欢迎访问ic37.com |
会员登录 免费注册
发布采购

CDK2307CILP64X 参数 Datasheet PDF下载

CDK2307CILP64X图片预览
型号: CDK2307CILP64X
PDF下载: 下载PDF文件 查看货源
内容描述: 双通道, 20/40/ 65 / 80MSPS , 12月13日位模拟数字转换器 [Dual, 20/40/65/80MSPS, 12/13-bit Analog-to-Digital Converters]
分类和应用: 转换器
文件页数/大小: 16 页 / 1142 K
品牌: CADEKA [ CADEKA MICROCIRCUITS LLC. ]
 浏览型号CDK2307CILP64X的Datasheet PDF文件第8页浏览型号CDK2307CILP64X的Datasheet PDF文件第9页浏览型号CDK2307CILP64X的Datasheet PDF文件第10页浏览型号CDK2307CILP64X的Datasheet PDF文件第11页浏览型号CDK2307CILP64X的Datasheet PDF文件第12页浏览型号CDK2307CILP64X的Datasheet PDF文件第14页浏览型号CDK2307CILP64X的Datasheet PDF文件第15页浏览型号CDK2307CILP64X的Datasheet PDF文件第16页  
Data Sheet  
The quality of the input clock is extremely important for The digital outputs can be set in tristate mode by setting  
the OE_N signal high.  
high-speed, high-resolution ADCs. The contribution to SNR  
from clock jitter with a full scale signal at a given frequency  
is shown in equation 1.  
Note that the out of range flags (ORNG) will behave differ-  
ently for 12-bit and 13-bit output. For 13-bit output ORNG  
will be set when digital output data are all ones or all  
zeros. For 12-bit output the ORNG flags will be set when  
all twelve bits are zeros or ones and when the thirteenth  
bit is equal to the rest of the bits.  
SNR  
= 20 log (2 π F εt)  
jitter  
IN  
where F is the signal frequency, and εt is the total rms  
IN  
jitter measured in seconds. The rms jitter is the total of all  
jitter sources including the clock generation circuitry, clock  
distribution and internal ADC circuitry.  
The CDK2307 employs digital offset correction. This means  
that the output code will be 4096 with the positive and  
negative inputs shorted together(zero differential). How-  
ever, small mismatches in parasitics at the input can cause  
this to alter slightly. The offset correction also results in  
possible loss of codes at the edges of the full scale range.  
With “NO” offset correction, the ADC would clip in one  
end before the other, in practice resulting in code loss at  
the opposite end. With the output being centered digitally,  
the output will clip, and the out of range flags will be set,  
before max code is reached. When out of range flags are  
set, the code is forced to all ones for over-range and all  
zeros for under-range.  
For applications where jitter may limit the obtainable per-  
formance, it is of utmost importance to limit the clock jit-  
ter. This can be obtained by using precise and stable clock  
references (e.g. crystal oscillators with good jitter specifi-  
cations) and make sure the clock distribution is well con-  
trolled. It might be advantageous to use analog power and  
ground planes to ensure low noise on the supplies to all  
circuitry in the clock distribution. It is of utmost importance  
to avoid crosstalk between the ADC output bits and the  
clock and between the analog input signal and the clock  
since such crosstalk often results in harmonic distortion.  
Data Format Selection  
The jitter performance is improved with reduced rise and  
fall times of the input clock. Hence, optimum jitter per-  
formance is obtained with LVDS or LVPECL clock with fast  
edges. CMOS and sine wave clock inputs will result in  
slightly degraded jitter performance.  
The output data are presented on offset binary form  
when DFRMT is low (connect to OVSS). Setting DFRMT  
high (connect to OVDD) results in 2’s complement output  
format. Details are shown in Table 1 on page 14.  
If the clock is generated by other circuitry, it should be  
retimed with a low jitter master clock as the last operation  
before it is applied to the ADC clock input.  
The data outputs can be used in three different configurations.  
Normal mode:  
All 13-bits are used. MSB is Dx_12 and LSB is Dx_0. This  
mode gives optimum performance due to reduced quanti-  
zation noise.  
Digital Outputs  
Digital output data are presented in a parallel CMOS form.  
The voltage on the OVDD pin sets the levels of the CMOS  
outputs. The output drivers are dimensioned to drive a  
wide range of loads for OVDD above 2.25V, but it is rec-  
ommended to minimize the load to ensure as low tran-  
sient switching currents and resulting noise as possible. In  
applications with a large fanout or large capacitive loads,  
it is recommended to add external buffers located close to  
the ADC chip.  
12-bit mode:  
The LSB is left unconnected such that only 12 bits are used.  
MSB is Dx_12 and LSB is Dx_1. This mode gives slightly  
reduced performance, due to increased quantization noise.  
Reduced full scale range mode:  
The full scale range is reduced from 2V to 1V which is  
pp  
pp  
equivalent to 6dB gain in the ADC frontend. MSB is Dx_11  
and LSB is Dx_0. Note that the codes will wrap around  
when exceeding the full scale range, and that out of range  
bits should be used to clamp output data. See section  
Reference Voltages for details. This mode gives slightly  
reduced performance.  
The timing is described in the Timing Diagram section.  
Note that the load or equivalent delay on CLK_EXT always  
should be lower than the load on data outputs to ensure  
sufficient timing margins.  
©2009 CADEKA Microcircuits LLC  
www.cadeka.com  
13  
 复制成功!