It would seem to me the function may not be the right approach. The reason I suspect this is that if you are going to build more than one copy of this product then each will have its own sensor. The table you showed in the first post of this thread is probably applicable to the first sensor that you have in hand. The second, third, fourth sensor .... would most likely have separate tables. (After all this is why the vendor of the sensor provided the "calibration table" in the first place. Right ??)

The tabular approach lends itself to an automation process of setting up each system as it is being built or installed/calibrated. (Hopefully you can get the sensor calibration data in electronic for from the vendor so that you don't have to type it in for each
system).

The tabular approach with discrete data points at say 50, 75, 100,......350, 375, 400 and a run time interpolation routine will also be better. The point for this reasoning is that you would like to devise a way to down load the "turned inside out" table of data for the system and store that data somewhere in a non-volatile manner. That could be done either in an IAP routine that erases and re-programms a page or two of FLASH or possibly using a serial EEPROM like the 93C46. In the latter case the solution of a full data table of 450->500 one or two byte values may not fit too well in a small and economical serial EEPROM.

Michael Karas