Hello,
We have tested ADC128S102 using a SPI XPRESS Master probe.
It seems that when we select another channel changing the adress, it requieres 3 consecutives reading cycles at the same adress to get a stable measurment.
On the first cycle I send for example the address 1, the data received on this first cycle corresponds to the address 0 (as described in datasheet SNAS298F page 15). On the second cycle whatever the address sent on this cycle, tha data received should correspond to the address 1 sent on the first cycle. However the data we receive does not correspond to the data that we expected. The gap between the expected value and the received value is big.
However if on the second cycle we send again the address 1 so that on the third cycle receive again the same value, then on the third cycle the data received corresponds exactly to the expected data (+/- 1LSB).
Furthermore if we continue sending the address 1 we find that the received data is stable and still correct (+/-1LSB).
What is surprising is that the second reading is far from the expected value. Neither on the datasheet nor on the aplication note, is mentioned that it is necessary to sample twice each address in order to get a correct (stable) value.
Do you confirm that 2 consecutives measurments of a channel are requirered to get stable measurement after an adress change ? If yaes, why?
Nota :
- The ADC128S102 is powered following the figure 39 of the datasheet (VA=3V3), the inputs are fitted with at least 1nF capacitor and 21,5ohm resistor.
- The SPI clock is 12,5MHz and our chronograms are equals to the data sheet one (page 7)
Thank you in advance for your help,
Looking forward to hearing from you,