• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 30 subscribers
  • Views 472 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM4F212H5QD - crystal versus external clock - spread spectrum

KyleH
Prodigy 160 points

I have a question regarding the main oscillator. We are planning to use a 16MHz crystal to generate TI clock, which is the same frequency as the crystal on the TI Eval board. However, we would like the ability to spread (using SSCG) this clock (we have an external part for doing this). One idea is to have a schematic/layout that supports either having the crystal or the external clock generator chip connect to the TI part via resistor pops/de-pops. If we used the external clock generator, we may want to spread the clock as much as 3%. 

 What are your thoughts regarding both the clock frequency and spread range as well as having a a non-standard PCB layout with regards to the crystal? I know there is a TI Stellaris app note with suggestions about the layout of the crystal circuit. How much can we safely deviate from this. Is it just a general suggestion, or is the Stellaris part particularly sensitive in this regard.

  • 0
    Guru 117855 points

    Fully admitted SWAG here: suggest consideration of 8MHz or lower freq. xtal - rather than your proposed 16MHz) as StellarisWare function, "SysCtlClockSet()" depends upon the "match" between your board's xtal and that specified w/in this function's xtal parameter.  (i.e. "SYSCTL_XTAL_8MHz").  My guess/hope for you is that the "hop" will be physically less (freq wise - not percentage wise) thru use of the lower freq. xtal - thus may not throw the MCU's timing elements/chains into disarray.  

    Suspect that you may be on the "ragged edge" of maker's knowledge/data bank - on this one...  Should yield to experimentation/tweaking - although you really should test over full spread: voltage/temp/xtal freqs...  (re: spread - no pun intended...)