Hi Akash, After looking further into this issue, the sawtooth waveform seems to be generated by a combination of two things here: [1] source resistance higher than the limit combined with [2] a decreasing input voltage where the internal sampling capacitor has been charged up to a higher voltage from the previous sample than the input voltage at the time of the present sample. When the the internal switch is turned on for the sampling time, the voltage measured at the input starts to rise, which makes sense theoretically (since the internal voltage would be higher than the input voltage, and current would start flowing outward from a higher potential to a lower potential). If the source resistance is low enough, I could see where additional drive strength would prevent this scenario from happening or at least minimize it. Regarding the source resistance (Rs) limit, it's mentioned in the app note mentioned above, but surprisingly not in the datasheet. I've requested that this limit (here, it's Rs < 10k Ohms for the minimum sample time equation) be included in the datasheet as well. Regards, James
↧
