TX BUFFER

录入:edatop.com    阅读:
设计的TX  BUFFER在电源噪声100M时候,眼图中jitter发生了分叉,如图所示
请教大神,为什么会出现这种分叉,在电源噪声比较小的时候未发生分叉。

Hi singerhaha:
If you only simulate TX buffer and the input data is ideal,
I guess the duty cycle is not ideal.
Could you check the output waveform?
mpig


hi mpig09,    thank you for your answer!But i don’t know the ralationship between my question with your answer.Could you explain more clearly!i am so eager for it
    thank you again.

Hi singerhaha:
Because I don't know your testbench, so I "guess" your add a 100MHz noise on VDD
and input a ideal clock to driver your TX buffer.
If there are multistage in the TX buffer, the noise may effect the output signal of each stage.
1. rising time / falling time will different of each buffer.
2. jitter may obvious on each output stage.
I suggest you to check each stage if the TX buffer has multistage in your design.
If I have any mistake, please let me know.
mpig


hi mpig09,    it's so kind of you!input data is random data,TX BUFFER consists of a predriver and main driver.I have checked both stages. What i cheked is differential signalings.Please let me know anything else should be cared during TX BUFEFER  design?
now the case is when i use a sine  noise,no 分叉!when i use a pulse noise ,分叉 still is there.What shoud i do now?


singerhaha  

jitter issues



   hi eeweiboy :
            could you tell me what helps result in the jitter issue?

nicer

Hi singerhaha:
If the problem is rising / falling time issue, you need to check the root cause at which stage.
==>Input a max. speed clock signal to your TX Buffer, and check which the difference between
rising time and falling time (the Trise - Tfall needs small)
If the problem is jitter, does the random data is ideal?
mpig


hi mpig09 :
             the random data is ideal !it's caused by random_data stream cell.random data's frequency is 1.6G, rising time/fall time is 40p.From this ,any problem?what confusing me is why a sine wave  no 分叉.


singerhaha

Hi singerhaha:
What mean "what confusing me is why a sine wave  no 分叉." ?
Does the waveform result meet your spec?
mpig

Hi singerhaha:
The waveform is p-n result?
Which view tool that you get the waveform?
If you use spice explorer,
you can use Tools --> Jitter vs time to get more detail info.
ex:  long time jitter
      long time duty variation
It will help to know the root cause.
If you  have any problem, please let me know.
mpig



   如果是加了电源噪声才出现分叉,看看是不是电源噪声引入的DJ。


hi bebe00
   
         加sine信号的噪声时是没有出现分叉,而加矩形波形的噪声就出现了分叉,为什么会这样?
         谢谢
singerhaha

申明:网友回复良莠不齐,仅供参考。如需专业解答,请学习本站推出的微波射频专业培训课程

上一篇:问下MOS器件栅极为什么用多晶硅而不直接用金属呢
下一篇:PFD的一些基础问题

射频和天线工程师培训课程详情>>

  网站地图