  
ORDER LPF  
RFB  
RFB  
1MΩ  
1MΩ  
REF SELECT  
AGND  
2.5V  
REF  
V3  
CLAMP  
CLAMP  
OS 2  
OS 1  
OS 0  
SECOND-  
ORDER LPF  
V3GND  
RFB  
RFB  
1MΩ  
1MΩ  
V4  
CLAMP  
CLAMP  
DOUT  
A
B
SECOND-  
ORDER LPF  
V4GND  
SERIAL  
RFB  
RFB  
1MΩ  
1MΩ  
DOUT  
8:1  
MUX  
PARALLEL/  
SERIAL  
RD/SCLK  
CS  
DIGITAL  
16-BIT  
SAR  
V5  
CLAMP  
CLAMP  
FILTER  
INTERFACE  
SECOND-  
ORDER LPF  
V5GND  
RFB  
RFB  
1MΩ  
1MΩ  
PAR/SER/BYTE SEL  
VDRIVE  
V6  
CLAMP  
CLAMP  
SECOND-  
ORDER LPF  
PARALLEL  
DB[15:0]  
V6GND  
RFB  
RFB  
1MΩ  
1MΩ  
AD7606  
V7  
CLAMP  
CLAMP  
SECOND-  
ORDER LPF  
V7GND  
CLK OSC  
RFB  
RFB  
1MΩ  
1MΩ  
BUSY  
CONTROL  
INPUTS  
V8  
CLAMP  
CLAMP  
SECOND-  
ORDER LPF  
FRSTDATA  
V8GND  
RFB  
1MΩ  
AGND  
CONVST A CONVST B RESET RANGE  
Figure 1.  
Rev. D  
Document Feedback  
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.  
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.  
Tel: 781.329.4700 ©2010–2017 Analog Devices, Inc. All rights reserved.  
Technical Support  
www.analog.com  
 
 
 
AD7606/AD7606-6/AD7606-4  
Data Sheet  
TABLE OF CONTENTS  
Features .............................................................................................. 1  
Analog Input............................................................................... 22  
ADC Transfer Function............................................................. 23  
Internal/External Reference...................................................... 24  
Typical Connection Diagram ................................................... 25  
Power-Down Modes .................................................................. 25  
Conversion Control ................................................................... 26  
Digital Interface.............................................................................. 27  
Applications....................................................................................... 1  
Functional Block Diagram .............................................................. 1  
Revision History ............................................................................... 2  
General Description......................................................................... 3  
Specifications..................................................................................... 4  
Timing Specifications .................................................................. 7  
Absolute Maximum Ratings.......................................................... 11  
Thermal Resistance .................................................................... 11  
ESD Caution................................................................................ 11  
Pin Configurations and Function Descriptions ......................... 12  
Typical Performance Characteristics ........................................... 17  
Terminology .................................................................................... 21  
Theory of Operation ...................................................................... 22  
Converter Details........................................................................ 22  
PAR  
Parallel Interface (  
/SER/BYTE SEL = 0).......................... 27  
/SER/BYTE SEL = 1, DB15 = 1)............... 27  
PAR  
PAR  
Parallel Byte (  
Serial Interface (  
/SER/BYTE SEL = 1)............................. 27  
Reading During Conversion..................................................... 28  
Digital Filter ................................................................................ 29  
Layout Guidelines....................................................................... 32  
Outline Dimensions....................................................................... 34  
Ordering Guide .......................................................................... 34  
REVISION HISTORY  
11/2017—Rev. C to Rev. D  
Changes to Features Section............................................................ 1  
Changes to Specifications Table Summary ................................... 3  
Deleted Endnote 1, Table 1; Renumbered Sequentially .............. 6  
Change to Table 6 ........................................................................... 14  
Changes to Typical Performance Characteristics Section......... 17  
Changes to Terminology Section.................................................. 21  
Changes to Ordering Guide .......................................................... 34  
1/2012—Rev. B to Rev. C  
Changes to Analog Input Ranges Section ................................... 22  
10/2011—Rev. A to Rev. B  
Changes to Input High Voltage (VINH) and Input Low Voltage  
(VINL) Parameters and Endnote 6, Table 2..................................... 4  
Changes to Table 3............................................................................ 7  
Changes to Table 4.......................................................................... 11  
Changes to Pin 32 Description, Table 6....................................... 13  
Changes to Analog Input Clamp Protection Section................. 22  
Changes to Typical Connection Diagram Section..................... 25  
8/2010—Rev. 0 to Rev. A  
Changes to Note 1, Table 2 .............................................................. 6  
5/2010—Revision 0: Initial Version  
Rev. D | Page 2 of 36  
 
Data Sheet  
AD7606/AD7606-6/AD7606-4  
GENERAL DESCRIPTION  
The AD76061/AD7606-6/AD7606-4 are 16-bit, simultaneous  
sampling, analog-to-digital data acquisition systems (DAS) with  
eight, six, and four channels, respectively. Each part contains  
analog input clamp protection, a second-order antialiasing filter,  
a track-and-hold amplifier, a 16-bit charge redistribution successive  
approximation analog-to-digital converter (ADC), a flexible  
digital filter, a 2.5 V reference and reference buffer, and high  
speed serial and parallel interfaces.  
all channels. The input clamp protection circuitry can tolerate  
voltages up to 16.5 V. The AD7606 has 1 MΩ analog input  
impedance regardless of sampling frequency. The single supply  
operation, on-chip filtering, and high input impedance eliminate  
the need for driver op amps and external bipolar supplies. The  
AD7606/AD7606-6/AD7606-4 antialiasing filter has a 3 dB cutoff  
frequency of 22 kHz and provides 40 dB antialias rejection when  
sampling at 200 kSPS. The flexible digital filter is pin driven, yields  
improvements in SNR, and reduces the 3 dB bandwidth.  
The AD7606/AD7606-6/AD7606-4 operate from a single 5 V  
supply and can accommodate 10 V and 5 V true bipolar input  
signals while sampling at throughput rates up to 200 kSPS for  
1 Patent pending.  
Rev. D | Page 3 of 36  
 
AD7606/AD7606-6/AD7606-4  
SPECIFICATIONS  
Data Sheet  
VREF = 2.5 V external/internal, AVCC = 4.75 V to 5.25 V, VDRIVE = 2.3 V to 5.25 V, fSAMPLE = 200 kSPS, TA = −40°C to +85°C, unless otherwise noted.  
Table 2.  
Parameter  
Test Conditions/Comments  
Min  
Typ  
Max  
Unit  
DYNAMIC PERFORMANCE  
Signal-to-Noise Ratio (SNR)1, 2  
fIN = 1 kHz sine wave unless otherwise noted  
Oversampling by 16; 10 V range; fIN = 130 Hz 94  
Oversampling by 16; 5 V range; fIN = 130 Hz 93  
No oversampling; 10 V Range  
No oversampling; 5 V range  
No oversampling; 10 V range  
No oversampling; 5 V range  
No oversampling; 10 V range  
No oversampling; 5 V range  
95.5  
94.5  
90  
89  
90  
89  
90.5  
90  
−107  
−108  
dB  
dB  
dB  
dB  
dB  
dB  
dB  
dB  
dB  
dB  
88.5  
87.5  
88  
Signal-to-(Noise + Distortion) (SINAD)1  
Dynamic Range  
87  
Total Harmonic Distortion (THD)1  
Peak Harmonic or Spurious Noise (SFDR)1  
Intermodulation Distortion (IMD)1  
Second-Order Terms  
Third-Order Terms  
Channel-to-Channel Isolation1  
−95  
fa = 1 kHz, fb = 1.1 kHz  
−110  
−106  
−95  
dB  
dB  
dB  
fIN on unselected channels up to 160 kHz  
ANALOG INPUT FILTER  
Full Power Bandwidth  
−3 dB, 10 V range  
−3 dB, 5 V range  
−0.1 dB, 10 V range  
−0.1 dB, 5 V range  
10 V Range  
23  
15  
10  
5
11  
15  
kHz  
kHz  
kHz  
kHz  
µs  
tGROUP DELAY  
5 V Range  
µs  
DC ACCURACY  
Resolution  
No missing codes  
16  
Bits  
Differential Nonlinearity1  
Integral Nonlinearity1  
Total Unadjusted Error (TUE)  
0.5  
0.5  
6
0.99  
2
LSB3  
LSB  
LSB  
10 V range  
5 V range  
12  
8
8
2
7
LSB  
LSB  
LSB  
ppm/°C  
ppm/°C  
LSB  
LSB  
LSB  
Positive Full-Scale Error1, 4  
External reference  
Internal reference  
External reference  
Internal reference  
10 V range  
32  
Positive Full-Scale Error Drift  
Positive Full-Scale Error Matching1  
Bipolar Zero Code Error1, 5  
5
32  
40  
6
5 V range  
10 V range  
16  
1
5 V range  
3
12  
LSB  
Bipolar Zero Code Error Drift  
Bipolar Zero Code Error Matching1  
Negative Full-Scale Error1, 4  
10 V range  
5 V range  
10 V range  
5 V range  
External reference  
Internal reference  
External reference  
Internal reference  
10 V range  
10  
5
1
µV/°C  
µV/°C  
LSB  
LSB  
LSB  
8
22  
6
8
8
4
8
5
16  
32  
LSB  
Negative Full-Scale Error Drift  
Negative Full-Scale Error Matching1  
ppm/°C  
ppm/°C  
LSB  
32  
40  
5 V range  
LSB  
Rev. D | Page 4 of 36  
 
Data Sheet  
AD7606/AD7606-6/AD7606-4  
Parameter  
Test Conditions/Comments  
Min  
Typ  
Max  
Unit  
ANALOG INPUT  
Input Voltage Ranges  
RANGE = 1  
RANGE = 0  
10  
5
V
V
Analog Input Current  
10 V; see Figure 31  
5 V; see Figure 31  
5.4  
2.5  
5
µA  
µA  
pF  
MΩ  
Input Capacitance6  
Input Impedance  
See the Analog Input section  
1
REFERENCE INPUT/OUTPUT  
Reference Input Voltage Range  
DC Leakage Current  
See the ADC Transfer Function section  
2.475  
2.5  
7.5  
2.49/  
2.505  
2.525  
1
V
µA  
pF  
V
Input Capacitance6  
REF SELECT = 1  
REFIN/REFOUT  
Reference Output Voltage  
Reference Temperature Coefficient  
LOGIC INPUTS  
10  
ppm/°C  
Input High Voltage (VINH  
)
0.7 × VDRIVE  
V
Input Low Voltage (VINL  
Input Current (IIN)  
Input Capacitance (CIN)6  
LOGIC OUTPUTS  
)
0.3 × VDRIVE  
2
V
µA  
pF  
5
Output High Voltage (VOH  
)
ISOURCE = 100 µA  
ISINK = 100 µA  
VDRIVE − 0.2  
V
V
µA  
pF  
Output Low Voltage (VOL  
)
0.2  
20  
Floating-State Leakage Current  
Floating-State Output Capacitance6  
Output Coding  
1
5
Twos complement  
CONVERSION RATE  
Conversion Time  
Track-and-Hold Acquisition Time  
Throughput Rate  
All eight channels included; see Table 3  
Per channel, all eight channels included  
4
1
µs  
µs  
kSPS  
200  
POWER REQUIREMENTS  
AVCC  
VDRIVE  
4.75  
2.3  
5.25  
5.25  
V
V
ITOTAL  
Digital inputs = 0 V or VDRIVE  
AD7606  
Normal Mode (Static)  
16  
14  
12  
22  
20  
17  
mA  
mA  
mA  
AD7606-6  
AD7606-4  
fSAMPLE = 200 kSPS  
AD7606  
AD7606-6  
Normal Mode (Operational)7  
20  
18  
15  
5
27  
24  
21  
8
mA  
mA  
mA  
mA  
µA  
AD7606-4  
Standby Mode  
Shutdown Mode  
2
6
Rev. D | Page 5 of 36  
AD7606/AD7606-6/AD7606-4  
Data Sheet  
Parameter  
Test Conditions/Comments  
Min  
Typ  
Max  
Unit  
Power Dissipation  
Normal Mode (Static)  
Normal Mode (Operational)7  
AD7606  
fSAMPLE = 200 kSPS  
AD7606  
AD7606-6  
AD7606-4  
80  
115.5  
mW  
100  
90  
75  
25  
10  
142  
126  
111  
42  
mW  
mW  
mW  
mW  
µW  
Standby Mode  
Shutdown Mode  
31.5  
1See the Terminology section.  
2 This specification applies when reading during a conversion or after a conversion. If reading during a conversion in parallel mode with VDRIVE = 5 V, SNR typically reduces by 1.5 dB  
and THD by 3 dB.  
3 LSB means least significant bit. With 5 V input range, 1 LSB = 152.58 µV. With 10 V input range, 1 LSB = 305.175 µV.  
4 These specifications include the full temperature range variation and contribution from the internal reference buffer but do not include the error contribution from  
the external reference.  
5 Bipolar zero code error is calculated with respect to the analog input voltage. See the Analog Input Clamp Protection section.  
6 Sample tested during initial release to ensure compliance.  
7 Operational power/current figure includes contribution when running in oversampling mode.  
Rev. D | Page 6 of 36  
 
 
 
 
 
 
Data Sheet  
AD7606/AD7606-6/AD7606-4  
TIMING SPECIFICATIONS  
AVCC = 4.75 V to 5.25 V, VDRIVE = 2.3 V to 5.25 V, VREF = 2.5 V external reference/internal reference, TA = TMIN to TMAX, unless otherwise noted.1  
Table 3.  
Limit at TMIN, TMAX  
(0.1 × VDRIVE and  
0.9 × VDRIVE  
Limit at TMIN, TMAX  
(0.3 × VDRIVE and  
0.7 × VDRIVE  
Logic Input Levels)  
Logic Input Levels)  
Parameter  
Min  
Typ  
Max  
Min  
Typ Max Unit  
Description  
PARALLEL/SERIAL/BYTE MODE  
tCYCLE  
1/throughput rate  
5
5
µs  
µs  
Parallel mode, reading during or after conversion; or  
serial mode: VDRIVE = 3.3 V to 5.25 V, reading during a  
conversion using DOUTA and DOUTB lines  
9.4  
Serial mode reading after a conversion; VDRIVE = 2.7 V  
9.7  
10.7 µs  
Serial mode reading after a conversion; VDRIVE = 2.3 V,  
DOUTA and DOUTB lines  
2
tCONV  
Conversion time  
3.45  
4
3
2
4.15  
3.45  
4
3
2
4.15 µs  
Oversampling off; AD7606  
Oversampling off; AD7606-6  
Oversampling off; AD7606-4  
Oversampling by 2; AD7606  
Oversampling by 4; AD7606  
Oversampling by 8; AD7606  
Oversampling by 16; AD7606  
Oversampling by 32; AD7606  
Oversampling by 64; AD7606  
µs  
µs  
7.87  
16.05  
33  
66  
133  
257  
9.1  
18.8  
39  
7.87  
16.05  
33  
66  
133  
257  
9.1  
µs  
18.8 µs  
39  
78  
158  
315  
100  
µs  
µs  
µs  
µs  
µs  
78  
158  
315  
100  
tWAKE-UP STANDBY  
STBY  
rising edge to CONVST x rising edge; power-up  
time from standby mode  
tWAKE-UP SHUTDOWN  
Internal Reference  
30  
13  
30  
13  
ms  
ms  
STBY  
rising edge to CONVST x rising edge; power-up  
time from shutdown mode  
STBY  
External Reference  
rising edge to CONVST x rising edge; power-up  
time from shutdown mode  
tRESET  
tOS_SETUP  
tOS_HOLD  
50  
20  
20  
50  
20  
20  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ms  
RESET high pulse width  
BUSY to OS x pin setup time  
BUSY to OS x pin hold time  
CONVST x high to BUSY high  
Minimum CONVST x low pulse  
Minimum CONVST x high pulse  
t1  
t2  
t3  
t4  
40  
45  
25  
25  
0
25  
25  
0
CS  
BUSY falling edge to falling edge setup time  
3
t5  
0.5  
25  
0.5  
25  
Maximum delay allowed between CONVST A, CONVST  
B rising edges  
t6  
ns  
ns  
CS  
Maximum time between last rising edge and BUSY  
falling edge  
Minimum delay between RESET low to CONVST x high  
t7  
25  
25  
PARALLEL/BYTE READ  
OPERATION  
t8  
0
0
0
0
ns  
ns  
CS RD  
to  
setup time  
hold time  
t9  
CS RD  
to  
t10  
RD  
low pulse width  
16  
21  
25  
32  
15  
22  
19  
24  
30  
37  
15  
22  
ns  
ns  
ns  
ns  
ns  
ns  
VDRIVE above 4.75 V  
VDRIVE above 3.3 V  
VDRIVE above 2.7 V  
VDRIVE above 2.3 V  
t11  
t12  
RD  
CS  
high pulse width  
high pulse width (see Figure 5); and  
CS  
RD  
linked  
Rev. D | Page 7 of 36  
 
 
AD7606/AD7606-6/AD7606-4  
Data Sheet  
Limit at TMIN, TMAX  
(0.1 × VDRIVE and  
0.9 × VDRIVE  
Limit at TMIN, TMAX  
(0.3 × VDRIVE and  
0.7 × VDRIVE  
Logic Input Levels)  
Logic Input Levels)  
Parameter  
Min  
Typ  
Max  
Min  
Typ Max Unit  
Description  
t13  
CS  
Delay from until DB[15:0] three-state disabled  
VDRIVE above 4.75 V  
16  
20  
25  
30  
19  
24  
30  
37  
ns  
ns  
ns  
ns  
VDRIVE above 3.3 V  
VDRIVE above 2.7 V  
VDRIVE above 2.3 V  
4
t14  
RD  
falling edge  
Data access time after  
VDRIVE above 4.75 V  
VDRIVE above 3.3 V  
VDRIVE above 2.7 V  
VDRIVE above 2.3 V  
16  
21  
25  
32  
19  
24  
30  
37  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
t15  
t16  
t17  
6
6
6
6
RD  
Data hold time after  
falling edge  
CS  
to DB[15:0] hold time  
22  
22  
CS  
Delay from rising edge to DB[15:0] three-state  
enabled  
SERIAL READ OPERATION  
fSCLK  
Frequency of serial read clock  
VDRIVE above 4.75 V  
VDRIVE above 3.3 V  
23.5  
17  
20  
15  
MHz  
MHz  
14.5  
11.5  
12.5 MHz  
VDRIVE above 2.7 V  
VDRIVE above 2.3 V  
10  
MHz  
t18  
CS  
Delay from until DOUTA/DOUTB three-state  
CS  
disabled/delay from until MSB valid  
15  
20  
30  
18  
23  
35  
ns  
ns  
ns  
VDRIVE above 4.75 V  
VDRIVE above 3.3 V  
VDRIVE = 2.3 V to 2.7 V  
Data access time after SCLK rising edge  
VDRIVE above 4.75 V  
VDRIVE above 3.3 V  
VDRIVE above 2.7 V  
VDRIVE above 2.3 V  
4
t19  
17  
23  
27  
34  
20  
26  
32  
39  
ns  
ns  
ns  
ns  
ns  
ns  
t20  
t21  
t22  
t23  
0.4 tSCLK  
0.4 tSCLK  
7
0.4 tSCLK  
0.4 tSCLK  
7
SCLK low pulse width  
SCLK high pulse width  
SCLK rising edge to DOUTA/DOUTB valid hold time  
22  
22  
ns  
CS  
rising edge to DOUTA/DOUTB three-state enabled  
FRSTDATA OPERATION  
t24  
CS  
Delay from falling edge until FRSTDATA three-  
state disabled  
15  
20  
25  
30  
18  
23  
30  
35  
ns  
ns  
ns  
ns  
ns  
VDRIVE above 4.75 V  
VDRIVE above 3.3 V  
VDRIVE above 2.7 V  
VDRIVE above 2.3 V  
t25  
CS  
Delay from falling edge until FRSTDATA high,  
serial mode  
15  
20  
25  
30  
18  
23  
30  
35  
ns  
ns  
ns  
ns  
VDRIVE above 4.75 V  
VDRIVE above 3.3 V  
VDRIVE above 2.7 V  
VDRIVE above 2.3 V  
t26  
RD  
falling edge to FRSTDATA high  
Delay from  
16  
20  
25  
30  
19  
23  
30  
35  
ns  
ns  
ns  
ns  
VDRIVE above 4.75 V  
VDRIVE above 3.3 V  
VDRIVE above 2.7 V  
VDRIVE above 2.3 V  
Rev. D | Page 8 of 36  
Data Sheet  
AD7606/AD7606-6/AD7606-4  
Limit at TMIN, TMAX  
(0.1 × VDRIVE and  
0.9 × VDRIVE  
Limit at TMIN, TMAX  
(0.3 × VDRIVE and  
0.7 × VDRIVE  
Logic Input Levels)  
Logic Input Levels)  
Parameter  
Min  
Typ  
Max  
Min  
Typ Max Unit  
Description  
t27  
RD  
falling edge to FRSTDATA low  
Delay from  
19  
24  
22  
29  
ns  
ns  
VDRIVE = 3.3 V to 5.25V  
VDRIVE = 2.3 V to 2.7V  
Delay from 16th SCLK falling edge to FRSTDATA low  
VDRIVE = 3.3 V to 5.25V  
t28  
17  
22  
24  
20  
27  
29  
ns  
ns  
ns  
VDRIVE = 2.3 V to 2.7V  
t29  
CS  
Delay from rising edge until FRSTDATA three-  
state enabled  
1 Sample tested during initial release to ensure compliance. All input signals are specified with tR = tF = 5 ns (10% to 90% of VDRIVE) and timed from a voltage level of 1.6 V.  
2 In oversampling mode, typical tCONV for the AD7606-6 and AD7606-4 can be calculated using ((N × tCONV) + ((N − 1) × 1 µs)). N is the oversampling ratio. For the AD7606-6,  
tCONV = 3 µs; and for the AD7606-4, tCONV = 2 µs.  
3 The delay between the CONVST x signals was measured as the maximum time allowed while ensuring a <10 LSB performance matching between channel sets.  
4 A buffer is used on the data output pins for these measurements, which is equivalent to a load of 20 pF on the output pins.  
Timing Diagrams  
t5  
CONVST A,  
CONVST B  
tCYCLE  
t2  
CONVST A,  
CONVST B  
t3  
tCONV  
t1  
BUSY  
t4  
CS  
t7  
tRESET  
RESET  
Figure 2. CONVST Timing—Reading After a Conversion  
t5  
CONVST A,  
CONVST B  
tCYCLE  
t2  
CONVST A,  
CONVST B  
t3  
tCONV  
t1  
BUSY  
t6  
CS  
t7  
tRESET  
RESET  
Figure 3. CONVST Timing—Reading During a Conversion  
Rev. D | Page 9 of 36  
 
 
 
 
 
 
AD7606/AD7606-6/AD7606-4  
Data Sheet  
CS  
t9  
t8  
t11  
t10  
t16  
t17  
RD  
t13  
t14  
V3  
t15  
V7  
DATA:  
INVALID  
DB[15:0]  
V1  
V2  
t27  
V4  
V8  
t26  
t29  
t24  
FRSTDATA  
CS  
RD  
Pulses  
Figure 4. Parallel Mode, Separate and  
t12  
CS AND RD  
t16  
t13  
t17  
DATA:  
V1  
V2  
V3  
V4  
V5  
V6  
V7  
V8  
DB[15:0]  
FRSTDATA  
CS  
RD  
Figure 5. and , Linked Parallel Mode  
CS  
t21  
t20  
SCLK  
t19  
t22  
DB1  
t23  
t18  
D
A,  
OUT  
DB15  
t25  
DB14  
DB13  
DB0  
t28  
D
B
OUT  
t29  
FRSTDATA  
Figure 6. Serial Read Operation (Channel 1)  
CS  
RD  
t8  
t9  
t10  
t11  
t16  
t17  
t13  
t15  
t14  
HIGH  
LOW  
BYTE V1  
HIGH  
BYTE V8  
LOW  
DATA: DB[7:0]  
FRSTDATA  
INVALID  
BYTE V1  
BYTE V8  
t26  
t29  
t27  
t24  
Figure 7. BYTE Mode Read Operation  
Rev. D | Page 10 of 36  
 
 
 
Data Sheet  
AD7606/AD7606-6/AD7606-4  
ABSOLUTE MAXIMUM RATINGS  
TA = 25°C, unless otherwise noted.  
THERMAL RESISTANCE  
Table 4.  
θJA is specified for the worst-case conditions, that is, a device  
soldered in a circuit board for surface-mount packages. These  
specifications apply to a 4-layer board.  
Parameter  
Rating  
AVCC to AGND  
VDRIVE to AGND  
−0.3 V to +7 V  
−0.3 V to AVCC + 0.3 V  
16.5 V  
−0.3 V to VDRIVE + 0.3 V  
−0.3 V to VDRIVE + 0.3 V  
−0.3 V to AVCC + 0.3 V  
10 mA  
Analog Input Voltage to AGND1  
Digital Input Voltage to AGND  
Digital Output Voltage to AGND  
REFIN to AGND  
Input Current to Any Pin Except Supplies1  
Operating Temperature Range  
B Version  
Table 5. Thermal Resistance  
Package Type  
θJA  
θJC  
Unit  
64-Lead LQFP  
45  
11  
°C/W  
ESD CAUTION  
−40°C to +85°C  
−65°C to +150°C  
150°C  
Storage Temperature Range  
Junction Temperature  
Pb/SN Temperature, Soldering  
Reflow (10 sec to 30 sec)  
Pb-Free Temperature, Soldering Reflow  
ESD (All Pins Except Analog Inputs)  
ESD (Analog Input Pins Only)  
240 (+0)°C  
260 (+0)°C  
2 kV  
7 kV  
1 Transient currents of up to 100 mA do not cause SCR latch-up.  
Stresses at or above those listed under Absolute Maximum  
Ratings may cause permanent damage to the product. This is a  
stress rating only; functional operation of the product at these  
or any other conditions above those indicated in the operational  
section of this specification is not implied. Operation beyond  
the maximum operating conditions for extended periods may  
affect product reliability.  
Rev. D | Page 11 of 36  
 
 
 
AD7606/AD7606-6/AD7606-4  
Data Sheet  
PIN CONFIGURATIONS AND FUNCTION DESCRIPTIONS  
64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49  
1
AV  
48  
AV  
CC  
CC  
ANALOG INPUT  
DECOUPLING CAP PIN  
POWER SUPPLY  
GROUND PIN  
PIN 1  
2
3
47 AGND  
46  
AGND  
OS 0  
REFGND  
4
5
45  
44  
43  
42  
41  
40  
39  
38  
37  
36  
OS 1  
OS 2  
REFCAPB  
REFCAPA  
REFGND  
REFIN/REFOUT  
AGND  
6
DATA OUTPUT  
PAR/SER/BYTE SEL  
AD7606  
7
STBY  
TOP VIEW  
DIGITAL OUTPUT  
DIGITAL INPUT  
(Not to Scale)  
8
RANGE  
9
AGND  
CONVST A  
REFERENCE INPUT/OUTPUT  
10  
11  
12  
13  
REGCAP  
CONVST B  
RESET  
RD/SCLK  
CS  
AV  
CC  
AV  
CC  
REGCAP  
BUSY 14  
35 AGND  
FRSTDATA 15  
34 REF SELECT  
33 DB15/BYTE SEL  
DB0  
16  
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32  
Figure 8. AD7606 Pin Configuration  
64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49  
1
AV  
48  
AV  
CC  
CC  
ANALOG INPUT  
DECOUPLING CAP PIN  
POWER SUPPLY  
GROUND PIN  
PIN 1  
2
3
47 AGND  
46  
AGND  
OS 0  
REFGND  
4
5
45  
44  
43  
42  
41  
40  
39  
38  
37  
36  
OS 1  
OS 2  
REFCAPB  
REFCAPA  
REFGND  
REFIN/REFOUT  
AGND  
6
DATA OUTPUT  
DIGITAL OUTPUT  
DIGITAL INPUT  
PAR/SER/BYTE SEL  
AD7606-6  
7
STBY  
TOP VIEW  
(Not to Scale)  
8
RANGE  
9
AGND  
CONVST A  
REFERENCE INPUT/OUTPUT  
10  
11  
12  
13  
REGCAP  
CONVST B  
RESET  
RD/SCLK  
CS  
AV  
CC  
AV  
CC  
REGCAP  
BUSY 14  
35 AGND  
FRSTDATA 15  
34 REF SELECT  
33 DB15/BYTE SEL  
DB0  
16  
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32  
Figure 9. AD7606-6 Pin Configuration  
Rev. D | Page 12 of 36  
 
Data Sheet  
AD7606/AD7606-6/AD7606-4  
64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49  
1
AV  
48  
AV  
CC  
CC  
ANALOG INPUT  
DECOUPLING CAP PIN  
POWER SUPPLY  
GROUND PIN  
PIN 1  
2
3
47 AGND  
46  
AGND  
OS 0  
REFGND  
4
5
45  
44  
43  
42  
41  
40  
39  
38  
37  
36  
OS 1  
OS 2  
REFCAPB  
REFCAPA  
REFGND  
REFIN/REFOUT  
AGND  
6
DATA OUTPUT  
PAR/SER/BYTE SEL  
AD7606-4  
7
STBY  
TOP VIEW  
DIGITAL OUTPUT  
DIGITAL INPUT  
(Not to Scale)  
8
RANGE  
9
AGND  
CONVST A  
REFERENCE INPUT/OUTPUT  
10  
11  
12  
13  
REGCAP  
CONVST B  
RESET  
RD/SCLK  
CS  
AV  
CC  
AV  
CC  
REGCAP  
BUSY 14  
35 AGND  
FRSTDATA 15  
34 REF SELECT  
33 DB15/BYTE SEL  
DB0  
16  
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32  
Figure 10. AD7606-4 Pin Configuration  
Table 6. Pin Function Descriptions  
Mnemonic  
AD7606-6 AD7606-4 Description  
Pin No.  
Type1  
AD7606  
AVCC  
1, 37, 38,  
48  
P
AVCC  
AVCC  
Analog Supply Voltage, 4.75 V to 5.25 V. This supply voltage is applied to  
the internal front-end amplifiers and to the ADC core. These supply pins  
should be decoupled to AGND.  
2, 26, 35,  
40, 41, 47  
P
AGND  
AGND  
AGND  
Analog Ground. These pins are the ground reference points for all analog  
circuitry on the AD7606. All analog input signals and external reference  
signals should be referred to these pins. All six of these AGND pins should  
connect to the AGND plane of a system.  
5, 4, 3  
DI  
DI  
OS [2:0]  
OS [2:0]  
OS [2:0]  
Oversampling Mode Pins. Logic inputs. These inputs are used to select the  
oversampling ratio. OS 2 is the MSB control bit, and OS 0 is the LSB control  
bit. See the Digital Filter section for more details about the oversampling  
mode of operation and Table 9 for oversampling bit decoding.  
Parallel/Serial/Byte Interface Selection Input. Logic input. If this pin is tied to  
a logic low, the parallel interface is selected. If this pin is tied to a logic high,  
the serial interface is selected. Parallel byte interface mode is selected when  
this pin is logic high and DB15/BYTE SEL is logic high (see Table 8).  
In serial mode, the RD/SCLK pin functions as the serial clock input. The  
DB7/DOUTA pin and the DB8/DOUTB pin function as serial data outputs. When  
the serial interface is selected, the DB[15:9] and DB[6:0] pins should be tied to  
ground.  
6
PAR/SER/  
BYTE SEL  
PAR/SER/  
BYTE SEL  
PAR/SER/  
BYTE SEL  
In byte mode, DB15, in conjunction with PAR/SER/BYTE SEL, is used to select  
the parallel byte mode of operation (see Table 8). DB14 is used as the HBEN  
pin. DB[7:0] transfer the 16-bit conversion results in two RD operations,  
with DB0 as the LSB of the data transfers.  
7
DI  
STBY  
STBY  
STBY  
Standby Mode Input. This pin is used to place the AD7606/AD7606-6/  
AD7606-4 into one of two power-down modes: standby mode or shutdown  
mode. The power-down mode entered depends on the state of the RANGE  
pin, as shown in Table 7. When in standby mode, all circuitry, except the on-  
chip reference, regulators, and regulator buffers, is powered down. When  
in shutdown mode, all circuitry is powered down.  
Rev. D | Page 13 of 36  
AD7606/AD7606-6/AD7606-4  
Data Sheet  
Mnemonic  
AD7606-6 AD7606-4 Description  
RANGE RANGE  
Pin No.  
Type1  
AD7606  
8
DI  
RANGE  
Analog Input Range Selection. Logic input. The polarity on this pin deter-  
mines the input range of the analog input channels. If this pin is tied to a  
logic high, the analog input range is 10 V for all channels. If this pin is tied to  
a logic low, the analog input range is 5 V for all channels. A logic change  
on this pin has an immediate effect on the analog input range. Changing  
this pin during a conversion is not recommended for fast throughput rate  
applications. See the Analog Input section for more information.  
9, 10  
DI  
CONVST A, CONVST A, CONVST A,  
Conversion Start Input A, Conversion Start Input B. Logic inputs. These  
logic inputs are used to initiate conversions on the analog input channels.  
For simultaneous sampling of all input channels, CONVST A and CONVST B  
can be shorted together, and a single convert start signal can be applied.  
Alternatively, CONVST A can be used to initiate simultaneous sampling: V1,  
V2, V3, and V4 for the AD7606; V1, V2, and V3 for the AD7606-6; and V1  
and V2 for the AD7606-4. CONVST B can be used to initiate simultaneous  
sampling on the other analog inputs: V5, V6, V7, and V8 for the AD7606;  
V4, V5, and V6 for the AD7606-6; and V3 and V4 for the AD7606-4. This is  
possible only when oversampling is not switched on. When the CONVST A or  
CONVST B pin transitions from low to high, the front-end track-and-hold  
circuitry for the respective analog inputs is set to hold.  
CONVST B  
CONVST B  
CONVST B  
11  
12  
DI  
DI  
RESET  
RESET  
RESET  
Reset Input. When set to logic high, the rising edge of RESET resets the  
AD7606/AD7606-6/AD7606-4. The device should receive a RESET pulse  
directly after power-up. The RESET high pulse should typically be 50 ns  
wide. If a RESET pulse is applied during a conversion, the conversion is  
aborted. If a RESET pulse is applied during a read, the contents of the  
output registers reset to all zeros.  
Parallel Data Read Control Input When the Parallel Interface Is Selected (RD)/  
Serial Clock Input When the Serial Interface Is Selected (SCLK). When both  
CS and RD are logic low in parallel mode, the output bus is enabled.  
In serial mode, this pin acts as the serial clock input for data transfers.  
The CS falling edge takes the DOUTA and DOUTB data output lines out  
of three-state and clocks out the MSB of the conversion result. The rising  
RD/SCLK  
RD/SCLK  
RD/SCLK  
edge of SCLK clocks all subsequent data bits onto the DOUTA and DOUT  
B
serial data outputs. For more information, see the Conversion Control  
section.  
13  
14  
DI  
CS  
CS  
CS  
Chip Select. This active low logic input frames the data transfer. When  
both CS and RD are logic low in parallel mode, the DB[15:0] output bus is  
enabled and the conversion result is output on the parallel data bus lines.  
In serial mode, CS is used to frame the serial read transfer and clock out  
the MSB of the serial output data.  
Busy Output. This pin transitions to a logic high after both CONVST A and  
CONVST B rising edges and indicates that the conversion process has started.  
The BUSY output remains high until the conversion process for all channels  
is complete. The falling edge of BUSY signals that the conversion data is  
being latched into the output data registers and is available to read after  
a Time t4. Any data read while BUSY is high must be completed before the  
falling edge of BUSY occurs. Rising edges on CONVST A or CONVST B have  
no effect while the BUSY signal is high.  
DO  
BUSY  
BUSY  
BUSY  
15  
DO  
FRSTDATA  
FRSTDATA  
FRSTDATA  
Digital Output. The FRSTDATA output signal indicates when the first channel,  
V1, is being read back on the parallel, byte, or serial interface. When the  
CS input is high, the FRSTDATA output pin is in three-state. The falling  
edge of CS takes FRSTDATA out of three-state. In parallel mode, the falling  
edge of RD corresponding to the result of V1 then sets the FRSTDATA pin  
high, indicating that the result from V1 is available on the output data bus.  
The FRSTDATA output returns to a logic low following the next falling edge  
of RD. In serial mode, FRSTDATA goes high on the falling edge of CS because  
this clocks out the MSB of V1 on DOUTA. It returns low on the 16th SCLK  
falling edge after the CS falling edge. See the Conversion Control section  
for more details.  
Rev. D | Page 14 of 36  
Data Sheet  
AD7606/AD7606-6/AD7606-4  
Mnemonic  
Pin No.  
Type1  
AD7606  
AD7606-6 AD7606-4 Description  
22 to 16  
DO  
DB[6:0]  
DB[6:0]  
DB[6:0]  
Parallel Output Data Bits, DB6 to DB0. When PAR/SER/BYTE SEL = 0, these  
pins act as three-state parallel digital input/output pins. When CS and RD  
are low, these pins are used to output DB6 to DB0 of the conversion result.  
When PAR/SER/BYTE SEL = 1, these pins should be tied to AGND. When  
operating in parallel byte interface mode, DB[7:0] outputs the 16-bit con-  
version result in two RD operations. DB7 (Pin 24) is the MSB; DB0 is the LSB.  
23  
24  
P
VDRIVE  
VDRIVE  
VDRIVE  
Logic Power Supply Input. The voltage (2.3 V to 5.25 V) supplied at this pin  
determines the operating voltage of the interface. This pin is nominally at the  
same supply as the supply of the host interface (that is, DSP and FPGA).  
Parallel Output Data Bit 7 (DB7)/Serial Interface Data Output Pin (DOUTA).  
When PAR/SER/BYTE SEL = 0, this pins acts as a three-state parallel digital  
input/output pin. When CS and RD are low, this pin is used to output DB7  
of the conversion result. When PAR/SER/BYTE SEL = 1, this pin functions  
as DOUTA and outputs serial conversion data (see the Conversion Control  
section for more details). When operating in parallel byte mode, DB7 is  
the MSB of the byte.  
Parallel Output Data Bit 8 (DB8)/Serial Interface Data Output Pin (DOUTB).  
When PAR/SER/BYTE SEL = 0, this pin acts as a three-state parallel digital  
input/output pin. When CS and RD are low, this pin is used to output  
DB8 of the conversion result. When PAR/SER/BYTE SEL = 1, this pin functions  
as DOUTB and outputs serial conversion data (see the Conversion Control  
section for more details).  
Parallel Output Data Bits, DB13 to DB9. When PAR/SER/BYTE SEL = 0, these  
pins act as three-state parallel digital input/output pins. When CS and RD  
are low, these pins are used to output DB13 to DB9 of the conversion result.  
When PAR/SER/BYTE SEL = 1, these pins should be tied to AGND.  
DO  
DB7/DOUT  
A
DB7/DOUT  
A
B
DB7/DOUT  
A
B
25  
DO  
DB8/DOUT  
DB[13:9]  
B
DB8/DOUT  
DB[13:9]  
DB8/DOUT  
DB[13:9]  
31 to 27  
32  
DO  
DO/DI  
DB14/  
HBEN  
DB14/  
HBEN  
DB14/  
HBEN  
Parallel Output Data Bit 14 (DB14)/High Byte Enable (HBEN). When PAR/  
SER/BYTE SEL = 0, this pin acts as a three-state parallel digital output pin.  
When CS and RD are low, this pin is used to output DB14 of the conversion  
result. When PAR/SER/BYTE SEL = 1 and DB15/BYTE SEL = 1, the AD7606/  
AD7606-6/AD7606-4 operate in parallel byte interface mode. In parallel  
byte mode, the HBEN pin is used to select whether the most significant byte  
(MSB) or the least significant byte (LSB) of the conversion result is output first.  
When HBEN = 1, the MSB is output first, followed by the LSB.  
When HBEN = 0, the LSB is output first, followed by the MSB.  
In serial mode, this pin should be tied to GND.  
33  
DO/DI  
DB15/  
BYTE SEL  
DB15/  
BYTE SEL  
DB15/  
BYTE SEL  
Parallel Output Data Bit 15 (DB15)/Parallel Byte Mode Select (BYTE SEL).  
When PAR/SER/BYTE SEL = 0, this pin acts as a three-state parallel digital  
output pin. When CS and RD are low, this pin is used to output DB15 of the  
conversion result. When PAR/SER/BYTE SEL = 1, the BYTE SEL pin is used to  
select between serial interface mode and parallel byte interface mode  
(see Table 8). When PAR/SER/BYTE SEL = 1 and DB15/BYTE SEL = 0, the  
AD7606 operates in serial interface mode. When PAR/SER/BYTE SEL = 1  
and DB15/BYTE SEL = 1, the AD7606 operates in parallel byte interface mode.  
34  
DI  
P
REF SELECT REF SELECT REF SELECT Internal/External Reference Selection Input. Logic input. If this pin is set to  
logic high, the internal reference is selected and enabled. If this pin is set to  
logic low, the internal reference is disabled and an external reference  
voltage must be applied to the REFIN/REFOUT pin.  
36, 39  
REGCAP  
REGCAP  
REGCAP  
Decoupling Capacitor Pin for Voltage Output from Internal Regulator.  
These output pins should be decoupled separately to AGND using a 1 μF  
capacitor. The voltage on these pins is in the range of 2.5 V to 2.7 V.  
Rev. D | Page 15 of 36  
AD7606/AD7606-6/AD7606-4  
Data Sheet  
Mnemonic  
AD7606-6 AD7606-4 Description  
Pin No.  
Type1  
AD7606  
42  
REF  
REFIN/  
REFOUT  
REFIN/  
REFOUT  
REFIN/  
REFOUT  
Reference Input (REFIN)/Reference Output (REFOUT). The on-chip reference  
of 2.5 V is available on this pin for external use if the REF SELECT pin is set to  
logic high. Alternatively, the internal reference can be disabled by setting  
the REF SELECT pin to logic low, and an external reference of 2.5 V can be  
applied to this input (see the Internal/External Reference section).  
Decoupling is required on this pin for both the internal and external  
reference options. A 10 μF capacitor should be applied from this pin to  
ground close to the REFGND pins.  
43, 46  
44, 45  
REF  
REF  
REFGND  
REFCAPA,  
REFCAPB  
REFGND  
REFCAPA,  
REFCAPB  
REFGND  
REFCAPA,  
REFCAPB  
Reference Ground Pins. These pins should be connected to AGND.  
Reference Buffer Output Force/Sense Pins. These pins must be connected  
together and decoupled to AGND using a low ESR, 10 μF ceramic capacitor.  
The voltage on these pins is typically 4.5 V.  
49  
AI  
V1  
V1  
V1  
Analog Input. This pin is a single-ended analog input. The analog input  
range of this channel is determined by the RANGE pin.  
50, 52  
AI GND  
V1GND,  
V2GND  
V1GND,  
V2GND  
V1GND,  
V2GND  
Analog Input Ground Pins. These pins correspond to Analog Input Pin V1  
and Analog Input Pin V2. All analog input AGND pins should connect to  
the AGND plane of a system.  
51  
AI  
V2  
V2  
V2  
Analog Input. This pin is a single-ended analog input. The analog input  
range of this channel is determined by the RANGE pin.  
53  
54  
AI/GND  
AI GND/ V3GND  
GND  
V3  
V3  
V3GND  
AGND  
AGND  
Analog Input 3. For the AD7606-4, this is an AGND pin.  
Analog Input Ground Pin. For the AD7606-4, this is an AGND pin.  
55  
56  
AI/GND  
AI GND/ V4GND  
GND  
V4  
AGND  
AGND  
AGND  
AGND  
Analog Input 4. For the AD7606-6 and the AD7606-4, this is an AGND pin.  
Analog Input Ground Pin. For the AD7606-6 and AD7606-4, this is an  
AGND pin.  
57  
58  
AI  
V5  
V4  
V3  
Analog Inputs. These pins are single-ended analog inputs. The analog  
input range of these channels is determined by the RANGE pin.  
Analog Input Ground Pins. All analog input AGND pins should connect to  
the AGND plane of a system.  
AI GND  
V5GND  
V4GND  
V3GND  
59  
60  
AI  
AI GND  
V6  
V6GND  
V5  
V5GND  
V4  
V4GND  
Analog Inputs. These pins are single-ended analog inputs.  
Analog Input Ground Pins. All analog input AGND pins should connect to  
the AGND plane of a system.  
61  
62  
AI/GND  
AI GND/ V7GND  
GND  
V7  
V6  
V6GND  
AGND  
AGND  
Analog Input Pins. For the AD7606-4, this is an AGND pin.  
Analog Input Ground Pins. For the AD7606-4, this is an AGND pin.  
63  
64  
AI/GND  
AI GND/ V8GND  
GND  
V8  
AGND  
AGND  
AGND  
AGND  
Analog Input Pin. For the AD7606-4 and AD7606-6, this is an AGND pin.  
Analog Input Ground Pin. For the AD7606-4 and AD7606-6, this is an  
AGND pin.  
1 P is power supply, DI is digital input, DO is digital output, REF is reference input/output, AI is analog input, GND is ground.  
Rev. D | Page 16 of 36  
Data Sheet  
AD7606/AD7606-6/AD7606-4  
TYPICAL PERFORMANCE CHARACTERISTICS  
Temperature range is from −40°C to +85°C. The AD7606 is functional up to 105°C with throughput rates < 160 kSPS. Specifications are  
guaranteed for the operating temperature range of −40°C to +85°C only.  
0
2.0  
AV , V  
CC DRIVE  
= 5V  
AV , V  
CC DRIVE  
= 5V  
INTERNAL REFERENCE  
±10V RANGE  
F
= 200kSPS  
SAMPLE  
= 25°C  
–20  
1.5  
T
A
F
F
= 200kSPS  
INTERNAL REFERENCE  
±10V RANGE  
SAMPLE  
= 1kHz  
–40  
IN  
1.0  
16,384 POINT FFT  
SNR = 90.17dB  
THD = –106.25dB  
–60  
0.5  
–80  
0
–100  
–120  
–140  
–160  
–180  
–0.5  
–1.0  
–1.5  
–2.0  
0
10k 20k 30k 40k 50k 60k 70k 80k 90k 100k  
INPUT FREQUENCY (Hz)  
0
10k  
20k  
30k  
CODE  
40k  
50k  
60k  
Figure 11. AD7606 FFT, ±10 V Range  
Figure 14. AD7606 Typical INL, ±10 V Range  
0
–20  
1.0  
0.8  
AV , V  
CC DRIVE  
INTERNAL REFERENCE  
±5V RANGE  
= 5V  
AV , V  
= 5V  
= 200kSPS  
CC DRIVE  
F
T
SAMPLE  
= 25°C  
A
F
F
= 200kSPS  
INTERNAL REFERENCE  
±10V RANGE  
SAMPLE  
= 1kHz  
0.6  
–40  
IN  
16,384 POINT FFT  
SNR = 89.48dB  
THD = –108.65dB  
0.4  
–60  
0.2  
–80  
0
–100  
–120  
–140  
–160  
–180  
–0.2  
–0.4  
–0.6  
–0.8  
–1.0  
0
10k 20k 30k 40k 50k 60k 70k 80k 90k 100k  
INPUT FREQUENCY (Hz)  
0
10k  
20k  
30k  
CODE  
40k  
50k  
60k  
Figure 12. AD7606 FFT Plot, ±± V Range  
Figure 1±. AD7606 Typical DNL, ±10 V Range  
0
–10  
–20  
–30  
–40  
2.0  
1.5  
AV , V  
CC DRIVE  
INTERNAL REFERENCE  
±10V RANGE  
= 5V  
AV , V  
CC DRIVE  
INTERNAL REFERENCE  
±5V RANGE  
= 5V  
F
T
F
= 11.5kSPS  
= 25°C  
= 133Hz  
F
T
= 200kSPS  
SAMPLE  
SAMPLE  
= 25°C  
A
A
1.0  
–50  
IN  
8192 POINT FFT  
OS BY 16  
SNR = 96.01dB  
THD = –108.05dB  
–60  
–70  
–80  
–90  
0.5  
0
–100  
–110  
–120  
–130  
–140  
–150  
–160  
–170  
–180  
–0.5  
–1.0  
–1.5  
–2.0  
0
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5  
FREQUENCY (kHz)  
0
8192 16,384 24,576 32,768 40,960 49,152 57,344 65,536  
CODE  
Figure 13. FFT Plot Oversampling By 16, ±10 V Range  
Figure 16. AD7606 Typical INL, ± ± V Range  
Rev. D | Page 17 of 36  
 
AD7606/AD7606-6/AD7606-4  
Data Sheet  
1.00  
10  
8
AV , V  
CC DRIVE  
INTERNAL REFERENCE  
±5V RANGE  
= 5V  
0.75  
0.50  
0.25  
0
PFS ERROR  
NFS ERROR  
F
T
= 200kSPS  
SAMPLE  
= 25°C  
6
A
4
2
0
–2  
–4  
–6  
–9  
–10  
–0.25  
–0.50  
–0.75  
–1.00  
10V RANGE  
AV , V  
= 5V  
CC DRIVE  
EXTERNAL REFERENCE  
0
8192 16,384 24,576 32,768 40,960 49,152 57,344 65,536  
CODE  
–40  
–25  
–10  
5
20  
35 50 65 80  
TEMPERATURE (°C)  
Figure 17. AD7606 Typical DNL, ±± V Range  
Figure 20. NFS and PFS Error Matching  
20  
15  
10  
5
10  
8
±10V RANGE  
±5V RANGE  
6
0
4
–5  
–10  
–15  
2
AV , V  
CC DRIVE  
= 5V  
F
= 200 kSPS  
SAMPLE  
= 25°C  
T
A
EXTERNAL REFERENCE  
SOURCE RESISTANCE IS MATCHED ON  
THE VxGND INPUT  
0
200kSPS  
AV , V  
= 5V  
CC DRIVE  
±10V AND ±5V RANGE  
EXTERNAL REFERENCE  
35 50 65 80  
TEMPERATURE (°C)  
–20  
–40  
–2  
–25  
–10  
5
20  
0
20k  
40k  
60k  
80k  
100k  
120k  
SOURCE RESISTANCE (Ω)  
Figure 18. NFS Error vs. Temperature  
Figure 21. PFS and NFS Error vs. Source Resistance  
20  
15  
10  
5
1.0  
0.8  
0.6  
0.4  
0.2  
0
0
±5V RANGE  
±10V RANGE  
5V RANGE  
–0.2  
–0.4  
–0.6  
–0.8  
–1.0  
–5  
–10  
–15  
10V RANGE  
200kSPS  
200kSPS  
AV , V  
= 5V  
AV , V  
= 5V  
CC DRIVE  
EXTERNAL REFERENCE  
CC DRIVE  
EXTERNAL REFERENCE  
–20  
–40  
–25  
–10  
5
20  
35 50 65 80  
–40  
–25  
–10  
5
20  
35 50 65 80  
TEMPERATURE (°C)  
TEMPERATURE (°C)  
Figure 19. PFS Error vs. Temperature  
Figure 22. Bipolar Zero Code Error vs. Temperature  
Rev. D | Page 18 of 36  
Data Sheet  
AD7606/AD7606-6/AD7606-4  
4
98  
96  
94  
92  
90  
88  
86  
84  
82  
80  
3
5V RANGE  
2
1
10V RANGE  
0
–1  
–2  
–3  
OS BY 64  
OS BY 32  
OS BY 16  
OS BY 8  
OS BY 4  
OS BY 2  
NO OS  
AV , V  
CC DRIVE  
= 5V  
F
CHANGES WITH OS RATE  
SAMPLE  
= 25°C  
200kSPS  
AV , V  
T
A
= 5V  
INTERNAL REFERENCE  
±5V RANGE  
CC DRIVE  
EXTERNAL REFERENCE  
35 50 65 80  
TEMPERATURE (°C)  
–4  
–40  
–25  
–10  
5
20  
10  
100  
1k  
10k  
100k  
INPUT FREQUENCY (Hz)  
Figure 23. Bipolar Zero Code Error Matching Between Channels  
Figure 26. SNR vs. Input Frequency for Different Oversampling Rates, ±5 V Range  
–40  
100  
98  
96  
94  
92  
90  
88  
±10V RANGE  
AV , V  
= +5V  
CC DRIVE  
–50  
–60  
F
= 200kSPS  
SAMPLE  
R
MATCHED ON Vx AND VxGND INPUTS  
SOURCE  
–70  
–80  
105k  
48.7kΩ  
23.7kΩ  
10kΩ  
5kΩ  
1.2kΩ  
100Ω  
51Ω  
–90  
86  
84  
82  
80  
OS BY 64  
OS BY 32  
OS BY 16  
OS BY 8  
OS BY 4  
OS BY 2  
NO OS  
–100  
–110  
–120  
AV , V  
CC DRIVE  
= 5V  
F
CHANGES WITH OS RATE  
SAMPLE  
= 25°C  
T
A
INTERNAL REFERENCE  
±10V RANGE  
0Ω  
1k  
10k  
100k  
10  
100  
1k  
10k  
100k  
INPUT FREQUENCY (Hz)  
INPUT FREQUENCY (Hz)  
Figure 24. THD vs. Input Frequency for Various Source Impedances,  
±±0 V Range  
Figure 27. SNR vs. Input Frequency for Different Oversampling Rates, ±±0 V Range  
–40  
–50  
±5V RANGE  
AV , V  
CC DRIVE  
INTERNAL REFERENCE  
AD7606 RECOMMENDED DECOUPLING USED  
= 5V  
AV , V  
= +5V  
CC DRIVE  
–60  
–70  
–50  
–60  
F
= 200kSPS  
SAMPLE  
R
MATCHED ON Vx AND VxGND INPUTS  
F
T
= 150kSPS  
SOURCE  
SAMPLE  
= 25°C  
A
INTERFERER ON ALL UNSELECTED CHANNELS  
–80  
–70  
–90  
±10V RANGE  
±5V RANGE  
–80  
–100  
–110  
–120  
–130  
–140  
105kΩ  
48.7kΩ  
23.7kΩ  
10kΩ  
5kΩ  
1.2kΩ  
100Ω  
51Ω  
–90  
–100  
–110  
–120  
0Ω  
1k  
10k  
100k  
0
20  
40  
60  
80  
100  
120  
140  
160  
INPUT FREQUENCY (Hz)  
NOISE FREQUENCY (kHz)  
Figure 25. THD vs. Input Frequency for Various Source Impedances,  
±5 V Range  
Figure 28. Channel-to-Channel Isolation  
Rev. D | Page 19 of 36  
 
AD7606/AD7606-6/AD7606-4  
Data Sheet  
100  
22  
20  
18  
16  
14  
12  
10  
8
98  
±10V RANGE  
96  
94  
±5V RANGE  
92  
90  
88  
86  
AV , V  
= 5V  
AV , V = 5V  
CC DRIVE  
84  
82  
80  
CC DRIVE  
= 25°C  
T
T = 25°C  
A
A
INTERNAL REFERENCE  
INTERNAL REFERENCE  
F VARIES WITH OS RATE  
SAMPLE  
F
SCALES WITH OS RATIO  
SAMPLE  
OFF  
OS2  
OS4  
OS8  
OS16  
OS32  
OS64  
NO OS  
OS2  
OS4  
OS8  
OS16  
OS32  
OS64  
OVERSAMPLING RATIO  
OVERSAMPLING RATIO  
Figure 29. Dynamic Range vs. Oversampling Rate  
Figure 32. Supply Current vs. Oversampling Rate  
2.5010  
2.5005  
2.5000  
2.4995  
2.4990  
2.4985  
2.4980  
140  
130  
120  
110  
100  
90  
AV  
= 5.25V  
CC  
AV  
AV  
= 5V  
CC  
CC  
±10V RANGE  
±5V RANGE  
= 4.75V  
80  
AV , V  
CC DRIVE  
INTERNAL REFERENCE  
AD7606 RECOMMENDED DECOUPLING USED  
= 5V  
70  
F
T
= 200kSPS  
SAMPLE  
= 25°C  
A
60  
–40  
–25  
–10  
5
20  
35  
50  
65  
80  
0
100 200 300 400 500 600 700 800 900 1000 1100  
AV NOISE FREQUENCY (kHz)  
TEMPERATURE (°C)  
CC  
Figure 30. Reference Output Voltage vs. Temperature for  
Different Supply Voltages  
Figure 33. PSRR  
8
6
AV , V  
CC DRIVE  
= 5V  
F
= 200kSPS  
SAMPLE  
4
2
0
–2  
–4  
–6  
–8  
–10  
+85°C  
+25°C  
–40°C  
–10  
–8  
–6  
–4  
–2  
0
2
4
6
8
10  
INPUT VOLTAGE (V)  
Figure 31. Analog Input Current vs. Temperature for Various Supply Voltages  
Rev. D | Page 20 of 36  
 
Data Sheet  
AD7606/AD7606-6/AD7606-4  
TERMINOLOGY  
Integral Nonlinearity  
Total Harmonic Distortion (THD)  
The maximum deviation from a straight line passing through  
the endpoints of the ADC transfer function. The endpoints of  
the transfer function are zero scale, at ½ LSB below the first  
code transition; and full scale, at ½ LSB above the last code  
transition.  
The ratio of the rms sum of the harmonics to the fundamental.  
For the AD7606/AD7606-6/AD7606-4, it is defined as  
THD (dB) =  
2
V22 +V32 +V42 +V52 +V62 +V72 +V82 +V9  
20log  
V1  
Differential Nonlinearity  
The difference between the measured and the ideal 1 LSB  
change between any two adjacent codes in the ADC.  
where:  
V1 is the rms amplitude of the fundamental.  
V2 to V9 are the rms amplitudes of the second through ninth  
harmonics.  
Peak Harmonic or Spurious Noise  
The ratio of the rms value of the next largest component in the  
ADC output spectrum (up to fS/2, excluding dc) to the rms value  
of the fundamental. Normally, the value of this specification is  
determined by the largest harmonic in the spectrum, but for  
ADCs where the harmonics are buried in the noise floor, it is  
determined by a noise peak.  
Bipolar Zero Code Error  
The deviation of the midscale transition (all 1s to all 0s) from  
the ideal, which is 0 V − ½ LSB.  
Bipolar Zero Code Error Match  
The absolute difference in bipolar zero code error between any  
two input channels.  
Positive Full-Scale Error  
The deviation of the actual last code transition from the ideal  
last code transition (10 V − 1½ LSB (9.99954) and 5 V − 1½ LSB  
(4.99977)) after bipolar zero code error is adjusted out. The  
positive full-scale error includes the contribution from the  
internal reference buffer.  
Intermodulation Distortion  
With inputs consisting of sine waves at two frequencies, fa and fb,  
any active device with nonlinearities creates distortion products  
at sum and difference frequencies of mfa nfb, where m, n = 0,  
1, 2, 3. Intermodulation distortion terms are those for which  
neither m nor n is equal to 0. For example, the second-order  
terms include (fa + fb) and (fa − fb), and the third-order terms  
include (2fa + fb), (2fa − fb), (fa + 2fb), and (fa − 2fb).  
The calculation of the intermodulation distortion is per the  
THD specification, where it is the ratio of the rms sum of the  
individual distortion products to the rms amplitude of the sum  
of the fundamentals expressed in decibels (dB).  
Positive Full-Scale Error Match  
The absolute difference in positive full-scale error between any  
two input channels.  
Negative Full-Scale Error  
The deviation of the first code transition from the ideal first  
code transition (−10 V + ½ LSB (−9.99984) and −5 V + ½ LSB  
(−4.99992)) after the bipolar zero code error is adjusted out.  
The negative full-scale error includes the contribution from the  
internal reference buffer.  
Power Supply Rejection Ratio (PSRR)  
Variations in power supply affect the full-scale transition but not  
the converters linearity. PSR is the maximum change in full-  
scale transition point due to a change in power supply voltage  
from the nominal value. The PSR ratio (PSRR) is defined as the  
ratio of the power in the ADC output at full-scale frequency, f,  
to the power of a 100 mV p-p sine wave applied to the ADCs  
Negative Full-Scale Error Match  
The absolute difference in negative full-scale error between any  
two input channels.  
Total Unadjusted Error (TUE)  
TUE is a comprehensive specification that includes the gain  
linearity and offset errors.  
Signal-to-(Noise + Distortion) Ratio  
The measured ratio of signal-to-(noise + distortion) at the  
output of the ADC. The signal is the rms amplitude of the  
fundamental. Noise is the sum of all nonfundamental signals  
up to half the sampling frequency (fS/2, excluding dc).  
V
DD and VSS supplies of Frequency fS.  
PSRR (dB) = 10 log (Pf/PfS)  
where:  
Pf is equal to the power at Frequency f in the ADC output.  
PfS is equal to the power at Frequency fS coupled onto the AVCC  
supply.  
Channel-to-Channel Isolation  
The ratio depends on the number of quantization levels in  
the digitization process: the more levels, the smaller the  
quantization noise.  
Channel-to-channel isolation is a measure of the level of crosstalk  
between all input channels. It is measured by applying a full-scale  
sine wave signal, up to 160 kHz, to all unselected input channels  
and then determining the degree to which the signal attenuates  
in the selected channel with a 1 kHz sine wave signal applied (see  
Figure 28).  
The theoretical signal-to-(noise + distortion) ratio for an ideal  
N-bit converter with a sine wave input is given by  
Signal-to-(Noise + Distortion) = (6.02 N + 1.76) dB  
Thus, for a 16-bit converter, the signal-to-(noise + distortion)  
is 98 dB.  
Rev. D | Page 21 of 36  
 
AD7606/AD7606-6/AD7606-4  
Data Sheet  
THEORY OF OPERATION  
Analog Input Clamp Protection  
CONVERTER DETAILS  
Figure 34 shows the analog input structure of the AD7606/  
AD7606-6/AD7606-4. Each analog input of the AD7606/  
AD7606-6/AD7606-4 contains clamp protection circuitry.  
Despite single ꢀ V supply operation, this analog input clamp  
protection allows for an input over voltage of up to ±±6.ꢀ V.  
The AD7606/AD7606-6/AD7606-4 are data acquisition systems  
that employ a high speed, low power, charge redistribution,  
successive approximation analog-to-digital converter (ADC)  
and allow the simultaneous sampling of eight/six/four analog input  
channels. The analog inputs on the AD7606/AD7606-6/AD7606-4  
can accept true bipolar input signals. The RANGE pin is used to  
select either ±±0 V or ±ꢀ V as the input range. The AD7606/  
AD7606-6/AD7606-4 operate from a single ꢀ V supply.  
R
FB  
1M  
Vx  
CLAMP  
CLAMP  
1MΩ  
VxGND  
The AD7606/AD7606-6/AD7606-4 contain input clamp  
protection, input signal scaling amplifiers, a second-order anti-  
aliasing filter, track-and-hold amplifiers, an on-chip reference,  
reference buffers, a high speed ADC, a digital filter, and high  
speed parallel and serial interfaces. Sampling on the AD7606/  
AD7606-6/AD7606-4 is controlled using the CONVST signals.  
SECOND-  
ORDER  
LPF  
R
FB  
Figure 34. Analog Input Circuitry  
Figure 3ꢀ shows the voltage vs. current characteristic of the  
clamp circuit. For input voltages of up to ±±6.ꢀ V, no current  
flows in the clamp circuit. For input voltages that are above ±±6.ꢀ V,  
the AD7606/AD7606-6/AD7606-4 clamp circuitry turns on.  
ANALOG INPUT  
Analog Input Ranges  
AV , V  
CC DRIVE  
= 5V  
30  
20  
T
= 25°C  
A
The AD7606/AD7606-6/AD7606-4 can handle true bipolar,  
single-ended input voltages. The logic level on the RANGE pin  
determines the analog input range of all analog input channels.  
If this pin is tied to a logic high, the analog input range is ±±0 V  
for all channels. If this pin is tied to a logic low, the analog input  
range is ±ꢀ V for all channels. A logic change on this pin has an  
immediate effect on the analog input range; however, there is  
typically a settling time of approximately 80 μs, in addition to  
the normal acquisition time requirement. The recommended  
practice is to hardwire the RANGE pin according to the desired  
input range for the system signals.  
10  
0
–10  
–20  
–30  
–40  
–50  
–20  
–15  
–10  
–5  
0
5
10  
15  
20  
During normal operation, the applied analog input voltage  
should remain within the analog input range selected via the  
RANGE pin. A RESET pulse must be applied after power up to  
ensure the analog input channels are configured for the range  
selected.  
SOURCE VOLTAGE (V)  
Figure 35. Input Protection Clamp Profile  
A series resistor should be placed on the analog input channels  
to limit the current to ±±0 mA for input voltages above ±±6.ꢀ V.  
In an application where there is a series resistance on an analog  
input channel, Vx, a corresponding resistance is required on the  
analog input GND channel, VxGND (see Figure 36). If there is  
no corresponding resistor on the VxGND channel, an offset  
error occurs on that channel. It is recommended that the input  
overvoltage clamp protection circuitry be used to protect the  
AD7606/AD7606-6/AD7606-4 against transient overvoltage  
events. It is not recommended to leave the AD7606/AD7606-6/  
AD7606-4 in a condition where the clamp protection circuitry  
is active in normal or power-down conditions for extended  
periods because this may degrade the bipolar zero code error  
performance of the AD7606/AD7606-6/AD7606-4.  
When in a power-down mode, it is recommended to tie the  
analog inputs to GND. Per the Analog Input Clamp Protection  
section, the overvoltage clamp protection is recommended for  
use in transient overvoltage conditions and should not remain  
active for extended periods. Stressing the analog inputs outside  
of the conditions mentioned here may degrade the bipolar zero  
code error and THD performance of the AD7606/AD7606-6/  
AD7606-4.  
Analog Input Impedance  
The analog input impedance of the AD7606/AD7606-6/  
AD7606-4 is ± MΩ. This is a fixed input impedance that does  
not vary with the AD7606 sampling frequency. This high analog  
input impedance eliminates the need for a driver amplifier in  
front of the AD7606/AD7606-6/AD7606-4, allowing for direct  
connection to the source or sensor. With the need for a driver  
amplifier eliminated, bipolar supplies (which are often a source  
of noise in a system) can be removed from the signal chain.  
Rev. D | Page 22 of 36  
 
 
 
 
 
 
Data Sheet  
AD7606/AD7606-6/AD7606-4  
R
R
hold (that is, the delay time between the external CONꢀST x  
signal and the track-and-hold actually going into hold) is well  
matched, by design, across all eight track-and-holds on one  
device and from device to device. This matching allows more  
than one AD7606/AD7606-6/AD7606-4 device to be sampled  
simultaneously in a system.  
FB  
AD7606  
ANALOG  
INPUT  
R
R
1M  
1MΩ  
Vx  
CLAMP  
CLAMP  
SIGNAL  
C
VxGND  
FB  
Figure 36. Input Resistance Matching on the Analog Input of the  
AD7606/AD7606-6/AD7606-4  
The end of the conversion process across all eight channels is  
indicated by the falling edge of BUSY; and it is at this point that the  
track-and-holds return to track mode, and the acquisition time  
for the next set of conversions begins.  
Analog Input Antialiasing Filter  
An analog antialiasing filter (a second-order Butterworth) is also  
provided on the AD7606/AD7606-6/AD7606-4. Figure 37 and  
Figure 38 show the frequency and phase response, respectively,  
of the analog antialiasing filter. In the ±± ꢀ range, the −3 dB  
frequency is typically 1± kHz. In the ±10 ꢀ range, the −3 dB  
frequency is typically 23 kHz.  
The conversion clock for the part is internally generated, and  
the conversion time for all channels is 4 μs on the AD7606,  
3 μs on the AD7606-6, and 2 μs on the AD7606-4. On the AD7606,  
the BUSY signal returns low after all eight conversions to indicate  
the end of the conversion process. On the falling edge of BUSY,  
the track-and-hold amplifiers return to track mode. New data  
can be read from the output register via the parallel, parallel  
byte, or serial interface after BUSY goes low; or, alternatively,  
data from the previous conversion can be read while BUSY is  
high. Reading data from the AD7606/AD7606-6/AD7606-4  
while a conversion is in progress has little effect on performance  
and allows a faster throughput to be achieved. In parallel mode  
at ꢀDRIꢀE > 3.3 , the SNR is reduced by ~1.± dB when reading  
during a conversion.  
5
0
±10V RANGE  
AV , V  
CC DRIVE  
= 5V  
–5  
–10  
–15  
–20  
–25  
–30  
–35  
–40  
F
T
= 200kSPS  
SAMPLE  
= 25°C  
±5V RANGE  
A
±10V RANGE 0.1dB  
3dB  
–40 10,303 24,365Hz  
+25 9619  
+85 9326  
23,389Hz  
22,607Hz  
±5V RANGE  
0.1dB  
–40 5225  
+25 5225  
+85 4932  
3dB  
ADC TRANSFER FUNCTION  
16,162Hz  
15,478Hz  
14,990Hz  
The output coding of the AD7606/AD7606-6/AD7606-4 is  
twos complement. The designed code transitions occur midway  
between successive integer LSB values, that is, 1/2 LSB and 3/2 LSB.  
The LSB size is FSR/6±,±36 for the AD7606. The ideal transfer  
characteristic for the AD7606/AD7606-6/AD7606-4 is shown  
in Figure 39.  
100  
1k  
10k  
100k  
INPUT FREQUENCY (Hz)  
Figure 37. Analog Antialiasing Filter Frequency Response  
18  
VIN  
10V  
VIN  
5V  
REF  
2.5V  
REF  
2.5V  
±10V CODE =  
× 32,768 ×  
16  
14  
12  
10  
8
±5V RANGE  
±10V RANGE  
±5V CODE =  
× 32,768 ×  
011...111  
011...110  
+FS – (–FS)  
216  
000...001  
000...000  
111...111  
LSB =  
6
4
2
100...010  
100...001  
100...000  
0
–2  
–4  
–6  
–8  
–FS + 1/2LSB 0V – 1/2LSB +FS – 3/2LSB  
AV , V  
= 5V  
= 200kSPS  
CC DRIVE  
ANALOG INPUT  
F
T
SAMPLE  
= 25°C  
A
+FS  
±10V RANGE +10V  
±5V RANGE +5V  
MIDSCALE –FS  
LSB  
305µV  
152µV  
0V  
0V  
–10V  
–5V  
10  
1k  
10k  
100k  
INPUT FREQUENCY (Hz)  
Figure 39. AD7606/AD7606-6/AD7606-4 Transfer Characteristics  
Figure 38. Analog Antialias Filter Phase Response  
The LSB size is dependent on the analog input range selected.  
Track-and-Hold Amplifiers  
The track-and-hold amplifiers on the AD7606/AD7606-6/  
AD7606-4 allow the ADC to accurately acquire an input sine wave  
of full-scale amplitude to 16-bit resolution. The track-and-hold  
amplifiers sample their respective inputs simultaneously on the  
rising edge of CONꢀST x. The aperture time for the track-and-  
Rev. D | Page 23 of 36  
 
 
 
 
 
AD7606/AD7606-6/AD7606-4  
Data Sheet  
Internal Reference Mode  
INTERNAL/EXTERNAL REFERENCE  
One AD7606/AD7606-6/AD7606-4 device, configured to operate  
in the internal reference mode, can be used to drive the remaining  
AD7606/AD7606-6/AD7606-4 devices, which are configured to  
operate in external reference mode (see Figure 42). The REFIN/  
REFOUT pin of the AD7606/AD7606-6/AD7606-4, configured  
in internal reference mode, should be decoupled using a 10 µF  
ceramic decoupling capacitor. The other AD7606/AD7606-6/  
AD7606-4 devices, configured in external reference mode,  
should use at least a 100 nF decoupling capacitor on their  
REFIN/REFOUT pins.  
The AD7606/AD7606-6/AD7606-4 contain an on-chip 2.5 V  
band gap reference. The REFIN/REFOUT pin allows access to  
the 2.5 V reference that generates the on-chip 4.5 V reference  
internally, or it allows an external reference of 2.5 V to be applied  
to the AD7606/AD7606-6/AD7606-4. An externally applied  
reference of 2.5 V is also gained up to 4.5 V, using the internal  
buffer. This 4.5 V buffered reference is the reference used by the  
SAR ADC.  
The REF SELECT pin is a logic input pin that allows the user to  
select between the internal reference and an external reference.  
If this pin is set to logic high, the internal reference is selected  
and enabled. If this pin is set to logic low, the internal reference  
is disabled and an external reference voltage must be applied  
to the REFIN/REFOUT pin. The internal reference buffer is  
always enabled. After a reset, the AD7606/AD7606-6/AD7606-4  
operate in the reference mode selected by the REF SELECT pin.  
Decoupling is required on the REFIN/REFOUT pin for both  
the internal and external reference options. A 10 µF ceramic  
capacitor is required on the REFIN/REFOUT pin.  
REFIN/REFOUT  
SAR  
REFCAPA  
BUF  
10µF  
REFCAPB  
2.5V  
REF  
Figure 40. Reference Circuitry  
The AD7606/AD7606-6/AD7606-4 contain a reference buffer  
configured to gain the REF voltage up to ~4.5 V, as shown in  
Figure 40. The REFCAPA and REFCAPB pins must be shorted  
together externally, and a ceramic capacitor of 10 μF applied to  
REFGND, to ensure that the reference buffer is in closed-loop  
operation. The reference voltage available at the REFIN/REFOUT  
pin is 2.5 V.  
AD7606  
AD7606  
AD7606  
REF SELECT  
REF SELECT  
REF SELECT  
REFIN/REFOUT  
REFIN/REFOUT  
REFIN/REFOUT  
100nF  
100nF  
100nF  
ADR421  
0.1µF  
When the AD7606/AD7606-6/AD7606-4 are configured in  
external reference mode, the REFIN/REFOUT pin is a high  
input impedance pin. For applications using multiple AD7606  
devices, the following configurations are recommended,  
depending on the application requirements.  
Figure 41. Single External Reference Driving Multiple AD7606/AD7606-6/  
AD7606-4 REFIN Pins  
V
DRIVE  
External Reference Mode  
AD7606  
AD7606  
AD7606  
REF SELECT  
REF SELECT  
REF SELECT  
One ADR421 external reference can be used to drive the  
REFIN/REFOUT pins of all AD7606 devices (see Figure 41).  
In this configuration, each REFIN/REFOUT pin of the  
AD7606/AD7606-6/AD7606-4 should be decoupled with at  
least a 100 nF decoupling capacitor.  
REFIN/REFOUT  
REFIN/REFOUT  
REFIN/REFOUT  
+
10µF  
100nF  
100nF  
Figure 42. Internal Reference Driving Multiple AD7606/AD7606-6/AD7606-4  
REFIN Pins  
Rev. D | Page 24 of 36  
 
 
 
 
Data Sheet  
AD7606/AD7606-6/AD7606-4  
The power-down mode is selected through the state of the  
TYPICAL CONNECTION DIAGRAM  
STBY  
RANGE pin when the  
pin is low. Table 7 shows the  
Figure 43 shows the typical connection diagram for the AD7606/  
AD7606-6/AD7606-4. There are four AVCC supply pins on the  
part, and each of the four pins should be decoupled using a 100 nF  
capacitor at each supply pin and a 10 µF capacitor at the supply  
source. The AD7606/AD7606-6/AD7606-4 can operate with the  
internal reference or an externally applied reference. In this  
configuration, the AD7606 is configured to operate with the  
internal reference. When using a single AD7606/AD7606-6/  
AD7606-4 device on the board, the REFIN/REFOUT pin  
should be decoupled with a 10 µF capacitor. Refer to the  
Internal/External Reference section when using an application  
with multiple AD7606/AD7606-6/AD7606-4 devices. The  
REFCAPA and REFCAPB pins are shorted together and  
decoupled with a 10 µF ceramic capacitor.  
configurations required to choose the desired power-down mode.  
When the AD7606/AD7606-6/AD7606-4 are placed in standby  
mode, the current consumption is 8 mA maximum and power-  
up time is approximately 100 µs because the capacitor on the  
REFCAPA and REFCAPB pins must charge up. In standby mode,  
the on-chip reference and regulators remain powered up, and  
the amplifiers and ADC core are powered down.  
When the AD7606/AD7606-6/AD7606-4 are placed in shutdown  
mode, the current consumption is 6 µA maximum and power-up  
time is approximately 13 ms (external reference mode). In shut-  
down mode, all circuitry is powered down. When the AD7606/  
AD7606-6/AD7606-4 are powered up from shutdown mode,  
a RESET signal must be applied to the AD7606/AD7606-6/  
AD7606-4 after the required power-up time has elapsed.  
The VDRIVE supply is connected to the same supply as the  
processor. The VDRIVE voltage controls the voltage value of the  
output logic signals. For layout, decoupling, and grounding  
hints, see the Layout Guidelines section.  
Table 7. Power-Down Mode Selection  
STBY  
Power-Down Mode  
Standby  
Shutdown  
RANGE  
0
0
1
0
After supplies are applied to the AD7606/AD7606-6/AD7606-4,  
a reset should be applied to the AD7606/AD7606-6/AD7606-4  
to ensure that it is configured for the correct mode of operation.  
POWER-DOWN MODES  
Two power-down modes are available on the AD7606/AD7606-6/  
STBY  
AD7606-4: standby mode and shutdown mode. The  
pin  
controls whether the AD7606/AD7606-6/AD7606-4 are in  
normal mode or in one of the two power-down modes.  
ANALOG SUPPLY  
VOLTAGE 5V1  
DIGITAL SUPPLY  
VOLTAGE +2.3V TO +5.25V  
+
1µF  
10µF  
100nF  
100nF  
2
AV  
V
DRIVE  
REFIN/REFOUT  
REFCAPA  
REGCAP  
CC  
PARALLEL  
INTERFACE  
+
DB0 TO DB15  
10µF  
REFCAPB  
REFGND  
CONVST A, CONVST B  
CS  
RD  
BUSY  
V1  
V1GND  
V2  
V2GND  
V3  
V3GND  
V4  
AD7606  
RESET  
OS 2  
OS 1  
OS 0  
OVERSAMPLING  
EIGHT ANALOG  
INPUTS V1 TO V8  
V4GND  
V5  
REF SELECT  
PAR/SER SEL  
V
DRIVE  
V5GND  
V6  
V6GND  
V7  
V7GND  
V8  
V8GND  
RANGE  
STBY  
V
DRIVE  
AGND  
1
DECOUPLING SHOWN ON THE AV PIN APPLIES TO EACH AV PIN (PIN 1, PIN 37, PIN 38, PIN 48).  
CC  
CC  
DECOUPLING CAPACITOR CAN BE SHARED BETWEEN AV  
PIN 37 AND PIN 38.  
CC  
2
DECOUPLING SHOWN ON THE REGCAP PIN APPLIES TO EACH REGCAP PIN (PIN 36, PIN 39).  
Figure 43. AD7606 Typical Connection Diagram  
Rev. D | Page 25 of 36  
 
 
 
 
AD7606/AD7606-6/AD7606-4  
Data Sheet  
transformers. In a 50 Hz system, this allows for up to 9° of phase  
compensation; and in a 60 Hz system, it allows for up to 10° of  
phase compensation.  
CONVERSION CONTROL  
Simultaneous Sampling on All Analog Input Channels  
The AD7606/AD7606-6/AD7606-4 allow simultaneous sampling  
of all analog input channels. All channels are sampled simul-  
taneously when both CONVST pins (CONVST A, CONVST B)  
are tied together. A single CONVST signal is used to control both  
CONVST x inputs. The rising edge of this common CONVST  
signal initiates simultaneous sampling on all analog input channels  
(V1 to V8 for the AD7606, V1 to V6 for the AD7606-6, and V1  
to V4 for the AD7606-4).  
This is accomplished by pulsing the two CONVST pins  
independently and is possible only if oversampling is not in use.  
CONVST A is used to initiate simultaneous sampling of the  
first set of channels (V1 to V4 for the AD7606, V1 to V3 for the  
AD7606-6, and V1 and V2 for the AD7606-4); and CONVST B  
is used to initiate simultaneous sampling on the second set of  
analog input channels (V5 to V8 for the AD7606, V4 to V6 for  
the AD7606-6, and V3 and V4 for the AD7606-4), as illustrated  
in Figure 44. On the rising edge of CONVST A, the track-and-  
hold amplifiers for the first set of channels are placed into hold  
mode. On the rising edge of CONVST B, the track-and-hold  
amplifiers for the second set of channels are placed into hold  
mode. The conversion process begins once both rising edges  
of CONVST x have occurred; therefore BUSY goes high on the  
rising edge of the later CONVST x signal. In Table 3, Time t5  
indicates the maximum allowable time between CONVST x  
sampling points.  
The AD7606 contains an on-chip oscillator that is used to  
perform the conversions. The conversion time for all ADC  
channels is tCONV. The BUSY signal indicates to the user when  
conversions are in progress, so when the rising edge of CONVST  
is applied, BUSY goes logic high and transitions low at the end  
of the entire conversion process. The falling edge of the BUSY  
signal is used to place all eight track-and-hold amplifiers back  
into track mode. The falling edge of BUSY also indicates that  
the new data can now be read from the parallel bus (DB[15:0]),  
the DOUTA and DOUTB serial data lines, or the parallel byte bus,  
DB[7:0].  
There is no change to the data read process when using two  
separate CONVST x signals.  
Simultaneously Sampling Two Sets of Channels  
Connect all unused analog input channels to AGND. The results  
for any unused channels are still included in the data read because  
all channels are always converted.  
The AD7606/AD7606-6/AD7606-4 also allow the analog input  
channels to be sampled simultaneously in two sets. This can be  
used in power-line protection and measurement systems to  
compensate for phase differences introduced by PT and CT  
V1 TO V4 TRACK-AND-HOLD  
ENTER HOLD  
V5 TO V8 TRACK-AND-HOLD  
ENTER HOLD  
t5  
CONVST A  
CONVST B  
BUSY  
AD7606 CONVERTS  
ON ALL 8 CHANNELS  
tCONV  
CS/RD  
V1  
V2  
V3  
V7  
V8  
DATA: DB[15:0]  
FRSTDATA  
Figure 44. AD7606 Simultaneous Sampling on Channel Sets While Using Independent CONVST A and CONVST B Signals—Parallel Mode  
Rev. D | Page 26 of 36  
 
 
Data Sheet  
AD7606/AD7606-6/AD7606-4  
DIGITAL INTERFACE  
The AD7606/AD7606-6/AD7606-4 provide three interface  
options: a parallel interface, a high speed serial interface, and  
a parallel byte interface. The required interface mode is selected  
RD  
signal is logic low, it enables the data conversion  
result from each channel to be transferred to the digital host  
(DSP, FPGA).  
When the  
PAR  
via the  
/SER/BYTE SEL and DB15/BYTE SEL pins.  
When there is only one AD7606/AD7606-6/AD7606-4 in  
a system/board and it does not share the parallel bus, data can  
be read using just one control signal from the digital host. The  
Table 8. Interface Mode Selection  
PAR/SER/BYTE SEL  
DB1±  
Interface Mode  
CS  
RD  
and  
In this case, the data bus comes out of three-state on the falling  
CS RD CS RD  
signal allows the data  
signals can be tied together, as shown in Figure 5.  
0
1
1
0
0
1
Parallel interface mode  
Serial interface mode  
Parallel byte interface mode  
edge of  
/
. The combined  
and  
to be clocked out of the AD7606/AD7606-6/AD7606-4 and to  
Operation of the interface modes is discussed in the following  
sections.  
CS  
be read by the digital host. In this case,  
data transfer of each data channel.  
is used to frame the  
PARALLEL INTERFACE (PAR/SER/BYTE SEL = 0)  
PARALLEL BYTE (PAR/SER/BYTE SEL = 1, DB1± = 1)  
Data can be read from the AD7606/AD7606-6/AD7606-4 via  
Parallel byte interface mode operates much like the parallel  
interface mode, except that each channel conversion result is read  
CS  
RD  
the parallel data bus with standard  
PAR  
and  
signals. To read the  
data over the parallel bus, the  
CS RD  
/SER/BYTE SEL pin should  
input signals are internally gated to  
RD  
out in two 8-bit transfers. Therefore, 16  
to read all eight conversion results from the AD7606. For the  
RD  
pulses are required  
be tied low. The  
and  
enable the conversion result onto the data bus. The data lines,  
CS  
AD7606-6, 12  
pulses are required; and on the AD7606-4,  
pulses are required to read all the channel results.  
To configure the AD7606/AD76706-6/AD7606-4 to operate in  
PAR  
DB15 to DB0, leave their high impedance state when both  
RD  
RD  
eight  
and  
are logic low.  
parallel byte mode, the  
/SER/BYTE SEL and BYTE SEL/  
AD7606  
INTERRUPT  
BUSY 14  
DB15 pins should be tied to logic high (see Table 8). In parallel  
byte mode, DB[7:0] are used to transfer the data to the digital  
host. DB0 is the LSB of the data transfer, and DB7 is the MSB of  
the data transfer. In parallel byte mode, DB14 acts as an HBEN  
pin. When DB14/HBEN is tied to logic high, the most  
significant byte (MSB) of the conversion result is output first,  
followed by the LSB of the conversion result. When DB14 is tied  
to logic low, the LSB of the conversion result is output first,  
followed by the MSB of the conversion result. The FRSTDATA  
pin remains high until the entire 16 bits of the conversion result  
from V1 are read from the AD7606/AD7606-6/AD7606-4.  
13  
12  
CS  
RD/SCLK  
DB[15:0]  
DIGITAL  
HOST  
[33:24]  
[22:16]  
Figure 4±. AD7606 Interface Diagram—One AD7606 Using the Parallel Bus,  
CS RD  
with and  
Shorted Together  
CS  
CS  
CS  
The rising edge of the  
the falling edge of the  
high impedance state.  
input signal three-states the bus, and  
input signal takes the bus out of the  
is the control signal that enables the  
data lines; it is the function that allows multiple AD7606/  
AD7606-6/ AD7606-4 devices to share the same parallel  
data bus.  
SERIAL INTERFACE (PAR/SER/BYTE SEL = 1)  
To read data back from the AD7606 over the serial interface, the  
CS  
RD  
signal  
The  
signal can be permanently tied low, and the  
PAR  
CS  
/SER/BYTE SEL pin must be tied high. The  
and SCLK  
can be used to access the conversion results as shown in Figure 4.  
A read operation of new data can take place after the BUSY  
signal goes low (see Figure 2); or, alternatively, a read operation  
of data from the previous conversion process can take place  
while BUSY is high (see Figure 3).  
signals are used to transfer data from the AD7606. The AD7606/  
AD7606-6/AD7606-4 have two serial data output pins, DOUTA  
and DOUTB. Data can be read back from the AD7606/AD76706-  
6/AD7606-4 using one or both of these DOUT lines. For the  
AD7606, conversion results from Channel V1 to Channel V4  
first appear on DOUTA, and conversion results from Channel V5  
to Channel V8 first appear on DOUTB. For the AD7606-6,  
conversion results from Channel V1 to Channel V3 first appear  
on DOUTA, and conversion results from Channel V4 to Channel  
V6 first appear on DOUTB. For the AD7606-4, conversion results  
from Channel V1 and Channel V2 first appear on DOUTA, and  
conversion results from Channels V3 and Channel V4 first  
appear on DOUTB.  
RD  
The  
results register. Applying a sequence of  
of the AD7606/AD7606-6/AD7606-4 clocks the conversion  
pin is used to read data from the output conversion  
RD RD  
pulses to the  
pin  
results out from each channel onto the Parallel Bus DB[15:0] in  
RD  
ascending order. The first  
falling edge after BUSY goes low  
RD  
clocks out the conversion result from Channel V1. The next  
falling edge updates the bus with the V2 conversion result, and so  
on. On the AD7606, the eighth falling edge of  
conversion result for Channel V8.  
RD  
clocks out the  
Rev. D | Page 27 of 36  
 
 
 
 
 
AD7606/AD7606-6/AD7606-4  
Data Sheet  
CS  
CS  
takes the bus out of three-state and clocks  
The falling edge takes the data output lines, DOUTA and DOUTB,  
The falling edge of  
out of three-state and clocks out the MSB of the conversion  
result. The rising edge of SCLK clocks all subsequent data bits  
out the MSB of the 16-bit conversion result. This MSB is valid  
CS  
on the first falling edge of the SCLK after the  
falling edge.  
CS  
onto the serial data outputs, DOUTA and DOUTB. The  
input  
The subsequent 15 data bits are clocked out of the AD7606/  
AD7606-6/AD7606-4 on the SCLK rising edge. Data is valid on  
the SCLK falling edge. To access each conversion result, 16 clock  
cycles must be provided to the AD7606/AD7606-6/AD7606-4.  
can be held low for the entire serial read operation, or it can be  
pulsed to frame each channel read of 16 SCLK cycles. Figure 46  
shows a read of eight simultaneous conversion results using two  
D
OUT lines on the AD7606. In this case, a 64 SCLK transfer is used  
The FRSTDATA output signal indicates when the first channel,  
CS  
to access data from the AD7606, and is held low to frame the  
entire 64 SCLK cycles. Data can also be clocked out using just  
one DOUT line, in which case it is recommended that DOUTA be  
used to access all conversion data because the channel data is  
output in ascending order. For the AD7606 to access all eight  
conversion results on one DOUT line, a total of 128 SCLK cycles  
CS  
V1, is being read back. When the input is high, the FRSTDATA  
output pin is in three-state. In serial mode, the falling edge of  
CS  
takes FRSTDATA out of three-state and sets the FRSTDATA  
pin high, indicating that the result from V1 is available on the  
DOUTA output data line. The FRSTDATA output returns to  
a logic low following the 16th SCLK falling edge. If all channels  
are read on DOUTB, the FRSTDATA output does not go high when  
V1 is being output on this serial data output pin. It goes high  
only when V1 is available on DOUTA (and this is when V5 is  
available on DOUTB for the AD7606).  
CS  
is required. These 128 SCLK cycles can be framed by one  
signal, or each group of 16 SCLK cycles can be individually  
CS  
framed by the  
signal. The disadvantage of using just one  
DOUT line is that the throughput rate is reduced if reading occurs  
after conversion. The unused DOUT line should be left unconnected  
in serial mode. For the AD7606, if DOUTB is to be used as a single  
READING DURING CONVERSION  
Data can be read from the AD7606/AD7606-6/AD7606-4 while  
BUSY is high and the conversions are in progress. This has little  
effect on the performance of the converter, and it allows a faster  
throughput rate to be achieved. A parallel, parallel byte, or serial  
read can be performed during conversions and when oversampling  
may or may not be in use. Figure 3 shows the timing diagram for  
reading while BUSY is high in parallel or serial mode. Reading  
during conversions allows the full throughput rate to be achieved  
when using the serial interface with VDRIVE above 4.75 V.  
D
OUT line, the channel results are output in the following order:  
V5, V6, V7, V8, V1, V2, V3, and V4; however, the FRSTDATA  
indicator returns low after V5 is read on DOUTB. For the AD7606-6  
and the AD7606-4, if DOUTB is to be used as a single DOUT line,  
the channel results are output in the following order: V4, V5, V6,  
V1, V2, and V3 for the AD7606-6; and V3, V4, V1, and V2 for  
the AD7606-4.  
Figure 6 shows the timing diagram for reading one channel of  
CS  
data, framed by the  
AD7606-4 in serial mode. The SCLK input signal provides the  
CS  
signal, from the AD7606/AD7606-6/  
Data can be read from the AD7606 at any time other than on  
the falling edge of BUSY because this is when the output data  
registers are updated with the new conversion data. Time t6, as  
outlined in Table 3, should be observed in this condition.  
clock source for the serial read operation. The  
goes low to  
access the data from the AD7606/AD7606-6/AD7606-4.  
CS  
64  
SCLK  
D
D
A
B
V1  
V5  
V2  
V6  
V3  
V7  
V4  
V8  
OUT  
OUT  
Figure 46. AD7606 Serial Interface with Two DOUT Lines  
Rev. D | Page 28 of 36  
 
 
Data Sheet  
AD7606/AD7606-6/AD7606-4  
tCYCLE  
DIGITAL FILTER  
CONVST A  
AND  
The AD7606/AD7606-6/AD7606-4 contain an optional digital  
first-order sinc filter that should be used in applications where  
slower throughput rates are used or where higher signal-to-noise  
ratio or dynamic range is desirable. The oversampling ratio of the  
digital filter is controlled using the oversampling pins, OS [2:0] (see  
Table 9). OS 2 is the MSB control bit, and OS 0 is the LSB control  
bit. Table 9 provides the oversampling bit decoding to select the  
different oversample rates. The OS pins are latched on the falling  
edge of BUSY. This sets the oversampling rate for the next  
conversion (see Figure 48). In addition to the oversampling  
function, the output result is decimated to 16-bit resolution.  
tCONV  
19µs  
CONVST B  
9µs  
4µs  
OS = 0 OS = 2 OS = 4  
BUSY  
t4  
t4  
t4  
CS  
RD  
DATA:  
DB[15:0]  
If the OS pins are set to select an OS ratio of eight, the next  
CONVST x rising edge takes the first sample for each channel,  
and the remaining seven samples for all channels are taken with  
an internally generated sampling signal. These samples are then  
averaged to yield an improvement in SNR performance. Table 9  
shows typical SNR performance for both the 10 V and the 5 V  
range. As Table 9 shows, there is an improvement in SNR as the  
OS ratio increases. As the OS ratio increases, the 3 dB frequency  
is reduced, and the allowed sampling frequency is also reduced.  
In an application where the required sampling frequency is  
10 kSPS, an OS ratio of up to 16 can be used. In this case, the  
application sees an improvement in SNR, but the input 3 dB  
bandwidth is limited to ~6 kHz.  
Figure 47. AD7606—No Oversampling, Oversampling × 2, and  
Oversampling × 4 While Using Read After Conversion  
Figure 47 shows that the conversion time extends as the over-  
sampling rate is increased, and the BUSY signal lengthens for the  
different oversampling rates. For example, a sampling frequency  
of 10 kSPS yields a cycle time of 100 µs. Figure 47 shows OS × 2  
and OS × 4; for a 10 kSPS example, there is adequate cycle time to  
further increase the oversampling rate and yield greater improve-  
ments in SNR performance. In an application where the initial  
sampling or throughput rate is at 200 kSPS, for example, and  
oversampling is turned on, the throughput rate must be reduced  
to accommodate the longer conversion time and to allow for the  
read. To achieve the fastest throughput rate possible when over-  
sampling is turned on, the read can be performed during the  
BUSY high time. The falling edge of BUSY is used to update the  
output data registers with the new conversion data; therefore, the  
reading of conversion data should not occur on this edge.  
The CONVST A and CONVST B pins must be tied/driven  
together when oversampling is turned on. When the over-  
sampling function is turned on, the BUSY high time for the  
conversion process extends. The actual BUSY high time  
depends on the oversampling rate that is selected: the higher the  
oversampling rate, the longer the BUSY high, or total conversion  
time (see Table 3).  
CONVST A  
AND  
CONVST B  
OVERSAMPLE RATE  
LATCHED FOR CONVERSION N + 1  
CONVERSION N  
CONVERSION N + 1  
BUSY  
tOS_HOLD  
tOS_SETUP  
OS x  
Figure 48. OS x Pin Timing  
Table 9. Oversample Bit Decoding  
Maximum Throughput  
OS  
SNR ± V Range SNR 10 V Range  
3 dB BW ± V Range 3 dB BW 10 V Range  
OS[2:0] Ratio  
(dB)  
(dB)  
(kHz)  
(kHz)  
CONVST Frequency (kHz)  
000  
001  
010  
011  
100  
101  
110  
111  
No OS  
2
4
8
16  
32  
64  
Invalid  
89  
90  
92  
93.6  
95  
96  
15  
15  
13.7  
10.3  
6
22  
22  
18.5  
11.9  
6
200  
100  
50  
91.2  
92.6  
94.2  
95.5  
96.4  
96.9  
25  
12.5  
6.25  
3.125  
96.7  
97  
3
1.5  
3
1.5  
Rev. D | Page 29 of 36  
 
 
 
 
AD7606/AD7606-6/AD7606-4  
Data Sheet  
1400  
1200  
1000  
800  
600  
400  
200  
0
Figure 49 to Figure 55 illustrate the effect of oversampling on  
the code spread in a dc histogram plot. As the oversample rate  
is increased, the spread of the codes is reduced.  
1000  
OVERSAMPLING BY 8  
= 25kSPS  
1263  
F
SAMPLE  
AV = 5V  
CC  
V
= 2.5V  
DRIVE  
NO OVERSAMPLING  
= 200kSPS  
928  
887  
783  
F
SAMPLE  
AV = 5V  
900  
800  
700  
600  
500  
400  
300  
200  
100  
0
CC  
V
= 2.5V  
DRIVE  
0
0
2
0
2
0
3
–3  
–2  
–1  
0
1
CODE (LSB)  
131  
–1  
97  
2
Figure ±2. Histogram of Codes—OS × 8 (Three Codes)  
0
3
2
3
1400  
OVERSAMPLING BY 16  
= 12.5kSPS  
–3  
–2  
0
1
1453  
F
SAMPLE  
AV = 5V  
CODE (LSB)  
1200  
1000  
800  
600  
400  
200  
0
CC  
V
= 2.5V  
DRIVE  
Figure 49. Histogram of Codes—No OS (Six Codes)  
1400  
OVERSAMPLING BY 2  
= 100kSPS  
F
SAMPLE  
AV = 5V  
1200  
1000  
800  
600  
400  
200  
0
CC  
1148  
V
= 2.5V  
DRIVE  
595  
804  
0
0
0
0
2
0
3
–3  
–2  
–1  
0
1
CODE (LSB)  
Figure ±3. Histogram of Codes—OS × 16 (Two Codes)  
80  
–1  
1600  
16  
2
0
0
0
3
OVERSAMPLING BY 32  
= 6.125kSPS  
1417  
F
–3  
–2  
0
1
SAMPLE  
AV = 5V  
1400  
1200  
1000  
800  
600  
400  
200  
0
CC  
CODE (LSB)  
V
= 2.5V  
DRIVE  
Figure ±0. Histogram of Codes—OS × 2 (Four Codes)  
1400  
OVERSAMPLING BY 4  
= 50kSPS  
1262  
F
SAMPLE  
AV = 5V  
1200  
1000  
800  
600  
400  
200  
0
CC  
631  
V
= 2.5V  
DRIVE  
764  
0
0
0
0
2
0
3
–3  
–2  
–1  
0
1
CODE (LSB)  
Figure ±4. Histogram of Codes—OS × 32 (Two Codes)  
1600  
OVERSAMPLING BY 64  
= 3kSPS  
1679  
19  
–1  
0
0
3
2
0
3
F
SAMPLE  
AV = 5V  
1400  
1200  
1000  
800  
600  
400  
200  
0
–3  
–2  
0
1
CC  
V
= 2.5V  
DRIVE  
CODE (LSB)  
Figure ±1. Histogram of Codes—OS × 4 (Four Codes)  
369  
0
0
0
0
2
0
3
–3  
–2  
–1  
0
1
CODE (LSB)  
Figure ±±. Histogram of Codes—OS × 64 (Two Codes)  
Rev. D | Page 30 of 36  
 
 
Data Sheet  
AD7606/AD7606-6/AD7606-4  
0
–10  
–20  
–30  
–40  
–50  
–60  
–70  
–80  
–90  
–100  
When the oversampling mode is selected for the AD7606/  
AD7606-6/AD7606-4, it has the effect of adding a digital filter  
function after the ADC. The different oversampling rates and  
the CONVST sampling frequency produce different digital filter  
frequency profiles.  
AV  
= 5V  
= 5V  
CC  
V
DRIVE  
= 25°C  
T
A
10V RANGE  
OS BY 16  
Figure 56 to Figure 61 show the digital filter frequency profiles for  
the different oversampling rates. The combination of the analog  
antialiasing filter and the oversampling digital filter can be used  
to eliminate and reduce the complexity of the design of any filter  
before the AD7606/AD7606-6/AD7606-4. The digital filtering  
combines steep roll-off and linear phase response.  
100  
1k  
10k  
100k  
1M  
10M  
0
AV  
= 5V  
CC  
FREQUENCY (Hz)  
V
= 5V  
DRIVE  
= 25°C  
–10  
–20  
–30  
–40  
–50  
–60  
–70  
–80  
–90  
T
A
Figure ±9. Digital Filter Response for OS 16  
10V RANGE  
OS BY 2  
0
–10  
–20  
–30  
–40  
–50  
–60  
–70  
–80  
–90  
–100  
AV  
= 5V  
= 5V  
CC  
V
DRIVE  
= 25°C  
T
A
10V RANGE  
OS BY 32  
100  
1k  
10k  
100k  
1M  
10M  
FREQUENCY (Hz)  
Figure ±6. Digital Filter Response for OS 2  
100  
1k  
10k  
100k  
1M  
10M  
0
–10  
–20  
–30  
–40  
–50  
–60  
–70  
–80  
–90  
–100  
AV  
= 5V  
= 5V  
CC  
FREQUENCY (Hz)  
V
DRIVE  
= 25°C  
T
A
Figure 60. Digital Filter Response for OS 32  
10V RANGE  
OS BY 4  
0
–10  
–20  
–30  
–40  
–50  
–60  
–70  
–80  
–90  
–100  
AV  
= 5V  
CC  
V
T
= 5V  
DRIVE  
= 25°C  
A
10V RANGE  
OS BY 64  
100  
1k  
10k  
100k  
1M  
10M  
FREQUENCY (Hz)  
Figure ±7. Digital Filter Response for OS 4  
0
–10  
–20  
–30  
–40  
–50  
–60  
–70  
–80  
–90  
–100  
100  
1k  
10k  
100k  
1M  
10M  
AV  
= 5V  
= 5V  
CC  
FREQUENCY (Hz)  
V
DRIVE  
= 25°C  
T
A
Figure 61. Digital Filter Response for OS 64  
10V RANGE  
OS BY 8  
100  
1k  
10k  
100k  
1M  
10M  
FREQUENCY (Hz)  
Figure ±8. Digital Filter Response for OS 8  
Rev. D | Page 31 of 36  
 
 
AD7606/AD7606-6/AD7606-4  
Data Sheet  
Figure 62 shows the recommended decoupling on the top layer  
of the AD7606 board. Figure 63 shows bottom layer decoupling,  
which is used for the four AVCC pins and the VDRIVE pin decoupling.  
Where the ceramic 100 nF caps for the AVCC pins are placed  
close to their respective device pins, a single 100 nF capacitor  
can be shared between Pin 37 and Pin 38.  
LAYOUT GUIDELINES  
The printed circuit board that houses the AD7606/AD7606-6/  
AD7606-4 should be designed so that the analog and digital  
sections are separated and confined to different areas of the board.  
At least one ground plane should be used. It can be common or  
split between the digital and analog sections. In the case of the  
split plane, the digital and analog ground planes should be  
joined in only one place, preferably as close as possible to the  
AD7606/AD7606-6/AD7606-4.  
If the AD7606/AD7606-6/AD7606-4 are in a system where  
multiple devices require analog-to-digital ground connections,  
the connection should still be made at only one point: a star  
ground point that should be established as close as possible to the  
AD7606/AD7606-6/AD7606-4. Good connections should be  
made to the ground plane. Avoid sharing one connection for  
multiple ground pins. Use individual vias or multiple vias to the  
ground plane for each ground pin.  
Avoid running digital lines under the devices because doing so  
couples noise onto the die. The analog ground plane should be  
allowed to run under the AD7606/AD7606-6/AD7606-4 to  
avoid noise coupling. Fast switching signals like CONVST A,  
CONVST B, or clocks should be shielded with digital ground  
to avoid radiating noise to other sections of the board, and they  
should never run near analog signal paths. Avoid crossover of  
digital and analog signals. Traces on layers in close proximity on  
the board should run at right angles to each other to reduce the  
effect of feedthrough through the board.  
Figure 62. Top Layer Decoupling REFIN/REFOUT,  
REFCAPA, REFCAPB, and REGCAP Pins  
The power supply lines to the AVCC and VDRIVE pins on the  
AD7606/AD7606-6/AD7606-4 should use as large a trace as  
possible to provide low impedance paths and reduce the effect  
of glitches on the power supply lines. Where possible, use supply  
planes and make good connections between the AD7606 supply  
pins and the power tracks on the board. Use a single via or multiple  
vias for each supply pin.  
Good decoupling is also important to lower the supply impedance  
presented to the AD7606/AD7606-6/AD7606-4 and to reduce  
the magnitude of the supply spikes. The decoupling capacitors  
should be placed close to (ideally, right up against) these pins  
and their corresponding ground pins. Place the decoupling  
capacitors for the REFIN/REFOUT pin and the REFCAPA and  
REFCAPB pins as close as possible to their respective AD7606/  
AD7606-6/AD7606-4 pins; and, where possible, they should be  
placed on the same side of the board as the AD7606 device.  
Figure 63. Bottom Layer Decoupling  
Rev. D | Page 32 of 36  
 
 
 
Data Sheet  
AD7606/AD7606-6/AD7606-4  
To ensure good device-to-device performance matching in  
a system that contains multiple AD7606/AD7606-6/AD7606-4  
devices, a symmetrical layout between the AD7606/AD7606-6/  
AD7606-4 devices is important.  
AVCC  
Figure 64 shows a layout with two AD7606/AD7606-6/AD7606-4  
devices. The AVCC supply plane runs to the right of both devices,  
and the VDRIVE supply track runs to the left of the two devices.  
The reference chip is positioned between the two devices, and  
the reference voltage track runs north to Pin 42 of U1 and south  
to Pin 42 of U2. A solid ground plane is used.  
U2  
These symmetrical layout principles can also be applied to a system  
that contains more than two AD7606/AD7606-6/AD7606-4  
devices. The AD7606/AD7606-6/AD7606-4 devices can be placed  
in a north-south direction, with the reference voltage located  
midway between the devices and the reference track running in  
the north-south direction, similar to Figure 64.  
U1  
Figure 64. Layout for Multiple AD7606 Devices—Top Layer and  
Supply Plane Layer  
Rev. D | Page 33 of 36  
 
AD7606/AD7606-6/AD7606-4  
OUTLINE DIMENSIONS  
Data Sheet  
12.20  
12.00 SQ  
11.80  
0.75  
0.60  
0.45  
1.60  
MAX  
64  
49  
1
48  
PIN 1  
10.20  
10.00 SQ  
9.80  
TOP VIEW  
(PINS DOWN)  
1.45  
1.40  
1.35  
0.20  
0.09  
7°  
3.5°  
0°  
0.08  
COPLANARITY  
16  
33  
0.15  
0.05  
SEATING  
17  
32  
PLANE  
VIEW A  
0.27  
0.22  
0.17  
0.50  
BSC  
LEAD PITCH  
VIEW A  
ROTATED 90° CCW  
COMPLIANT TO JEDEC STANDARDS MS-026-BCD  
Figure 65. 64-Lead Low Profile Quad Flat Package [LQFP]  
(ST-64-2)  
Dimensions shown in millimetres  
ORDERING GUIDE  
Model1, 2, 3  
AD7606BSTZ  
AD7606BSTZ-RL  
AD7606BSTZ-6  
AD7606BSTZ-6RL  
AD7606BSTZ-4  
AD7606BSTZ-4RL  
EVAL-AD7606SDZ  
EVAL-AD7606-6SDZ  
EVAL-AD7606-4SDZ  
EVAL-SDP-CB1Z  
Temperature Range  
−40°C to +85°C  
−40°C to +85°C  
−40°C to +85°C  
−40°C to +85°C  
−40°C to +85°C  
−40°C to +85°C  
Package Description  
Package Option  
64-Lead Low Profile Quad Flat Package [LQFP]  
64-Lead Low Profile Quad Flat Package [LQFP]  
64-Lead Low Profile Quad Flat Package [LQFP]  
64-Lead Low Profile Quad Flat Package [LQFP]  
64-Lead Low Profile Quad Flat Package [LQFP]  
64-Lead Low Profile Quad Flat Package [LQFP]  
Evaluation Board for the AD7606  
Evaluation Board for the AD7606-6  
Evaluation Board for the AD7606-4  
Evaluation Controller Board  
ST-64-2  
ST-64-2  
ST-64-2  
ST-64-2  
ST-64-2  
ST-64-2  
1 Z = RoHS Compliant Part.  
2 The EVAL-AD7606SDZ, EVAL-AD7606-6SDZ, and EVAL-AD7606-4SDZ can be used as standalone evaluation boards or in conjunction with the EVAL-SDP-CB1Z for  
evaluation/demonstration purposes.  
3 The EVAL-SDP-CB1Z allows the PC to control and communicate with all Analog Devices, Inc., evaluation boards ending in the SDZ designator.  
Rev. D | Page 34 of 36  
 
 
Data Sheet  
NOTES  
AD7606/AD7606-6/AD7606-4  
Rev. D | Page 35 of 36  
AD7606/AD7606-6/AD7606-4  
NOTES  
Data Sheet  
©2010–2017 Analog Devices, Inc. All rights reserved. Trademarks and  
registered trademarks are the property of their respective owners.  
D08479-0-11/17(D)  
Rev. D | Page 36 of 36  

相关型号:

EVAL-AD7606BFMCZ

8-Channel DAS Bipolar Input, Simultaneous Sampling ADC
ADI

EVAL-AD7606EDZ

8-/6-/4-Channel DAS with 16-Bit,Bipolar Input,Simultaneous Sampling ADC
ADI

EVAL-AD7606SDZ

8-/6-/4-Channel DAS with 16-Bit, Bipolar Input, Simultaneous Sampling ADC
ADI

EVAL-AD7607EDZ

8-Channel DAS with 14-Bit, Bipolar Input, Simultaneous Sampling ADC
ADI

EVAL-AD7609EDZ

8-Channel Differential DAS with 18-Bit
ADI

EVAL-AD7610CB

16-Bit, 250 kSPS, Unipolar/Bipolar Programmable Input PulSAR ADC
ADI

EVAL-AD7621CB2

16-Bit, 2 LSB INL, 3 MSPS PulSAR® ADC
ADI

EVAL-AD7625FMCZ

16-Bit, 6 MSPS, PulSAR Differential ADC
ADI

EVAL-AD7626EDZ

16-Bit, 10 MSPS, PulSAR Differential ADC
ADI

EVAL-AD7626FMCZ

16-Bit, 10 MSPS, PulSAR Differential ADC
ADI

EVAL-AD7631CBZ

18-Bit, 250 kSPS, Differential Programmable Input PulSAR® ADC
ADI

EVAL-AD7641CB

18-Bit 250/670 kSPS PulSAR Bipolar Programmable Inputs ADC
ADI