Problem with output signal in simulink

hello, i have created 3 filters with fdatool, a low pass (20-250 hz), a band pass (250-2000 Hz) and a high pass (2000-20000 Hz9, export these 3 filters and put them in simulink to test the 3 filters The problem is that when introducing a 100 Hz signal it gives me a signal of 1 [V] in the low pass and 0.2 in the band pass and 1 [V] in the highpass. The expected response is that it delivers 1 [V] in the band pass and the other signals are attenuated ( Figure 1). Attached images of the configurations

Figure 1:

other settings:

Complete filters:

[Mathieu NOE]

hello again Juan

did you test your filters first - a matlab script to do bode plots ?

[Juan Chehin]

Hi, I designed it with the fdatools and it gave me a correct graphic

[Juan Chehin]

[Paul]

I don't have an explanation but have two comments:

a. I think the expected response would be to deliver 1V in the low pass and be attenuated in the bandpass and highpass.

b. It looks like the the low pass filter has a sample frequency of 600 Hz. But the Sine wave source is sampled at 1000 Hz. This combination will result in undesirable shifts in the effective input to the filter and so the response won't be what you're expecting. What are the sample frequencies (Fs) of the other two filters?

[Juan Chehin]

Fs for band pass is 4400, and for high pass 44000

[Juan Chehin]

In the end I solved it as follows: Calculate twice the maximum signal treated (2 * 20000) and take that as the Fs for all filters. For the scope sample time I put (1/44000). The Fs must be greater than twice the maximum frequency treated (for Nyquist), adopt it as 44000.