Hi Rolek, If I understand your question, you want to take two consecutive measurements on the same channel so that you can get longer settling times. I would need to check with the designer to know for certain, but my guess is that it will behave the same as any other combination; it will open and then reclose the same channel in this specific case. Regardless if it is a NOP or temporary break, it will not make much difference in the overall settling times. Most of the charge transfer needed to settle is a result of the MUXout node and the input sampling capacitor of the ADC. If the input voltage is about the same between the two samples, then the first sample will charge all of the input capacitances to very close (ideally to 1/2LSB) to the final value. When the ADC converts the first time, about 10-20% of the charge on the input sampling cap will be lost, and this must be recharged on the next sample. This charge will dominate between the two samples, including any break-make action of the multiplexor. For all of the channel sequencing modes, you have full control over the conversion rate based upon the frequency of /CS. For the channel that you want longer settling time, if you extended the /CS time by keeping this signal low for a longer period, this will directly increase the sampling time and improve overall settling. Either approach should help, but increasing the /CS period for a specific channel will provide the best improvement. Regards, Keith N. Precision ADC Applications
↧