One slight mistake I noticed is the temperature slope, which depends on current. At 0.1mA, -2mV/C is about right, but at 10mA, it’s closer to 1.5mV. Vishay’s datasheet shows this, as does a recent HackADay post:

https://hackaday.com/2018/04/16/two-cent-temperature-sensors/ ]]>

https://www.seventransistorlabs.com/Images/CD4001_TR.ckt

It uses this model,

https://people.rit.edu/lffeee/RIT_Models_For_LTSPICE.txt

although curiously I found it necessary to swap the NRD/NRS parameters for P/N to get the right characteristic.

The transistor model gives good enough agreement with a typical part (well within the spread of normal parts), and the ESD diodes are even modeled based on a real part.

]]>