A full range speaker?

I have systems that are 3 dB down at 45 Hz, 24 Hz and 20 Hz, the latter augmented by a pair of powered subs that are 3 dB down at 16 Hz, which is well below what is audible (although it can certainly be felt).  The latter are used only for video, not for normal audio listening.

I'd draw the line at 30 Hz for realistic reproduction of orchestral  music, but I find the system with the highest roll off the best for listening to strings and unamplified music (ironically they take up the most space as they are electrostatic panels).

Do you miss much by not being able to go below 30 dB at a listenable volume?  Probably not unless you are a video fan and have a system able to reproduce it all.  And much of the time you aren't hearing the fundamental note but rather the higher harmonics anyway.

It is fun to sometimes put on some of the relatively few full size organ recordings with some content below 20 Hz but you can't really hear anything down there, you feel it, and may hear the overtones.  Not a huge number of organs that can reproduce 16Hz anyway - 32 foot pipes take up a lot of space.

Since Asked.. For those that care...

Inaudible High-Frequency Sounds Affect Brain Activity: Hypersonic Effect
https://journals.physiology.org/doi/full/10.1152/jn.2000.83.6.3548

There are two factors that may have some bearing on this issue.
First, it has been suggested that infrasonic exposure may possibly have an adverse effect on human health suggesting that the biological sensitivity of human beings may not be parallel with the “conscious” audibility of air vibration.

Second, the natural environment, such as tropical rain forests, usually contains sounds that are extremely rich in HFCs over 100 kHz. From an anthropogenetic point of view, the sensory system of human beings exposed to a natural environment would stand a good chance of developing some physiological sensitivity to HFCs. It is premature to conclude that consciously inaudible high-frequency sounds have no effect on the physiological state of listeners.

To measure human physiological responses to HFCs, we selected two noninvasive techniques: analysis of electroencephalogram (EEG) and positron emission tomography (PET) measurements of the regional cerebral blood flow (rCBF).  EEG has excellent time resolution, is sensitive to the state of human brain functioning, and places fewer physical and mental constraints on subjects...On the other hand, PET provides us with detailed spatial information on the neuroanatomical substrates of brain activity. Combining these two techniques with psychological assessments, we provide evidence herein that inaudible high-frequency sounds have a significant effect on humans.

http://www.enjoythemusic.com/magazine/manufacture/0114/the_world_beyond_20khz.htm

To fully meet the requirements of human auditory perception I believe that a sound system must cover the frequency range of about 15Hz to at least 40kHz (some say 80kHz or more)...

https://www.researchgate.net/publication/290557870_Design_and_Evaluation_of_Electronic_Circuit_for_P...
  A plasma speaker produces sound through an electrical arc.  It works by heating air, causing nearby air molecules to vibrate and expand. This releases a pressure wave, which is heard as sound...  The primary advantage of a plasma speaker for sound reproduction is that there is no diaphragm and therefore no mechanical mass. This means a plasma speaker has the potential to reproduce sound with high fidelity up to 150 kHz.

@ggc
I mean this sincerely, thanks for posting those links. I’ll have a good read with and try to keep an open mind.

What would you set as the lower limit of music reproduction for a speaker to be called full range?

~~~~~~~~~~~~~~~
If you look at the link above
giving the fq range of all classical instruments, 
You will see 
60hz -2k hz 
is the  low/high of 90% of the orchestra.
Now I just hada  experience of a  96db wide band driver.
I found ita  bit thin and female vocals not accurate.
I found the davidlouis 91db far superior in sonics  vs a  10x's more expensive wide  bander,


Seems each wide bad offers something different, just like xover type speakers , each is completely different from the others.

The little DavidLouis 4 incher delivers all the body in this 60hz-2k hz that i need ina  midrange. 
But to answer your Q. 40hz is the very bottom, 
20hz-40hz is nothing but a  fantasy, These fq;'s are so rare in classical makes this area nothing but a  empty bag.
Same for any fq's over say 12k,, nothing but a fantasy.
The only fq's that really amtter in any speaker,  horn, wide band, low db sens etc, plannras, ribbons, etc etc , A-Z speakers, any speaker that was ever made....
The only fq's that count is the 60hz -say 3khz tops, really 2k. 
But i'll  extend graciously to 3k.
This is where classical music  is performed.

We should all drop the fq range specs of speakers as a  distraction,
The only thing that really counts, Is how does the speaker voice the 60hz -3k hz. 
thats it. 
This is why I feel my Frankenstein really voices these fq's as a  grade A/B speaker. 
The weak link in the Frankenstein is the  davidLouis in the 200hz-1khz range, , only here hasa  slight edge on the fq's. 
Bottom is sweet, top is sweet, Its in the middle fq's where there is a slight *edge*, Which is why I give my Frankenstein's a  A/B grading. , 
Beats Vandersteens, Class C speakers. Thats for sure. 
Beats Wilsons as well. Class B/C speakers. 

millercarbon. Like you, I do think that inaudible upper frequencies matter; however, I have a hypothetical question: Is it possible that we do not necessarily know that our brains perceive this, and could these frequencies be used for directional location, i.e. placement of instruments.

That's a part of it for sure. My understanding is the ear has four times as many hair cells devoted to sensing frequencies above what we consider audible, as detect audible frequencies.  

This seeming paradox is resolved by the fact the so-called audible frequencies are all tested using sine waves. Something that really does not exist in nature. ggc is right, we evolved to find our way around a natural environment. It would make more sense to look there than at unnatural sine waves.  

In nature- which includes musical instruments- there are lots of things (like ggc's rain forests) that produce ultrasonic sounds. But unlike the test sine wave that exists unnaturally on its own, these are all mixed in with and riding on top of lower frequency fundamentals. When a leopard growls the powerful menacing part of the growl is down low. But the part that tells us how far and in which direction is for certain a lot higher in frequency.  

High frequencies attenuate with distance a lot faster than low frequencies. The frequency itself is enough to point us in the right direction. The balance of frequencies however can be used to tell us how far. It is to me just plain nuts to think we evolved over a billion years by staying alive, and yet somehow managed to do so never knowing how far away we are from being eaten. This is nuts. Absolutely nuts. 

For sure this stuff happens unconsciously. Has to. If we had to sit around thinking and analyzing we would be eaten alive. There is no time. There isn't even time for some of this to be processed in the cerebral cortex. Studies show that when shown an image of a snake the visual response time is faster than it would take neurons to go from the eye to the visual cortex. The response is, in other words, reflexive. There is no reason to think the same does not happen with sounds. Do predators hunt only in daylight?