Swastik here from Asiczen Technologies Pvt Ltd, Bhubaneswar, India. We are working on a Japanese MNC's project for monitoring precise body temperature with 32 individual thermistors.
We made a custom board for LTC2984 where we used two (2) 16 channel mux to switch between thermistor channels. As of now we used LTC channel 19 as "Sense Resistor" and 20 as "Thermistor Custom Steinhart-Hart". We modified our firmware to switch between thermistors and set coefficient accordingly.
To test our setup, we dip the thermistor nodes into a highly precise temperature bath with a constant temperature. But we are getting a temperature delta of 0.5 degree Celsius to 1.2 degree Celsius.
To debug it more, rather than connecting to a thermistor, we connected a fixed 10K resistor. We are getting a fixed 10K value in the development board. But in our custom board we are getting some resistance between 9.6 K to 9.7 K.
Is there any other way to measure or debug why we are getting less resistance when using a fixed 10K resistors and also is there any existing procedure for LTC to check and eliminate noise in our soldering that we are getting while using a muxed setup?
Thanks and Regards,