- Ask a related questionWhat is a related question?A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
Tool/software:
Hi Experts, I have the LMK05318B on custom board and it works very well.
XO OCXO 24.576MHz
Ref1 GPS PPS
Ref2 IEEE1588 PPS optional
out1 25MHz
out2 10MHz
out7 PPS
When Ref1 is validated the output starts folowing. In one minute LOFL_DPLL goes low and after a while we have also DPLL phase lock, LOPL_DPLL low.
Tuning word history update HIST bit flashes periodic. R14 now is mostly 0x00 except for HIST bit.
I have adjusted the OCXO +0.1 PPM by purpose to see the difference between holdover and freerun.
But when i interrupt gps reception the LMK always falls to XO ppm offset instead of calculated averaged historical offset.
Expected the LMK holds the locked freq offset as long as no ref input in valid.
What am i doing wrong?
R185 0x00B9F5
R186 0x00BA01
R187 0x00BB00
R188 0x00BC00
R189 0x00BD00
R190 0x00BE00
R252 0x00FC2D
Hi Octo,
Just to clarify, are you setting the OCXO to +0.1ppm during the whole duration of this test? If I'm understanding correctly on the output of the LMK05318B you're seeing:
Initial startup (LOFL = 1): +0.1 ppm
Around a minute later (LOFL = 0): 0 ppm (locked to 1pps input)
History accumulation: 0 ppm (locked to 1pps input)
1pps reference is lost (LOFL = 1): 0.1 ppm (holdover)
It looks like you have intermediate updates enabled so during holdover the device should automatically use the latest free run word. If intermediate updates were not enabled, you would generally need to wait for at least 3 Tavg periods plus the delay timer to ensure that the history is valid (see Figure 9-20 in the datasheet). Do you see any difference with different intermediate update intervals? I can also see if I can replicate this behavior on an EVM early next week.
Regards,
Connor
