Thanks for contributing to this thread. I have a couple of filter questions.
1. I decided to go analog active crossover for my subs, LR4 @ 80 Hz. I have read that digital and analog filters behave the same so does that mean that an analog RC filter @ 80 Hz has less precision than an RC filter at higher frequency?
Not necessarily, they may be more lossy than one at a higher frequency. The active crossover is the more reliable way to go about that. The idea that analog and digital filters work the same is, however, rather suspect. In analog filters, the result is always IIR, and there will be phase shift. (Well, except for very high-frequency filters used in some RF devices called "SAW" filters which you're not going to be using for subs.)
Digital filters can be IIR, but the impulse response will be different. This is neither good nor bad, it just is. It is also possible to make an FIR (finite impulse response) filter, in which case the filters can be pure time delay (no phase shift beyond time delay the same at all frequencies, i.e. phi = 2*pi*t*f where t is delay in seconds and f is in hertz, and phi is the time delay resulting from a constant delay). However, FIR filters with high frequency resolution (as are required with very low frequencies) require agonizingly long filters, and hence that is usually reserved for crossovers not in the sub-100Hz range for a variety of reasons. The problem with an IIR filter in digital form is that it trades off length (i.e. computational load) quite heavily for increased need for world length. Win some, lose some.
2. Do 2 subs @ 80 Hz solve the "narrow" and "shallow" issue you mention for one sub?
If each is fed by one channel, and they are located to either side, yes, to the extent that's possible with 2 channel signals. One of the effects of LP's is that out-of-phase rumble is usually the more common rumble mode, and this can also create a sensation of width.
3. I have been using the FR script developed by
@JP to EQ my cart/TT using FIR filters. I have been using minimum phase filters because I read somewhere that cart non linearities are minimum phase. Is that correct or is it more complicated than that?
Thanks
A minimum phase inverse filter will correct both frequency response and phase shift of another minimum phase filter, so that is indeed probably a good choice. The problem comes if there is any non-minimum-phase contribution. This is where things like room correction and loudspeaker correction can get very complicated very fast.