A correct filter type to get rid of noise is a low pass filter. Very basicly it takes the previous value(s) and calculates new value(s). So if the previous value is 0 and the next one is 100 it will probaly come out with something like 50, 50.
In your vi you take a single sample of several parallel channels into one array and present it as input to the filter vi. But this vi takes it as a sequence (in time) of one single reading and tries to compare them to each other.
You might rather want something like a sliding average.
Initialize an array (of e.g. ten) with zeros and in a loop replace one element after the other with a new reading. At the end restart from the beginning. Take an average of the readings and the result will reflect
the last (ten) samples.
Gabi
7.1 -- 2013
CLA