Open In App

Clock Skew in synchronous digital circuit systems

Improve
Improve
Like Article
Like
Save
Share
Report

In Synchronous circuits where all the logic elements share the same clock signal, it becomes imperative to design these elements as close to the clock source as possible because a system-on-chip, FPGA, CPLD contain Billions of transistors. Even though these distances are minute due to their sheer number there is a propagation delay which leads to the clock signal arriving at different parts of the chip at different times. This is called Clock Skew. 

In Digital Circuit Design a ” Sequentially Adjacent ” circuit is one where if a pulse emitted from a common source is supposed to arrive at the same time. Using this definition we can write a mathematical expression for clock skew as 

 

Sequentially Adjacent Circuit

 

Non-Sequentially Adjacent Circuit.

Ta(Time of arrival of clock pulse at component a) 
Tb(Time of arrival of clock pulse at component b) 

Then, 
 

Clock skew Ts = Ta - Tb 

Factors causing Clock Skew : 
 

  • Interconnect Length 
     
  • Temperature Variations 
     
  • Capacitive Coupling 
     
  • Material Imperfections 
     
  • Differences in input capacitance on the clock inputs 
     

Types of Clock Skew : 
 

  • Positive Skew – 
    This occurs when the receiving register receives the clock pulse later than it is required. 
     
  • Negative Skew – 
    This occurs when the receiving register receives the clock pulse earlier than required. 
     

 

Types of Clock Skews

 


Last Updated : 22 Dec, 2022
Like Article
Save Article
Previous
Next
Share your thoughts in the comments
Similar Reads