cancel
Showing results for
Show  only  | Search instead for
Did you mean:
Contributor
658 Views
Registered: ‎07-03-2018

## Internal "clock" divider latency

Hi everyone,

When I create 2 signals with 10ms period from clock 62.5MHz and 122.88MHz, I saw that the distance between 2 signals is not constant after running a long time. Anyone can explain this issue? I thought that it will remain the distance between 2 signals because it has the same period (10ms)?

Ps: With clock 62.5MHz, I have 625000/2 cycles for high level '1' and 625000/2 cycles for low level '0'

With clock 122.88MHz, I have 1228800/2 cycles for high level '1' and 1228800/2 cycles for low level '0'

Thank you,

Yurivn

1 Solution

Accepted Solutions
Explorer
582 Views
Registered: ‎06-25-2014

How are your 2 clocks generated? If they are from 2 crystals, then you are seeing the ppm differences between them. If using an MMCM then you will probably see the actual frequency is not the same as the requested.

4 Replies
Moderator
592 Views
Registered: ‎08-08-2017

Hi @yurivn

Can you elaborate more on distance between two signals ?  May be you can attach the Simulation or ILA screenshot explaning the same.

If i presume correctly you mean the phase difference between two signals is not constant throughout , Is it the case?

Did your observations based on Simulation or Hardware results ?

-------------------------------------------------------------------------------------------------------------------------------
Reply if you have any queries, give kudos and accept as solution
-------------------------------------------------------------------------------------------------------------------------------
Teacher
585 Views
Registered: ‎07-09-2009

Did you mean the two frequencies to be integer divisions of each other ?

Thinking of the findamental sine wave you have, 122.88, and 62.5 , they are not integer divisions of each other,

If they were 122.88 and 61.44, as used in telecom , then they would stay in sync,

Two different frequencies not integer division related are goign to beet, as you have seen.

<== If this was helpful, please feel free to give Kudos, and close if it answers your question ==>
Explorer
583 Views
Registered: ‎06-25-2014

How are your 2 clocks generated? If they are from 2 crystals, then you are seeing the ppm differences between them. If using an MMCM then you will probably see the actual frequency is not the same as the requested.

Contributor
492 Views
Registered: ‎07-03-2018

A problem is 2 clocks are generated from 2 different clock source, so it makes a different between 2 clocks after a long time running.