I have a tolerance statistics question: It is probably best if I describe it with an example, even though I don't have one specificlly in mind. It just bothers me that I either forgot how to do this or was never taught.
Suppose I have a sensor measuring something. For my example I have a thermocouple in water reading 211F. The tolerance on the sensor is
+/-2F.My understanding is that for the T/C that I have a normal distribution with three times the standard deviation being 2F centered around 211F. That is to say that I have a 99.73% chance that the true reading is within +/-2F of my indicated value. Am I on the right page here?
I add another thermocouple, it reads 213F with the same tolerance. I know the water is uniform in temp.
Now with my second sensor included in the mix, the most probable true value would become 212F (the avg of the two sensors as they have the same PDF, just centered different).
My question is how do i compute my new standard deviation of the combination of the two sensors? I want to know my new tolerance. Or am I way off shore in my thinking.
thanks
-josh