How important are frequency extremes to authentic sound reproduction?

With due respect, this must  be the weirdest thread I've ever seen, Loki. Is this by invitation only or are other members allowed to interject and offer their humble contributions to the subject?

Scroll down about five threads and Richard explains

Thanks Jeff. 

Please feel free to contribute Tony and Jeff (and everyone else)! The original comments were threatening to derail the original post about speakers so we felt it better to bring them out of the Hifi forum. I've asked for my first, rather messy attempt to be expunged!

Apologies, Loki. Back on topic...

With regard to frequency extremes, the issue of constructive and destructive interference has not been discussed. It certainly is a real issue on the bottom end, where low frequency waves can cancel or enhance each other and alter the music produced in a listener's room, esp with rear ported speakers. So it seems ultrasonic frequencies could have the same effect, except that, to my knowledge, those frequencies aren't recorded.

Analog tape records to 15k, well above the frequencies produced by musical instruments. CD up to 20k, beyond the range of anyone's hearing, but for what purpose? Are there some reverberation frequencies in the 15-20k range that are recorded and reproduced by speakers, and then become useful to the listener?

Joe: the bottom end is a curious one to get right. When using my Allen and Heath/MC2/Tannoy PA rig it is the visceral bass from the subs which is key to the party. The problem is the frequencies being reproduced. It can sound quite boomy in the booth, but there are nodes of constructive and destructive waves on the dance floor. Sometimes the bass is best heard outside the venue! The subs are set to handle everything below 70hz and I tend to parametrically equalize out anything above 12khz. The latter is for two reasons: a) it's the high frequencies which fry your ears at high volumes and b) to conserve amp power for the frequencies which are most useful in a disco setting. However, what I have noticed is that with the high frequency adjustment, a certain sense of space can be lost.

And herein lies my point. When looking at the specs of headphones, amps, and speakers whilst auditioning, I tend to prefer the ones with the widest frequency responses. My headphones apparently go as low as 18hz and as high as 35khz. The bass is clear, deep and visceral, even though they're cans.  I wonder whether what I  am hearing is a sort of frequency 'headroom', rather than the frequencies themselves, that by using equipment which goes into supra and sub aural frequencies the audible spectrum is freed up?

And yet, against that theory, the back of my old NAD 3120 had a switch which toggled between a) normal 20hz-20khz gently rolling filtered output to enhance musicality and power output and b) 'lab' unfiltered output. The latter went low and was impressive for techno, albeit lacking grunt but the former was just that much more cohesive, and powerful making sense emotionally if not as free of the box as the other.

And thus the gentle art of compromise once more won the day!

Instruments like violins have frequencies higher than 30kHz. My understanding is that if you have two tones - one at for example 25 kHz and one at 26 kHz you should hear the interference beat at 1 kHz, even although the human ear does not respond to such high frequencies. I think this is why digital at 96kHz sounds better than the 44.1 kHz stuff, but I have no proof.

Just found this from an old Tannoy brochure:

Always an innovator where quality sound

reproduction is concerned, Tannoy has continued

this tradition by being at the forefront of the

development of WideBand™ technology. All

Prestige models except Autograph Mini are

compatible with Tannoy’s SuperTweeter™designs, providing the opportunity to extend

high-frequency response to above 50kHz,

thereby providing all of the bandwidth required for

today’s wide bandwidth digital recording formats.

In fact the SuperTweeter™only starts working at

a point that is close to what is generally

considered the limit of the audible frequency

band for most adults. The Dual Concentric™in the main loudspeaker, still working to its full

frequency specifications, continues to provide the

heart of the musical information as a coherent

point source. However, the extreme high

frequencies are then resolved by the

SuperTweeter™to provide incredible, wide

bandwidth detail and enlivening the performance

by increasing the tonal accuracy of individual

instruments for more natural and true to life

music reproduction.

By extending the high frequency capability of the

complete loudspeaker system out to beyond

50kHz, this in turn corrects the time and phase

response at the upper end of audibility. This

delivers enhanced accuracy and spaciousness,

improved clarity within the essential mid band

area and even an enhancement of the definition

and impact of low frequencies. Benefits are heard

clearly with any source material, regardless of

bandwidth, therefore ensuring that a Tannoy

SuperTweeter™is an essential accessory.

It was bollox then and is still bollox now.

Unless the than sampling rate is above 44 for CD standard or 48 then the supeweetter  is only processing recording noise, rather than musical innformation.

However musical information stops at 14 or 16 k Hz, which was understood in analogue days [when the microphones - still used stopped at 20 k - worked on analogue mixing desks and analogue tape recorders that stopped at 16 k], and is still regarded as rated as the gold standard of recording. Don't confuse analogue with high-resolution digital ... Or bat-hearing with musical information ...

ATB from George

Like George, I tend to look at the efficacy of a supertweeter pragmatically and say that if no sounds are recorded above a given frequency then what value is there to having a speaker capable of reproducing frequencies beyond that extreme.

On the other hand, a typical two-way speaker has a crossover in the 2-3 kHz range, leaving the tweeter the task of handling all frequencies above that. A tall order, perhaps. Couldn't the value of adding a supertweeter be that it is actually more adept at handling those extreme (recorded) upper frequencies than a conventional tweeter? Rather than the unrealistic task of adding unrecorded supersonic frequencies the supertweeter simply provides better controlled sound at the uppermost ranges of the recording, thereby enhancing the tweeter?

Interesting Joe: and is the super-tweeter's capability enhanced by the extreme headroom it then has above audible/musical sounds?

And what of atmosphere? How is our perception of music affected by the dynamic-acoustic environment of the performance? Whether or not we can hear them in a conscious way, those frequencies still exist and presumably interact with those we can hear. For example I can feel sub-bass which I can't hear, why should there not be something similar going on at the other end of the audio spectrum?

Perhaps that's what Tannoy mean by:

By extending the high frequency capability of the

complete loudspeaker system out to beyond

50kHz, this in turn corrects the time and phase

response at the upper end of audibility. This

delivers enhanced accuracy and spaciousness,

improved clarity within the essential mid band

area and even an enhancement of the definition

and impact of low frequencies. Benefits are heard

clearly with any source material, regardless of

bandwidth, therefore ensuring that a Tannoy

SuperTweeter™is an essential accessory.

While there is no doubt that some speakers do not address phase coherence very well across the audible frequency range, some speaker designs do!

Conventional cross-overs with their filtering networks can disturb phase coherence, and one specific problem with a break between mid-bass driver and high frequency driver between 2000 and 3000 Hz is that this falls into a part of the human hearing that is very acute. 

An advantage of the BMR unit used in the Ovator speakers is that the cross-over is much lower in pitch than is required for convention high frequency drivers, away from this very acute part of human hearing. 

So the argument about a super-tweeter improving the high frequency performance of a loudspeaker in some way such as apparent coherency, would depend on how successful the design of the speaker was in the first place.

ATB from George

It is hard to judge super tweeters as a whole. Individual standalone models cross over into the audible range at very different frequency levels, some as high at 18kHz (Tannoy) and some as low as 6kHz (Townshend, Audiosmile).

Similarly, speakers with built-in super tweeters use them first and foremost for providing treble up to 20kHz. The Harbeth SHL5+, for example, crosses over to its tweeter at 3.3kHz, and then to its super tweeter at 12kHz.

In either case, stand alone or built-in, what a super tweeter delivers at ultrasonic frequencies is almost certainly less important (unless you are a bat or a dolphin) than what it delivers up to 20kHz.

Hook

So what about adding a stand alone supertweeter to an existing speaker system? In the case of a supertweeter designed into the speaker, there is a cross over. Would a supertweeter added to a conventional speaker compete with the original tweeter and possibly cause destructive interference/ coherency issues?

The Townshend and Tannoy supertweeters were both designed to be add-ons Joe.

This is from the Townshend website: