It looks to me like the problem is that your sub-VIs are in loops waiting for Stop to be pressed. A subVI does not provide its outputs to the calling VI until the subVI is complete. Your subVIs won't be complete until you press Stop on the subVI front panels. So your calling VI won't get any data from the subVIs until then.
This feature is part of the very foundation of the dataflow paradigm of LabView. A node (function, structure, or subVI, etc.) will not start to execute until it gets data on every wire going to it. A node won't provide any outputs until the node's execution is complete.
Review the section in the LabView User's Manual, Chapter 5, Section titled "Block Diagram Data Flow".
You can solve your problem in a number of ways. Here are a few.
1. Remove the loops from your subVIs. This is a quick fix, but it makes your subVIs less usable standalone.
2. Add an input to the subVIs: a boolean called Wait for Stop. If Wait for Stop is TRUE, stay in the loop and wait for Stop to be pressed. If it's FALSE, execute the loop only once. Wire this new input to a FALSE constant in your calling VI and default it to TRUE on the subVI front panel so it's TRUE by default when running stand-alone. This is probably the way I'd go to fix your problem.
3. Use global variables: write globals in your subVIs, read globals in you calling VI. Use a global for each inidcator you want to pass to the calling VI. I don't recommend this: as in any language, you should use globals sparingly and be careful of race conditions.
4. Use control references in your calling VI to read the data on the subVI indicators. I don't recommend this: it's more complicated than you need for your application.
In options 3 and 4, since the loops repreatedly reading the data will still be in the subVIs, you'll need to put the reading of the globals or the control references in a nested loop. The main loop in your calling VI will execute only once until all subVIs are stopped. That's another basic part of LabView that's important to understand. In a loop, everything in the loop executes once per iteration, and the loop doesn't go to the next iteration until everything in the loop is done.
I have to warn you about the way you're using arrays. Whenever you call Insert Into Array or Build Array, LabView needs to create a copy of the array. As the array continues to grow, the operation gets slower and slower. It would be much more efficient if you could save a fixed number of readings and using Replace Array Subset to create a circular buffer in the array.
Another warning: you're only saving the data to a file at the end of the program. If the PC crashes or someone presses the Abort button, you won't save any data. Especially since you seem to want to take a reading only every second, you have plenty of time to open the file, write one record, and close the file on each iteration.
