[circularduck]

I've had a Titan now for a couple of years, but recently I purchased a Behringer Feedback Destroyer to try and smooth out the in room responce.  This led me to measuring, and re-measuring a lot more then I ever did in the past.  I noticed what a profound effect changing the filters and the phase had on the frequency response, especially above the crossover point (which I'm limited to 90 Hz only).

All of this led me to finally thinking a little more in depth at what the filters do.  I believe I am correct in saying that the filters only effect the slope of the rolloff above the filter points.  Essentially this leaves the slope of the speakers the same.  I am using the sub out from my receiver to the left input of the Titan.  I had always assumed (up to now) that the only real options you had were to max out the filters on the Titan, or set one filter at the same crossover point as the receiver, or set both filters at the crossover point of the receiver.  So my question is, what are the reasons to use the filters at frequencies other than the receivers crossover point?

One other question I had was about the phase adjustment.  I mentioned earlier how this had a great effect on changing the frequency response above the crossover point.  I had a huge null at about 110 Hz followed by a large spike.  By adjusting the phase, I was able to get rid of the spike and move the null (which shrunk) to 100 Hz.  What I was doing was trying to get rid of the null as much as possible, so I adjusted the phase while playing a test tone at the null point so that it would be minimized.  My question here is, if I change the phase to get the smoothest FR, is that the correct phase?  What is the best way to properly set the phase?

Sorry for the long winded questions.

[Mike Dzurko]

First, I suggest optimizing the controls on the Titan itself before beginning to use an external EQ. The EQ should be a final tuning step.
Each low pass filter has a 12db/oct roll off and is used to set the upper cutoff point of the subwoofer. Normally, you will set both filters at the same frequency for a 24db/oct roll off. In some systems it may be advantageous to be able to independently adjust these filters for the optimum blend. Generally it is recommended to set these controls at about the same point as the crossover you are using from your processor. With test EQ and a means to measure, it is possible to use these filters to fine tune the integration between the sub and mains.

The phase is variable from 0 - 180 degrees. Correct phase is dependent on positioning of the sub relative to the main and the listening position. The correct phase setting will yield the greatest level of bass. A null, indicates an out of phase condition at the crossover point. So, in your case, the correct phase adjustment fills in that null.

Again, only after getting the best blend with the subwoofer controls, would I start to use external EQ. EQ is most effective at reducing peaks caused by room modes. Rarely can you fill in a null with EQ, you will usually just over drive the sub.

[BradJudy]

While you might not have the equipment to measure like this yourself, here's an example of the impact proper phase can have on response.  My sub only has a 0/180 phase switch instead of a continuously variable dial, but I was lucky that it worked for my current positioning.  For many it might be that 90 degrees would be the best setting.

Take a look at the graphs here: http://www.bradjudy.com/audioblog/2005/08/30/the-a-difference-a-phase-makes/

[circularduck]

Thanks Mike and Brad,

I have basically been taking the approach you recommend Mike, trying to get the sub set-up before eq'ing.  I am quite limited in my placement possibilities, I have only about a foot in each direction of where my sub is that I can possibly move it.  Similarly I don't have too much room to move my speakers either.

I have an interesting situation where my speakers produce a peak at about 125 Hz, and the sub actually helps tame that peak, so if I set both filters at 90 Hz, I get that peak appearing.  I will be trying more things out later this week.

I have been mindful of trying to boost valleys, I only boosted to a maximum of 3dB when the boost actually did something.  For the most part though, I was just trying to tame the peaks.

As for the phase, should I be testing with pink noise, or sine waves of specific frequencies?

Thanks for all the help so far!

[BradJudy]

You should use waves at frequencies that cover the crossover region (where the sub and mains overlap and can either constructively or destructively interfere with each other). Because pink noise contains so many frequencies, it is difficult to tell if a small region of them (e.g. 80-100Hz) is too low.  Another option is using music that contains strong sounds in the crossover region.  100Hz is pretty common in a lot of music and you could probably hear the best phase setting using a track with strong 100Hz notes.  

Also, it's easiest to find the right place if you have another person adjust the phase knob while you're in the listening seat.  Running back and forth and trying to remember what sounded best is hard.