Thank you for your kind remarks. I appreciate the close scrutiny you have given my previous post but I think you are at a stage of system design where many basics of your system should be considered and evaluated before the issues of my post need be considered - if ever. My remarks were more in the nature of optimizing a well-developed system and discovering how that system interacts with itself than with a fundamental design critique.
However, I would like to offer some thoughts that are relevant to the issues you’ve raised.
ONE BASS HORN OR TWO
Several of your questions essentially revolve around a basic issue to your system design – if you have space for (mid)bass horns should you try for a single horn with a larger mouth and lower cutoff frequency or two smaller horns running in stereo.
This was a question I was going to pose to Romy in the design phase of his midbass horns. He originally considered using the horns into a range much higher than he does now and where both stereo imaging and localization effects might be less acceptable. He rather quickly found the strengths of his midbass horns were in a lower range.
If you end up with a pair of bass horns next to each other it may not matter too much whether you design for two or a single horn as you might get the positive effects of a single horn but you may also get unwanted cancellation of out-of phase bass signals. Summing the bass signals electrically does no good – it merely electrically cancels the out-of phase signals instead of acoustically canceling them. Fortunately, as I mentioned in my earlier post, we are somewhat protected against out-of-phase deep bass signals by a variety of recording practices over the years. So closely placed bass horns might not be so bad.
Furthermore, I am not on the strongest theoretical ground here as I am not sure if the separation of the horns should be considered the separation of their acoustical centers (normally true) or their perimeter closeness (relevant for some analytical approaches). In analyzing woofer drivers the size of the driver is very small compared to bass wavelengths but for the large mouth of the bass horns this distinction may be important. The only practical experience I have had of speakers this size is of large panel speakers. When panel speakers are placed together the effect of the larger baffle size and reduced dipole cancellation effects were of prime importance. So I would need to listen to knowledgeable people with practical experience of large horns or be prepared to research the issue myself.
It comes down to a choice between extending the sound quality that you value in horns or attempting to preserve the stereo signal in a frequency range (below 100hz) with limited stereo information both as recorded and as perceived. Of course, there are alternative approaches.
PRESERVING STEREO WITH A MONO BASS HORN
There is a technique for fully reproducing stereo while using a low frequency mono speaker along with a regular stereo pair. This technique can be used with a subwoofer or a midbass speaker where the prime signal is carried by the mono speaker and the more limited stereo information is reproduced by the stereo pair.
It uses both the sum and the difference of the left and right signals.
SUMMING BASS SIGNALS:
Be careful here. As I mentioned above, if you only sum the bass channels you will cancel the out-of-phase signals electrically before they reach the loudspeaker. That is really not what you want. You want to preserve the stereo out-of-phase information so that it reaches the listener and propagates throughout the room to create the reflection/reverberation pattern you have designed or accepted in the room. If you ever pushed the mono button on a pre-amp or receiver and noticed that the bass went flat you have an example of the problem with a simple summing. I have sometimes preferred using just one channel or the other instead of summing to mono.
THE SUM/DIFFERENCE TECHNIQUE
Create both a sum and a difference signal from the original left and right signals:
Sum = Left + Right = (L+R) (mono)
Difference = Left - Right = (L-R) (stereo info)
A central woofer or bass horn carries the monaural sum signal which is the principal component in the bass.
The stereo speakers carry the difference signals which acoustically combine with the sum signal from the mono horn to recreate the stereo left/right signal at the listener.
Mono Horn + Left speaker = (L+R) + (L-R) = 2L (i.e. the true left channel signal)
Mono Horn + Right speaker = (L+R) – (L-R) = 2R (true right channel signal)
(Note the change of polarity on the right channel difference signal)
Consider how this could be used in Romy’s system,
His midbass horns would both carry the identical sum (L+R) signal so a single horn could have been used instead.
Romy’s bass towers in the new spaced, time-aligned positions would carry not only the ULF but also the stereo signals (L-R) to complement the monaural midbass horn, which should acoustically sum at the listener to the original stereo signals.
By the way, this is an old technique. Alan Blumlein used a variant in one part of his original patent on stereo. He used specially wound transformers to create the sum/difference signals. Today we use op-amps or circuits with differential stages like a Darlington (long-tailed) pair.
FINAL THOUGHT ON AMBIENCE
You wondered about trying to reproduce ambience/environmental sounds at an elevated level. I agree completely with you that sounds like applause, reverberation and general room noise which are part of the recorded sound in the audible bandwidth should be reproduced at the same level as the music. I don’t know how to do it otherwise without creating ambience effects artificially.
I interpret Romy’s findings to relate only to SUBAUDIBLE sounds – sound which lies below our general low frequency hearing. These subaudible sounds need to be quite loud to be even be classified as above the threshold of hearing. I suspect Romy is finding a range of loudness for these sounds which is barely at or below literal hearing but can be perceived or felt. I am not sure if he is compensating for low frequency roll-off in the recording or discovering a way of enhancing the sense of space or acoustic volume with a perceptional trick of hyper-elevating the subaudible sounds. He seems to be convinced that this technique enhances the music reproduction even beyond the sense of acoustic space. I have experienced the beneficial effects of deep bass reproduction and as Romy and others have indicated even without music the awareness of the acoustical space is enhanced. These experiences involved reproduction of the subaudible signal at the same level as the audible bandwidth so Romy’s technique of hyper-elevating the subaudible is new to me.