I highly recommend that you eliminate use of the dynamic data type as much as possible entirely. While it might seem easier, it can cause all sorts of problems because it attempts to adapt to the type of data it thinks you want, often in ways you don't expect.
If you post your code, we can probably help sort out your problems. It's reasonable to use either waveforms or arrays, depending on which is more convenient for your application, but you probably want to be consistent throughout (don't read some channels as waveforms and others as arrays unless there's a good reason for doing so).
