JAJSD61 データシート

JAJSD61 May 2017 ADC32RF42

PRODUCTION DATA.

9 Detailed Description

9.1 Overview

The ADC32RF42 is a dual, 14-bit, 1.5-GSPS, analog-to-digital converter (ADC) followed by a multi-band digital down-converter (DDC) that can be bypassed, and a back-end JESD204B digital interface.

The ADCs are preceded by an input buffer and on-chip termination to provide a uniform input impedance over a large input frequency range. Furthermore, an internal differential clamping circuit provides first-level protection against overvoltage conditions. Each ADC channel is internally interleaved two times and equipped with background, analog and digital, and interleaving correction.

The on-chip DDC enables single- or dual-band internal processing to pre-select and filter smaller bands of interest and also reduces the digital output data traffic. Each DDC is equipped with up to three independent,
16-bit numerically-controlled oscillators (NCOs) for phase coherent frequency hopping; the NCOs can be controlled through the SPI or GPIO pins. The ADC32RF42 also provides three different power detectors on-chip with alarm outputs in order to support external automatic gain control (AGC) loops.

The processed data are passed into the JESD204B interface where the data are framed, encoded, serialized, and output on one to four lanes per channel, depending on the ADC sampling rate and decimation. The CLKIN, SYSREF, and SYNCB inputs provide the device clock and the SYSREF and SYNCB signals to the JESD204B interface that are used to derive the internal local frame and local multiframe clocks and establish the serial link. All features of the ADC32RF42 are configurable through the SPI.

9.2 Functional Block Diagram

9.3 Feature Description

9.3.1 Analog Inputs

The ADC32RF42 analog signal inputs are designed to be driven differentially. The analog input pins have internal analog buffers that drive the sampling circuit. The ADC32RF42 provides on-chip, differential termination to minimize reflections. The buffer also helps isolate the external driving circuit from the internal switching currents of the sampling circuit, thus resulting in a more constant SFDR performance across input frequencies.

The common-mode voltage of the signal inputs is internally biased to CM using the 32.5-Ω termination resistors that allow for ac-coupling of the input drive network. Figure 33 and Figure 34 show SDD11 at the analog inputs from dc to 5 GHz with a 100-Ω reference impedance.

ADC32RF42 equvlnt_inpt_imdnce_sbas747.gif

Figure 33. Equivalent Input Impedance

Figure 34. SDD11 Over the Input Frequency Range

The input impedance of analog inputs can also be modelled as parallel combination of equivalent resistance and capacitance. Figure 35 and Figure 36 show how equivalent impedance (C_IN and R_IN) vary over frequency.

Figure 35. Differential Input Capacitance vs
Input Frequency

Figure 36. Differential Input Resistance vs Input Frquency

Each input pin (INP, INM) must swing symmetrically between (CM + 0.3375 V) and (CM – 0.3375 V), resulting in a 1.35-V_PP (default) differential input swing. As shown in Figure 37, the input sampling circuit has a 3-dB bandwidth that extends up to approximately 3.2 GHz.

Figure 37. Input Bandwidth with a 100-Ω Source Resistance

9.3.1.1 Input Clamp Circuit

The ADC32RF42 analog inputs include an internal, differential clamp for overvoltage protection. As shown in Figure 38 and Figure 39, the clamp triggers for any input signals at approximately 600 mV above the input common-mode voltage, effectively limiting the maximum input signal to approximately 2.4 V_PP.

When the clamp circuit conducts, the maximum differential current flowing through the circuit (via input pins) must be limited to 20 mA.

Figure 38. Clamp Circuit in the ADC32RF42

Figure 39. Clamp Response Timing Diagram

9.3.2 Clock Input

The ADC32RF42 sampling clock input includes internal 100-Ω differential termination along with on-chip biasing. The clock input is recommended to be ac-coupled externally. The input bandwidth of the clock input is approximately 3 GHz; the smith chart of Figure 40 shows a clock input impedance with a 100-Ω reference impedance.

Figure 40. SDD11 of the Clock Input

The analog-to-digital converter (ADC) aperture jitter is a function of the clock amplitude applied to the pins. Figure 41 shows the equivalent aperture jitter for input frequencies at a 1-GHz and a 2-GHz input (f_S = 1.5 GSPS). Depending on the clock frequency, a matching circuit can be designed in order to maximize the clock amplitude.

Figure 41. Equivalent Aperture Jitter vs Input Clock Amplitude

9.3.3 SYSREF Input

The SYSREF signal is a periodic signal that is sampled by the ADC32RF42 device clock and is used to align the boundary of the local multiframe clock inside the data converter. SYSREF is also used to reset critical blocks [such as the clock divider for the interleaved ADCs, numerically-controlled oscillators (NCOs), decimation filters and so forth].

The SYSREF input requires external biasing. Furthermore, SYSREF must be established before the SPI registers are programmed. A programmable delay on the SYSREF input, as shown in Figure 42, is available to help with skew adjustment when the sampling clock and SYSREF are not provided from the same source.

Figure 42. SYSREF Internal Circuit Diagram

9.3.3.1 Using SYSREF

The ADC32RF42 uses SYSREF information to reset the clock divider, the NCO phase, and the LMFC counter of the JESD interface. The device provides flexibility to provide SYSREF information either from dedicated pins or through SPI register bits. As Figure 43 shows, SYSREF is asserted by a low-to-high transition on the SYSREF pins or a 0-to-1 change in the ASSERT SYSREF REG bit when using SPI registers.

ADC32RF42 sysref_dvdr_countr_sbas747.gif

Figure 43. Using SYSREF to Reset the Clock Divider, the NCO, and the LMFC Counter

The ADC32RF42 samples the SYSREF signal on the input clock rising edge. Required setup and hold time are listed in the Timing Requirements table. Table 1 shows that the input clock divider gets reset each time that SYSREF is asserted, whereas the NCO phase and the LMFC counter of the JESD interface are reset on each SYSREF assertion after disregarding the first two assertions.

Table 1. Asserting SYSREF

SYSREF ASSERTION INDEX	ACTION
SYSREF ASSERTION INDEX	INPUT CLOCK DIVIDER	NCO PHASE	LMFC COUNTER
1	Gets reset	Does not get reset	Does not get reset
2	Gets reset	Does not get reset	Does not get reset
3	Gets reset	Gets reset	Gets reset
4 and onwards	Gets reset	Gets reset	Gets reset

The SESREF use-cases can be classified broadly into two categories:

SYSREF is applied as aperiodic multi-shot pulses.

Figure 44 shows a case when only a counted number of pulses are applied as SYSREF to the ADC.

Alternatively, the SYSREF buffer can be powered down with the PDN SYSREF bit.

Figure 44. SYSREF Used as Aperiodic, Finite Number of Pulses

After the first SYSREF pulse is applied, allow the DLL in the clock path to settle by waiting for the t_DLL time (> 40 µs) before applying the second pulse. During this time, mask the SYSREF going to the input clock divider by setting the MASK CLKDIV SYSREF bit so that the divider output phase remains stable. The NCO phase and LMFC counter are reset on the third SYSREF pulse. After the third SYSREF pulse, the SYSREF going to the NCO and JESD block can be disabled by setting the MASK NCO SYSREF bit to avoid any unwanted resets.

SYSREF is applied as a periodic pulse.

Figure 45 shows how SYSREF can be applied as a continuous periodic waveform.

ADC32RF42 sysref_priodic_waveform_sbas747.gif

t_SYSREF is a period of the SYSREF waveform.

Alternatively, the SYSREF buffer can be powered down using the PDN SYSREF bit.

Figure 45. SYSREF Used as a Periodic Waveform

After applying the SYSREF signal, DLL must be allowed to lock, and the NCO phase and LMFC counter must be allowed to reset by waiting for at least the t_DLL (40 µs) + 2 × t_SYSREF time. Then, the SYSREF going to the NCO and JESD can be masked by setting the MASK NCO SYSREF register bit.

9.3.3.2 Frequency of the SYSREF Signal

Equation 1 describes that when SYSREF is a periodic signal, its frequency is required to be a sub-harmonic of the internal local multi-frame clock (LMFC) frequency. The LMFC frequency is determined by the selected decimation, frames per multi-frame setting (K), samples per frame (S), and device input clock frequency.

Equation 1. SYSREF = LMFC / N

where

N is an integer value (1, 2, 3, and so forth)

In order for the interleaving correction engine to synchronize properly, the SYSREF frequency must also be a multiple of f_S / 64. Table 2 provides a summary of the valid LMFC clock settings.

Table 2. SYSREF and LMFC Clock Frequency

OPERATING MODE	LMFS SETTING	LMFC CLOCK FREQUENCY	SYSREF FRQUENCY
Bypass mode	42810	f_S⁽¹⁾ / (10 × K)	f_S / [N × LCM⁽²⁾ (64, 10 × K⁽³⁾)]
Bypass mode	4222	f_S / (2 × K)	f_S / [N × LCM (64, 2 × K)]
Decimation	Various	f_S / (D × S⁽⁴⁾ × K)	f_S / [N × LCM (64, D⁽⁵⁾ × S × K)]

(1) f_S = sampling (device) clock frequency.

(2) LCM = least-common multiple.

(3) K = number of frames per multi-frame.

(4) S = samples per frame.

(5) D = decimation ratio.

The SYSREF signal is recommended to be a low-frequency signal less than 5 MHz in order to reduce coupling to the signal path both on the printed circuit board (PCB) as well as internal to the device.

Example 1: f_S = 1.5 GSPS, Bypass Mode (LMFS = 42810), K = 16

SYSREF = 1.5 GSPS / LCM (64, 10 × 16) / N = 4.6875 MHz / N

Operate SYSREF at 2.34375 MHz (effectively divide-by-640, N = 2)

Example 2: f_S = 1.5 GSPS, Divide-by-8 (LMFS = 8411), K = 16

SYSREF = 2.6 GSPS / LCM (4 ,64, 16) = 40.625 MHz / N

Operate SYSREF at 2.539063 MHz (effectively divide-by-1024, N = 16)

For proper device operation, disable the SYSREF signal after the JESD synchronization is established.

9.3.4 DDC Block

The ADC32RF42 provides a sophisticated on-chip, digital down converter (DDC) block that can be controlled through SPI register settings and the general-purpose input/output (GPIO) pins. The DDC block supports two basic operating modes: receiver (RX) mode with single- or dual-band DDC and wide-bandwidth observation receiver mode.

Figure 46 shows that each ADC channel is followed by two DDC chains consisting of the digital filter along with a complex digital mixer with a 16-bit numerically-controlled oscillator (NCO). The NCOs allow accurate frequency tuning within the Nyquist zone prior to the digital filtering. One DDC chain is intended for supporting a dual-band DDC configuration in receiver mode and the second DDC chain supports the wide-bandwidth output option for the observation configuration. At any given time, either the single-band DDC, the dual-band DDC, or the wideband DDC can be enabled. Furthermore, three different NCO frequencies can be selected on that path and are quickly switched using the SPI or the GPIO pins to enable wide-bandwidth observation in a multi-band application.

NOTE:

Red traces show SYSREF going to the NCO blocks.

Figure 46. DDC Chains Overview (One ADC Channel Shown)

Additionally, the decimation filter block provides the option to convert the complex output back to real format at twice the decimated, complex output rate. The filter response with a real output is identical to a complex output. The band is centered in the middle of the Nyquist zone (mixed with f_OUT / 4) based on a final output data rate of f_OUT.

9.3.4.1 Operating Mode: Receiver

In receiver mode (and as shown in Figure 47), the DDC block can be configured to single- or dual-band operation. Both DDC chains use the same decimation filter setting and the available options are discussed in the Decimation Filters section. The decimation filter setting also directly affects the interface rate and number of lanes of the JESD204B interface.

ADC32RF42 operating_mode_receiver_sbas844.gif

NOTE:

Red traces show SYSREF going to the NCO blocks.

Figure 47. Decimation Filter Option for Single- or Dual-Band Operation

9.3.4.2 Operating Mode: Wide-Bandwidth Observation Receiver

This mode is intended for using a DDC with a wide bandwidth output, but for multiple bands. Figure 48 shows that this mode uses a single DDC chain where up to three NCOs can be used to perform wide-bandwidth observation in a multi-band environment. The three NCOs can be switched dynamically using either the GPIO pins or an SPI command. All three NCOs operate continuously to ensure phase continuity; however, when the NCO is switched, the output data are invalid until the decimation filters are completely flushed with data from the new band.

ADC32RF42 operating_mode_wide_bw_rcvr_sbas844.gif

NOTE:

Red traces show SYSREF going to the NCO blocks.

Figure 48. Decimation Filter Implementation for Single-Band and Wide-Bandwidth Mode

9.3.4.3 Decimation Filters

The stop-band rejection of the decimation filters is approximately 90 dB with a pass-band bandwidth of approximately 80%. Table 3 gives an overview of the pass-band bandwidth depending on decimation filter setting and ADC sampling rate.

Table 3. Decimation Filter Summary and Maximum Available Output Bandwidth

DECIMATION SETTING	NO. OF DDCS AVAILABLE PER CHANNEL	NOMINAL PASSBAND GAIN	BANDWIDTH		ADC SAMPLE RATE = N MSPS		ADC SAMPLE RATE = 2.6 GSPS
DECIMATION SETTING	NO. OF DDCS AVAILABLE PER CHANNEL	NOMINAL PASSBAND GAIN	3 dB (%)	1 dB (%)	OUTPUT RATE (MSPS) PER BAND	OUTPUT BANDWIDTH (MHz) PER BAND	COMPLEX OUTPUT RATE (MSPS) PER BAND	OUTPUT BANDWIDTH (MHz) PER BAND
Divide-by-4 complex	1	–0.4 dB	90.9	86.8	N / 4 complex	0.4 × N / 2	650	520
Divide-by-6 complex	1	–0.65 dB	90.6	86.1	N / 6 complex	0.4 × N / 3	433.3	346.64
Divide-by-8 complex	2	–0.27 dB	91.0	86.8	N / 8 complex	0.4 × N / 4	325	260
Divide-by-9 complex	2	–0.45 dB	90.7	86.3	N / 9 complex	0.4 × N / 4.5	288.9	231.12
Divide-by-10 complex	2	–0.58 dB	90.7	86.3	N / 10 complex	0.4 × N / 5	260	208
Divide-by-12 complex	2	–0.55 dB	90.7	86.4	N / 12 complex	0.4 × N / 6	216.7	173.36
Divide-by-16 complex	2	–0.42 dB	90.8	86.4	N / 16 complex	0.4 × N / 8	162.5	130

Figure 49 shows a dual-band example with a divide-by-8 complex.

Figure 49. Dual-Band Example

The decimation filter responses normalized to the ADC sampling clock are illustrated in Figure 49 to Figure 64.

As shown in Figure 50, each figure contains the filter pass-band, transition bands, and alias bands. The x-axis in Figure 50 shows the offset frequency (after the NCO frequency shift) normalized to the ADC sampling clock frequency.

For example, in the divide-by-4 complex, the output data rate is an f_S / 4 complex with a Nyquist zone of f_S / 8 or 0.125 × f_S. The transition band is centered around 0.125 × f_S and the alias transition band is centered at 0.375 × f_S. The alias bands that alias on top of the wanted signal band are centered at 0.25 × f_S and 0.5 × f_S (and are colored in red).

The decimation filters of the ADC32RF42 provide greater than 90-dB attenuation for the alias bands.

Figure 50. Interpretation of the Decimation Filter Plots

9.3.4.3.1 Divide-by-4

Peak-to-peak pass-band ripple: approximately 0.22 dB

Figure 51. Filter Response Decimate-by-4

Figure 52. Filter Response Decimate-by-4 (Zoomed)

9.3.4.3.2 Divide-by-6

Peak-to-peak pass-band ripple: approximately 0.38 dB

Figure 53. Filter Response Decimate-by-6

Figure 54. Filter Response Decimate-by-6 (Zoomed)

9.3.4.3.3 Divide-by-8

Peak-to-peak pass-band ripple: approximately 0.25 dB

Figure 55. Filter Response Decimate-by-8

Figure 56. Filter Response Decimate-by-8 (Zoomed)

9.3.4.3.4 Divide-by-9

Peak-to-peak pass-band ripple: approximately 0.39 dB

Figure 57. Filter Response Decimate-by-9

Figure 58. Filter Response Decimate-by-9 (Zoomed)

9.3.4.3.5 Divide-by-10

Peak-to-peak pass-band ripple: approximately 0.39 dB

Figure 59. Filter Response Decimate-by-10

Figure 60. Filter Response Decimate-by-10 (Zoomed)

9.3.4.3.6 Divide-by-12

Peak-to-peak pass-band ripple: approximately 0.36 dB

Figure 61. Filter Response Decimate-by-12

Figure 62. Filter Response Decimate-by-12 (Zoomed)

9.3.4.3.7 Divide-by-16

Peak-to-peak pass-band ripple: approximately 0.29 dB

Figure 63. Filter Response Decimate-by-16

Figure 64. Filter Response Decimate-by-16 (Zoomed)

9.3.4.4 Digital Multiplexer (MUX)

The ADC32RF42 supports a mode where the output data of the ADC channel A can be routed internally to the digital blocks of both channel A and channel B. Figure 65 shows how ADC channel B can be powered down. In this manner, the ADC32RF42 can be configured as a single-channel ADC with up to four independent DDC chains or two wideband DDC chains. All decimation filters and JESD204B format configurations are identical to the two ADC channel operation.

Figure 65. Digital Multiplexer Option

9.3.4.5 Numerically-Controlled Oscillators (NCOs) and Mixers

The ADC32RF42 is equipped with three independent, complex NCOs per ADC channel. Equation 2 describes how the oscillator generates a complex exponential sequence.

Equation 2. x[n] = e^–jωn

where

frequency (ω) is specified as a signed number 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.

Each ADC channel has two DDCs. The first DDC has three NCOs and the second DDC has one NCO. The first DDC can dynamically select one of the three NCOs based on the GPIO pin or SPI selection. In wide-bandwidth mode (lower decimation factors 4 and 6), there can only be one DDC for each ADC channel. The NCO frequencies can be programmed independently through the DDCx, NCO[4:1], and the MSB and LSB register settings.

Equation 3 provides the 16-bit register value that sets the NCO frequency setting:

Equation 3. ADC32RF42 nco_frequency_eq_sbas747.gif

where

x = 0, 1
y = 1 to 4

For example:

If f_S = 1.5 GSPS, then the NCO register setting = 38230 (decimal).

Thus, Equation 4 defines f_NCO:

Equation 4. ADC32RF42 eq_fnco_sbas844.gif

Any register setting changes that occur after the JESD204B interface is operational results in a non-deterministic NCO phase. If a deterministic phase is required, the JESD204B interface must be reinitialized after changing the register setting.

In bypass mode (when decimation filters are not used), the NCOs are powered down in order to avoid creating unwanted spurs.

9.3.5 NCO Switching

The first DDC (DDC0) on each ADC channel provides three different NCOs that can be used for phase-coherent frequency hopping. This feature is available in both single-band and dual-band mode, but only affects DDC0.

The NCOs can be switched through an SPI control or by using the GPIO pins with the register configurations shown in Table 4 for channel A (50xxh) and channel B (58xxh). The assignment of which GPIO pin to use for INSEL0 and INSEL1 is done based on Table 5, using registers 5438h and 5C38h. The NCO selection is done based on the logic selection on the GPIO pins; see Table 6 and Figure 66.

Table 4. NCO Register Configurations

REGISTER	ADDRESS	DESCRIPTION
NCO CONTROL THROUGH GPIO PINS
NCO SEL pin	500Fh, 580Fh	Selects the NCO control through the SPI (default) or a GPIO pin.
INSEL0, INSEL1	5438h, 5C38h	Selects which two GPIO pins are used to control the NCO.
NCO CONTROL THROUGH SPI CONTROL
NCO SEL pin	500Fh, 580Fh	Selects the NCO control through the SPI (default) or a GPIO pin.
NCO SEL	5010h, 5810h	Selects which NCO to use for DDC0.

Table 5. GPIO Pin Assignment

INSELx[1:0] (Where x = 0 or 1)	GPIO PIN SELECTED
00	GPIO4
01	GPIO1
10	GPIO3
11	GPIO2

Table 6. NCO Selection

NCO SEL[1]	NCO SEL[0]	NCO SELECTED
0	0	NCO1
0	1	NCO2
1	0	NCO3
1	1	n/a

Figure 66. NCO Switching from GPIO and SPI

9.3.6 SerDes Transmitter Interface

Each 12.5-Gbps serializer, deserializer (SerDes) LVDS transmitter output requires ac-coupling between the transmitter and receiver. Terminate the differential pair as shown in Figure 67 with 100-Ω resistance (that is, two 50-Ω resistors) as close to the receiving device as possible to avoid unwanted reflections and signal degradation.

ADC32RF42 cml_serdes_transmtr_intrfc_sbas747.gif

Figure 67. External Serial JESD204B Interface Connection

9.3.7 Eye Diagrams

Figure 68 and Figure 69 show the serial output eye diagrams of the ADC32RF42 at 5.0 Gbps and 12 Gbps against the JESD204B mask.

Figure 68. Data Eye at 5 Gbps

Figure 69. Data Eye at 12 Gbps

9.3.8 Alarm Outputs: Power Detectors for AGC Support

The GPIO pins can be configured as alarm outputs for channels A and B. The ADC32RF42 supports three different power detectors (an absolute peak power detector, crossing detector, and RMS power detector) as well as fast overrange from the ADC. The power detectors operate off the full-rate ADC output prior to the decimation filters.

9.3.8.1 Absolute Peak Power Detector

In this detector mode, the peak is computed over eight samples of the ADC output. Next (as illustrated in Figure 70 and Figure 71), the peak for a block of N samples (N × S`) is computed over a programmable block length and then compared against a threshold to either set or reset the peak detector output. There are two sets of thresholds and each set has two thresholds for hysteresis. The programmable DWELL-time counter is used for clearing the block detector alarm output.

ADC32RF42 abslt_pk_pwr_dtctr_sbas747.gif

Figure 70. Peak Power Detector Implementation

ADC32RF42 abslt_pk_pwr_dtctr_tmg_dgm_sbas747.gif

Figure 71. Peak Power Detector Timing Diagram

Table 7 shows the register configurations required to set up the absolute peak power detector. The detector operates in the f_S / 8 clock domain; one peak sample is calculated over eight actual samples.

The automatic gain control (AGC) modes can be configured separately for channel A (54xxh) and channel B (5Cxxh), although some registers are common in 54xxh (such as the GPIO pin selection).

Table 7. Registers Required for the Peak Power Detector

REGISTER	ADDRESS	DESCRIPTION
PKDET EN	5400, 5C00h	Enables peak detector
BLKPKDET	5401h, 5402h, 5403h, 5C01h, 5C02h, 5C03h	Sets the block length N of number of samples (S`). Number of actual ADC samples is 8x this value: N is 17 bits: 1 to 2¹⁶.
BLKTHHH, BLKTHHL, BLKTHLH, BLKTHLL	5407h, 5408h, 5409h, 540Ah, 5C07h, 5C08h, 5C09h, 5C0Ah	Sets the different thresholds for the hysteresis function values from 0 to 256 (where 256 is equivalent to the peak amplitude). For example: if BLKTHHH is to –2 dBFS from peak, 10^{(–2 / 20)} × 256 = 203, then set 5407h and 5C07h = CBh.
DWELL	540Bh, 540Ch, 5C0Bh, 5C0Ch	When the computed block peak crosses the upper thresholds BLKTHHH or BLKTHLH, the peak detector output flags are set. In order to be reset, the computed block peak must remain continuously lower than the lower threshold (BLKTHHL or BLKTHLL) for the period specified by the DWELL value. This threshold is 16 bits and is specified in terms of f_S / 8 clock cycles.
OUTSEL GPIO[4:1]	5432h, 5433h, 5434h, 5435h	Connects the BLKPKDETH, BLKPKDETL alarms to the GPIO pins; common register.
IODIR	5437h	Selects the direction for the four GPIO pins; common register.
RESET AGC	542Bh, 5C2Bh	After configuration, reset the AGC module to start operation.

9.3.8.2 Crossing Detector

In this detector mode the peak is computed over eight samples of the ADC output. Next, the peak for a block of N samples (N × S`) is computed over a programmable block length and then the peak is compared against two sets of programmable thresholds (with hysteresis). The crossing detector counts how many f_S / 8 clock cycles that the block detector outputs are set high over a programmable time period and compares the counter value against the programmable thresholds. Figure 72 and Figure 73 show how the alarm outputs are updated at the end of the time period, routed to the GPIO pins, and held in that state through the next cycle. Alternatively, a 2-bit format can be used but (because the ADC32RF42 has four GPIO pins available) this feature uses all four pins for a single channel.

Figure 72. Crossing Detector Implementation

ADC32RF42 crssng_dtctr_tmg_dgm_sbas747.gif

Figure 73. Crossing Detector Timing Diagram

Table 8 shows the register configurations required to set up the crossing detector. The detector operates in the
f_S / 8 clock domain. The AGC modes can be configured separately for channel A (54xxh) and channel B (5Cxxh), although some registers are common in 54xxh (such as the GPIO pin selection).

Table 8. Registers Required for the Crossing Detector Operation

REGISTER	ADDRESS	DESCRIPTION
PKDET EN	5400h, 5C00h	Enables peak detector
BLKPKDET	5401h, 5402h, 5403h, 5C01h, 5C02h, 5C03h	Sets the block length N of number of samples (S`). Number of actual ADC samples is 8x this value: N is 17 bits: 1 to 2¹⁶.
BLKTHHH, BLKTHHL, BLKTHLH, BLKTHLL	5407h, 5408h, 5409h, 540Ah, 5C07h, 5C08h, 5C09h, 5C0Ah	Sets the different thresholds for the hysteresis function values from 0 to 256 (where 256 is equivalent to the peak amplitude). For example: if BLKTHHH is to –2 dBFS from peak, 10^{(–2 / 20)} × 256 = 203, then set 5407h and 5C07h = CBh.
FILT0LPSEL	540Dh, 5C0Dh	Select block detector output or 2-bit output mode as the input to the interrupt identification register (IIR) filter.
TIMECONST	540Eh, 540Fh, 5C0Eh, 5C0Fh	Sets the crossing detector time period for N = 0 to 15 as 2N × f_S / 8 clock cycles. The maximum time period is 32768 × f_S / 8 clock cycles (approximately 174 µs at 2.6 GSPS).
FIL0THH, FIL0THL, FIL1THH, FIL1THL	540Fh-5412h, 5C0Fh-5C12h, 5416h-5419h, 5C16h-5C19h	Comparison thresholds for the crossing detector counter. These thresholds are 16-bit thresholds in 2.14-signed notation. A value of 1 (4000h) corresponds to 100% crossings, a value of 0.125 (0800h) corresponds to 12.5% crossings.
DWELLIIR	541Dh, 541Eh, 5C1Dh, 5C1Eh	DWELL counter for the IIR filter hysteresis.
IIR0 2BIT EN, IIR1 2BIT EN	5413h, 54114h, 5C13h, 5C114h	Enables 2-bit output format for the crossing detector.
OUTSEL GPIO[4:1]	5432h, 5433h, 5434h, 5435h	Connects the IIRPKDET0, IIRPKDET1 alarms to the GPIO pins; common register.
IODIR	5437h	Selects the direction for the four GPIO pins; common register.
RESET AGC	542Bh, 5C2Bh	After configuration, reset the AGC module to start operation.

9.3.8.3 RMS Power Detector

In this detector mode the peak power is computed for a block of N samples over a programmable block length and then compared against two sets of programmable thresholds (with hysteresis).

The RMS power detector circuit shown in Figure 74 provides configuration options. The RMS power value (1 or 2 bit) can be output onto the GPIO pins. In 2-bit output mode, two different thresholds are used whereas the 1-bit output provides one threshold together with hysteresis.

Figure 74. RMS Power Detector Implementation

Table 9 shows the register configurations required to set up the RMS power detector. The detector operates in the f_S / 8 clock domain. The AGC modes can be configured separately for channel A (54xxh) and channel B (5Cxxh), although some registers are common in 54xxh (such as the GPIO pin selection).

Table 9. Registers Required for Using the RMS Power Detector Feature

REGISTER	ADDRESS	DESCRIPTION
RMSDET EN	5420h, 5C20h	Enables RMS detector
PWRDETACCU	5421h, 5C21h	Programs the block length to be used for RMS power computation. The block length is defined in terms of f_S / 8 clocks. The block length can be programmed as 2^M with M = 0 to 16.
PWRDETH, PWRDETL	5422h, 5423h, 5424h, 5425h, 5C22h, 5C23h, 5C24h, 5C25h	The computed average power is compared against these high and low thresholds. One LSB of the thresholds represents 1 / 2¹⁶. For example: is PWRDETH is set to –14 dBFS from peak, [10^{(–14 / 20)}]² × 2¹⁶ = 2609, then set 5422h, 5423h, 5C22h, 5C23h = 0A31h.
RMS2BIT EN	5427h, 5C27h	Enables 2-bit output format for the RMS detector output.
OUTSEL GPIO[4:1]	5432h, 5433h, 5434h, 5435h	Connects the PWRDET alarms to the GPIO pins; common register.
IODIR	5437h	Selects the direction for the four GPIO pins; common register.
RESET AGC	542Bh, 5C2Bh	After configuration, reset the AGC module to start operation.

9.3.8.4 GPIO AGC MUX

The GPIO pins can be used to control the NCO in wideband DDC mode or as alarm outputs for channel A and B. The GPIO pins can be configured as shown in Figure 75 through the SPI control to output the alarm from the peak power (1 bit), crossing detector (1 or 2 bit), faster overrange, or the RMS power output.

The programmable output MUX allows connecting any signal (including the NCO control) to any of the four GPIO pins. These pins can be configured as outputs (AGC alarm) or inputs (NCO control) through SPI programming.

Figure 75. GPIO Output MUX Implementation

9.3.9 Power-Down Mode

The ADC32RF42 provides a lot of configurability for the power-down mode. Power-down can be enabled using the PDN pin or the SPI register writes.

9.3.10 ADC Test Pattern

The ADC32RF42 provides several different options to output test patterns instead of the actual output data of the ADC in order to simplify the serial interface and system debug of the JESD204B digital interface link. Figure 76 shows the output data path.

Figure 76. Test Pattern Generator Implementation

9.3.10.1 Digital Block

The ADC test pattern replaces the actual output data of the ADC. The test patterns listed in Table 10 are available when the DDC is enabled and located in register 37h of the decimation filter page. When programmed, the test patterns are output for each converter (M) stream. The number of converter streams per channel increases by 2 when complex (I, Q) output or dual-band DDC is selected. The test patterns can be synchronized for both ADC channels using the SYSREF signal.

Additionally, a 12-bit ramp test pattern is available in DDC bypass mode.

NOTE

The number of converters increases in dual-band DDC mode and with a complex output.

Table 10. Test Pattern Options (Register 37h)

BIT	NAME	DEFAULT	DESCRIPTION
7-4	TEST PATTERN	0000	Test pattern outputs on channel A and B. 0000 = Normal operation using ADC output data 0001 = Outputs all 0s 0010 = Outputs all 1s 0011 = Outputs toggle pattern: output data are an alternating sequence of 10101010101010 and 01010101010101 0100 = Output digital ramp: output data increment by one LSB every clock cycle from code 0 to 65535 0110 = Single pattern: output data are a custom pattern 1 (75h and 76h) 0111 Double pattern: output data alternate between custom pattern 1 and custom pattern 2 1000 = Deskew pattern: output data are AAAAh 1001 = SYNC pattern: output data are FFFFh

9.3.10.2 Transport Layer

The transport layer maps the ADC output data into 8-bit octets and constructs the JESD204B frames using the LMFS parameters. Tail bits or 0's are added when needed. Alternatively, as described in Table 11, the JESD204B long transport layer test pattern can be substituted instead of the ADC data with the JESD frame.

Table 11. Transport Layer Test Mode EN (Register 01h)

BIT	NAME	DEFAULT	DESCRIPTION
4	TESTMODE EN	0	Generates long transport layer test pattern mode according to section 5.1.6.3 of the JESD204B specification. 0 = Test mode disabled 1 = Test mode disabled

9.3.10.3 Link Layer

The link layer contains the scrambler and the 8b, 10b encoding of any data passed on from the transport layer. Additionally, the link layer also handles the initial lane alignment sequence that can be manually restarted.

The link layer test patterns are intended for testing the quality of the link (jitter testing and so forth). Table 12 lists the test pattern options.

Table 12. Link Layer Test Mode (Register 03h)

BIT	NAME	DEFAULT	DESCRIPTION
7-5	LINK LAYER TESTMODE	000	Generates a pattern according to section 5.3.3.8.2 of the JESD204B document. 000 = Normal ADC data 001 = D21.5 (high-frequency jitter pattern) 010 = K28.5 (mixed-frequency jitter pattern) 011 = Repeat the initial lane alignment (generates a K28.5 character and repeats lane alignment sequences continuously) 100 = 12-octet random pattern (RPAT) jitter pattern

Furthermore, a 2¹⁵ pseudo-random binary sequence (PRBS) can be enabled by setting up a custom test pattern (AAAAh) in the ADC section and running AAAAh through the 8b, 10b encoder with scrambling enabled.

9.4 Device Functional Modes

9.4.1 Device Configuration

The ADC32RF42 can be configured using a serial programming interface, as described in the Serial Interface section. In addition, the device has one dedicated parallel pin (PDN) for controlling the power-down modes.

9.4.2 JESD204B Interface

The ADC32RF42 supports device subclass 1 with a maximum output data rate of 12.5 Gbps for each serial transmitter.

An external SYSREF signal is used to align all internal clock phases and the local multiframe clock to a specific sampling clock edge. This alignment allows synchronization of multiple devices in a system and minimizes timing and alignment uncertainty. Figure 77 shows how the SYNCB input is used to control the JESD204B SerDes blocks.

Depending on the ADC sampling rate, the JESD204B output interface can be operated with one, two, or four lanes per ADC channel. The JESD204B setup and configuration of the frame assembly parameters is controlled through the SPI interface.

Figure 77. JESD Signal Overview

The JESD204B transmitter block shown in Figure 78 consists of the transport layer, the data scrambler, and the link layer. The transport layer maps the ADC output data into the selected JESD204B frame data format and manages if the ADC output data or test patterns are transmitted. The link layer performs the 8b, 10b data encoding as well as the synchronization and initial lane alignment using the SYNC input signal. Optionally, data from the transport layer can be scrambled.

Figure 78. JESD Digital Block Implementation

9.4.2.1 JESD204B Initial Lane Alignment (ILA)

The receiving device starts the initial lane alignment process by deasserting the SYNCB signal. The SYNCB signal can be issued using the SYNCB input pins or by setting the proper SPI bits. When a logic low is detected on the SYNCB input (as shown in Figure 79), the ADC32RF42 starts transmitting comma (K28.5) characters to establish the code group synchronization.

When synchronization completes, the receiving device reasserts the SYNCB signal and the ADC32RF42 starts the initial lane alignment sequence with the next local multiframe clock boundary. The ADC32RF42 transmits four multiframes, each containing K frames (K is SPI programmable). Each of the multiframes contains the frame start and end symbols. The second multiframe also contains the JESD204 link configuration data.

ADC32RF42 jesd204b_lane_algnmnt_sbas844.gif

Figure 79. JESD Internal Timing Information

9.4.2.2 JESD204B Frame Assembly

The JESD204B standard defines the following parameters:

F is the number of octets per frame clock period
L is the number of lanes per link
M is the number of converters for the device
S is the number of samples per frame

9.4.2.3 JESD204B Frame Assembly in Bypass Mode

Table 13 lists the available JESD204B formats and valid ranges for the ADC32RF42. The ranges are limited by the SerDes line rate and the maximum ADC sample frequency. Table 14 shows the sample alignment for the bypass modes on the different lanes.

Table 13. JESD Mode Options: Bypass Mode

DECIMATION SETTING (Complex)	OUTPUT RESOLUTION (Bits)	L	M	F	S	12-BIT MODE	PLL MODE	JESD MODE0	JESD MODE1	JESD MODE2	MAX f_CLK (Gsps)	RATIO [f_SerDes / f_CLK (Gbps / GSPS)]
Bypass	12⁽¹⁾	4	2	8	10	3	16x	3	0	0	1.5	8
Bypass	14	4	2	2	2	0	20x	1	0	0	1.25	10

(1) In full rate output, the two LSBs are truncated to a 12-bit output.

Table 14. JESD Sample Lane Alignments: Bypass Mode⁽¹⁾

OUTPUT LANE	LMFS = 4222		LMFS = 42810
DA0	A₀[13:6]	A₀[5:0], 00	A₀[11:4]	A₀[3:0], A₁[11:8]	A₁[7:0]	A₂[11:4]	A₂[3:0], A₃[11:8]	A₃[7:0]	A₄[11:4]	A₄[3:0], 0000
DA1	A₁[13:6]	A₁[5:0], 00	A₅[11:4]	A₅[3:0], A₆[11:8]	A₆[7:0]	A₇[11:4]	A₇[3:0], A₈[11:8]	A₈[7:0]	A₉[11:4]	A₉[3:0], 0000
DA2	—	—	—	—	—	—	—	—	—	—
DA3	—	—	—	—	—	—	—	—	—	—
DB0	B₀[13:6]	B₀[5:0], 00	B₀[11:4]	B₀[3:0], B₁[11:8]	B₁[7:0]	B₂[11:4]	B₂[3:0], B₃[11:8]	B₃[7:0]	B₄[11:4]	B₄[3:0], 0000
DB1	B₁[13:6]	B₁[5:0], 00	B₅[11:4]	B₅[3:0], B₆[11:8]	B₆[7:0]	B₇[11:4]	B₇[3:0], B₈[11:8]	B₈[7:0]	B₉[11:4]	B₉[3:0], 0000
DB2	—	—	—	—	—	—	—	—	—	—
DB3	—	—	—	—	—	—	—	—	—	—

(1) Blue shading indicates channel A and yellow shading indicates channel B.

9.4.2.4 JESD204B Frame Assembly with Decimation (Single-Band DDC): Complex Output

Table 15 lists the available JESD204B interface formats and valid ranges for the ADC32RF42 with decimation (single-band DDC) when using a complex output format. The ranges are limited by the SerDes line rate and the maximum ADC sample frequency. Table 16 shows the sample alignment on the different lanes.

Table 15. JESD Mode Options: Single-Band Complex Output

DECIMATION SETTING (Complex)	NUMBER OF ACTIVE DDCS	L	M	F	S	PLL MODE	JESD MODE0	RATIO [f_SerDes / f_CLK (Gbps / GSPS)]
Divide-by-4	1 per channel	4	4	2	1	20x	1	5
Divide-by-4	1 per channel	2	4	4	1	40x	2	2.5
Divide-by-6	1 per channel	4	4	2	1	20x	1	3.33
Divide-by-6	1 per channel	2	4	4	1	40x	2	6.66
Divide-by-8	1 per channel	4	4	2	1	20x	1	2.5
Divide-by-8	1 per channel	2	4	4	1	40x	2	5
Divide-by-9	1 per channel	4	4	2	1	20x	1	2.22
Divide-by-9	1 per channel	2	4	4	1	40x	2	4.44
Divide-by-10	1 per channel	4	4	2	1	20x	1	2
Divide-by-10	1 per channel	2	4	4	1	40x	2	4
Divide-by-12	1 per channel	2	4	4	1	40x	1	3.33
Divide-by-16	1 per channel	2	4	4	1	40x	2	2.5

Table 16. JESD Sample Lane Alignments: Single-Band Complex Output⁽¹⁾

OUTPUT LANE	LMFS = 4421 20x		LMFS = 4421 40x		LMFS = 4442				LMFS = 2441
DA0	AI₀ [15:8]	AI₀ [7:0]
DA1	AQ₀ [15:8]	AQ₀ [7:0]	AI₀ [15:8]	AI₀ [7:0]	AI₀ [15:8]	AI₀ [7:0]	AI₁ [15:8]	AI₁ [7:0]	AI₀ [15:8]	AI₀ [7:0]	AQ₀ [15:8]	AQ₀ [7:0]
DA2			AQ₀ [15:8]	AQ₀ [7:0]	AQ₀ [15:8]	AQ₀ [7:0]	AQ₁ [15:8]	AQ₁ [7:0]
DA3
DB0	BI₀ [15:8]	BI₀ [7:0]
DB1	BQ₀ [15:8]	BQ₀ [7:0]	BI₀ [15:8]	BI₀ [7:0]	BI₀ [15:8]	BI₀ [7:0]	BI₁ [15:8]	BI₁ [7:0]	BI₀ [15:8]	BI₀ [7:0]	BQ₀ [15:8]	BQ₀ [7:0]
DB2			BQ₀ [15:8]	BQ₀ [7:0]	BQ₀ [15:8]	BQ₀ [7:0]	BQ₁ [15:8]	BQ₁ [7:0]
DB3

9.4.2.5 JESD204B Frame Assembly with Decimation (Single-Band DDC): Real Output

Table 17 lists the available JESD204B formats and valid ranges for the ADC32RF42 with decimation (single-band DDC) when using real output format. The ranges are limited by the SerDes line rate and the maximum ADC sample frequency.

Table 17. JESD Mode Options: Single-Band Real Output (Wide Bandwidth)

DECIMATION SETTING (Complex)	NUMBER OF ACTIVE DDCS	L	M	F	S	PLL MODE	JESD MODE0	JESD MODE1	JESD MODE2	RATIO [f_SerDes / f_CLK (Gbps / GSPS)]
Divide-by-4 (Divide-by-2 real)	1 per channel	4	2	1	1	20x	1	1	0	5
		4	2	2	2	20x	1	0	0	10
		2	2	2	1	40x	0	0	1
		2	2	4	2	40x	2	0	0
Divide-by-6 (Divide-by-3 real)	1 per channel	4	2	1	1	20x	1	1	0	3.33
		4	2	2	2	20x	1	0	0	6.66
		2	2	2	1	40x	0	0	1
		2	2	4	2	40x	2	0	0

9.4.2.6 JESD204B Frame Assembly with Decimation (Single-Band DDC): Real Output

Table 18 lists the available JESD204B formats and valid ranges for the ADC32RF42 with decimation (dual-band DDC) when using a complex output format. Table 19 shows the sample alignment on the different lanes.

Table 18. JESD Mode Options: Single-Band Real Output

DECIMATION SETTING (Complex)	NUMBER OF ACTIVE DDCS	L	M	F	S	PLL MODE	JESD MODE0	JESD MODE1	JESD MODE2	RATIO [f_SerDes / f_CLK (Gbps / GSPS)]
Divide-by-8 (Divide-by-4 real)	1 per channel	4	2	1	1	20x	1	1	0	2.5
		4	2	2	2	20x	1	0	0	2.5
		2	2	2	1	40x	0	0	1	5
		2	2	4	2	40x	2	0	0	5
Divide-by-9 (Divide-by-4.5 real)	1 per channel	4	2	1	1	20x	1	1	0	2.22
		4	2	2	2	20x	1	0	0	2.22
		2	2	2	1	40x	0	0	1	4.44
		2	2	4	2	40x	2	0	0	4.44
Divide-by-10 (Divide-by-5 real)	1 per channel	4	2	1	1	20x	1	1	0	2
		4	2	2	2	20x	1	0	0	2
		2	2	2	1	40x	0	0	1	4
		2	2	4	2	40x	2	0	0	4
Divide-by-12 (Divide-by-6 real)	1 per channel	2	2	2	1	40x	0	0	1	3.33
Divide-by-12 (Divide-by-6 real)	1 per channel	2	2	4	2	40x	2	0	0	3.33
Divide-by-16 (Divide-by-8 real)	1 per channel	2	2	2	1	40x	0	0	1	2.5
Divide-by-16 (Divide-by-8 real)	1 per channel	2	2	4	2	40x	2	0	0	2.5

Table 19. JESD Sample Lane Assignment: Single-Band Real Output⁽¹⁾

OUTPUT LANE	LMFS = 4211	LMFS = 4222		LMFS = 2221		LMFS = 2242
DA0	A₀[15:8]	A₀[15:8]	A₀[7:0]
DA1	A₀[7:0]	A₁[15:8]	A₁[7:0]	A₀ [15:8]	A₀[7:0]	A₀[15:8]	A₀[7:0]	A₁[15:8]	A₁[7:0]
DB0	B₀[15:8]	B₀[15:8]	B₀[7:0]
DB1	B₀[7:0]	B₁[15:8]	B₁[7:0]	B₀[15:8]	B₀[7:0]	B₀[15:8]	B₀[7:0]	B₁[15:8]	B₁[7:0]

9.4.2.7 JESD204B Frame Assembly with Decimation (Dual-Band DDC): Complex Output

Table 20 lists the available JESD204B formats and valid ranges for the ADC32RF42 with decimation (dual-band DDC) when using a complex output format. The ranges are limited by the SerDes line rate and the maximum ADC sample frequency. Table 21 shows the sample alignment on the different lanes.

Table 20. JESD Mode Options: Dual-Band Complex Output

DECIMATION SETTING (Complex)	NUMBER OF ACTIVE DDCS	L	M	F	S	PLL MODE	JESD MODE0	RATIO [f_SerDes / f_CLK (Gbps / GSPS)]
Divide-by-8	2 per channel	8	8	2	1	20x	1	2.5
Divide-by-8	2 per channel	4	8	4	1	40x	2	5
Divide-by-9	2 per channel	8	8	2	1	20x	1	2.22
Divide-by-9	2 per channel	4	8	4	1	40x	2	4.44
Divide-by-10	2 per channel	8	8	2	1	20x	1	2
Divide-by-10	2 per channel	4	8	4	1	40x	2	4
Divide-by-12	2 per channel	4	8	4	1	40x	2	3.33
Divide-by-16	2 per channel	4	8	4	1	40x	2	2.5

Table 21. JESD Sample Lane Assignment: Dual-Band Complex Output⁽¹⁾

OUTPUT LANE	LMFS = 8821		LMFS = 4841
DA0	A1₀[15:8]	A1₀[7:0]
DA1	A1Q₀[15:8]	A1Q₀[7:0]	A1I₀[15:8]	A1I₀[7:0]	A1Q₀[15:8]	A1Q₀[7:0]
DA2	A2I₀[15:8]	A2I₀[7:0]	A2I₀[15:8]	A2I₀[7:0]	A2Q₀[15:8]	A2Q₀[7:0]
DA3	A2Q₀[15:8]	A2Q₀[7:0]
DB0	B1I₀[15:8]	B1I₀[7:0]
DB1	B1Q₀[15:8]	B1Q₀[7:0]	B1I₀[15:8]	B1I₀[7:0]	B1Q₀[15:8]	B1Q₀[7:0]
DB2	B2I₀[15:8]	B2I₀[7:0]	B2I₀[15:8]	B2I₀[7:0]	B2Q₀[15:8]	B2Q₀[7:0]
DB3	B2Q₀[15:8]	B2Q₀[7:0]

(1) Blue and green shading indicates the two bands for channel A; yellow and orange shading indicates the two bands for channel B.

9.4.2.8 JESD204B Frame Assembly with Decimation (Dual-Band DDC): Real Output

Table 22 lists the available JESD204B formats and valid ranges for the ADC32RF42 with decimation (dual-band DDC) when using real output format. The ranges are limited by the SerDes line rate and the maximum ADC sample frequency. Table 23 shows the sample alignment on the different lanes.

Table 22. JESD Mode Options: Dual-Band Real Output

DECIMATION SETTING (Complex)	NUMBER OF ACTIVE DDCS	L	M	F	S	PLL MODE	JESD MODE0	JESD MODE1	JESD MODE2	RATIO [f_SerDes / f_CLK (Gbps / GSPS)]
Divide-by-8 (Divide-by-4 real)	2 per channel	8	4	1	1	20x	1	1	0	2.5
		8	4	2	2	20x	1	0	0	2.5
		4	4	2	1	40x	0	0	1	5
		4	4	4	2	40x	2	0	0	5
Divide-by-9 (Divide-by-4.5 real)	2 per channel	8	4	1	1	20x	1	1	0	2.22
		8	4	2	2	20x	1	0	0	2.22
		4	4	2	1	40x	0	0	1	4.44
		4	4	4	2	40x	2	0	0	4.44
Divide-by-10 (Divide-by-5 real)	2 per channel	8	4	1	1	20x	1	1	0	2
		8	4	2	2	20x	1	0	0	2
		4	4	2	1	40x	0	0	1	4
		4	4	4	2	40x	2	0	0	4
Divide-by-12 (Divide-by-6 real)	2 per channel	4	4	2	1	40x	0	0	1	3.33
Divide-by-12 (Divide-by-6 real)	2 per channel	4	4	4	2	40x	2	0	0	3.33
Divide-by-16 (Divide-by-8 real)	2 per channel	4	4	2	1	40x	0	0	1	2.5
Divide-by-16 (Divide-by-8 real)	2 per channel	4	4	4	2	40x	2	0	0	2.5

Table 23. JESD Sample Lane Assignment: Dual-Band Complex Output⁽¹⁾

OUTPUT LANE	LMFS = 8411	LMFS = 8422		LMFS = 4421		LMFS = 4442
DA0	A1₀[15:8]	A1₀[15:8]	A1₀[7:0]
DA1	A1₀[7:0]	A1₁[15:8]	A1₁[7:0]	A1₀[15:8]	A1₀[7:0]	A1₀[15:8]	A1₀[7:0]	A1₁[15:8]	A1₁[7:0]
DA2	A2₀[15:8]	A2₀[15:8]	A2₀[7:0]	A2₀[15:8]	A2₀[7:0]	A2₀[15:8]	A2₀[7:0]	A2₁[15:8]	A2₁[7:0]
DA3	A2₀[7:0]	A2₁[15:8]	A2₁[7:0]
DB0	B1₀[15:8]	B1₀[15:8]	B1₀[7:0]
DB1	B1₀[7:0]	B1₁[15:8]	B1₁[7:0]	B1₀[15:8]	B1₀[7:0]	B1₀[15:8]	B1₀[7:0]	B1₁[15:8]	B1₁[7:0]
DB2	B2₀[15:8]	B2₀[15:8]	B2₀[7:0]	B2₀[15:8]	B2₀[7:0]	B2₀[15:8]	B2₀[7:0]	B2₁[15:8]	B2₁[7:0]
DB3	B2₀[7:0]	B2₁[15:8]	B2₁[7:0]

9.4.3 Serial Interface

The ADC has a set of internal registers that can be accessed by the serial interface formed by the SEN (serial interface enable), SCLK (serial interface clock), and SDIN (serial interface data) pins. Serially shifting bits into the device is enabled when SEN is low. Figure 80 shows that SDIN serial data are latched at every SCLK rising edge when SEN is active (low). Table 24 shows that the interface can function with SCLK frequencies from 20 MHz down to low speeds (of a few hertz) and also with a non-50% SCLK duty cycle.

The SPI access described in Table 25 uses 24 bits consisting of eight register data bits, 12 register address bits, and four special bits to distinguish between read/write, page and register, and individual channel access.

ADC32RF42 serial_interface_dgm_sbas747.gif

Figure 80. SPI Timing Diagram

Table 24. SPI Timing Information

		MIN	TYP	MAX	UNIT
f_SCLK	SCLK frequency (equal to 1 / t_SCLK)	1		20	MHz
t_SLOADS	SEN to SCLK setup time	50			ns
t_SLOADH	SCLK to SEN hold time	50			ns
t_DSU	SDIN setup time	10			ns
t_DH	SDIN hold time	10			ns
t_SDOUT	Delay between SCLK falling edge to SDOUT		10		ns

Table 25. SPI Input Description

SPI BIT	DESCRIPTION	OPTIONS
R/W bit	Read/write bit	0 = SPI write 1 = SPI read back
M bit	SPI bank access	0 = Analog SPI bank (master) 1 = All digital SPI banks (main digital, interleaving, decimation filter, JESD digital, and so forth)
P bit	Digital page selection bit	0 = Page access 1 = Register access
CH bit	SPI access for a specific channel of the JESD digital page	0 = Channel B 1 = Channel A
ADDR[11:0]	SPI address bits	—
DATA[7:0]	SPI data bits	—

Figure 81 shows the SDOUT timing when data are read back from a register. Data are placed on the SDOUT bus at the SCLK falling edge so that the data can be latched at the SCLK rising edge by the external receiver.

Figure 81. SDOUT Timing

9.4.3.1 Serial Register Write: Analog Bank

The internal register of the ADC32RF42 analog bank (Figure 82) can be programmed by:

Driving the SEN pin low.
Initiating a serial interface cycle selecting the page address of the register whose content must be written. To select the master page: write address 0012h with 04h. To select the ADC page: write address 0011h with FFh.
Writing the register content. When a page is selected, multiple registers located in the same page can be programmed.

Figure 82. SPI Write Timing Diagram for the Analog Bank

9.4.3.2 Serial Register Readout: Analog Bank

Contents of the registers located in the two pages of the analog bank (Figure 83) can be readback by:

Driving the SEN pin low.
Selecting the page address of the register whose content must be read. Master page: write address 0012h with 04h. ADC page: write address 0011h with FFh.
Setting the R/W bit to 1 and writing the address to be read back.
Reading back the register content on the SDOUT pin. When a page is selected, the contents of multiple registers located in same page can be readback.

Figure 83. SPI Read Timing Diagram for the Analog Bank

9.4.3.3 Serial Register Write: Digital Bank

The digital bank contains seven pages (offset corrector page for channel A and B; digital gain page for channel A and B; main digital page for channel A and B; and JESD digital page). Figure 84 shows the timing for the individual page selection. The registers located in the pages of the digital bank can be programmed by:

Driving the SEN pin low.
Setting the M bit to 1 and specifying the page with the desired register. There are seven pages in digital bank. These pages can be selected by appropriately programming register bits DIGITAL BANK PAGE SEL, located in addresses 002h, 003h, and 004h, using three consecutive SPI cycles. Addressing in a SPI cycle begins with 4xxx when selecting a page from digital bank because the M bit must be set to 1.
- To select the offset corrector page channel A: write address 4004h with 61h, 4003h with 00h, and 4002h with 00h.
- To select the offset corrector page channel B: write address 4004h with 61h, 4003h with 01h, and 4002h with 00h.
- To select the digital gain page channel A: write address 4004h with 61h, 4003h with 00h, and 4002h with 05h.
- To select the digital gain page channel B: write address 4004h with 61h, 4003h with 01h, and 4002h with 05h.
- To select the main digital page channel A: write address 4004h with 68h, 4003h with 00h, and 4002h with 00h.
- To select the main digital page channel B: write address 4004h with 68h, 4003h with 01h, and 4002h with 00h.
- To select the JESD digital page: write address 4004h with 69h, 4003h with 00h, and 4002h with 00h.

Figure 84. SPI Write Timing Diagram for Digital Bank Page Selection

Writing into the desired register by setting both the M bit and P bit to 1. Write register content. When a page is selected, multiple writes into the same page can be done. As shown in Figure 85, addressing in an SPI cycle begins with 6xxx when selecting a page from the digital bank because the M bit must be set to 1.

The JESD digital page is common for both channels. The CH bit can be used to distinguish between two channels when programming registers in the JESD digital page. When CH = 0, registers are programmed for channel B; when CH = 1, registers are programmed for channel A. Thus, an SPI cycle to program registers for channel B begins with 6xxx and channel A begins with 7xxx.

ADC32RF42 dgtal_bank_reg_write_sbas747.gif

Figure 85. SPI Write Timing Diagram for Digital Bank Register Write

9.4.3.4 Serial Register Readout: Digital Bank

Readback of the register in one of the digital banks (as shown in Figure 86) can be accomplished by:

Driving the SEN pin low.
Selecting the page in the digital page: follow step 2 in the Serial Register Write: Digital Bank section.
Set the R/W, M, and P bits to 1, select channel A or channel B, and write the address to be read back.
- JESD digital page: use the CH bit to select channel B (CH = 0) or channel A (CH = 1).
Read back the register content on the SDOUT pin. When a page is selected, multiple read backs from the same page can be done.

ADC32RF42 digital_bank_serial_read_tim_sbas747.gif

Figure 86. SPI Read Timing Diagram for the Digital Bank

9.4.3.5 Serial Register Write: Decimation Filter and Power Detector Pages

The decimation filter and power detector pages are special pages that accept direct addressing. The sampling clock and SYSREF signal are required to properly configure the decimation settings. Figure 87 shows that registers located in these pages can be programmed in one SPI cycle.

Drive the SEN pin low.
Directly write to the decimation filter or power detector pages. To program registers in these pages, set M = 1 and CH = 1. Additionally, address bit A[10] selects the decimation filter page (A[10] = 0) or the power detector page (A[10] = 1). Address bit A[11] selects channel A (A[11] = 0) or channel B (A[11] = 1).
- Decimation filter page: write address 50xxh for channel A or 58xxh for channel B.
- Power detector page: write address 54xxh for channel A or 5Cxxh for channel B.

Example: Writing address 5001h with 02h selects the decimation filter page for channel A and programs decimation factor of divide-by-8 (complex output).

ADC32RF42 dgtl_bank_srl_rgstr_write_timing_sbas747.gif

Figure 87. SPI Write Timing Diagram for the Decimation and Power Detector Pages

9.5 Register Maps

The ADC32RF42 contains two main SPI banks. The analog SPI bank provides access to the ADC core and the digital SPI bank controls the digital blocks (including the serial JESD interface). Figure 88 and Figure 89 provide a conceptual view of the SPI registers inside the ADC32RF42. The analog SPI bank contains the master and ADC pages. The digital SPI bank is divided into multiple pages (the main digital, digital gain, decimation filter, JESD digital, and power detector pages). STOPPED HERE

ADC32RF42 serial_reg_2stepprog_sbas747.gif

1. In general, SPI writes are completed in two steps. The first step is to access the necessary page. The second step is to program the desired register in that page. When a page is accessed, the registers in that page can be programmed multiple times.

2. Registers in the decimation filter page and the power detector page can be directly programmed in one SPI cycle.

3. The CH bit is a don't care bit and is recommended to be kept at 0.

Figure 88. SPI Registers, Two-Step Addressing

ADC32RF42 serial_reg_direct_prog_sbas747.gif

1. Registers in the decimation filter page and the power detector page can be directly programmed in one SPI cycle.

2. To program registers in the decimation filter page, aet M = 1, CH = 1, A[10] = 0, and A[11] = 0 or 1 for channel A or B. Addressing begins at 50xx for channel A and 58xx for channel B.

3. To program registers in power detector page, set M = 1, CH = 1, A[10] = 1, and A[11] = 0 or 1 for channel A or B. Addressing begins at 54xx for channel A and 5Cxx for channel B.

Figure 89. SPI Registers: Direct Addressing

Table 26 lists the register map for the ADC32RF42.

Table 26. Register Map

REGISTER ADDRESS A[11:0] (Hex)	REGISTER DATA
REGISTER ADDRESS A[11:0] (Hex)	7	6	5	4	3	2	1	0
GENERAL REGISTERS AND PAGE SELECTION
000	RESET	0	0	0	0	0	0	RESET
002	DIGITAL BANK PAGE SEL[7:0]
003	DIGITAL BANK PAGE SEL[15:8]
004	DIGITAL BANK PAGE SEL[23:16]
010	0	0	0	0	0	0	0	3 or 4 WIRE
011	ADC PAGE SEL
012	0	0	0	0	0	MASTER PAGE SEL	0	0
MASTER PAGE (M = 0)
020	0	0	0	PDN SYSREF	0	0	PDN CHB	GLOBAL PDN
032	0	0	INCR CM IMPEDANCE	0	0	0	0	0
039	0	ALWAYS WRITE 1	0	ALWAYS WRITE 1	0	0	PDN CHB EN	SYNC TERM DIS
03C	0	SYSREF DEL EN	0	0	0	0	SYSREF DEL[4:3]
03D	0	0	0	0	0	JESD OUTPUT SWING
05A	SYSREF DEL[2:0]			0	0	0	0	0
057	0	0	0	SEL SYSREF REG	ASSERT SYSREF REG	0	0	0
058	0	0	SYNCB POL	0	0	0	0	0
ADC PAGE (FFh, M = 0)
03F	0	0	0	0	0	SLOW SP EN1	0	0
042	0	0	0	SLOW SP EN2	0	0	ALWAYS WRITE 1	ALWAYS WRITE 1
Offset Corr Page Channel A (610000h, M = 1)
68	FREEZE OFFSET CORR	0	ALWAYS WRITE 1	0	0	DIS OFFSET CORR	ALWAYS WRITE 1	0
Offset Corr Page Channel B (610100h, M = 1)
68	FREEZE OFFSET CORR	0	ALWAYS WRITE 1	0	0	DIS OFFSET CORR	ALWAYS WRITE 1	0
Digital Gain Page Channel A (610005, M = 1)
0A6	0	0	0	0	DIGITAL GAIN
Digital Gain Page Channel B (610105, M = 1)
0A6	0	0	0	0	DIGITAL GAIN
Main Digital Page Channel A (680000h, M = 1)
000	0	0	0	0	0	0	0	DIG CORE RESET GBL
0A2	0	0	0	0	NQ ZONE EN	NYQUIST ZONE
Main Digital Page Channel B (680100h, M = 1)
0A2	0	0	0	0	NQ ZONE EN	NYQUIST ZONE
JESD DIGITAL PAGE (690000h, M = 1)
001	CTRL K	0	0	TESTMODE EN	0	LANE ALIGN	FRAME ALIGN	TX LINK DIS
002	SYNC REG	SYNC REG EN	0	0	12BIT MODE		JESD MODE0
003	LINK LAYER TESTMODE			LINK LAY RPAT	LMFC MASK RESET	JESD MODE1	JESD MODE2	RAMP 12BIT
004	0	0	0	0	0	0	REL ILA SEQ
006	SCRAMBLE EN	0	0	0	0	0	0	0
007	0	0	0	FRAMES PER MULTIFRAME (K)
016	LANE 0		LANE 1		LANE 2		LANE 3
017	0	0	0	0	LANE0 POL	LANE1 POL	LANE2 POL	LANE3 POL
032	SEL EMP LANE 0						0	0
033	SEL EMP LANE 1						0	0
034	SEL EMP LANE 2						0	0
035	SEL EMP LANE 3						0	0
036	80X MODE EN	CMOS SYNCB	0	0	0	0	0	0
037	0	0	0	0	0	0	PLL MODE
03C	0	0	0	0	0	0	0	EN CMOS SYNCB
03E	0	MASK CLKDIV SYSREF	MASK NCO SYSREF	0	0	0	0	0
SPECIAL PAGE CHANNEL A (6A0100h, M = 1)
019	0	0	0	ALWAYS WRITE 1		0	0	0
SPECIAL PAGE CHANNEL B (6A0000h, M = 1)
019	0	0	0	ALWAYS WRITE 1		0	0	0
DECIMATION FILTER PAGE (Direct Addressing, 16-Bit Address, 5000h for Channel A and 5800h for Channel B)
000	0	0	0	0	0	0	0	DDC EN
001	0	0	0	0	DECIM FACTOR
002	0	0	0	0	0	0	0	DUAL BAND EN
005	0	0	0	0	0	0	0	REAL OUT EN
006	0	0	0	0	0	0	0	DDC MUX
007	DDC0 NCO1 LSB
008	DDC0 NCO1 MSB
009	DDC0 NCO2 LSB
00A	DDC0 NCO2 MSB
00B	DDC0 NCO3 LSB
00C	DDC0 NCO3 MSB
00D	DDC1 NCO4 LSB
00E	DDC1 NCO4 MSB
00F	0	0	0	0	0	0	0	NCO SEL PIN
010	0	0	0	0	0	0	NCO SEL
011	0	0	0	0	0	0	LMFC RESET MODE
014	0	0	0	0	0	0	0	DDC0 6DB GAIN
016	0	0	0	0	0	0	0	DDC1 6DB GAIN
01E	0	DDC DET LAT			0	0	0	0
01F	0	0	0	0	0	0	0	WBF 6DB GAIN
020	0	0	0	0	0	0	0	ALWAYS WRITE 1
033	CUSTOM PATTERN1[7:0]
034	CUSTOM PATTERN1[15:8]
035	CUSTOM PATTERN2[7:0]
036	CUSTOM PATTERN2[15:8]
037	0	0	0	0	TEST PATTERN SEL
038	TEST PATTERN DDC2 Q-DATA				TEST PATTERN DDC2 I -DATA
039	0	0	0	0	0	0	0	USE COMMON TEST PATTERN
03A	0	0	0	0	0	0	TEST PAT RES	TP RES EN
POWER DETECTOR PAGE (Direct Addressing, 16-Bit Address, 5400h for Channel A and 5C00h for Channel B)
000	0	0	0	0	0	0	0	PKDET EN
001	BLKPKDET [7:0]
002	BLKPKDET [15:8]
003	0	0	0	0	0	0	0	BLKPKDET [16]
007	BLKTHHH
008	BLKTHHL
009	BLKTHLH
00A	BLKTHLL
00B	DWELL[7:0]
00C	DWELL[15:8]
00D	0	0	0	0	0	0	0	FILT0LPSEL
00E	0	0	0	0	TIMECONST
00F	FIL0THH[7:0]
010	FIL0THH[15:8]
011	FIL0THL[7:0]
012	FIL0THL[15:8]
013	0	0	0	0	0	0	0	IIR0 2BIT EN
016	FIL1THH[7:0]
017	FIL1THH[15:8]
018	FIL1THL[7:0]
019	FIL1THL[15:8]
01A	0	0	0	0	0	0	0	IIR1 2BIT EN
01D	DWELLIIR[7:0]
01E	DWELLIIR[15:8]
020	0	0	0	0	0	0	0	IIR0 2BIT EN
021	0	0	0	PWRDETACCU
022	PWRDETH[7:0]
023	PWRDETH[15:8]
024	PWRDETL[7:0]
025	PWRDETL[15:8]
POWER DETECTOR PAGE (continued)
027	0	0	0	0	0	0	0	RMS 2BIT EN
02B	0	0	0	RESET AGC	0	0	0	0
037	0	0	0	0	IODIR GPIO4	IODIR GPIO3	IODIR GPIO2	IODIR GPIO1
038	0	0	INSEL1		0	0	INSEL0
POWER DETECTOR PAGE (Direct Addressing, 16-Bit Address, 5400h)
032	OUTSEL GPIO1
033	OUTSEL GPIO2
034	OUTSEL GPIO3
035	OUTSEL GPIO4

9.5.1 Example Register Writes

This section provides three different example register writes. Table 27 describes a global power-down register write, Table 28 describes the register writes when the scrambler is enabled, and Table 29 describes the register writes for 8x decimation for channels A and B (complex output, 1 DDC mode) with the NCO set to 1.56 GHz (f_S = 2.6 GSPS) and the JESD format configured to LMFS = 4421.

Table 27. Global Power-Down

ADDRESS	DATA	COMMENT
12h	04h	Set the master page
20h	01h	Set the global power-down

Table 28. Scrambler Enable

ADDRESS	DATA	COMMENT
4004h	69h	Select the digital JESD page
4003h	00h	Select the digital JESD page
6006h	80h	Scrambler enable, channel A
7006h	80h	Scrambler enable, channel B

Table 29. 8x Decimation for Channel A and B

ADDRESS	DATA	COMMENT
4004h	68h	Select the main digital page for channel A
4003h	00h	Select the main digital page for channel A
6000h	01h	Issue a digital reset for channel A
6000h	00h	Clear the digital for reset channel A
4003h	01h	Select the main digital page for channel B
6000h	01h	Issue a digital reset for channel B
6000h	00h	Clear the digital reset for channel B
4004h	69h	Select the digital JESD page
4003h	00h	Select the digital JESD page
6002h	01h	Set JESD MODE0 = 1, channel A
7002h	01h	Set JESD MODE0 = 1, channel B
5000h	01h	Enable the DDC, channel A
5001h	02h	Set decimation to 8x complex
5007h	9Ah	Set the LSB of DDC0, NCO1 to 9Ah (f_NCO = 1.56GHz, f_S = 2.6 GSPS)
5008h	99h	Set the MSB of DDC0, NCO1 to 99h (f_NCO = 1.56 GHz, f_S = 2.6 GSPS)
5014h	01h	Enable the 6-dB digital gain of DDC0
5801h	02h	Set decimation to 8x complex
5807h	9Ah	Set the LSB of DDC0, NCO1 to 9Ah (f_NCO = 1.56 GHz, f_S = 2.6 GSPS)
5808h	99h	Set the MSB of DDC0, NCO1 to 99h (f_NCO = 1.56 GHz, f_S = 2.6 GSPS)
5814h	01h	Enable the 6-dB digital gain of DDC0

9.5.2 Register Descriptions

Table 30. ADC32RF42 Access Type Codes

Access Type	Code	Description
R	R	Read
R-W	R/W	Read or Write
W	W	Write
-n		Value after reset or the default value

9.5.2.1 General Registers

9.5.2.1.1 Register 000h (address = 000h), General Registers

Figure 90. Register 000h

7	6	5	4	3	2	1	0
RESET	0	0	0	0	0	0	RESET
R/W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 31. Register 000h Field Descriptions

Bit	Field	Type	Reset	Description
7	RESET	R/W	0h	0 = Normal operation 1 = Internal software reset, clears back to 0
6-1	0	W	0h	Must write 0
0	RESET	R/W	0h	0 = Normal operation⁽¹⁾ 1 = Internal software reset, clears back to 0

(1) Both bits (7, 0) must be set simultaneously to perform a reset.

9.5.2.1.2 Register 002h (address = 002h), General Registers

Figure 91. Register 002h

7	6	5	4	3	2	1	0
DIGITAL BANK PAGE SEL[7:0]
R/W-0h

Table 32. Register 002h Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DIGITAL BANK PAGE SEL[7:0]	R/W	0h	Program the JESD BANK PAGE SEL[23:0] bits to access the desired page in the digital bank. 610000h = Offset corr page channel A selected 610100h = Offset corr page channel B selected 610005h = Digital gain page channel A selected 610105h = Digital gain page channel B selected 680000h = Main digital page channel A selected 680100h = Main digital page channel B selected 690000h = JESD digital page selected

9.5.2.1.3 Register 003h (address = 003h), General Registers

Figure 92. Register 003h

7	6	5	4	3	2	1	0
DIGITAL BANK PAGE SEL[15:8]
R/W-0h

Table 33. Register 003h Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DIGITAL BANK PAGE SEL[15:8]	R/W	0h	Program the JESD BANK PAGE SEL[23:0] bits to access the desired page in the digital bank. 610000h = Offset corr page channel A selected 610100h = Offset corr page channel B selected 610005h = Digital gain page channel A selected 610105h = Digital gain page channel B selected 680000h = Main digital page channel A selected 680100h = Main digital page channel B selected 690000h = JESD digital page selected

9.5.2.1.4 Register 004h (address = 004h), General Registers

Figure 93. Register 004h

7	6	5	4	3	2	1	0
DIGITAL BANK PAGE SEL[23:16]
R/W-0h

Table 34. Register 004h Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DIGITAL BANK PAGE SEL[23:16]	R/W	0h	Program the JESD BANK PAGE SEL[23:0] bits to access the desired page in the digital bank. 610000h = Offset corr page channel A selected 610100h = Offset corr page channel B selected 610005h = Digital gain page channel A selected 610105h = Digital gain page channel B selected 680000h = Main digital page channel A selected 680100h = Main digital page channel B selected 690000h = JESD digital page selected

9.5.2.1.5 Register 010h (address = 010h), General Registers

Figure 94. Register 010h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	3 or 4 WIRE
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 35. Register 010h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	3 or 4 WIRE	R/W	0h	0 = 4-wire SPI (default) 1 = 3-wire SPI where SDIN become input or output

9.5.2.1.6 Register 011h (address = 011h), General Registers

Figure 95. Register 011h

7	6	5	4	3	2	1	0
ADC PAGE SEL
R/W-0h

Table 36. Register 011h Field Descriptions

Bit	Field	Type	Reset	Description
7-0	ADC PAGE SEL	R/W	0h	00000000 = Normal operation, ADC page is not selected 11111111 = ADC page is selected; MASTER PAGE SEL must be set to 0

9.5.2.1.7 Register 012h (address = 012h), General Registers

Figure 96. Register 012h

7	6	5	4	3	2	1	0
0	0	0	0	0	MASTER PAGE SEL	0	0
W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h	W-0h	W-0h

Table 37. Register 012h Field Descriptions

Bit	Field	Type	Reset	Description
7-3	0	W	0h	Must write 0
2	MASTER PAGE SEL	R/W	0h	0 = Normal operation 1 = Selects the master page address; ADC PAGE must be set to 0
1-0	0	W	0h	Must write 0

9.5.3 Master Page (M = 0)

9.5.3.1 Register 020h (address = 020h), Master Page

Figure 97. Register 020h

7	6	5	4	3	2	1	0
0	0	0	PDN SYSREF	0	0	PDN CHB	GLOBAL PDN
W-0h	W-0h	W-0h	R/W-0h	W-0h	R/W-0h	R/W-0h	R/W-0h

Table 38. Register 020h Field Descriptions

Bit	Field	Type	Reset	Description
7-5	0	W	0h	Must write 0
4	PDN SYSREF	R/W	0h	This bit powers down the SYSREF input buffer. 0 = Normal operation 1 = SYSREF input capture buffer is powered down and further SYSREF input pulses are ignored
3-2	0	W	0h	Must write 0
1	PDN CHB	R/W	0h	This bit powers down channel B. 0 = Normal operation 1 = Channel B is powered down
0	GLOBAL PDN	R/W	0h	This bit enables the global power-down. 0 = Normal operation 1 = Global power-down enabled

9.5.3.2 Register 032h (address = 032h), Master Page

Figure 98. Register 032h

7	6	5	4	3	2	1	0
0	0	INCR CM IMPEDANCE	0	0	0	0	0
W-0h	W-0h	R/W-0h	W-0h	W-0h	W-0h	W-0h	W-0h

Table 39. Register 032h Field Descriptions

Bit	Field	Type	Reset	Description
7-6	0	W	0h	Must write 0
5	INCR CM IMPEDANCE	R/W	0h	Only use this bit when analog inputs are dc-coupled to the driver. 0 = VCM buffer directly drives the common point of biasing resistors. 1 = VCM buffer drives the common point of biasing resistors with > 5 kΩ
4-0	0	W	0h	Must write 0

9.5.3.3 Register 039h (address = 039h), Master Page

Figure 99. Register 039h

7	6	5	4	3	2	1	0
0	ALWAYS WRITE 1	0	ALWAYS WRITE 1	0	0	PDN CHB EN	SYNC TERM DIS
W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h	R/W-0h	R/W-0h

Table 40. Register 039h Field Descriptions

Bit	Field	Type	Reset	Description
7	0	W	0h	Must write 0
6	ALWAYS WRITE 1	W	0h	Always set this bit to 1
5	0	W	0h	Must write 0
4	ALWAYS WRITE 1	W	0h	Always set this bit to 1
3-2	0	W	0h	Must write 0
1	PDN CHB EN	R/W	0h	This bit enables the power-down control of channel B through the SPI in register 20h. 0 = PDN control disabled 1 = PDN control enabled
0	SYNC TERM DIS	R/W	0h	This bit disables the on-chip, 100-Ω termination resistors on the SYNCB input. 0 = On-chip, 100-Ω termination enabled 1 = On-chip, 100-Ω termination disabled

9.5.3.4 Register 03Ch (address = 03Ch), Master Page

Figure 100. Register 03Ch

7	6	5	4	3	2	1	0
0	SYSREF DEL EN	0	0	0	0	SYSREF DEL[4:3]
W-0h	R/W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 41. Register 03Ch Field Descriptions

Bit	Field	Type	Reset	Description
7	0	W	0h	Must write 0
6	SYSREF DEL EN	R/W	0h	This bit allows an internal delay to be added to the SYSREF input. 0 = SYSREF delay disabled 1 = SYSREF delay enabled through register settings [3Ch (bits 1-0), 5Ah (bits 7-5)]
5-2	0	W	0h	Must write 0
1-0	SYSREF DEL[4:3]	R/W	0h	When the SYSREF delay feature is enabled (3Ch, bit 6) the delay can be adjusted in 25-ps steps; the first step is 175 ps. The PVT variation of each 25-ps step is ±10 ps. The 175-ps step is ±50 ps; see Table 43.

9.5.3.5 Register 05Ah (address = 05Ah), Master Page

Figure 101. Register 05Ah

7	6	5	4	3	2	1	0
SYSREF DEL[2:0]			0	0	0	0	0
W-0h	R/W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h

Table 42. Register 05Ah Field Descriptions

Bit	Field	Type	Reset	Description
7	SYSREF DEL2	W	0h	When the SYSREF delay feature is enabled (3Ch, bit 6) the delay can be adjusted in 25-ps steps; the first step is 175 ps. The PVT variation of each 25-ps step is ±10 ps. The 175-ps step is ±50 ps; see Table 43.
6	SYSREF DEL1	R/W
5	SYSREF DEL0	W
4-0	0	W	0h	Must write 0

Table 43. SYSREF DEL[2:0] Bit Settings

STEP	SETTING	STEP (NOM)	TOTAL DELAY (NOM)
1	01000	175 ps	175 ps
2	00111	25 ps	200 ps
3	00110	25 ps	225 ps
4	00101	25 ps	250 ps
5	00100	25 ps	275 ps
6	00011	25 ps	300 ps

9.5.3.6 Register 03Dh (address = 3Dh), Master Page

Figure 102. Register 03Dh

7	6	5	4	3	2	1	0
0	0	0	0	0	JESD OUTPUT SWING
W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 44. Register 03Dh Field Descriptions

Bit	Field	Type	Reset	Description
7-3	0	W	0h	Must write 0
2-0	JESD OUTPUT SWING	R/W	0h	These bits select the output amplitude, V_OD (mV_PP), of the JESD transmitter for all lanes. 0 = 860 mV_PP 1= 810 mV_PP 2 = 770 mV_PP 3 = 745 mV_PP 4 = 960 mV_PP 5 = 930 mV_PP 6 = 905 mV_PP 7 = 880 mV_PP

9.5.3.7 Register 057h (address = 057h), Master Page

Figure 103. Register 057h

7	6	5	4	3	2	1	0
0	0	0	SEL SYSREF REG	ASSERT SYSREF REG	0	0	0
W-0h	W-0h	W-0h	R/W-0h	R/W-0h	W-0h	W-0h	W-0h

Table 45. Register 057h Field Descriptions

Bit	Field	Type	Reset	Description
7-5	0	W	0h	Must write 0
4	SEL SYSREF REG	R/W	0h	Set this bit to use the SPI register to assert SYSREF. 0 = SYSREF is asserted by device pins 1 = SYSREF can be asserted by the ASSERT SYSREF REG register bit Other bits = 0
3	ASSERT SYSREF REG	R/W	0h	SYSREF can be asserted using this bit. Ensure that the SEL SYSREF REG register bit is set high before using this bit; see the Using SYSREF section. 0 = SYSREF is logic low 1 = SYSREF is logic high
2-0	0	W	0h	Must write 0

9.5.3.8 Register 058h (address = 058h), Master Page

Figure 104. Register 058h

7	6	5	4	3	2	1	0
0	0	SYNCB POL	0	0	0	0	0
W-0h	W-0h	R/W-0h	W-0h	W-0h	W-0h	W-0h	W-0h

Table 46. Register 058h Field Descriptions

Bit	Field	Type	Reset	Description
7-6	0	W	0h	Must write 0
5	SYNCB POL	R/W	0h	This bit inverts the SYNCB polarity. 0 = Polarity is not inverted; this setting matches the timing diagrams in this document and is the proper setting to use 1 = Polarity is inverted
4-0	0	W	0h	Must write 0

9.5.4 ADC Page (FFh, M = 0)

9.5.4.1 Register 03Fh (address = 03Fh), ADC Page

Figure 105. Register 03Fh

7	6	5	4	3	2	1	0
0	0	0	0	0	SLOW SP EN1	0	0
W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h	W-0h	W-0h

Table 47. Register 03Fh Field Descriptions

Bit	Field	Type	Reset	Description
7-3	0	W	0h	Must write 0
2	SLOW SP EN1	R/W	0h	This bit must be enabled for clock rates below 2.5 GSPS. 0 = ADC sampling rates are faster than 2.5 GSPS 1 = ADC sampling rates are slower than 2.5 GSPS
1-0	0	W	0h	Must write 0

9.5.4.2 Register 042h (address = 042h), ADC Page

Figure 106. Register 042h

7	6	5	4	3	2	1	0
0	0	0	SLOW SP EN2	0	0	ALWAYS WRITE 1	ALWAYS WRITE 1
W-0h	W-0h	W-0h	R/W-0h	W-0h	W-0h	W-0h	W-0h

Table 48. Register 042h Field Descriptions

Bit	Field	Type	Reset	Description
7-5	0	W	0h	Must write 0
4	SLOW SP EN2	R/W	0h	This bit must be enabled for clock rates below 2.5 GSPS. 0 = ADC sampling rates are faster than 2.5 GSPS 1 = ADC sampling rates are slower than 2.5 GSPS
3-2	0	W	0h	Must write 0
1-0	ALWAYS WRITE 1	W	1h	Always set this bit to 1

9.5.5 Offset Corr Page Channel A (610000h, M = 1)

9.5.5.1 Register 068h (address = 068h), Offset Corr Page Channel A

Figure 107. Register 068h

7	6	5	4	3	2	1	0
FREEZE OFFSET CORR	0	ALWAYS WRITE 1	0	0	DIS OFFSET CORR	ALWAYS WRITE 1	0
R/W-0h	W-0h	R/W-0h	W-0h	W-0h	R/W-0h	R/W-0h	R/W-0h

Table 49. Register 068h Field Descriptions

Bit	Field	Type	Reset	Description
7	FREEZE OFFSET CORR	R/W	0h	Use this bit and bits 5 and 1 to freeze the offset estimation process of the offset corrector; see the Using DC Coupling in the ADC32RF42 section. 011 = Apply this setting after powering up the device 111 = Offset corrector is frozen, does not estimate offset anymore, and applies the last computed value. Others = Do not use
6	0	W	0h	Must write 0
5	ALWAYS WRITE 1	R/W	0h	Always set this bit to 1 for the offset correction block to work properly.
4-3	0	W	0h	Must write 0
2	DIS OFFSET CORR	R/W	0h	0 = Offset correction block works and removes f_S/8, f_S/4, 3f_S/8, and f_S/2 spurs 1 = Offset correction block is disabled
1	ALWAYS WRITE 1	R/W	0h	Always set this bit to 1 for the offset correction block to work properly.
0	0	W	0h	Must write 0

9.5.6 Offset Corr Page Channel B (610100h, M = 1)

9.5.6.1 Register 068h (address = 068h), Offset Corr Page Channel B

Figure 108. Register 068h

7	6	5	4	3	2	1	0
FREEZE OFFSET CORR	0	ALWAYS WRITE 1	0	0	DIS OFFSET CORR	ALWAYS WRITE 1	0
R/W-0h	W-0h	R/W-0h	W-0h	W-0h	R/W-0h	R/W-0h	R/W-0h

Table 50. Register 068h Field Descriptions

Bit	Field	Type	Reset	Description
7,5,1	FREEZE OFFSET CORR	R/W	0h	Use this bit and bits 5 and 1 to freeze the offset estimation process of the offset corrector; see the Using DC Coupling in the ADC32RF42 section. 011 = Apply this setting after powering up the device 111 = Offset corrector is frozen, does not estimate offset anymore, and applies the last computed value. Others = Do not use
6	0	W	0h	Must write 0
5	ALWAYS WRITE 1	R/W	0h	Always set this bit to 1 for the offset correction block to work properly.
4-3	0	W	0h	Must write 0
2	DIS OFFSET CORR	R/W	0h	0 = Offset correction block works and removes f_S/8, f_S/4, 3f_S/8, and f_S/2 spurs 1 = Offset correction block is disabled
1	ALWAYS WRITE 1	R/W	0h	Always set this bit to 1 for the offset correction block to work properly.
0	0	W	0h	Must write 0

9.5.7 Digital Gain Page (610005h, M = 1 for Channel A and 610105h, M = 1 for Channel B)

9.5.7.1 Register 0A6h (address = 0A6h), Digital Gain Page

Figure 109. Register 0A6h

7	6	5	4	3	2	1	0
0	0	0	0	DIGITAL GAIN
W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 51. Register 0A6h Field Descriptions

Bit	Field	Type	Reset	Description
7-4	0	W	0h	Must write 0
3-0	DIGITAL GAIN	R/W	0h	These bits apply a digital gain to the ADC data (before the DDC) up to 11 dB. 0000 = Default 0001 = 1 dB 1011 = 11 dB Others = Do not use

9.5.8 Main Digital Page Channel A (680000h, M = 1)

9.5.8.1 Register 000h (address = 000h), Main Digital Page Channel A

Figure 110. Register 000h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	DIG CORE RESET GBL
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 52. Register 000h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	DIG CORE RESET GBL	R/W	0h	Pulse this bit (0 →1 →0) to reset the digital core (applies to both channel A and B). All Nyquist zone settings take effect when this bit is pulsed.

9.5.8.2 Register 0A2h (address = 0A2h), Main Digital Page Channel A

Figure 111. Register 0A2h

7	6	5	4	3	2	1	0
0	0	0	0	NQ ZONE EN	NYQUIST ZONE
W-0h	W-0h	W-0h	W-0h	R/W-0h	R/W-0h

Table 53. Register 0A2h Field Descriptions

Bit	Field	Type	Reset	Description
7-4	0	W	0h	Must write 0
3	NQ ZONE EN	R/W	0h	This bit allows for specification of the operating Nyquist zone. 0 = Nyquist zone specification disabled 1 = Nyquist zone specification enabled
2-0	NYQUIST ZONE	R/W	0h	These bits specify the operating Nyquist zone for the analog correction loop. Set the NQ ZONE EN bit before programming these bits. For example, at a 1.5-GSPS chip clock, the first Nyquist zone is from dc to 750 MHz, the second Nyquist zone is from 750 MHz to 1.5 GHz, and so on. 000 = First Nyquist zone (dc – f_S / 2) 001 = Second Nyquist zone (f_S / 2 – f_S) 010 = Third Nyquist zone 011 = Fourth Nyquist zone

9.5.9 Main Digital Page Channel B (680100h, M = 1)

9.5.9.1 Register 0A2h (address = 0A2h), Main Digital Page Channel B

Figure 112. Register 0A2h

7	6	5	4	3	2	1	0
0	0	0	0	NQ ZONE EN	NYQUIST ZONE
W-0h	W-0h	W-0h	W-0h	R/W-0h	R/W-0h

Table 54. Register 0A2h Field Descriptions

Bit	Field	Type	Reset	Description
7-4	0	W	0h	Must write 0
3	NQ ZONE EN	R/W	0h	This bit allows for specification of the operating Nyquist zone. 0 = Nyquist zone specification disabled 1 = Nyquist zone specification enabled
2-0	NYQUIST ZONE	R/W	0h	These bits specify the operating Nyquist zone for the analog correction loop. Set the NQ ZONE EN bit before programming these bits. For example, at a 2.6-GSPS chip clock, first Nyquist zone is from dc to 1.3 GHz, the second Nyquist zone is from 1.3 GHz to 2.6 GHz, and so on. 000 = First Nyquist zone (dc – f_S / 2) 001 = Second Nyquist zone (f_S / 2 – f_S) 010 = Third Nyquist zone 011 = Fourth Nyquist zone

9.5.10 JESD Digital Page (690000h, M = 1)

9.5.10.1 Register 001h (address = 001h), JESD Digital Page

Figure 113. Register 001h

7	6	5	4	3	2	1	0
CTRL K	0	0	TESTMODE EN	0	LANE ALIGN	FRAME ALIGN	TX LINK DIS
R/W-0h	W-0h	W-0h	R/W-0h	W-0h	R/W-0h	R/W-0h	R/W-0h

Table 55. Register 001h Field Descriptions

Bit	Field	Type	Reset	Description
7	CTRL K	R/W	0h	This bit is the enable bit for the number of frames per multiframe. 0 = Default is five frames per multiframe 1 = Frames per multiframe can be set in register 07h
6-5	0	R/W	0h	Must write 0
4	TESTMODE EN		0	This bit generates a long transport layer test pattern mode according to section 5.1.6.3 of the JESD204B specification. 0 = Test mode disabled 1 = Test mode enabled
3	0	W	0h	Must write 0
2	LANE ALIGN	R/W	0h	This bit inserts a lane alignment character (K28.3) for the receiver to align to the lane boundary per section 5.3.3.5 of the JESD204B specification. 0 = Normal operation 1 = Inserts lane alignment characters
1	FRAME ALIGN	R/W	0h	This bit inserts a frame alignment character (K28.7) for the receiver to align to the frame boundary per section 5.3.35 of the JESD204B specification. 0 = Normal operation 1 = Inserts frame alignment characters
0	TX LINK DIS	R/W	0h	This bit disables sending the initial link alignment (ILA) sequence when SYNC is deasserted. 0 = Normal operation 1 = ILA disabled

9.5.10.2 Register 002h (address = 002h ), JESD Digital Page

Figure 114. Register 002h

7	6	5	4	3	2	1	0
SYNC REG	SYNC REG EN	0	0	12BIT MODE		JESD MODE0
R/W-0h	R/W-0h	W-0h	W-0h	R/W-0h		R/W-0h

Table 56. Register 002h Field Descriptions

Bit	Field	Type	Reset	Description
7	SYNC REG	R/W	0h	This bit provides SYNC control through the SPI. 0 = Normal operation 1 = ADC output data are replaced with K28.5 characters
6	SYNC REG EN	R/W	0h	This bit is the enable bit for SYNC control through the SPI. 0 = Normal operation 1 = SYNC control through the SPI is enabled (ignores the SYNCB input pins)
5-4	0	W	0h	Must write 0
3-2	12BIT MODE	R/W	0h	This bit enables the 12-bit output mode for more efficient data packing. 00 = Normal operation, 14-bit output 01, 10 = Unused 11 = High-efficient data packing enabled
1-0	JESD MODE0	R/W	0h	These bits select the configuration register to configure the correct LMFS frame assemblies for different decimation settings; see the JESD frame assembly tables in the JESD204B Frame Assembly section. 00 = 0 01 = 1 10 = 2 11 = 3

9.5.10.3 Register 003h (address = 003h), JESD Digital Page

Figure 115. Register 003h

7	6	5	4	3	2	1	0
LINK LAYER TESTMODE			LINK LAY RPAT	LMFC MASK RESET	JESD MODE1	JESD MODE2	RAMP 12BIT
R/W-0h			R/W-0h	R/W-0h	R/W-1h	R/W-0h	R/W-0h

Table 57. Register 003h Field Descriptions

Bit	Field	Type	Reset	Description
7-5	LINK LAYER TESTMODE	R/W	0h	These bits generate a pattern according to section 5.3.3.8.2 of the JESD204B document. 000 = Normal ADC data 001 = D21.5 (high-frequency jitter pattern) 010 = K28.5 (mixed-frequency jitter pattern) 011 = Repeat initial lane alignment (generates a K28.5 character and repeats lane alignment sequences continuously) 100 = 12-octet RPAT jitter pattern
4	LINK LAY RPAT	R/W	0h	This bit changes the running disparity in a modified RPAT pattern test mode (only when link layer test mode = 100). 0 = Normal operation 1 = Changes disparity
3	LMFC MASK RESET	R/W	0h	0 = Normal operation
2	JESD MODE1	R/W	1h	These bits select the configuration register to configure the correct LMFS frame assemblies for different decimation settings; see the JESD frame assembly tables in the JESD204B Frame Assembly section
1	JESD MODE2	R/W	0h	These bits select the configuration register to configure the correct LMFS frame assemblies for different decimation settings; see the JESD frame assembly tables in the JESD204B Frame Assembly section
0	RAMP 12BIT	R/W	0h	This bit enables the RAMP test pattern for 12-bit mode only (LMFS = 42810). 0 = Normal data output 1 = Digital output is the RAMP pattern

9.5.10.4 Register 004h (address = 004h), JESD Digital Page

Figure 116. Register 004h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	REL ILA SEQ
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 58. Register 004h Field Descriptions

Bit	Field	Type	Reset	Description
7-2	0	W	0h	Must write 0
1-0	REL ILA SEQ	R/W	0h	These bits delay the generation of the lane alignment sequence by 0, 1, 2, or 3 multiframes after the code group synchronization. 00 = 0 multiframe delays 01 = 1 multiframe delay 10 = 2 multiframe delays 11 = 3 multiframe delays

9.5.10.5 Register 006h (address = 006h), JESD Digital Page

Figure 117. Register 006h

7	6	5	4	3	2	1	0
SCRAMBLE EN	0	0	0	0	0	0	0
R/W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h

Table 59. Register 006h Field Descriptions

Bit	Field	Type	Reset	Description
7	SCRAMBLE EN	R/W	0h	This bit is the scramble enable bit in the JESD204B interface. 0 = Scrambling disabled 1 = Scrambling enabled
6-0	0	W	0h	Must write 0

9.5.10.6 Register 007h (address = 007h), JESD Digital Page

Figure 118. Register 007h

7	6	5	4	3	2	1	0
0	0	0	FRAMES PER MULTIFRAME (K)
W-0h	W-0h	W-0h	R/W-0h

Table 60. Register 007h Field Descriptions

Bit	Field	Type	Reset	Description
7-5	0	W	0h	Must write 0
4-0	FRAMES PER MULTIFRAME (K)	R/W	0h	These bits set the number of multiframes. Actual K is the value in hex + 1 (that is, 0Fh is K = 16).

9.5.10.7 Register 016h (address = 016h), JESD Digital Page

Figure 119. Register 016h

7	6	5	4	3	2	1	0
LANE 0		LANE 1		LANE 2		LANE 03
R/W-0h		R/W-0h		R/W-0h		R/W-0h

Table 61. Register 016h Field Descriptions

Bit	Field	Type	Reset	Description
7-6	LANE 0	R/W	0h	For 80x mode: set these bits as 70h. Also set the 80X MODE EN register bit. For 40x mode: set these bits as 70h. For 20x mode: these bits can be used to swap the data on output lanes as shown in Table 62 and Figure 120.
5-4	LANE 1	R/W	0h
3-2	LANE 2	R/W	0h
1-0	LANE 3	R/W	0h

Table 62. Swapping Data on Output Lanes for 20x Serialization

REGISTER BIT LANE 0	OUTPUT LANE0 CARRIES	REGISTER BIT LANE 1	OUTPUT LANE1 CARRIES	REGISTER BIT LANE 2	OUTPUT LANE2 CARRIES	REGISTER BIT LANE 3	OUTPUT LANE3 CARRIES
00	D0	00	D1	00	D2	00	D3
01	D1	01	D2	01	D3	01	D0
10	D2	10	D3	10	D0	10	D1
11	D3	11	D0	11	D1	11	D2

Figure 120. Output Lane Multiplexer

9.5.10.8 Register 017h (address = 017h), JESD Digital Page

Figure 121. Register 017h

7	6	5	4	3	2	1	0
0	0	0	0	Lane0 POL	Lane1 POL	Lane2 POL	Lane3 POL
W-0h	R/W-0h	R/W-0h	R/W-0h	W-0h	W-0h	W-0h	W-0h

Table 63. Register 017h Field Descriptions

Bit	Field	Type	Reset	Description
7	0	W	0h	Must write 0
6-4	0	R/W	0h	Must write 0
3-0	Lane[3:0] POL	W	0h	These bits set the polarity of the individual JESD output lanes. 0 = Polarity as given in the pinout (noninverted) 1 = Inverts polarity (positive, P, or negative, M)

9.5.10.9 Register 032h-035h (address = 032h-035h), JESD Digital Page

Figure 122. Register 032h

7	6	5	4	3	2	1	0
SEL EMP LANE 0						0	0
R/W-0h						W-0h	W-0h

Figure 123. Register 033h

7	6	5	4	3	2	1	0
SEL EMP LANE 1						0	0
R/W-0h						W-0h	W-0h

Figure 124. Register 034h

7	6	5	4	3	2	1	0
SEL EMP LANE 2						0	0
R/W-0h						W-0h	W-0h

Figure 125. Register 035h

7	6	5	4	3	2	1	0
SEL EMP LANE 3						0	0
R/W-0h						W-0h	W-0h

Table 64. Register 032h-035h Field Descriptions

Bit	Field	Type	Reset	Description
7-2	SEL EMP LANE	R/W	0h	These bits select the amount of de-emphasis for the JESD output transmitter. The de-emphasis value in dB is measured as the ratio between the peak value after the signal transition to the settled value of the voltage in one bit period. 0 = 0 dB 1 = –1 dB 3 = –2 dB 7 = –4.1 dB 15 = –6.2 dB 31 = –8.2 dB 63 = –11.5 dB
1-0	0	W	0h	Must write 0

9.5.10.10 Register 036h (address = 036h), JESD Digital Page

Figure 126. Register 036h

7	6	5	4	3	2	1	0
80X MODE EN	CMOS SYNCB	0	0	0	0	0	0
R/W-0h	R/W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h

Table 65. Register 036h Field Descriptions

Bit	Field	Type	Reset	Description
7	80X MODE EN	R/W	0h	This bit enables the 80x mode. 0 = 80x mode disable 1 = 80x mode enable
6	CMOS SYNCB	R/W	0h	This bit enables single-ended control of SYNCB using the GPIO4 pin (pin 63). The differential SYNCB input is ignored. Set the EN CMOS SYNC register bit to make this bit effective. When programming this bit, keep the CH bit set to 1. 0 = Differential SYNCB input 1 = Single-ended SYNCB input using pin 63
5-0	0	W	0h	Must write 0

9.5.10.11 Register 037h (address = 037h), JESD Digital Page

Figure 127. Register 037h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	PLL MODE
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 66. Register 037h Field Descriptions

Bit	Field	Type	Reset	Description
7-2	0	W	0h	Must write 0
1-0	PLL MODE	R/W	0h	These bits select the PLL multiplication factor; see the JESD tables in the JESD204B Frame Assembly section for settings. 00 = 20x mode 01 = 16x mode 10 = 40x mode (write register 16h with 70h) 11 = 80x mode (the 40X_80X MODE bit in register 16h must also be set)

9.5.10.12 Register 03Ch (address = 03Ch), JESD Digital Page

Figure 128. Register 03Ch

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	EN CMOS SYNCB
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 67. Register 03Ch Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
5	EN CMOS SYNCB	R/W	0h	0 = Default 1 = This bit enables the control of the SYNC request from the CMOS SYNCB register bit. When programming this bit, keep the CH bit set to 1.

9.5.10.13 Register 03Eh (address = 03Eh), JESD Digital Page

Figure 129. Register 03Eh

7	6	5	4	3	2	1	0
0	MASK CLKDIV SYSREF	MASK NCO SYSREF	0	0	0	0	0
W-0h	R/W-0h	R/W-0h	W-0h	W-0h	W-0h	W-0h	W-0h

Table 68. Register 03Eh Field Descriptions

Bit	Field	Type	Reset	Description
7	0	W	0h	Must write 0
6	MASK CLKDIV SYSREF	R/W	0h	Use this bit to mask the SYSREF going to the input clock divider. 0 = Input clock divider is reset when SYSREF is asserted (that is, when SYSREF transitions from low to high) 1 = Input clock divider ignores SYSREF assertions
5	MASK NCO SYSREF	R/W	0h	Use this bit to mask the SYSREF going to the NCO in the DDC block and LMFC counter of the JESD interface. 0 = NCO phase and LMFC counter are reset when SYSREF is asserted (that is, when SYSREF transitions from low to high) 1 = NCO and LMFC counter ignore SYSREF assertions
4-0	0	W	0h	Must write 0

9.5.11 Special Page Channel A

Channel A (6A0100h, M = 1)

9.5.11.1 Register 019h (address = 019h), Special Page Channel A

Figure 130. Register 019h

7	6	5	4	3	2	1	0
0	0	0	ALWAYS WRITE 1		0	0	0
W-0h	W-0h	W-0h	R/W-0h		W-0h	W-0h	W-0h

Table 69. Register 019h Field Descriptions

Bit	Field	Type	Reset	Description
7-5	0	W	0h	Must write 0
4-3	W-0h	R/W	0h	Always write this bit as 1.
2-0	0	W	0h	Must write 0

9.5.12 Special Page Channel B

Channel B (6A0000h, M = 1)

9.5.12.1 Register 019h (address = 019h), Special Page Channel B

Figure 131. Register 019h

7	6	5	4	3	2	1	0
0	0	0	ALWAYS WRITE 1		0	0	0
W-0h	W-0h	W-0h	R/W-0h		W-0h	W-0h	W-0h

Table 70. Register 019h Field Descriptions

Bit	Field	Type	Reset	Description
7-5	0	W	0h	Must write 0
4-3	W-0h	R/W	0h	Always write this bit as 1.
2-0	0	W	0h	Must write 0

9.5.13 Decimation Filter Page

Direct Addressing, 16-Bit Address, 5000h for Channel A, 5800h for Channel B

9.5.13.1 Register 000h (address = 000h), Decimation Filter Page

Figure 132. Register 000h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	DDC EN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 71. Register 000h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	DDC EN	R/W	0h	This bit enables the decimation filter and disables the bypass mode. 0 = Bypass mode (DDC disabled) 1 = Decimation filter enabled

9.5.13.2 Register 001h (address = 001h), Decimation Filter Page

Figure 133. Register 001h

7	6	5	4	3	2	1	0
0	0	0	0	DECIM FACTOR
W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 72. Register 001h Field Descriptions

Bit	Field	Type	Reset	Description
7-4	0	W	0h	Must write 0
3-0	DECIM FACTOR	R/W	0h	These bits configure the decimation filter setting. 0010 = Divide-by-4 complex 0101 = Divide-by-6 complex 0111 = Divide-by-8 complex 1000 = Divide-by-9 complex 1001 = Divide-by-10 complex 1010 = Divide-by-12 complex 1100 = Divide-by-16 complex Others = Not used

9.5.13.3 Register 002h (address = 2h), Decimation Filter Page

Figure 134. Register 002h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	DUAL BAND EN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 73. Register 002h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	DUAL BAND EN	R/W	0h	This bit enables the dual-band DDC filter for the corresponding channel. 0 = Single-band DDC 1 = Dual-band DDC

9.5.13.4 Register 005h (address = 005h), Decimation Filter Page

Figure 135. Register 005h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	REAL OUT EN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 74. Register 005h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	REAL OUT EN	R/W	0h	This bit converts the complex output to real output at 2x the output rate. 0 = Complex output format 1 = Real output format

9.5.13.5 Register 006h (address = 006h), Decimation Filter Page

Figure 136. Register 006h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	DDC MUX
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 75. Register 006h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	DDC MUX	R/W	0h	This bit connects the DDC to the alternate channel ADC to enable up to four DDCs with one ADC and completely turn off the other ADC channel. 0 = Normal operation 1 = DDC block takes input from the alternate ADC

9.5.13.6 Register 007h (address = 007h), Decimation Filter Page

Figure 137. Register 007h

7	6	5	4	3	2	1	0
DDC0 NCO1 LSB
R/W-0h

Table 76. Register 007h Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DDC0 NCO1 LSB	R/W	0h	These bits are the LSB of the NCO frequency word for NCO1 of DDC0 (band 1). The LSB represents f_S / (2¹⁶), where f_S is the ADC sampling frequency.

9.5.13.7 Register 008h (address = 008h), Decimation Filter Page

Figure 138. Register 008h

7	6	5	4	3	2	1	0
DDC0 NCO1 MSB
R/W-0h

Table 77. Register 008h Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DDC0 NCO1 MSB	R/W	0h	These bits are the MSB of the NCO frequency word for NCO1 of DDC0 (band 1). The LSB represents f_S / (2¹⁶), where f_S is the ADC sampling frequency.

9.5.13.8 Register 009h (address = 009h), Decimation Filter Page

Figure 139. Register 009h

7	6	5	4	3	2	1	0
DDC0 NCO2 LSB
R/W-0h

Table 78. Register 009h Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DDC0 NCO2 MSB	R/W	0h	These bits are the LSB of the NCO frequency word for NCO2 of DDC0 (band 1). The LSB represents f_S / (2¹⁶), where f_S is the ADC sampling frequency.

9.5.13.9 Register 00Ah (address = 00Ah), Decimation Filter Page

Figure 140. Register 00Ah

7	6	5	4	3	2	1	0
DDC0 NCO2 MSB
R/W-0h

Table 79. Register 00Ah Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DDC0 NCO2 MSB	R/W	0h	These bits are the MSB of the NCO frequency word for NCO2 of DDC0 (band 1). The LSB represents f_S / (2¹⁶), where f_S is the ADC sampling frequency.

9.5.13.10 Register 00Bh (address = 00Bh), Decimation Filter Page

Figure 141. Register 00Bh

7	6	5	4	3	2	1	0
DDC0 NCO3 LSB
R/W-0h

Table 80. Register 00Bh Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DDC0 NCO3 LSB	R/W	0h	These bits are the LSB of the NCO frequency word for NCO3 of DDC0 (band 1). The LSB represents f_S / (2¹⁶), where f_S is the ADC sampling frequency.

9.5.13.11 Register 00Ch (address = 00Ch), Decimation Filter Page

Figure 142. Register 00Ch

7	6	5	4	3	2	1	0
DDC0 NCO3 MSB
R/W-0h

Table 81. Register 00Ch Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DDC0 NCO3 MSB	R/W	0h	These bits are the MSB of the NCO frequency word for NCO3 of DDC0 (band 1). The LSB represents f_S / (2¹⁶), where f_S is the ADC sampling frequency.

9.5.13.12 Register 00Dh (address = 00Dh), Decimation Filter Page

Figure 143. Register 00Dh

7	6	5	4	3	2	1	0
DDC1 NCO4 LSB
R/W-0h

Table 82. Register 00Dh Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DDC1 NCO4 LSB	R/W	0h	These bits are the LSB of the NCO frequency word for NCO4 of DDC1 (band 2, only when dual-band mode is enabled). The LSB represents f_S / (2¹⁶), where f_S is the ADC sampling frequency.

9.5.13.13 Register 00Eh (address = 00Eh), Decimation Filter Page

Figure 144. Register 00Eh

7	6	5	4	3	2	1	0
DDC1 NCO4 MSB
R/W-0h

Table 83. Register 00Eh Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DDC1 NCO4 MSB	R/W	0h	These bits are the MSB of the NCO frequency word for NCO4 of DDC1 (band 2, only when dual-band mode is enabled). The LSB represents f_S / (2¹⁶), where f_S is the ADC sampling frequency.

9.5.13.14 Register 00Fh (address = 00Fh), Decimation Filter Page

Figure 145. Register 00Fh

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	NCO SEL PIN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 84. Register 00Fh Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	NCO SEL PIN	R/W	0h	This bit enables NCO selection through the GPIO pins. 0 = NCO selection through SPI (see address 0h10) 1 = NCO selection through GPIO pins

9.5.13.15 Register 010h (address = 010h), Decimation Filter Page

Figure 146. Register 010h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	NCO SEL
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 85. Register 010h Field Descriptions

Bit	Field	Type	Reset	Description
7-2	0	W	0h	Must write 0
1-0	NCO SEL	R/W	0h	These bits enable NCO selection through register setting. 00 = NCO1 selected for DDC 1 01 = NCO2 selected for DDC 1 10 = NCO3 selected for DDC 1

9.5.13.16 Register 011h (address = 011h), Decimation Filter Page

Figure 147. Register 011h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	LMFC RESET MODE
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 86. Register 011h Field Descriptions

Bit	Field	Type	Reset	Description
7-2	0	W	0h	Must write 0
1-0	LMFC RESET MODE	R/W	0h	These bits reset the configuration for all DDCs and NCOs. 00 = All DDCs and NCOs are reset with every LMFC RESET 01 = Reset with first LMFC RESET after DDC start. Afterwards, reset only when analog clock dividers are resynchronized. 10 = Reset with first LMFC RESET after DDC start. Afterwards, whenever analog clock dividers are resynchronized, use two LMFC resets. 11 = Do not use an LMFC reset at all. Reset the DDCs only when a DDC start is asserted and afterwards continue normal operation. Deterministic latency is not ensured.

9.5.13.17 Register 014h (address = 014h), Decimation Filter Page

Figure 148. Register 014h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	DDC0 6DB GAIN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 87. Register 014h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	DDC0 6DB GAIN	R/W	0h	This bit scales the output of DDC0 by 2 (6 dB) to compensate for real-to-complex conversion and image suppression. This scaling does not apply to the high-bandwidth filter path (divide-by-4 and -6); see register 1Fh. 0 = Normal operation 1 = 6-dB digital gain is added

9.5.13.18 Register 016h (address = 016h), Decimation Filter Page

Figure 149. Register 016h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	DDC1 6DB GAIN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 88. Register 016h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	DDC1 6DB GAIN	R/W	0h	This bit scales the output of DDC0 by 2 (6 dB) to compensate for real-to-complex conversion and image suppression. This scaling does not apply to the high-bandwidth filter path (divide-by-4 and -6); see register 1Fh. 0 = Normal operation 1 = 6-dB digital gain is added

9.5.13.19 Register 01Eh (address = 01Eh), Decimation Filter Page

Figure 150. Register 01Eh

7	6	5	4	3	2	1	0
0	DDC DET LAT			0	0	0	0
W-0h	R/W-0h			W-0h	W-0h	W-0h	W-0h

Table 89. Register 01Eh Field Descriptions

Bit	Field	Type	Reset	Description
7	0	W	0h	Must write 0
6-4	DDC DET LAT	R/W	0h	These bits ensure deterministic latency depending on the decimation setting used; see Table 90.
3-0	0	W	0h	Must write 0

Table 90. DDC DET LAT Bit Settings

SETTING	COMPLEX DECIMATION SETTING
10h	Divide-by-24, -32 complex
20h	Divide-by-16, -18, -20 complex
40h	Divide-by-by 6, -12 complex
50h	Divide-by-4, -8, -9, -10 complex

9.5.13.20 Register 01Fh (address = 01Fh), Decimation Filter Page

Figure 151. Register 01Fh

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	WBF 6DB GAIN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 91. Register 01Fh Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	WBF 6DB GAIN	R/W	0h	This bit scales the output of the wide bandwidth DDC filter by 2 (6 dB) to compensate for real-to-complex conversion and image suppression. This setting only applies to the high-bandwidth filter path (divide-by-4 and -6). 0 = Normal operation 1 = 6-dB digital gain is added

9.5.13.21 Register 020h (address = 020h), Decimation Filter Page

Figure 152. Register 20h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	ALWAYS WRITE 1
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 92. Register 020h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	ALWAYS WRITE 1	R/W	0h	Always set this bit to 1

9.5.13.22 Register 033h-036h (address = 033h-036h), Decimation Filter Page

Figure 153. Register 033h

7	6	5	4	3	2	1	0
CUSTOM PATTERN1[7:0]
R/W-0h

Figure 154. Register 034h

7	6	5	4	3	2	1	0
CUSTOM PATTERN1[15:8]
R/W-0h

Figure 155. Register 035h

7	6	5	4	3	2	1	0
CUSTOM PATTERN2[7:0]
R/W-0h

Figure 156. Register 036h

7	6	5	4	3	2	1	0
CUSTOM PATTERN2[15:8]
R/W-0h

Table 93. Register 033h-036h Field Descriptions

Bit	Field	Type	Reset	Description
7-0	CUSTOM PATTERN	R/W	0h	These bits set the custom test pattern in address 33h, 34h, 35h, or 36h.

9.5.13.23 Register 037h (address = 037h), Decimation Filter Page

Figure 157. Register 037h

7	6	5	4	3	2	1	0
0	0	0	0	TEST PATTERN SEL
W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 94. Register 037h Field Descriptions

Bit	Field	Type	Reset	Description
7-3	0	W	0h	Must write 0
3-0	TEST PATTERN SEL	R/W	0h	These bits select the test pattern output on the channel. 0000 = Normal operation using ADC output data 0001 = Outputs all 0s 0010 = Outputs all 1s 0011 = Outputs toggle pattern: output data are an alternating sequence of 10101010101010 and 01010101010101 0100 = Output digital ramp: output data increment by one LSB every clock cycle from code 0 to 16384 0110 = Single pattern: output data are custom pattern 1 (75h and 76h) 0111 = Double pattern: output data alternate between custom pattern 1 and custom pattern 2 1000 = Deskew pattern: output data are AAAAh 1001 = SYNC pattern: output data are FFFFh

9.5.13.24 Register 038h (address = 038h), Decimation Filter Page

Figure 158. Register 038h

7	6	5	4	3	2	1	0
TEST PATTERN DDC2 Q-DATA				TEST PATTERN DDC2 I-DATA
R/W-0h				R/W-0h

Table 95. Register 038h Field Descriptions

Bit	Field	Type	Reset	Description
7-4	TEST PATTERN DDC2 Q-DATA	R/W	0h	These bits select the test patten for the Q stream of the DDC2. 0000 = Normal operation using ADC output data 0001 = Outputs all 0s 0010 = Outputs all 1s 0011 = Outputs toggle pattern: output data are an alternating sequence of 10101010101010 and 01010101010101 0100 = Output digital ramp: output data increment by one LSB every clock cycle from code 0 to 65535 0110 = Single pattern: output data are a custom pattern 1 (75h and 76h) 0111 Double pattern: output data alternate between custom pattern 1 and custom pattern 2 1000 = Deskew pattern: output data are AAAAh 1001 = SYNC pattern: output data are FFFFh
3-0	TEST PATTERN DDC2 I-DATA	R/W	0h	These bits select the test patten for the I stream of the DDC2. 0000 = Normal operation using ADC output data 0001 = Outputs all 0s 0010 = Outputs all 1s 0011 = Outputs toggle pattern: output data are an alternating sequence of 10101010101010 and 01010101010101 0100 = Output digital ramp: output data increment by one LSB every clock cycle from code 0 to 65535 0110 = Single pattern: output data are a custom pattern 1 (75h and 76h) 0111 Double pattern: output data alternate between custom pattern 1 and custom pattern 2 1000 = Deskew pattern: output data are AAAAh 1001 = SYNC pattern: output data are FFFFh

9.5.13.25 Register 039h (address = 039h), Decimation Filter Page

Figure 159. Register 039h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	USE COMMON TEST PATTERN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 96. Register 039h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	USE COMMON TEST PATTERN	R/W	0h	0 = Each data stream sends test patterns programmed by bits[3:0] of register 37h. 1 = Test patterns are individually programmed for the I and Q stream of each DDC using the TEST PATTERN DDCx y-DATA register bits (where x = 1 or 2 and y = I or Q).

9.5.13.26 Register 03Ah (address = 03Ah), Decimation Filter Page

Figure 160. Register 03Ah

7	6	5	4	3	2	1	0
0	0	0	0	0	0	TEST PAT RES	TP RES EN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h	R/W-0h

Table 97. Register 03Ah Field Descriptions

Bit	Field	Type	Reset	Description
7-2	0	W	0h	Must write 0
1	TEST PAT RES	R/W	0h	Pulsing this bit resets the test pattern. The test pattern reset must be enabled first (bit D0). 0 = Normal operation 1 = Reset the test pattern
0	TP RES EN	R/W	0h	This bit enables the test pattern reset. 0 = Reset disabled 1 = Reset enabled

9.5.14 Power Detector Page

9.5.14.1 Register 000h (address = 000h), Power Detector Page

Figure 161. Register 000h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	PKDET EN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 98. Register 000h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	PKDET EN	R/W	0h	This bit enables the peak power and crossing detector. 0 = Power detector disabled 1 = Power detector enabled

9.5.14.2 Register 001h-002h (address = 001h-002h), Power Detector Page

Figure 162. Register 001h

7	6	5	4	3	2	1	0
BLKPKDET [7:0]
R/W-0h

Figure 163. Register 002h

7	6	5	4	3	2	1	0
BLKPKDET [15:8]
R/W-0h

Table 99. Register 001h-002h Field Descriptions

Bit	Field	Type	Reset	Description
7-0	BLKPKDET	R/W	0h	This register specifies the block length in terms of number of samples (S`) used for peak power computation. Each sample S` is a peak of 8 actual ADC samples. This parameter is a 17-bit value directly in linear scale. In decimation mode, the block length must be a multiple of a divide-by-4 or -6 complex: length = 5 × decimation factor. The divide-by-8 to -32 complex: length = 10 × decimation factor.

9.5.14.3 Register 003h (address = 003h), Power Detector Page

Figure 164. Register 003h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	BLKPKDET[16]
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 100. Register 003h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	BLKPKDET[16]	R/W	0h	This register specifies the block length in terms of number of samples (S`) used for peak power computation. Each sample S` is a peak of 8 actual ADC samples. This parameter is a 17-bit value directly in linear scale. In decimation mode, the block length must be a multiple of a divide-by-4 or -6 complex: length = 5 × decimation factor. The divide-by-8 to -32 complex: length = 10 × decimation factor.

9.5.14.4 Register 007h-00Ah (address = 007h-00Ah), Power Detector Page

Figure 165. Register 007h

7	6	5	4	3	2	1	0
BLKTHHH
R/W-0h

Figure 166. Register 008h

7	6	5	4	3	2	1	0
BLKTHHL
R/W-0h

Figure 167. Register 009h

7	6	5	4	3	2	1	0
BLKTHLH
R/W-0h

Figure 168. Register 00Ah

7	6	5	4	3	2	1	0
BLKTHLL
R/W-0h

Table 101. Register 007h-00Ah Field Descriptions

Bit	Field	Type	Reset	Description
7-0	BLKTHHH BLKTHHL BLKTHLH BLKTHLL	R/W	0h	These registers set the four different thresholds for the hysteresis function threshold values from 0 to 256 (2TH), where 256 is equivalent to the peak amplitude. Example: BLKTHHH is set to –2 dBFS from peak: 10^{(-2 / 20) × 256} = 203, then set 5407h, 5C07h = CBh.

9.5.14.5 Register 00Bh-00Ch (address = 00Bh-00Ch), Power Detector Page

Figure 169. Register 00Bh

7	6	5	4	3	2	1	0
DWELL[7:0]
R/W-0h

Figure 170. Register 00Ch

7	6	5	4	3	2	1	0
DWELL[15:8]
R/W-0h

Table 102. Register 00Bh-00Ch Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DWELL	R/W	0h	DWELL time counter. When the computed block peak crosses the upper thresholds BLKTHHH or BLKTHLH, the peak detector output flags are set. In order to be reset, the computed block peak must remain continuously lower than the lower threshold (BLKTHHL or BLKTHLL) for the period specified by the DWELL value. This threshold is 16 bits, is specified in terms of f_S / 8 clock cycles, and must be set to 0 for the crossing detector. Example: if f_S = 3 GSPS, f_S / 8 = 375 MHz, and DWELL = 0100h then the DWELL time = 2⁹ / 375 MHz = 1.36 µs.

9.5.14.6 Register 00Dh (address = 00Dh), Power Detector Page

Figure 171. Register 00Dh

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	FILT0LPSEL
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 103. Register 00Dh Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	FILT0LPSEL	R/W	0h	This bit selects either the block detector output or 2-bit output as the input to the IIR filter. 0 = Use the output of the high comparators (HH and HL) as the input of the IIR filter 1 = Combine the output of the high (HH and HL) and low (LH and LL) comparators to generate a 3-level input to the IIR filter (–1, 0, 1)

9.5.14.7 Register 00Eh (address = 00Eh), Power Detector Page

Figure 172. Register 00Eh

7	6	5	4	3	2	1	0
0	0	0	0	TIMECONST
W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 104. Register 00Eh Field Descriptions

Bit	Field	Type	Reset	Description
7-4	0	W	0h	Must write 0
3-0	TIMECONST	R/W	0h	These bits set the crossing detector time period for N = 0 to 15 as 2^N × f_S / 8 clock cycles. The maximum time period is 32768 × f_S / 8 clock cycles (approximately 100 µs at 2.6 GSPS).

9.5.14.8 Register 00Fh, 010h-012h, and 016h-019h (address = 00Fh, 010h-012h, and 016h-019h), Power Detector Page

Figure 173. Register 00Fh

7	6	5	4	3	2	1	0
FIL0THH[7:0]
R/W-0h

Figure 174. Register 010h

7	6	5	4	3	2	1	0
FIL0THH[15:8]
R/W-0h

Figure 175. Register 011h

7	6	5	4	3	2	1	0
FIL0THL[7:0]
R/W-0h

Figure 176. Register 012h

7	6	5	4	3	2	1	0
FIL0THL[15:8]
R/W-0h

Figure 177. Register 016h

7	6	5	4	3	2	1	0
FIL1THH[7:0]
R/W-0h

Figure 178. Register 017h

7	6	5	4	3	2	1	0
FIL1THH[15:8]
R/W-0h

Figure 179. Register 018h

7	6	5	4	3	2	1	0
FIL1THL[7:0]
R/W-0h

Figure 180. Register 019h

7	6	5	4	3	2	1	0
FIL1THL[15:8]
R/W-0h

Table 105. Register 00Fh, 010h, 011h, 012h, 016h, 017h, 018h, and 019h Field Descriptions

Bit	Field	Type	Reset	Description
7-0	FIL0THH FIL0THL FIL1THH FIL1THL	R/W	0h	Comparison thresholds for the crossing detector counter. This threshold is 16 bits in 2.14 signed notation. A value of 1 (4000h) corresponds to 100% crossings, a value of 0.125 (0800h) corresponds to 12.5% crossings.

9.5.14.9 Register 013h-01Ah (address = 013h-01Ah), Power Detector Page

Figure 181. Register 013h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	IIR0 2BIT EN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Figure 182. Register 01Ah

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	IIR1 2BIT EN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 106. Register 013h and 01Ah Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	IIR0 2BIT EN IIR1 2BIT EN	R/W	0h	This bit enables 2-bit output format of the IIR0 and IIR1 output comparators. 0 = Selects 1-bit output format 1 = Selects 2-bit output format

9.5.14.10 Register 01Dh-01Eh (address = 01Dh-01Eh), Power Detector Page

Figure 183. Register 01Dh

7	6	5	4	3	2	1	0
DWELLIIR[7:0]
R/W-0h

Figure 184. Register 01Eh

7	6	5	4	3	2	1	0
DWELLIIR[15:8]
R/W-0h

Table 107. Register 01Dh-01Eh Field Descriptions

Bit	Field	Type	Reset	Description
7-0	DWELLIIR	R/W	0h	DWELL time counter for the IIR output comparators. When the IIR filter output crosses the upper thresholds FIL0THH or FIL1THH, the IIR peak detector output flags are set. In order to be reset, the output of the IIR filter must remain continuously lower than the lower threshold (FIL0THL or FIL1THL) for the period specified by the DWELLIIR value. This threshold is 16 bits and is specified in terms of f_S / 8 clock cycles. Example: if f_S = 2.6 GSPS, f_S / 8 = 325 MHz, and DWELLIIR = 0100h, then the DWELL time = 2⁹ / 325 MHz = 1.57 µs.

9.5.14.11 Register 020h (address = 020h), Power Detector Page

Figure 185. Register 020h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	RMSDET EN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 108. Register 020h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	RMSDET EN	R/W	0h	This bit enables the RMS power detector. 0 = Power detector disabled 1 = Power detector enabled

9.5.14.12 Register 021h (address = 021h), Power Detector Page

Figure 186. Register 021h

7	6	5	4	3	2	1	0
0	0	0	PWRDETACCU
W-0h	W-0h	W-0h	R/W-0h

Table 109. Register 021h Field Descriptions

Bit	Field	Type	Reset	Description
7-5	0	W	0h	Must write 0
4-0	PWRDETACCU	R/W	0h	These bits program the block length to be used for RMS power computation. The block length is defined in terms of f_S / 8 clocks and can be programmed as 2M, where M = 0 to 16.

9.5.14.13 Register 022h-025h (address = 022h-025h), Power Detector Page

Figure 187. Register 022h

7	6	5	4	3	2	1	0
PWRDETH[7:0]
R/W-0h

Figure 188. Register 023h

7	6	5	4	3	2	1	0
PWRDETH[15:8]
R/W-0h

Figure 189. Register 024h

7	6	5	4	3	2	1	0
PWRDETL[7:0]
R/W-0h

Figure 190. Register 025h

7	6	5	4	3	2	1	0
PWRDETL[15:8]
R/W-0h

Table 110. Register 022h-025h Field Descriptions

Bit	Field	Type	Reset	Description
7-0	PWRDETH[15:0] PWRDETL[15:0]	R/W	0h	The computed average power is compared against these high and low thresholds. One LSB of the thresholds represents 1 / 2¹⁶. Example: if PWRDETH is set to –14 dBFS from peak, (10^{(–14 / 20)})² × 2¹⁶ = 2609, then set 5422h, 5423h, 5C22h, 5C23h = 0A31h.

9.5.14.14 Register 027h (address = 027h), Power Detector Page

Figure 191. Register 027h

7	6	5	4	3	2	1	0
0	0	0	0	0	0	0	RMS 2BIT EN
W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	W-0h	R/W-0h

Table 111. Register 027h Field Descriptions

Bit	Field	Type	Reset	Description
7-1	0	W	0h	Must write 0
0	RMS 2BIT EN	R/W	0h	This bit enables 2-bit output format on the RMS output comparators. 0 = Selects 1-bit output format 1 = Selects 2-bit output format

9.5.14.15 Register 02Bh (address = 02Bh), Power Detector Page

Figure 192. Register 02Bh

7	6	5	4	3	2	1	0
0	0	0	RESET AGC	0	0	0	0
W-0h	W-0h	W-0h	R/W-0h	W-0h	W-0h	W-0h	W-0h

Table 112. Register 02Bh Field Descriptions

Bit	Field	Type	Reset	Description
7-5	0	W	0h	Must write 0
4	RESET AGC	R/W	0h	After configuration, the AGC module must be reset and then brought out of reset to start operation. 0 = Clear AGC reset 1 = Set AGC reset Example: set 542Bh to 10h and then to 00h.
3-0	0	W	0h	Must write 0

9.5.14.16 Register 037h (address = 037h), Power Detector Page

Figure 193. Register 037h

7	6	5	4	3	2	1	0
0	0	0	0	IODIR GPIO4	IODIR GPIO3	IODIR GPIO2	IODIR GPIO1
W-0h	W-0h	W-0h	W-0h	R/W-0h	R/W-0h	R/W-0h	R/W-0h

Table 113. Register 037h Field Descriptions

Bit	Field	Type	Reset	Description
7-4	0	W	0h	Must write 0
3-0	IODIRGPIO[4:1]	R/W	0h	These bits select the output direction for the GPIO[4:1] pins. 0 = Input (for the NCO control) 1 = Output (for the AGC alarm function)

9.5.14.17 Register 038h (address = 038h), Power Detector Page

Figure 194. Register 038h

7	6	5	4	3	2	1	0
0	0	INSEL1		0	0	INSEL0
W-0h	W-0h	R/W-0h		R/W-0h	R/W-0h	R/W-0h

Table 114. Register 038h Field Descriptions

Bit	Field	Type	Reset	Description
7-6	0	W	0h	Must write 0
5-4	INSEL1	R/W	0h	These bits select which GPIO pin is used for the INSEL1 bit. 00 = GPIO4 01 = GPIO1 10 = GPIO3 11 = GPIO2 See the NCO Switching section for details.
3-2	0	W	0h	Must write 0
1-0	INSEL0	R/W	0h	These bits select which GPIO pin is used for the INSEL0 bit. 00 = GPIO4 01 = GPIO1 10 = GPIO3 11 = GPIO2 See the NCO Switching section for details.

9.5.14.18 Power Detector Page (Direct Addressing, 16-Bit Address, 5400h)

9.5.14.18.1 Register 032h-035h (address = 032h-035h), Power Detector Page

Figure 195. Register 032h

7	6	5	4	3	2	1	0
OUTSEL GPIO1
R/W-0h

Figure 196. Register 033h

7	6	5	4	3	2	1	0
OUTSEL GPIO2
R/W-0h

Figure 197. Register 034h

7	6	5	4	3	2	1	0
OUTSEL GPIO3
R/W-0h

Figure 198. Register 035h

7	6	5	4	3	2	1	0
OUTSEL GPIO4
R/W-0h

Table 115. Register 032h-035h Field Descriptions

Bit	Field	Type	Reset	Description
7-0	OUTSEL GPIOx	R/W	0h	These bits set the function or signal for each GPIO pin. 0 = IIR PK DET0[0] of channel A 1 = IIR PK DET0[1] of channel A (2-bit mode) 2 = IIR PK DET1[0] of channel A 3 = IIR PK DET1[1] of channel A (2-bit mode) 4 = BLKPKDETH of channel A 5 = BLKPKDETL of channel A 6 = PWR Det[0] of channel A 7 = PWR Det[1] of channel A (2-bit mode) 8 = FOVR of channel A 9-17 = Repeat outputs 0-8 but for channel B instead

REGISTER BIT LANE 0	OUTPUT LANE0 CARRIES	REGISTER BIT LANE 1	OUTPUT LANE1 CARRIES	REGISTER BIT LANE 2	OUTPUT LANE2 CARRIES	REGISTER BIT LANE 3	OUTPUT LANE3 CARRIES
00	D0	00	D1	00	D2	00	D3
01	D1	01	D2	01	D3	01	D0
10	D2	10	D3	10	D0	10	D1
11	D3	11	D0	11	D1	11	D2

REGISTER BIT LANE 0	OUTPUT LANE0 CARRIES	REGISTER BIT LANE 1	OUTPUT LANE1 CARRIES	REGISTER BIT LANE 2	OUTPUT LANE2 CARRIES	REGISTER BIT LANE 3	OUTPUT LANE3 CARRIES
00	D0	00	D1	00	D2	00	D3
01	D1	01	D2	01	D3	01	D0
10	D2	10	D3	10	D0	10	D1
11	D3	11	D0	11	D1	11	D2

9 Detailed Description

9.1 Overview

9.2 Functional Block Diagram

9.3 Feature Description

9.3.1 Analog Inputs

9.3.1.1 Input Clamp Circuit

9.3.2 Clock Input

9.3.3 SYSREF Input

9.3.3.1 Using SYSREF

Table 1. Asserting SYSREF

9.3.3.2 Frequency of the SYSREF Signal

Table 2. SYSREF and LMFC Clock Frequency

9.3.4 DDC Block

9.3.4.1 Operating Mode: Receiver

9.3.4.2 Operating Mode: Wide-Bandwidth Observation Receiver

9.3.4.3 Decimation Filters

Table 3. Decimation Filter Summary and Maximum Available Output Bandwidth

9.3.4.3.1 Divide-by-4

9.3.4.3.2 Divide-by-6

9.3.4.3.3 Divide-by-8

9.3.4.3.4 Divide-by-9

9.3.4.3.5 Divide-by-10

9.3.4.3.6 Divide-by-12

9.3.4.3.7 Divide-by-16

9.3.4.4 Digital Multiplexer (MUX)

9.3.4.5 Numerically-Controlled Oscillators (NCOs) and Mixers

9.3.5 NCO Switching

Table 4. NCO Register Configurations

Table 5. GPIO Pin Assignment

Table 6. NCO Selection

9.3.6 SerDes Transmitter Interface

9.3.7 Eye Diagrams

9.3.8 Alarm Outputs: Power Detectors for AGC Support

9.3.8.1 Absolute Peak Power Detector

Table 7. Registers Required for the Peak Power Detector

9.3.8.2 Crossing Detector

Table 8. Registers Required for the Crossing Detector Operation

9.3.8.3 RMS Power Detector

Table 9. Registers Required for Using the RMS Power Detector Feature

9.3.8.4 GPIO AGC MUX

9.3.9 Power-Down Mode

9.3.10 ADC Test Pattern

9.3.10.1 Digital Block

Table 10. Test Pattern Options (Register 37h)

9.3.10.2 Transport Layer

Table 11. Transport Layer Test Mode EN (Register 01h)

9.3.10.3 Link Layer

Table 12. Link Layer Test Mode (Register 03h)

9.4 Device Functional Modes

9.4.1 Device Configuration

9.4.2 JESD204B Interface

9.4.2.1 JESD204B Initial Lane Alignment (ILA)

9.4.2.2 JESD204B Frame Assembly

9.4.2.3 JESD204B Frame Assembly in Bypass Mode

Table 13. JESD Mode Options: Bypass Mode

Table 14. JESD Sample Lane Alignments: Bypass Mode(1)

9.4.2.4 JESD204B Frame Assembly with Decimation (Single-Band DDC): Complex Output

Table 15. JESD Mode Options: Single-Band Complex Output

Table 16. JESD Sample Lane Alignments: Single-Band Complex Output(1)

9.4.2.5 JESD204B Frame Assembly with Decimation (Single-Band DDC): Real Output

Table 17. JESD Mode Options: Single-Band Real Output (Wide Bandwidth)

9.4.2.6 JESD204B Frame Assembly with Decimation (Single-Band DDC): Real Output

Table 18. JESD Mode Options: Single-Band Real Output

Table 19. JESD Sample Lane Assignment: Single-Band Real Output(1)

9.4.2.7 JESD204B Frame Assembly with Decimation (Dual-Band DDC): Complex Output

Table 20. JESD Mode Options: Dual-Band Complex Output

Table 21. JESD Sample Lane Assignment: Dual-Band Complex Output(1)

9.4.2.8 JESD204B Frame Assembly with Decimation (Dual-Band DDC): Real Output

Table 22. JESD Mode Options: Dual-Band Real Output

Table 23. JESD Sample Lane Assignment: Dual-Band Complex Output(1)

9.4.3 Serial Interface

Table 24. SPI Timing Information

Table 25. SPI Input Description

9.4.3.1 Serial Register Write: Analog Bank

9.4.3.2 Serial Register Readout: Analog Bank

9.4.3.3 Serial Register Write: Digital Bank

9.4.3.4 Serial Register Readout: Digital Bank

9.4.3.5 Serial Register Write: Decimation Filter and Power Detector Pages

9.5 Register Maps

Table 26. Register Map

Table 14. JESD Sample Lane Alignments: Bypass Mode⁽¹⁾

Table 16. JESD Sample Lane Alignments: Single-Band Complex Output⁽¹⁾

Table 19. JESD Sample Lane Assignment: Single-Band Real Output⁽¹⁾

Table 21. JESD Sample Lane Assignment: Dual-Band Complex Output⁽¹⁾

Table 23. JESD Sample Lane Assignment: Dual-Band Complex Output⁽¹⁾

REGISTER BIT LANE 0	OUTPUT LANE0 CARRIES	REGISTER BIT LANE 1	OUTPUT LANE1 CARRIES	REGISTER BIT LANE 2	OUTPUT LANE2 CARRIES	REGISTER BIT LANE 3	OUTPUT LANE3 CARRIES
00	D0	00	D1	00	D2	00	D3
01	D1	01	D2	01	D3	01	D0
10	D2	10	D3	10	D0	10	D1
11	D3	11	D0	11	D1	11	D2