On page 19 of the ADAU1701 specification it describes the hardware components values required for proper crystal operation. The crystal load capacitance is recommended to be 18pF. The calculation for the load capacitance is shown to be the series combination of C1 & C2 plus the stray capacitance, where the stray capacitance is 2pF - 5pF. C1 & C2 normally are of the same value so per my calculations a 33pF would produce a C-load of 16.5pF + C-stray; or about 18.5pF if C-stray is small. If C-stray is 5pF then the values for C1& C2 should be 24pF, which is not a standard value but I would likely choose a 27pF. The application drawing shows a 22pF. Can you explain the reasoning behind this component value selection? When we had C1 & C2 = 22pF the crystal was a little low in frequency so we were able to get the crystal on the correct frequency by using a 33pF.
Another item a little different from the evaluation board schematic is the crystal type we are using is an AT-49 type from Abricom (ABLS-12.288MHZ-B4-T). The evaluation board schematic is using a ABM3B-12.288MHZ-10-1-U-T which is an ABM3B series and has a drive level specification of 10uW TYP and 100uW max. The AT-49 crystal we are using has all the exact specs except for a higher drive level: 100uW TYP and 1mW max. Are there any other precautions we would need to take in order to assure proper crystal starup? Would the series resistor need to be lowered to allow for a higher drive level?