Let's look at a few things.
1. The statistics Express VI does a lot of things, but it does not do what you expect it to do in your VI. When configured as you have for the Maximum value, it determines the maximum value of all the data wired to Signals on each call. Since you are only wiring one scalar value at a time, the result is always the same as the input. This VI is designed to work with waveforms or arrays containing multiple values. The only reason it appears to work is that the external comparison to the -3 dB value stops calling the VI when you have passed that point.
2. The reverse ramp is required due to 1. If you ramp up, then it will stop at the last value which is the highest frequency, not the cutoff frequency.
3. I showed an FFT in my previous post but that is misleading. It is taking the spectrum of the frequency response curve. This is not the same as the frequency response of the filter. I think I may have been thinking of a more typical filter where feeding it a broadband signal such as noise or a chirp and then taking the spectrum of the result is a standard way of evaluating filters. With this special filter implementation where you feed it a signal one component at a time, that does not work.
4. You can do the calculation of the cutoff frequency after the loop. Take the frequency response curve (that you are feeding to the FFT) and find where it crosses the -3 dB threshold. Due to the randomness in the filter output, a single measurement will have some error.
Since this is an academic assignment, I am not posting the code which will do these things. Your instructor may be monitoring the Forum!
Lynn
