I am working in a board with a ADC10D1500 and the supply is designed based on "ADC0xD1520RB Reference Board".
THE ADC requires 1.9V that are generated from a 12V voltage supply. In a first step ~2.25v are generated from the 12V using a LM25576 and, afterthat, the 1.9V are generated by a LP38513 regulator.
The fact is that I measure a distortion at the output of the ADC at a frequency ~300KHz that matchs with the frequency of the LM25576. I have measured again the SINAD of the ADC with an external voltage source driving the linear regulator and that noise seems to be removed (switching regulator was still working but its output was not used as input for the linar regulator).
C194 seems to be a very low impedance for a 300KHz noise. Any idea about how to reduce the ripple or improve the isolation? Maybe test with a different choke (L2)?
I have been checking the datasheet of the LM25576 and I see that the value chosen for L1 (figure 9 of the datasheet, L10 in the attached schematic) , 10uH, gives a ripple in the current ~0.65App (equation 6 in the datasheet ) while it is stated that it should be "less than twice the minimum load current or 0.5App to ensure CCM operation". Would it be a good idea using a larger value of L here (and, hence, a larger Cramp too) in order to recuce the ripple and the coupled noise to the ADC? Is there any drawback to this modification?
Thanks in advance,
Sergio
