NAN comparison in expressions and limit tests works as follows
(NAN > x) == false
(NAN == x) == false
(NAN < x) == false
(NAN == NAN) == true
To summarize, a NAN is neither greater or less than another number and is only equal to another NAN.
If you want NAN to fail and measurement > some_limit to pass, then just use the > (GT) operator.
If you want NAN to also pass, then you need to check for it separately. You can do this dozens of ways. Here are a few:
a) Use a pass/fail step with an expression like: Locals.x > 10 || Locals.x == NAN
b) Use two limit steps, one to check for NAN and one to check the limit. Use preconditions to specify the NAN check only runs for the NAN value and the limit check only runs for non-NAN values. This will ensure
the measurement makes it to the report.
c) If NAN is your instruments way of returning "resistance too high to measure", then you could transform the NAN into an INF in the data source expression and then use a GT limit. Example data source expression:
Step.Result.Numeric == NAN ? INF : Step.Result.Numeric