LabVIEW
Error in NI's 'Threshold 1-D array'?
Your VI diagram is password protected, thus rather useless.
Can you elaborate what you get when "it does not work" (e.g. wrong result, no result, error message, etc.).
I've never had a problem with the plain vanilla "threshold array" included with LabVIEW. Of course it only finds the first zero crossing that goes from negative to positive so you need some extra coding to make it find all the others, e.g. in a loop increment the start index to the rounded up fractional index output to find the next zero with positive slope. Repeat until the end of the input array is reached. etc. Then repeat with the array multiplied by (-1) to get all the zeroes with negative original slope.
Can you elaborate what you get when "it does not work" (e.g. wrong result, no result, error message, etc.).
I've never had a problem with the plain vanilla "threshold array" included with LabVIEW. Of course it only finds the first zero crossing that goes from negative to positive so you need some extra coding to make it find all the others, e.g. in a loop increment the start index to the rounded up fractional index output to find the next zero with positive slope. Repeat until the end of the input array is reached. etc. Then repeat with the array multiplied by (-1) to get all the zeroes with negative original slope.
OK, the only known problem for "threshold array" is documented in the online help: "... and it might return incorrect data if threshold y is less than the value at start index". Does this apply to your data?
I don't think it has anything to do with the number of calls on the diagram, but is an inherent limitation. The array must start with values below the desired threshold.
Your "fixed" VI basically searches the array until the data goes from negative to positive, then starts the "threshold array" at the index of the last found negative value, ensuring that the limitation does not apply. For some reason it jumps through quite a few uneccessary hoops doing so. All the stuff with bundling is not needed, it does not make sense. I have included
a cleaned-up version below. See if this works for you. (llb contains both VIs, I also did a few minor changes to the main vi for simplicity).
As a more general comment, I think there might be better ways to determine a phase shift. How does your data typically look like? (e.g. amount of noise, presence of other frequencies, DC offsets, etc. all will cause serious problems with this simplistic approach).
I don't think it has anything to do with the number of calls on the diagram, but is an inherent limitation. The array must start with values below the desired threshold.
Your "fixed" VI basically searches the array until the data goes from negative to positive, then starts the "threshold array" at the index of the last found negative value, ensuring that the limitation does not apply. For some reason it jumps through quite a few uneccessary hoops doing so. All the stuff with bundling is not needed, it does not make sense. I have included
a cleaned-up version below. See if this works for you. (llb contains both VIs, I also did a few minor changes to the main vi for simplicity).
As a more general comment, I think there might be better ways to determine a phase shift. How does your data typically look like? (e.g. amount of noise, presence of other frequencies, DC offsets, etc. all will cause serious problems with this simplistic approach).
