Your info is inexact. Low frequency roll-off with ported enclosures is not 3 db but rather 24 dB per octave below the box tuning frequency. So for example, with a box tuned at 50 Hz, then the response would be at least -24 dB at 25 Hz and the speaker would be far from qualifying as a subwoofer. Also, driving a woofer below the box tuning frequency provides no protection to the unloaded driver with a great possibility of the cone moving beyond its X-max. That would necessitate repair or replacement.
You are absolutely correct! Thanks for letting me know that I communicated poorly.
My tuning frequencies are typically in the lower 20's to upper teens, so below that point the rolloff is indeed 24 dB per octave.
Let me try to explain where that "3 dB per octave" comes from:
My subwoofer design philosophy focuses on room interaction, and at least two authors (Martin Colloms and a woofer designer whose name slips my mind) have published curves indicating that "typical" room gain from boundary reinforcement is approximately 3 dB per octave south of about 80 Hz or so. (Briefly what happens is, reflections from room boundaries which are within 1/4 wavelength of the sub combine with its output approximately in-phase, rather than in semi-random phase. The longer the wavelengths, the more room boundaries included.) So if we want sub(s) + room to be "approximately flat" (as a general trend; ignoring the inevitable room interaction peaks and dips), then we want the sub's response to be the approximate inverse of room gain. That would be, approximately -3 dB per octave across the bass region. Of course this is not true in every room, but imo it's a valid starting point.
If we use a "normal" vented box tuning, and our vented box is dead flat to 20 Hz, typical room gain would result in about +6 dB at 20 Hz (relative to 80 Hz), which will sound boomy. Of course EQ can fix that.
If we use a sealed box the rolloff will (usually) be slower than 3 dB per octave above resonance and (always) faster than 3 dB per octave below resonance. Again, EQ can fix that.
So mine is by no means the only valid approach, but it does theoretically result in less need for EQ to "fix" the sub's in-room response.
...you could get a complete HSU VTF-15H MK2 sub around $1000 shipped. It's close in performance to SVS PB3000, but having a bit deeper bass at expense of having a bit less than pb3000 mid-bass power. (2-3db at most)
The one-port-open frequency response curves for the
Hsu VTF-15H Mk2 display the sort of gently-downward-sloping trend I'm talking about, which might be called a "room gain compensation" curve. I used to be a dealer for Dr. Hsu, and agree with BoredSysAdmin about the VTF-15H Mk2 being a killer deal.
Like the Hsu, the kits on Speaker Hardware's website also have two pluggable ports.
My preferred method of dealing with the inevitable room-interaction peaks and dips is to use multiple subwoofers intelligently distributed, which is an acoustic solution to what is basically an acoustic problem, and which (at the risk of oversimplifying) makes EQ even more effective, if it's needed. I have been manufacturing a four-sub system called the Swarm since 2006, using Earl Geddes' ideas with his permission. (As anecdotal evidence of some basic competence on my part, The Swarm has received three Editor's Choice awards, three Golden Ear awards, and one Product of the Year award, from The Absolute Sound. Robert E. Greene
reviewed the Swarm in 2015.) Regarding the concept, last time I checked Gene and Matt were still advocating distributed mult-sub systems, though not mine. There are more cost-effective solutions than the Swarm, such as a quartet of the big Hsu's.
So I agree with everything you said, and thank you for pointing out that what I wrote came across as completely incorrect. Hopefully I've cleared that up a bit, but let me know if anything is still unclear.
