By default, we set the critically damped filter as the factory default for all GEMAC sensors with an adjustable digital filter (sensors in the Reference, Classic, and BasicLINE series). The reason for this is the versatility of this filter. It has a response time or delay that is four times shorter than that of the Butterworth filter (see figure below). The Butterworth filter, on the other hand, has higher attenuation even at frequencies slightly above the set cutoff frequency.

So when should each filter be used?

As a general rule, users should start by testing with the filter’s default settings, since a shorter response time is usually a key requirement. If the signal response—that is, the ramp signal—is “too noisy,” the cutoff frequency should first be lowered until the response time limit is reached. The response time of the filter response for the critically damped filter in [s] is 1/set cutoff frequency [Hz]. If the response time is not relevant, a cutoff frequency of 0.1 Hz should be tested, which is the lowest adjustable cutoff frequency.
The Butterworth filter should only be tested and used in applications where no input shocks—or only minor and infrequent ones—are expected, and where NO control loop is controlled by the sensor, as the filter response exhibits overshoot in the signal response to input steps. The Butterworth filter is particularly useful when very low-frequency, disruptive vibrations are present in the application. In this case, the significantly greater separation between the passband and stopband is specifically utilized.