At issue with sub tuning article by Bob Pariseau.

[CoryW]

At issue with the point made by this gentleman. At question was this statement :

“A typical Crossover implementation will roll into effect at a rate of, "-12dB per octave". 12dB, you'll recall from my discussion of Balancing Speaker Volume Trims with an SPL Meter”

He states that a loudspeaker with a published low of say 50hz will still be putting out very noticeable volumes at frequencies far lower than we realize. He specified that a crossover of 100hz ( or double the speakers low spec) is a good place to start. This is very sound thinking from what I understand.

Where I take issue is, I utilize loudspeakers that all have very steep roll off at 24dB per octave. Would you think that the same logic would imply that this effect of having to double the crossover point is somewhat negated, and the system crossovers may be set to numbers far closer to the speaker’s published specs. I’m going to try and ping the author, but I’ve found value in posting the minutia here as well. It’s a terrific read. I like his methodology.

 

[ski2xblack]

You are not accounting for different types of speaker enclosures and their inherent acoustic roll off, and what happens when they're fed frequencies below that.

Sealed alignments roll off at 12db/octave (typically), but reflex alignments roll off at 24 db/octave, due to the output from the port going out of phase with the active driver causing cancellation. It's still a good idea to cross ported mains well above those frequencies, as it's easier to get a smoother transition to your subs by doing so, so in the case of ported speakers I would agree with the author.

Another reason to keep in mind with reflex enclosures is that at frequencies where the active driver and port are out of phase, the speaker will exhibit excessive cone motion even though it's not increasing in output due to cancellation with the port. This will wildly increase modulation distortion and potentially damage the driver. So, if you take your ported speaker with output to 50 Hz, and send it a signal, high passed at 50 Hz with a 12db/oct slope, the speaker will indeed get significant signal in the lower frequencies, resulting in the problems described.

 

[CoryW]

Thanks for your time. Your points are well taken. Like the way you think. My really big guys, my RBH T1/r is sealed with the 24dB per octave roll off. It’s low is 50hZ. I seem to like it the best at 80hz. I’m lucky with my subs as they cover up to 180hZ and set in a typical stereo configuration, so localization won’t be an issue. I’ve ordered the test dvd he suggested.

I’m starting to mess with a smart phone app by RTA. If you could point me to useful books you may know if. I want to deep dive. If I can read a packet trace on source routed token ring with beaconing , I think I can learn a bit more.

 

[ski2xblack]

The speakers you have are outliers then. I don't know how they achieved such a steep roll off with a sealed alignment without other filters applied, but whatever.

I applaud your following the maxim "If they fit through the door they're not too big" with your speaker choice.

 

[ryanosaur]

"If they fit through the door they're not too big"

THIS!!!!

 

[ski2xblack]

@CoryW Toole's book is the best recommendation, as well as the articles here at Audioholics that he's consulted on.