Can you even sense 115db at 3.5hz assuming that Bosso has that much output? How clean could it possibly be?
Still no info from Bosso as to what max SPL he gets at 3.5Hz with his subs and exactly how many units are involved.
I don't think its about being able to hear single digits with your ears. I think its about reproduction to real world levels. A fictitious but plausible example would be If a door slam produces 3.5Hz @ 90dB and the hair on the back of your neck twitches, or your stomach feels strange for a fraction of a second, that should be the reproduction goal. The Fletcher-Munson curve threshold is meaningless when talking about the broadband experience.
The Doctor has it right. F/M had no practical way to conduct their experiment to 20Hz and below. The curve is theoretical, based on human reactions to higher frequencies. Also, the tests were strictly auditory. Single digits exist constantly in every day life, but certainly not in triple digits dBSPL. They are perceived, not heard. As I attempted to point out earlier, consulting firms are hired to design a car door mechanism, seals, etc, that results in the perception of quality when the door is closed. The perception graphs for a higher quality sound all de-emphasize the upper frequencies and emphasize the subsonic (to zero) frequencies. Of course, it should go without saying that closing a car door does not result in SPL of 125dB @ 3Hz, but it has been proven many times that elimination of the single digits of that event (or even a relatively slight drop in the SD range) is immediately perceived as a lesser quality automobile. This tells me that not only are humans able to perceive single digits, they are able to perceive small changes in level in that range.
How clean? THD is irrelevant at 3Hz. H2, H3, H4 and H5 are inaudible. THD may actually boost output in the first two octaves. I'm certain it does, but not certain as to whether or not it alters perception. I suspect is does not.
What level? I have a hump at 5Hz in my particular room, which is close to 4,000 cubes, that reflects approximately 18dB of gain from 4-6Hz. I imagine that this is probably not unique to my room. The graphs I posted were generated using the QSC amplifiers, which have a 2nd order HP at 5Hz, which would attenuated the output by 3dB at 5Hz and roll off from there at 12dB/octave plus 12dB per octave natural roll off.
Knowing these particulars, and knowing the anechoic FR capability of my system, allows me to extrapolate a fairly close guess of maximum (amplifier limited) peak output at 3-5Hz to be about 115dB.
Playback of full spectrum of natural events will not require anything near 115dB at 3Hz. Movie soundtracks that have SD content (such as the WOTW lightning strikes) are usually encoded at -20dBSF in that area, requiring 105dB at playback.
As I've posted before, I have a 20Hz HP switch, so I can A/B the lightning strikes with and without SD playback by looping the scene and switching the HP in and out. There is no question as to the difference in perception. (Keep in mind that the SD content of the lightning strikes is only in 2 of the strikes. The rest of them are higher in frequency.
It's also worth noting that the inverse square law does not strictly apply to ELF. SD sound waves can travel much farther than higher frequency sound waves without losing intensity. This may be perceived as room and/or boundary gain when graphed at the listening position, when in fact, it's more of a loss of intensity through the inverse square law at higher frequencies.
Free lunch from corner loading: Corner loading, which is the same thing as horn loading, is said to excite the highest number of frequencies, which may result in peaks/nulls at the LP when using a single subwoofer.
Even if this results in your having to use EQ to lower the input signal at the center frequencies of the peaks, it's the same thing as adding headroom by the amount of cut applied at those peaks. When using multiple subs, it's easier to move them to spots that result in a smoother in-room response, but such peaks in the single digits range (such as the one I have at 4-5Hz, which is pure free lunch from any perspective) are absolutely irrelevant. The relevant fact is that you have playback in the first 2 octaves of the Dolby LFE spec, not the amount of THD or whether or not the response is flat in that range.
I'm curious as to how Bosso has done the cooling system.
To further clarify, he mounted a heatsink to the driver and then another aluminum heatink that is exposed to the outside, with a fan stirring air inside? I'm somewhat lost, his web page doesn't exactly describe it too clearly.
Heatsinking the driver to the outside of the box of a sealed system is an interesting idea. Active cooling of that heatsink also very interesting. Bosso probably won't want to discuss this in detail if he intended to market a system using such techniques.
Once again, Dr. Vinc is correct, although it's not so much that I may wish to market the idea as it is that I would like to be specific in posting the benefits of such a method. This requires temperature probes on the driver motors and in the enclosure to monitor inside ambient temperature change, etc as well as doing the same tests without the cooling system.
It's a unique system that took years to get right and it will take some time before I show the details. I have little doubt that the idea is a valid one for any alignment since sub drivers are so inefficient that increasing the thermal handling by a few dB is significant.