Hi Everyone,
in our current application we use the AD7790 to sample NTC10K temperature sensors. The reference is connected to VDD (3.3 V) wich has about 20 mV ripple.
The datasheet states:
In applications where the excitation (voltage or current) for the transducer on the analog input also drives the reference voltage for the part, the effect of the low frequency noise in the excitation source will be removed because the application is ratiometric.
Since this is the case I didn't worry about the ripple. However the temperature reading is off by ~ 1K sometimes. Better input filtering seems to fix the issue on first glance.
Is this plausible to you? Can you provide a transferfunction containing the reference voltage?
Best wishes
Blobbi

