This was already discussed in another thread.
SDS100: - What is the difference with all the filters in the sds100 scanner?
Below is an attempt to put it together. Ubbe provided the details about the filter functions, which was very helpful for me since before I had no idea what they did. Even given that, it is still most often easiest for me to just try each one to see if it helps.
There will be some on these forums that will be able to understand the filters without much explanation, and others with little radio knowledge which can make it very nebulous for them. It all has to do with experience.
It is possible for a strong signal near the frequency you intend to monitor, to cause interference. Without either filter applied, the interfering signal could be anywhere within the IF bandpass limits, which is plus and minus 5 MHZ from the tuned frequency's converted IF frequency with Filters Off. If you were able to determine the frequency of the interfering signal, it may help with selecting the correct filter to use.
The filters shift the IF passband so that instead of the tuned frequency's converted IF frequency being +5 MHz and -5 MHz from the 10 MHz passband limits, it is either -1 MHz and +9 MHz (NORMAL FILTER}, -9 MHZ and +1 MHz (INVERTED), -3 MHz and +7 MHz (WIDE NORMAL), -7 MHZ and +3 MHz (WIDE INVERTED) from the passband limits. This shift could filter out the interfering signal.
Determining the frequency of the interfering signal can be difficult to do, but not impossible. It could be an internally generated signal by the scanner, or any external interference from an actual transmitted signal, or any nearby electrical or electronic device. So trial and error is much easier since there are only 4 filters to try. AUTO setting will apparently attempt to detect the best filter and apply that one, but as has been pointed out by others, this will slow down scanning. As mentioned above, OFF leaves the IF passband at +-5 MHz. Global applies a filter setting to all programmed systems, unless a system is programmed independently with a particular filter.
Also as mentioned by Ubbe, IFX could solve the problem since it shifts the IF passband to a different IF center frequency.




