Sensor validations

[Scott Cytacki]

Since the beginning the sensor code has supported sensor validations.  I think these 

have been misunderstood and not used correctly.   Perhaps we shouldn't use them at all 

and let the student (and teacher) figure out on their own that they have attached the 

wrong sensor or have a switch in the wrong place. 

These validations are currently used with Force, Pressure, and Current sensors. In a 

Force experiment you might need a sensor that has a range of +/- 40N.  Or you might 

need a sensor with a resolution of 0.01 N.  With a Vernier sensor you need to flip a 

switch for each of those cases. With pressure if you are measuring atmospheric 

pressure you'll want high resolution around 100kPa.  If you are measuring gas pressure

you want to handle lower pressure.

The sensor code attempts to be vendor agnostic so it doesn't let the author specify a 

particular sensor or switch position.  Instead it lets the author require a certain resolution 

and a certain range.   However, the Rails Portal does not expose these requirements 

directly to authors.  Instead the portal presents a list of sensors with predefined

requirements.

Looking at Force sensors, the portal gives the author 2 choices:

- Force (5N) --  max: 4.0N, min: -4.0N, resolution: 0.01N

- Force (50N) --   max: 40.0N, min: -40.0N, resolution: 0.1N

Here are some options from Vernier and Pasco

- Vernier Force 10N switch position -- +/- 10N, resolution 0.01N

- Vernier Force 50N switch position -- +/- 50N, resolution 0.05N

- Pasco Force -- +/- 50N, resolution: 0.03N

- Pasco High Res Force -- +/- 50N, resolution: 0.002 N

If the author chooses Force (5N), then the following sensors meet its requirements:

Vernier Force 10N switch position and Pasco High Res Force.  

This leads to confusion because authors think Force (5N) is referring to some specific 

sensor or position of the switch on the sensor.  I'd guess in many cases the author 

doesn't need 0.01N.  Perhaps they only need 0.03N or better yet 0.05N.   If the 

experiment works with a 0.05N resolution then any of the above sensors will work. 

There are various ways to improve the situation around sensor validations, which 

one is the best for authors and students is not clear.  At least I wanted to document 

the current situation.

Scott