Analog Devices AD9233 Evaluation Board AD9233-80EBZ AD9233-80EBZ Fiche De Données

Codes de produits
AD9233-80EBZ
Page de 44
   
AD9233
 
Rev. A | Page 19 of 44 
A third option is to ac-couple a differential LVDS signal to the 
sample clock input pins, as shown in Figure 48. The 
/
 family of clock 
drivers offers excellent jitter performance.  
054
92
-050
100Ω
0.1µF
0.1µF
0.1µF
0.1µF
50Ω*
LVDS DRIVER
50Ω*
CLK
CLK
*50Ω RESISTORS ARE OPTIONAL
CLK–
CLK+
ADC
AD9233
CLOCK
INPUT
CLOCK
INPUT
AD951x
 
Figure 48. Differential LVDS Sample Clock 
In some applications, it is acceptable to drive the sample clock 
inputs with a single-ended CMOS signal. In such applications, 
directly drive CLK+ from a CMOS gate, while bypassing the 
CLK− pin to ground with a 0.1 μF capacitor. Although the 
CLK+ input circuit supply is AVDD (1.8 V), this input is 
designed to withstand input voltages up to 3.6 V, making the 
selection of the drive logic voltage very flexible. When driving 
CLK+ with a 1.8 V CMOS signal, it is required to bias the  
CLK− pin with a 0.1 μF capacitor in parallel with a 39 kΩ 
resistor (see Figure 49). The 39 kΩ resistor is not required when 
driving CLK+ with a 3.3 V CMOS signal (see Figure 50). 
0549
2-
05
1
CLOCK
INPUT
0.1µF
0.1µF
0.1µF
39
AD951x
CMOS DRIVER
50
Ω
*
OPTIONAL
100Ω
*50
Ω RESISTOR IS OPTIONAL
CLK–
CLK+
ADC
AD9233
VCC
1kΩ
1kΩ
 
Figure 49. Single-Ended 1.8 V CMOS Sample Clock 
054
92-
052
CLOCK
INPUT
0.1µF
0.1µF
0.1µF
VCC
AD951x
CMOS DRIVER
50
Ω*
OPTIONAL
100Ω
*50
Ω RESISTOR IS OPTIONAL
CLK–
CLK+
ADC
AD9233
1kΩ
1kΩ
 
Figure 50. Single-Ended 3.3 V CMOS Sample Clock 
Clock Duty Cycle 
Typical high speed ADCs use both clock edges to generate a 
variety of internal timing signals. As a result, these ADCs may 
be sensitive to clock duty cycle. Commonly, a ±5% tolerance is 
required on the clock duty cycle to maintain dynamic perform-
ance characteristics.  
The AD9233 contains a DCS that retimes the nonsampling, or 
falling edge, providing an internal clock signal with a nominal 
50% duty cycle. This allows a wide range of clock input duty 
cycles without affecting the performance of the AD9233. Noise 
and distortion performance are nearly flat for a wide range of 
duty cycles when the DCS is on, as shown in Figure 31.  
Jitter in the rising edge of the input is still of paramount 
concern and is not reduced by the internal stabilization circuit. 
The duty cycle control loop does not function for clock rates 
less than 20 MHz nominally. The loop has a time constant 
associated with it that needs to be considered in applications 
where the clock rate can change dynamically, which requires a 
wait time of 1.5 μs to 5 μs after a dynamic clock frequency 
increase (or decrease) before the DCS loop is relocked to the 
input signal. During the time the loop is not locked, the DCS 
loop is bypassed, and the internal device timing is dependant 
on the duty cycle of the input clock signal. In such an application, 
it can be appropriate to disable the duty cycle stabilizer. In all 
other applications, enabling the DCS circuit is recommended to 
maximize ac performance.  
The DCS can be enabled or disabled by setting the SDIO/DCS 
pin when operating in the external pin mode (see Table 10), or 
via the SPI, as described in the Table 15
 
Table 10. Mode Selection (External Pin Mode) 
Voltage at Pin 
SCLK/DFS 
SDIO/DCS 
AGND 
Binary (default) 
DCS disabled 
AVDD Twos 
complement 
DCS enabled (default)  
 
JITTER CONSIDERATIONS 
High speed, high resolution ADCs are sensitive to the quality of 
the clock input. The degradation in SNR at a given input 
frequency (F
IN
) due to jitter (t
J
) is calculated as 
SNR = −20 log (2π × F
IN
 × t
J
In the equation, the rms aperture jitter (t
J
) represents the root-
mean-square of all jitter sources, which include the clock input, 
analog input signal, and ADC aperture jitter specification. IF 
undersampling applications are particularly sensitive to jitter, as 
shown in Figure 51. 
70
65
60
55
50
45
40
1
10
100
1000
05
49
2-
04
6
S
NR (
d
Bc)
INPUT FREQUENCY (MHz)
3.00ps
0.05ps
MEASURED
PERFORMANCE
0.20ps
0.5ps
1.0ps
1.50ps
2.00ps
2.50ps
 
Figure 51. SNR vs. Input Frequency and Jitter