The ADS58J64 is a low-power, wide-bandwidth, 14bit, 1-GSPS, quad-channel, telecom receiver device.
The ADS58J64 supports a JESD204B serial interface
with data rates up to 10 Gbps with one lane per
channel. The buffered analog input provides uniform
input impedance across a wide frequency range and
minimizessample-and-holdglitchenergy.The
ADS58J64 provides excellent spurious-free dynamic
range (SFDR) over a large input frequency range with
very low power consumption. The digital signal
processing block includes complex mixers followed
by low-pass filters with decimate-by-2 and -4 options
supporting up to a 200-MHz receive bandwidth. The
ADS58J64 also supports a 14-bit, 500-MSPS output
in burst mode, making the device suitable for a digital
pre-distortion (DPD) observation receiver.
The JESD204B interface reduces the number of
interface lines, thus allowing high system integration
density.Aninternalphase-lockedloop(PLL)
multiplies the incoming analog-to-digital converter
(ADC) sampling clock to derive the bit clock that is
used to serialize the 14-bit data from each channel.
Device Information
PART NUMBERPACKAGEBODY SIZE (NOM)
ADS58J64VQFN (72)10.00 mm × 10.00 mm
(1) For all available packages, see the orderable addendum at
the end of the data sheet.
Simplified Block Diagram
(1)
2Applications
•Multi-Carrier GSM Cellular Infrastructure
•Multi-Carrier Multi-Mode Cellular Infrastructure
•Telecommunications Receivers
•Telecom DPD Observation Receivers
1
Base Stations
Base Stations
An IMPORTANT NOTICE at the end of this data sheet addresses availability, warranty, changes, use in safety-critical applications,
intellectual property matters and other important disclaimers. PRODUCTION DATA.
NOTE: Page numbers for previous revisions may differ from page numbers in the current version.
Changes from Original (January 2017) to Revision APage
•Changed Sample to Sampling in third Features bullet .......................................................................................................... 1
•Changed Bandwitdth: 250 MHz to Sample Rate: 500 MSPS in fourth Features bullet ......................................................... 1
•Added Input 3-dB Bandwidth bullet to Features section........................................................................................................ 1
•Changed plot and SNR and SFDR conditions of Figure 9 .................................................................................................. 13
•Added for loading trims to description of bit 1 in Register 64h Field Descriptions .............................................................. 45
•Changed select to set in description of bits 7-0 in Register 8Dh Field Descriptions and Register 8Eh Field Descriptions. 45
•Changed select to set in description of bits 7-0 in Register 8Fh Field Descriptions and Register 90h Field Descriptions.. 46
•Added Others: Do not use to Description column of Register 71h Field Descriptions and Register 72h Field
•Added Valid only when CTRL_LID = 1 to description of bits 7-4 in Register 2Dh Field Descriptions ................................. 57
•Changed Description column of Register 41h Field Descriptions........................................................................................ 61
•Changed 1 : to 3 : and added Others: Do not use to Description column of Register 42h Field Descriptions ................... 61
•Changed description of bits 7-0 in Register 07h Field Descriptions..................................................................................... 65
•Changed description of bits 7-0 in Register 08h Field Descriptions ................................................................................... 65
NC22, 23—No connection
PDN50I/O
RES49—Reserved pin, connect to GND
RESET48IHardware reset; active high. This pin has an internal 10-kΩ pulldown resistor.
SCLK6ISerial interface clock input. This pin has an internal 10-kΩ pulldown resistor.
SDIN5ISerial interface data input. This pin has an internal 10-kΩ pulldown resistor.
SDOUT11O1.8-V logic serial interface data output
SEN7ISerial interface enable. This pin has an internal 10-kΩ pullup resistor to DVDD.
IDifferential analog input pin for channel D, internal bias via a 2-kΩ resistor to V
IDifferential clock input pin for the ADC with internal 100-Ω differential termination, requires external ac coupling
IExternal SYSREF input, requires dc coupling and external termination
Power down. This pin can be configured via an SPI register setting. This pin has an internal 10-kΩ pulldown
resistor.
Trigger-ready output for burst mode for channels A and B. This pin can be configured via SPI to a TRDY signal for
all four channels in burst mode, and can be left open if not used.
Trigger-ready output for burst mode for channels C and D. This pin can be configured via SPI to a TRDY signal for
all four channels in burst mode, and can be left open if not used.
Manual burst mode trigger input for channels A and B. This pin can be configured via SPI to a manual trigger input
signal for all four channels in burst mode, and can be connected to GND if not used. This pin has an internal 10kΩ pulldown resistor.
Manual burst mode trigger input for channels C and D. This pin can be configured via SPI to a manual trigger
input signal for all four channels in burst mode, and can be connected to GND if not used. This pin has an internal
10-kΩ pulldown resistor.
OJESD204B serial data output pin for channel A
OJESD204B serial data output pin for channel B
OJESD204B serial data output pin for channel C
OJESD204B serial data output pin for channel D
Synchronization input pin for JESD204B port channels A and B. This pin can be configured via SPI to a SYNCb
I
signal for all four channels. This pin has an internal differential termination of 100 Ω.
Synchronization input pin for JESD204B port channels C and D. This pin can be configured via SPI to a SYNCb
I
signal for all four channels. This pin has an internal differential termination of 100 Ω..
over operating free-air temperature range (unless otherwise noted)
AVDD19–0.32.1
Supply voltage
Voltage between AGND and DGND–0.30.3V
Voltage applied to input pins
Storage temperature, T
(1) Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are stress ratings
only, which do not imply functional operation of the device at these or any other conditions beyond those indicated under RecommendedOperating Conditions. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability.
over operating free-air temperature range (unless otherwise noted)
MINNOMMAXUNIT
AVDD191.81.92
Supply voltage range
Analog inputs
Clock inputs
Temperature
(1) Assumes system thermal design meets the TJspecification.
(2) Prolonged use above this junction temperature can increase the device failure-in-time (FIT) rate.
(3) The recommended maximum temperature at the PCB footprint thermal pad assumes the junction-to-package bottom thermal resistance,
R
consumption is 2.5 W.
= 0.2°C/W, the thermal resistance of the device thermal pad connection to the PCB footprint is negligible, and the device power
θJC(bot)
AVDD1.11.151.2
DVDD1.11.151.2
V
IOVDD1.11.151.2
Differential input voltage range1.1V
PP
Input common-mode voltage (VCM)1.3V
Input clock frequency, device clock frequency4001000MHz
Sine wave, ac-coupled1.5
Input clock amplitude differential
(V
– V
CLKM
)
CLKP
LVPECL, ac-coupled1.6
LVDS, ac-coupled0.7
V
PP
Input device clock duty cycle, default after reset45%50%55%
104.5
(1)
(2)
ºC
(3)
Operating free-air, T
Operating junction, T
A
J
Specified maximum, measured at the device footprint thermal
pad on the printed circuit board, T
typical values are at TA= 25°C, full temperature range is from T
mode 8: 2x decimation with burst mode output, 50% clock duty cycle, AVDD19 = 1.9 V, AVDD = DVDD = 1.15 V, –1-dBFS
differential input, and fIN= 190 MHz (unless otherwise noted)
PARAMETERTEST CONDITIONSMINTYPMAX UNIT
GENERAL
ADC sampling rate1 GSPS
Resolution14Bits
POWER SUPPLY
AVDD19 1.9-V analog supply1.851.91.95V
AVDD1.15-V analog supply1.11.151.2V
DVDD1.15-V digital supply1.11.151.2V
I
AVDD19
I
AVDD
I
DVDD
PdisTotal power dissipation
ANALOG INPUTS
ISOLATION
1.9-V analog supply current100-MHz, full-scale input on all four channels618mA
1.15-V analog supply current100-MHz, full-scale input on all four channels415mA
Mode 8, 100 MHz, full-scale input on all four
channels
Mode 3, 100 MHz, full-scale input on all four
1.15-V digital supply current
channels
Mode 0 and 2, 100 MHz, full-scale input on all four
channels
Mode 1, 4, 6, and 7, 100 MHz, full-scale input on
all four channels
Mode 8, 100 MHz, full-scale input on all four
channels
Mode 3, 100 MHz, full-scale input on all four
channels
Mode 0 and 2, 100 MHz, full-scale input on all four
channels
Mode 1, 4, 6, and 7, 100 MHz, full-scale input on
CLKINP and CLKINM pins are connected to the
internal biasing voltage through a 5-kΩ resistor
0.7V
6.6 AC Performance
typical values are at TA= 25°C, full temperature range is from T
mode 8: 2x decimation with burst mode output, 50% clock duty cycle, AVDD19 = 1.9 V, AVDD = DVDD = 1.15 V, –1-dBFS
differential input, and fIN= 190 MHz (unless otherwise noted)
typical values are at TA= 25°C, full temperature range is from T
mode 8: 2x decimation with burst mode output, 50% clock duty cycle, AVDD19 = 1.9 V, AVDD = DVDD = 1.15 V, –1-dBFS
differential input, and fIN= 190 MHz (unless otherwise noted)
PARAMETERTEST CONDITIONSMINTYPMAXUNIT
DIGITAL INPUTS (RESET, SCLK, SEN, SDIN, PDN, TRIGAB, TRIGCD)
V
IH
V
IL
I
IH
I
IL
High-level input voltageAll digital inputs support 1.2-V and 1.8-V logic levels0.8V
Low-level input voltageAll digital inputs support 1.2-V and 1.8-V logic levels0.4V
typical values are at TA= 25°C, full temperature range is from T
mode 8: 2x decimation with burst mode output, 50% clock duty cycle, AVDD19 = 1.9 V, AVDD = DVDD = 1.15 V, –1-dBFS
differential input, and fIN= 190 MHz (unless otherwise noted)
SAMPLE TIMING CHARACTERISTICS
Aperture delay0.550.92ns
Aperture delay matching between two channels on the same device±100ps
Aperture delay matching between two devices at the same temperature and supply
voltage
Aperture jitter100fSrms
Wake-up time
Data latency: ADC
sample to digital
output
t
SU_SYSREF
t
H_SYSREF
Setup time for SYSREF, referenced to input clock rising edge350900ps
Hold time for SYSREF, referenced to input clock rising edge100ps
JESD OUTPUT INTERFACE TIMING CHARACTERISTICS
Unit interval100ps
Serial output data rate10Gbps
Total jitter for BER of 1E-15 and lane rate = 10 Gbps24ps
Random jitter for BER of 1E-15 and lane rate = 10 Gbps0.95ps rms
Deterministic jitter for BER of 1E-15 and lane rate = 10 Gbps8.8ps, pk-pk
tR, t
F
Data rise time, data fall time: rise and fall times measured from 20% to 80%,
differential output waveform, 2.5 Gbps ≤ bit rate ≤ 10 Gbps
Global power-down10ms
Pin power-down (fast power-down)5µs
Burst mode116
typical values are at TA= 25°C, full temperature range is from T
1 GSPS, mode 8: 2x decimation with burst mode output, 50% clock duty cycle, AVDD19 = 1.9 V, AVDD = DVDD = 1.15 V,
–1-dBFS differential input, and fIN= 190 MHz (unless otherwise noted)
typical values are at TA= 25°C, full temperature range is from T
1 GSPS, mode 8: 2x decimation with burst mode output, 50% clock duty cycle, AVDD19 = 1.9 V, AVDD = DVDD = 1.15 V,
–1-dBFS differential input, and fIN= 190 MHz (unless otherwise noted)
typical values are at TA= 25°C, full temperature range is from T
1 GSPS, mode 8: 2x decimation with burst mode output, 50% clock duty cycle, AVDD19 = 1.9 V, AVDD = DVDD = 1.15 V,
–1-dBFS differential input, and fIN= 190 MHz (unless otherwise noted)
The ADS58J64 is a quad-channel device with a complex digital down-converter (DDC) and digital decimation to
allow flexible signal processing to suit different usage cases. Each channel is composed of two interleaved
analog-to-digital converters (ADCs) sampling at half the input clock rate. The 2x interleaved data are decimated
by 2 to provide a processing gain of 3 dB. The decimation filter can be configured as low pass (default) or high
pass. The half-rate (with regards to the input clock) data are available on the output, in burst mode (DDC mode =
8) as a stream of high (14-bit) and low (9-bit) resolution samples. Burst mode can be enabled by device
programming along with other options (such as the number of high- and low-resolution samples, and trigger
mode as either automatic or pin-controlled). In default mode, the device operates in DDC mode 0, where the
input is mixed with a constant frequency of –fS/ 4 and is given out as complex IQ. The different operational
modes modes of the ADS58J64 are listed in Table 1.
The ADS58J64 analog signal inputs are designed to be driven differentially. The analog input pins have internal
analog buffers that drive the sampling circuit. As a result of the analog buffer, the input pins present a highimpedance input across a very wide frequency range to the external driving source that enables great flexibility in
the external analog filter design as well as excellent 50-Ω matching for RF applications. The buffer also helps
isolate the external driving circuit from the internal switching currents of the sampling circuit, resulting in a more
constant SFDR performance across input frequencies. The common-mode voltage of the signal inputs is
internally biased to 1.3 V using 2-kΩ resistors to allow for ac-coupling of the input drive network. Each input pin
(INP, INM) must swing symmetrically between (VCM + 0.275 V) and (VCM – 0.275 V), resulting in a 1.1-V
(default) differential input swing. The input sampling circuit has a 3-dB bandwidth that extends up to 1000 MHz.
7.3.2 Recommended Input Circuit
In order to achieve optimum ac performance, the following circuitry (shown in Figure 43) is recommended at the
analog inputs.
PP
Figure 43. Analog Input Driving Circuit
7.3.3 Clock Input
The clock inputs of the ADS58J64 supports LVDS and LVPECL standards. The CLKP, CLKM inputs have an
internal termination of 100 Ω. The clock inputs must be ac-coupled because the input pins are self-biased to a
common-mode voltage of 0.7 V, as shown in Figure 44 and Figure 45.
The ADS58J64 has two stages of digital decimation filters, as shown in Figure 46. The first stage is mandatory
and decimates by 2, and can be configured as either a low-pass or high-pass filter. The second stage decimation
supports real to complex quadrature demodulation and decimation by 2 or 4. After decimation, the complex
signal can be converter back to a real signal through digital quadrate modulation at a frequency of f
f
is the sample frequency after decimation.
OUT
Optionally, a burst mode output can be used to output the decimate-by-2 data directly.
The four channels can be configured as pairs (A, B and C, D) to either burst or decimation mode. If all four
channels are in decimation mode, then the decimation setting must be the same decimation for all four channels.
All modes of operation and the maximum bandwidth provided at a sample rate of 491.52 MSPS and
368.64 MSPS are listed in Table 1. The first stage decimation filter prior to the 16-bit numerically controlled
oscillator (NCO) is a noise suppression filter with 45% pass-band bandwidth relative to the input sample rate,
less than 0.2-dB ripple, and approximately 40-dB stop-band attenuation. This filter is only used to reduce the
ADC output rate from 1 GSPS to 500 MSPS prior to the second stage decimation filter or burst mode. Some
analog filtering of other Nyquist zones after the first stage decimation filter is expected to be required.
The second stage filter has a pass-band bandwidth of 81.4% relative to the output sample rate, supporting a
200-MHz bandwidth with a 245.76-MSPS complex output rate.
/ 4, where
OUT
Figure 46. ADS58J64 Channel (1 of 4) Block Diagram
7.4.1.1 Numerically Controlled Oscillators (NCOs) and Mixers
The ADS58J64 is equipped with a complex numerically-controlled oscillator. The oscillator generates a complex
exponential sequence: x[n] = e
jωn
. The frequency (ω) is specified by the 16-bit register setting. The complex
exponential sequence is multiplied by the real input from the ADC to mix the desired carrier down to 0 Hz.
The NCO frequency setting is set by the 16-bit register value, NCO_FREQ[n]:
(1)
7.4.1.2 Decimation Filter
The ADS58J64 has two decimation filters (decimate-by-2) in the data path. The first stage of the decimation filter
is non-programmable and is used in all functional modes. The second stage of decimation, available in DDC
mode 2 and 6, can be used to obtain noise and linearity improvement for low bandwidth applications.
7.4.1.2.1 Stage-1 Filter
The first stage filter is used for decimation of the 2x interleaved data from f
CLK
to f
/ 2. The frequency
CLK
response and pass-band ripple of the first stage decimation filter are shown in Figure 47 and Figure 48,
respectively.
Input clock rate = 1 GHz
Figure 47. Decimation Filter Response vs Frequency
The second stage filter is used for decimating the data from a sample rate of f
CLK
/ 2 to f
/ 4. The frequency
CLK
www.ti.com
response and pass-band ripple of the second stage filter are shown in Figure 49 and Figure 50, respectively.
Input clock rate (f
Figure 49. Decimation Filter Response vs Frequency
) = 1 GHz
CLK
Input clock rate (f
) = 1 GHz
CLK
Figure 50. Decimation Filter Pass-Band Ripple vs
Frequency
7.4.1.3 Mode 0: Decimate-by-4 With IQ Outputs and fS/ 4 Mixer
In mode 0, the DDC block includes a fixed frequency ±fS/ 4 complex digital mixer preceding the second stage
decimation filters. The IQ passband is approximately ±100 MHz centered at fS/ 8 or 3fS/ 8, as shown in
7.4.1.4 Mode 1: Decimate-by-4 With IQ Outputs and 16-Bit NCO
In mode 1, the DDC block includes a 16-bit frequency resolution complex digital mixer preceding the second
stage decimation filters, as shown in Figure 52.
Figure 52. Operating Mode 1
7.4.1.5 Mode 2: Decimate-by-4 With Real Output
In mode 2, the DDC block cascades two decimate-by-2 filters. Each filter can be configured as low pass (LP) or
high pass (HP) to allow down conversion of different frequency ranges, as shown in Table 2. The LP, HP and
HP, LP output spectra are inverted as shown in Figure 53.
Figure 53. Operating in Mode 2
Table 2. ADS58J64 Operating Mode 2 Down-Converted Frequency Ranges
7.4.1.6 Mode 3: Decimate-by-2 Real Output With Frequency Shift
In mode 3, the DDC block includes a 16-bit complex NCO digital mixer followed by a fS/ 4 mixer with a real
output to center the band at fS/ 4. The NCO must be set to a value different from ±fS/ 4, or else the samples are
zeroed as shown in Figure 54.
Figure 54. Operating Mode 3
7.4.1.7 Mode 4: Decimate-by-4 With Real Output
In mode 4, the DDC block includes a 16-bit complex NCO digital mixer preceding the second stage decimation
filter. The signal is then mixed with f
/ 4 to generate a real output, as shown in Figure 55. The bandwidth