1
\$\begingroup\$

I am not a very experienced firmware developer, so I came across this doubt when I was reading the datasheet of a microcontroller:

In my application, I will use a crystal to generate a clock signal of 26 MHz, the same frequency I want in my system's clock bus. Reading the section that explain about the PLL module, I found the table below:

Table from the manufacturer datasheet

In the first line, it is shown a use case where the output frequency is the same as the input frequency.

So, my question is: Is there any advantage to make use of a PLL module with frequency gain equals to one, as my source clock is already in the desired frequency?

\$\endgroup\$

2 Answers 2

2
\$\begingroup\$

The use case for that would be for the PLL cleaning up the clock signal, i.e. lowering jitter. If you already have a good clock source, then there is no real benefit.

\$\endgroup\$
1
  • 1
    \$\begingroup\$ If your XTAL oscillator environment is so noisy the jitter needs to be run thru a PLL (on the same chip) for zero-crossing cleanup, then I suggest you not bother. Clean up the environment instead. \$\endgroup\$ Commented Jan 26, 2018 at 5:21
0
\$\begingroup\$

Is there any advantage to make use of a PLL module with frequency gain equals to one

Maybe not in your circuit but if you want a clock signal that is in quadrature to an existing clock signal then you can use a PLL. Even if the existing clock frequency moved +/- 20% (or more), the PLL clock would remain at 90 degrees to the original. I am describing a type I phase detecting PLL: -

enter image description here

An EXOR is a type I phase detector and if used within a PLL loop it would naturally align the output to be mid-range on the detector i.e. \$\pi/2\$ or 90 degrees. Quite useful in some radio receivers.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.