background image

AD9776A/AD9778A/AD9779A 

 

 

Rev. B | Page 48 of 56 

OPTIMIZING THE DATA INPUT TIMING 

The AD9776A/AD9778A/AD9779A have on-chip circuitry that 
enables the user to optimize the input data timing by adjusting 
the relationship between the DATACLK output and DCLK_SMP 
(the internal clock that samples the input data). This optimization 
is made by a sequence of 3-wire interface register read and write 
operations. The timing optimization can be done under strict 
control of the user, or the device can be programmed to maintain a 
configurable timing margin automatically. This function is only 
available when the input data is referenced to the DATACLK 
output. Each of these methods is detailed in the following section. 

Figure 86 shows the circuitry that detects sample timing errors 
and adjusts the data interface timing. The DCLK_SMP signal is 
the internal clock used to latch the input data. Ultimately, it is 
the rising edge of this signal that needs to be centered in the 
valid sampling period of the input data. This is accomplished by 
adjusting the time delay, t

D

, which changes the DATACLK 

timing and, as a result, the arrival time of the input data with 
respect to DCLK_SMP. 

TIMING
ERROR IRQ

D

Q

Q

D

CLK

CLK

DCLK_SMP

PD1[0]

Δ

t

M

Δ

t

D

Δ

t

M

DATACLK
DELAY[3:0]

TIMING
MARGIN[3:0]

DATACLK

TIMING
ERROR TYPE

TIMING

ERROR

DETECTION

06

45

2-

4

02

 

Figure 86. Timing Error Detection and Optimization Circuitry 

The error detect circuitry works by creating two sets of sampled 
data (referred to as the margin test data) in addition to the 
actual sampled data used in the device data path. One set of 
sampled data is latched before the actual data sampling point. 
The other set of sampled data is latched after the actual data 
sampling point. If the margin test data match the actual data, 
the sampling is considered valid and no error is declared. If 
there is a mismatch between the actual data and the margin test 
data, an error is declared.  

The Data Timing Margin[3:0] variable determines how much 
before and after the actual data sampling point the margin test 
data are latched. Therefore, the data timing margin variable 
determines how much setup and hold margin the interface 
needs for the data timing error IRQ to remain inactive (show 
error free operation). Therefore, the timing error IRQ is set 
whenever the setup and hold margins drop below the Data 
Timing Margin[3:0] value and does not necessarily indicate that 
the data latched into the device is incorrect.  

In addition to setting the data timing error IRQ, the data timing 
error type bit is indicated when an error occurs. The data timing 
error type bit is set low to indicate a hold error and high to 
indicate a setup error. Figure 87 shows a timing diagram of the 
data interface and the status of the data timing error type bit.  

Δ

t

M

Δ

t

M

DATA

TIMING ERROR = 0

Δ

t

M

Δ

t

M

DATA

TIMING ERROR = 1
DATA TIMING ERROR TYPE = 1

Δ

t

M

Δ

t

M

DATA

DELAYED

DATA

SAMPLING

ACTUAL

SAMPLING

INSTANT

DELAYED

CLOCK

SAMPLING

TIMING ERROR = 1
DATA TIMING ERROR TYPE = 0

06

45

2-

4

03

 

Figure 87. Timing Diagram of Margin Test Data 

Automatic Timing Optimization 

When automatic timing optimization mode is enabled  
(Register 0x03, Bit 7 = 1), the device continuously monitors  
the data timing error IRQ and data timing error type bits. The 
DATACLK Delay[3:0] is increased if a setup error is detected and 
decreased if a hold error is detected. The value of the DATACLK 
Delay[3:0] setting currently in use can be read back by the user.  

Manual Timing Optimization 

When the device is operating in manual timing optimization 
mode (Register 0x03, Bit 7 = 0), the device does not alter the 
DATACLK Delay[3:0] value from what is programmed by the 
user. By default, the DATACLK delay enable bit is inactive. This 
bit must be set high for the DATACLK Delay[3:0] value to be 
realized. The delay (in absolute time) when programming 
DATACLK delay between 00000 and 11111 varies from about 
700 ps to about 6.5 ns. The typical delays per increment over 
temperature are shown in Table 29

Table 29. Data Delay Line Typical Delays Over Temperature 

Delay 

−40°C +25°C +85°C Unit 

Zero Code Delay (Delay Upon 

Enabling Delay Line) 

630 700 740 ps 

Average Unit Delay  

175 

190 

210 

ps 

 

When the device is placed into manual mode, the error 
checking logic is activated. If the IRQs are enabled, an interrupt 
is generated if a setup/hold violation is detected. One error 
check operation is performed per device configuration. Any 
change to the Data Timing Margin[3:0] or DATACLK 
Delay[3:0] values triggers a new error check operation. 

 

Summary of Contents for AD9776A

Page 1: ...TION The AD9776A AD9778A AD9779A are dual 12 14 16 bit high dynamic range digital to analog converters DACs that provide a sample rate of 1 GSPS permitting a multicarrier generation up to the Nyquist...

Page 2: ...SB Transfers 27 3 Wire Interface Register Map 28 Interpolation Filter Architecture 33 Interpolation Filter Bandwidth Limits 37 Inverse Sinc Filter 38 Sourcing the DAC Sample Clock 39 Direct Clocking 3...

Page 3: ...alibration with Memory Section 41 Changes to Set and Forget Device Option Section 41 Added Table 26 41 Changes to Internal Reference Section 43 Changed Transmit Path Gain and Offset Correction Heading...

Page 4: ...DAC REFCLK REFCLK OUT1_P OUT1_N AUX1_P AUX1_N AUX2_P AUX2_N OUT2_P OUT2_N GAIN GAIN GAIN GAIN 16 BIT Q DAC 2 SINC 1 I LATCH DELAY LINE Q LATCH P2D 15 0 P1D 15 0 SYNC_O SYNC_I DATACLK 2 2 2 n fDAC 8 n...

Page 5: ...998 1 998 1 998 1 998 1 998 1 998 mA Output Compliance Range Source 0 1 6 0 1 6 0 1 6 V Output Compliance Range Sink 0 8 1 6 0 8 1 6 0 8 1 6 V Output Resistance 1 1 1 M Auxiliary DAC Monotonicity Guar...

Page 6: ...IDTH 100 100 mV Input Differential Hysteresis VIDTHH VIDTHL 20 mV Receiver Differential Input Impedance RIN 80 120 LVDS Input Rate Additional limits on fSYNC_I apply see description of Register 0x05 B...

Page 7: ...modulation 297 DACCLK cycles Inverse Sync 18 DACCLK cycles 3 WIRE INTERFACE Maximum Clock Rate SCLK 40 MHz Minimum Pulse Width High tPWH 12 5 ns Minimum Pulse Width Low tPWL 12 5 ns Setup Time tDS SD...

Page 8: ...LATION DISTORTION IMD fDAC 200 MSPS fOUT 50 MHz 87 87 91 dBc fDAC 400 MSPS fOUT 60 MHz 80 85 85 dBc fDAC 400 MSPS fOUT 80 MHz 75 81 81 dBc fDAC 800 MSPS fOUT 100 MHz 75 80 81 dBc NOISE SPECTRAL DENSIT...

Page 9: ...3 V to 0 3 V I120 VREF IPTAT AGND 0 3 V to AVDD33 0 3 V OUT1_P OUT1_N OUT2_P OUT2_N AUX1_P AUX1_N AUX2_P AUX2_N AGND 1 0 V to AVDD33 0 3 V P1D 15 0 P2D 15 0 DGND 0 3 V to DVDD33 0 3 V DATACLK TXENABL...

Page 10: ...DOMAIN NC NO CONNECT 06452 002 NOTES 1 FOR OPTIMAL THERMAL PERFORMANCE THE EXPOSED PAD SHOULD BE SOLDERED TO THE GROUND PLANE FOR THE 100 LEAD THERMALLY ENHANCED TQFP PACKAGE Figure 3 AD9776A Pin Conf...

Page 11: ...SDIO 3 Wire Interface Port Data Input Output 68 SCLK 3 Wire Interface Port Clock Pin No Mnemonic Description 69 CSB 3 Wire Interface Port Chip Select Bar 70 RESET Reset Active High 71 IRQ Interrupt R...

Page 12: ...THE EXPOSED PAD SHOULD BE SOLDERED TO THE GROUND PLANE FOR THE 100 LEAD THERMALLY ENHANCED TQFP PACKAGE Figure 4 AD9778A Pin Configuration Table 8 AD9778A Pin Function Descriptions Pin No Mnemonic Des...

Page 13: ...Data Input Output 68 SCLK 3 Wire Interface Port Clock 69 CSB 3 Wire Interface Port Chip Select Bar 70 RESET Reset Active High Pin No Mnemonic Description 71 IRQ Interrupt Request 72 AGND Analog Groun...

Page 14: ...SHOULD BE SOLDERED TO THE GROUND PLANE FOR THE 100 LEAD THERMALLY ENHANCED TQFP PACKAGE Figure 5 AD9779A Pin Configuration Table 9 AD9779A Pin Function Descriptions Pin No Mnemonic Description Pin No...

Page 15: ...Interface Port Data Input Output 68 SCLK 3 Wire Interface Port Clock 69 CSB 3 Wire Interface Port Chip Select Bar 70 RESET Reset Active High Pin No Mnemonic Description 71 IRQ Interrupt Request 72 AGN...

Page 16: ...ATA 200MSPS fDATA 250MSPS 06452 007 Figure 8 AD9779A In Band SFDR vs fOUT 1 Interpolation 100 50 0 100 fOUT MHz SFDR dBc 90 80 70 60 20 40 60 80 fDATA 160MSPS fDATA 200MSPS fDATA 250MSPS 06452 008 Fig...

Page 17: ...fOUT MHz SFDR dBc 90 80 70 60 10 20 30 40 fDATA 50MSPS fDATA 100MSPS fDATA 125MSPS 06452 013 Figure 14 AD9779A Out of Band SFDR vs fOUT 8 Interpolation 100 50 0 40 fOUT MHz SFDR dBc 90 80 70 60 10 20...

Page 18: ...TA 200MSPS fDATA 100MSPS 06452 019 Figure 20 AD9779A Third Order IMD vs fOUT 4 Interpolation fOUT MHz IMD dBc fDATA 75MSPS fDATA 125MSPS fDATA 100MSPS 90 100 80 70 60 50 450 425 400 375 350 325 300 27...

Page 19: ...TEN 20dB EXT REF DC COUPLED 06452 023 Figure 26 AD9779A Single Tone 4 Interpolation fDATA 100 MSPS fOUT 30 MHz STOP 400 0MHz SWEEP 1 203s 601 pts VBW 20kHz START 1 0MHz RES BW 20kHz REF 0dBm PEAK Log...

Page 20: ...22 88 MSPS fDAC 4 Modulation 55 90 85 80 75 70 65 60 0 260 240 220 200 180 160 140 120 100 80 60 40 20 fOUT MHz ACLR dBc 06452 301 3dBFS PLL DISABLED 6dBFS PLL DISABLED 0dBFS PLL ENABLED 0dBFS PLL DIS...

Page 21: ...52 036 Figure 39 AD9778A In Band SFDR vs fOUT 2 Interpolation 90 0 250 fOUT MHz ACLR dBc 70 80 60 25 50 75 100 125 150 175 200 225 FIRST ADJACENT CHANNEL SECOND ADJACENT CHANNEL THIRD ADJACENT CHANNEL...

Page 22: ...at 6 dBFS fDATA 200 MSPS 0 4 0 4096 CODE INL 12 BIT LSB 0 4 512 1024 2560 2048 1536 3072 3584 0 3 0 2 0 1 0 0 1 0 2 0 3 06452 041 Figure 44 AD9776A Typical INL 0 20 0 4096 CODE DNL 12 BIT LSB 2048 0...

Page 23: ...0 78 05 77 73 dBm 87 67 90 73 90 41 LOWER dBc 75 30 77 99 77 50 dBm 87 97 90 66 90 17 UPPER REF 25 29dBm AVG log 10dB PAVG 10 W1 S2 ATTEN 4dB 06452 046 Figure 49 AD9776A Single Carrier W CDMA 4 Interp...

Page 24: ...tart of the output transition In Band Spurious Free Dynamic Range SFDR In band SFDR is the difference in decibels between the peak amplitude of the output signal and the peak spurious signal between d...

Page 25: ...D9776A AD9778A AD9779A PLL are wider than those for the AD9776 AD9778 AD9779 This means that the AD9776A AD9778A AD9779A PLL remains in lock in a given range over a wider temperature range than the AD...

Page 26: ...ransfer is a read or write the number of bytes in the data transfer and the starting register address for the first byte of the data transfer The first eight SCLK rising edges of each communication cy...

Page 27: ...SB Multibyte data transfers in MSB first format start with an instruction byte that includes the register address of the most significant data byte Subsequent data bytes should follow from high addres...

Page 28: ...YNC_I enable SYNC_O enable SYNC_O triggering edge Clock State 4 0 0x00 PLL Control 0x08 08 PLL Band Select 5 0 PLL VCO Drive 1 0 0xE7 0x09 09 PLL enable PLL VCO Divide Ratio 1 0 PLL Loop Divide Ratio...

Page 29: ...Table 19 for filter modes 0000 0x01 1 DATACLK Delay 4 Sets MSB of delay of REFCLK input to DATACLK output 0 0x01 0 Zero stuffing enable 0 zero stuffing off 1 zero stuffing on 0 0x02 7 Data format 0 t...

Page 30: ...the value of the delay line of the SYNC_O signal The delay of SYNC_O is relative to REFCLK The delay line resolution is 80 ps per step 00000 nominal delay 00001 adds 80 ps delay to SYNC_O 00010 adds 1...

Page 31: ...oltage at VCO control voltage input readback only A value of 011 indicates the VCO centered in its frequency range 000 0x0A 4 0 PLL Loop Bandwidth 4 0 Controls the bandwidth of the PLL filter Increasi...

Page 32: ...timing violation occurred on the input data port The IRQ is latched This bit is cleared when the Interrupt register is read 0 0x19 6 Sync timing error IRQ Read only Active high indicates a timing viol...

Page 33: ...H 45 245 H 12 H 44 0 H 13 H 43 408 H 14 H 42 0 H 15 H 41 650 H 16 H 40 0 H 17 H 39 1003 H 18 H 38 0 H 19 H 37 1521 H 20 H 36 0 H 21 H 35 2315 H 22 H 34 0 H 23 H 33 3671 H 24 H 32 0 H 25 H 31 6642 H 26...

Page 34: ...n be seen in Figure 60 4 8 3 6 2 4 1 2 DC 1 1 3 2 5 3 7 7 5 3 1 2 4 6 4 0 8 06452 086 Figure 60 Nyquist Zones Figure 57 Figure 58 and Figure 59 show the low pass response of the digital filters with n...

Page 35: ...filter response In dual channel mode the devices expect the real and imaginary components of a complex signal at Digital Input Port 1 and Digital Input Port 2 I and Q respectively The DAC outputs the...

Page 36: ...AC 2 8 0 55 0 5 0 45 8 0x09 fDAC 2 shifted 7 0 4875 0 4375 0 3875 8 0x0A 3fDAC 8 6 0 425 0 375 0 343 8 0x0B 3fDAC 8 shifted 5 0 3625 0 3125 0 2625 8 0x0C fDAC 4 4 0 3 0 25 0 2 8 0x0D fDAC 4 shifted 3...

Page 37: ...HIFTED f DAC 8 SHIFTED f DAC 4 SHIFTED 3 f DAC 8 06452 087 Figure 70 Shifted Bandwidths Accessible with the Filter Architecture With this filter architecture a signal placed anywhere in the spectrum i...

Page 38: ...by setting the inverse sinc enable bit Bit 3 in Register 0x02 Mixing Sequence fDAC 2 I I I I I 0 0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 0 4 5 0 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 45 0 50 f fSAMPLE dB 0...

Page 39: ...Figure 72 Clock Multiplier Circuit The clock multiplier circuit operates such that the VCO outputs a frequency fVCO equal to the REFCLK input signal frequency multiplied by N1 N2 N2 N1 f f REFCLK VCO...

Page 40: ...658 100111 39 1564 1639 100110 38 1555 1606 100101 37 1521 1600 100100 36 1514 1575 100011 35 1480 1553 100010 34 1475 1529 100001 33 1439 1505 100000 32 1435 1489 PLL Lock Ranges Over Temperature 40...

Page 41: ...Band Is in the Higher Range 32 to 62 If System Startup Temperature Is Set PLL Band as Follows 40 C to 30 C Set PLL band readback band 3 30 C to 10 C Set PLL band readback band 2 10 C to 15 C Set PLL...

Page 42: ...ve Circuit If a clean sine clock is available it can be transformer coupled to REFCLK as shown in Figure 73 Use of a CMOS or TTL clock is also acceptable for lower sample rates It can be routed throug...

Page 43: ...a DAC output full scale current of 20 mA Because the gain error is a linear function of this resistor a high precision resistor improves gain matching to the internal matching specification of the dev...

Page 44: ...current IREFERENCE to the auxiliary DAC reference current is 16 67 mA with the auxiliary DAC gain set to full scale 10 bit values 3 wire interface Register 0x0D and 3 wire interface Register 0x11 Thi...

Page 45: ...ther auxiliary DAC It may take practice before an effective algorithm is achieved Using the AD9776A AD9778A AD9779A evaluation board the LO feedthrough can typically be adjusted down to the noise floo...

Page 46: ...e data for both DACs is received on the Port 1 input bus P1D 15 0 I and Q data samples are interleaved and are sampled on the rising edges of DATACLK Along with the data a framing signal must be suppl...

Page 47: ...fer into the device 06452 309 tSREFCLK tHREFCLK tS_SYNC tH_SYNC SYNC_I REFCLK DATA Figure 84 Input Data Port Timing Data Referenced to REFCLK fDACCLK fREFCLK Note that even though the setup and hold t...

Page 48: ...rgin the interface needs for the data timing error IRQ to remain inactive show error free operation Therefore the timing error IRQ is set whenever the setup and hold margins drop below the Data Timing...

Page 49: ...to suppress outgoing pulses if the incoming SYNC_I frequency is greater than DACCLK 32 Extra pulses can be suppressed by the ratios listed in Table 30 The SYNC_I frequency can be lower than DACCLK 32...

Page 50: ...the REFCLK signal SYNC_I is sampled at the DACCLK rate This means that the rising edge of the SYNC_I signal must occur after the hold time of the preceding DACCLK rising edge not the preceding REFCLK...

Page 51: ...250 fDATA MSPS POWER W 0 08 25 50 75 100 125 150 175 200 225 8 INTERPOLATION 4 INTERPOLATION 2 INTERPOLATION 1 INTERPOLATION 0 06 0 04 0 02 06452 079 Figure 93 Power Dissipation Clock 1 8 V Supply I...

Page 52: ...TxDACs can be placed in sleep or power down mode In sleep mode the TxDAC output is turned off thus reducing power dissipation The reference remains powered on however so that recovery from sleep mode...

Page 53: ...k The spectral purity of the clock directly affects the device per formance A low noise low jitter clock source is required All necessary connections to the evaluation board are shown in more detail i...

Page 54: ...software window is shown in Figure 102 The evaluation board also comes populated with the ADL537x modulator to allow for the evaluation of an RF subsystem Complete details on the evaluation board and...

Page 55: ...TQFP_EP SV 100 1 Dimensions shown in millimeters ORDERING GUIDE Model Temperature Range Package Description Package Option AD9776ABSVZ1 40 C to 85 C 100 Lead Thin Quad Flat Package Exposed Pad TQFP_E...

Page 56: ...AD9776A AD9778A AD9779A Rev B Page 56 of 56 NOTES 2007 2008 Analog Devices Inc All rights reserved Trademarks and registered trademarks are the property of their respective owners D06452 0 9 08 B...

Reviews: