LMK05318B-Q1: LMK05318B-Q1: 1PPS support

Hi Esakki,

A 1PPS input is available for PRIREF pin. I am working on datasheet updates to clarify that SECREF should not be used for 1PPS.

The local oscillator must be routed to the XO input pin and must fulfill the frequency and accuracy noted by this spreadsheet:

5078.1pps phase detection calculator.xlsx

A low frequency and high accuracy TCXO (such as 12.8 MHz +/- 2.5 ppm) routed to the XO input ensures the 1PPS phase detector can properly validate the input.

Please refer to this previous post for more details on the phase detector:

https://e2e.ti.com/support/clock-timing-group/clock-timing-internal/f/clock-timing---internal-forum/1500646/lmk05318b-q1-generating-1pps-signal/

Regards,

Jennifer

Hi Esakki,

Sorry, I didn't realize I shared a TI-internal link. Here is the key information I wanted to share:

""

• The validation of the 1PPS input depends primarily on the error between the XO input frequency and the 1PPS input. The 1PPS input signal is validated using a phase validation detector which checks the edge of the input signal. The detector is timed by the clock cycles of the XO input which has a maximum counter value of 63. I have attached a spreadsheet which can help you determine if your XO input and 1PPS input setup is valid by frequency error.
  • (see 1pps calculator spreadsheet)
• Per datasheet, the XO input frequency must have a fractional relationship with the VCO1 frequency to avoid integer boundary spurs (IBS) on the outputs. An XO frequency of 19.2MHz or 12.8MHz is acceptable.
  1. 
• We have done most of our 1PPS testing with 12.8MHz XO input frequency. A lower XO input frequency means a larger XO input frequency error is allowed, as shown in the calculator:
  • 
• A higher XO input frequency means even less frequency error is allowed:
  • 

""

Regards,

Jennifer