A part of a fatigue testing data acquisition program requires peak detection. I've read the other peak questions on here, but mine seems to be different.
Each cycle, I collect n points. Within that, there are x peaks. The data between peaks 1 and 2 is extracted, potentially saved to disk, and always inserted into a circular buffer. This continues for each peak found. Of course, there is always data left after the last peak that does not get included into a cycle. So, I shift register it back to the next data collection loop, and append it to the beginning of the next set of data.
My problem is that peak detect doesn't seem to find the peak if it's the first point in a data set. This seems to happen regardless of settings that I use. I c
an't change threshold values, and any width seems to have the same result - skipping the first point. Consequently, all the data that got held over from the last cycle gets discarded because the first peak is ignored.
Anyone have any ideas? I'm reinitializing the peak detection system each cycle because if I don't, it screws up the location indicies. I may put together a picture later, the whole problem makes more sense graphically.
Thanks!
