I have heard those Tannoy super tweeters - but only on larger Tannoy's.

As far as I know they only come in at about 20 KHz - which is way beyond my hearing (16 KHz when last tested - about 2 years ago).

Funny thing is - they do make a slight difference to the sound - even to me. God only knows why....

maxg:

I have heard those Tannoy super tweeters - but only on larger Tannoy's.

As far as I know they only come in at about 20 KHz - which is way beyond my hearing (16 KHz when last tested - about 2 years ago).

Funny thing is - they do make a slight difference to the sound - even to me. God only knows why....

That is interesting. Did you get to hear one of the big Prestige line Tannoy's? This is on my list of just gotta hear speakers. 

I'd like to compare the super tweeter to the CT125, which I know to be killer. If you can improve on that, more power to you.

maxg:

I have heard those Tannoy super tweeters - but only on larger Tannoy's.

As far as I know they only come in at about 20 KHz - which is way beyond my hearing (16 KHz when last tested - about 2 years ago).

Funny thing is - they do make a slight difference to the sound - even to me. God only knows why....

Here's a guess. Every driver on a speaker interferes with the other drivers, this is unavoidable. So, even though you can't hear the "direct" sound coming out above 20khz from the supertweeter, you hear the result of it's interference on the other driver, probably a regular tweeter. 

Mallette:

Mark:

>The acoustic side is not the same as the electrical side. 

After I sent that it occured to me that, properly recorded, any such interactions would be audible on a CD even if caused by harmonics beyond the range of a CD. 

Is that what you are saying? 

As you know, I have come to believe through experiments to which you've been privy that the limitations of 16/44.1 are not as severe as what we've been served up had us believing.  OTOH, there ARE audible differences in this sampling rate compared to higher resolutions. 

In my case, I know I have little after 13khz in one ear and barely 15khz in the other, so I remain mystified as to why I can hear these differences. 

Regards,

Dave

Test 1.

Two speakers:. One with 20khz level response, the other with a rolloff beginning at 15khz. Using an oscillator you confirm that you can only hear up to 14khz on either speaker.  

Test 2.

One pair of speakers with level response to 20khz. Add 2 amplifiers. One that rolls off beginning at 15khz, the other has flat response to 250khz. They are otherwise the same circuit.  Prediction: you will easily hear the difference between these amplifiers, even though you have previously shown that you can't hear above 15khz. 

This seems like a good place to ask this.  I once asked it 20 years ago in college... well...actually 25 years ago (*** time flies!)

If you have a tweeter that will play to 20khz and another that will play to 40khz ( just using round numbers for reference)

Then it seems to me the 40khz tweeter will oscilate faster because it goes higher, right?

So, if the 40khz tweeter can move faster, have a quicker motor & all that stuff...  would it also then follow that the 40khz tweeter would be able to "fill in" the sound more accurately at 15khz where the 20khz tweeter, not being as "quick" might not fill in as accurately??

In other words, it's like painting a picture with a 4" brush verses a 1" brush.  The 1" (being the 40khz tweeter) will allow you to fine tune more detail than the 4" brush would.

The gist of my logic would be that although you might not hear specifically above 15 or 20khz, that which you would directly hear would (or might be?) more accurate with a tweeter capable of playing to a higher frequency.

??

mdeneen:

Mallette:

Mark:

>The acoustic side is not the same as the electrical side. 

After I sent that it occured to me that, properly recorded, any such interactions would be audible on a CD even if caused by harmonics beyond the range of a CD. 

Is that what you are saying? 

As you know, I have come to believe through experiments to which you've been privy that the limitations of 16/44.1 are not as severe as what we've been served up had us believing.  OTOH, there ARE audible differences in this sampling rate compared to higher resolutions. 

In my case, I know I have little after 13khz in one ear and barely 15khz in the other, so I remain mystified as to why I can hear these differences. 

Regards,

Dave

 

Test 1.

Two speakers:. One with 20khz level response, the other with a rolloff beginning at 15khz. Using an oscillator you confirm that you can only hear up to 14khz on either speaker.  

Test 2.

One pair of speakers with level response to 20khz. Add 2 amplifiers. One that rolls off beginning at 15khz, the other has flat response to 250khz. They are otherwise the same circuit.  Prediction: you will easily hear the difference between these amplifiers, even though you have previously shown that you can't hear above 15khz. 

Mark,

The substitution on the amp side is interesting.

I think I've heard and appreciated differences between tweeters producing sound at frequencies beyond my hearing, but I don't understand why that should be the case.

What sort of thing is going on there?

Coytee:

This seems like a good place to ask this.  I once asked it 20 years ago in college... well...actually 25 years ago (*** time flies!)

If you have a tweeter that will play to 20khz and another that will play to 40khz ( just using round numbers for reference)

Then it seems to me the 40khz tweeter will oscilate faster because it goes higher, right?

So, if the 40khz tweeter can move faster, have a quicker motor & all that stuff...  would it also then follow that the 40khz tweeter would be able to "fill in" the sound more accurately at 15khz where the 20khz tweeter, not being as "quick" might not fill in as accurately??

In other words, it's like painting a picture with a 4" brush verses a 1" brush.  The 1" (being the 40khz tweeter) will allow you to fine tune more detail than the 4" brush would.

The gist of my logic would be that although you might not hear specifically above 15 or 20khz, that which you would directly hear would (or might be?) more accurate with a tweeter capable of playing to a higher frequency.

??

Ask yourself the same question about a woofer.  You may begin to intuit that loudspeaker drivers are bandpass devices. Their response is shaped like an inverted "U". They have their lowest distortion, and highest power output in the center of their passband. To be low distortion at both 15khz and 40khz, such a driver would have to have a response like 10khz to 60khz or something on those lines. To get very HF response the moving mass must become lighter and lighter. To get very good LF response the moving part must become larger and larger. You see the problem---"lighter and lighter" is inconsistent with "larger and larger". Hence, the inverted "U" shape of the pass band.

>I think I've heard and appreciated differences between tweeters producing sound at frequencies beyond my hearing, but I don't understand why that should be the case.

What sort of thing is going on there?

I am with you, DD.  Mark has confirmed the phenomenon, but we still don't know WHY.  It would effectly prove the point about 22khz being too low a cut off point for Redbook if you did this experiment with an amp that did not pass anything beyond 22khz and one that did and were able to clearly tell the difference. 

Sure would like to understand the why of it...

Dave

What the heck is a "super" tweeter? A tweeter that goes 4kHz to 20kHz, or something that goes 8kHz to 40kHz? Some tweeters get the "super" designation from their marketing folks, but there isn't much "super" about them -- the Eminence APT driver that BEC uses is a good example of that. It's a good tweeter, but it's just a tweeter for cry'in out loud. When I think "super tweeter" I think about "bullet" tweeters -- very narrow dispersion -- coming at you like a laser beam out to 100kHz.:) Anyways, a tweeter that goes to 40kHz should have better transient response than one that goes out to 20kHz, so it's not necessarily the extended frequency response one is perceiving (especially if you can't hear past 15kHz). There is also the issue of attenuation -- if the tweeter you substitute or add on is a dB or two higher, you can easily be faked into thinking it's going higher. Filter slope is a big factor too -- if you add a "super" tweeter that goes out to 40kHz to one below it that goes out to 20kHz -- and you do it using a first order filter crossing over at 20kHz -- there is a load of overlapping energy between 10kHz and 20kHz. So again, the difference being heard probably doesn't have much to do with having FR out to 40kHz. Finally, don't forget about the power response and its role in what we hear -- the ratio of on and off-axis energy. 

Leave your speakers alone unless you can test and verify and/or know what you're doing. Replacing caps in a crossover is one thing, changing/adding driver/horns is something else entirely.

"One pair of speakers with level response to 20khz. Add 2 amplifiers. One that rolls off beginning at 15khz, the other has flat response to 250khz. They are otherwise the same circuit.  Prediction: you will easily hear the difference between these amplifiers, even though you have previously shown that you can't hear above 15khz."

Right, but it's not necessarily related directly to FR. 

Mallette:

Sure would like to understand the why of it...

Dave

Narrow band circuits have poor transient response by definition. Wide band circuits are "faster." Faster circuits give the sense of more air, more liveness. Back in the vintage era, Stu Hegaman somewhat pioneered the wideband amp with his Citation designs for HK. Once you heard the Citation 11 preamp (I think it was the 11), there was no going back to the narrow band ways. The Redbook spec was crap when it was conceived and has been frozen at the crap level for 30 years. Meanwhile, analog gear has advanced for those 30 years, so good LPs on an analog system make CDs sound like a mechanical music box. 

Mallette:

I am with you, DD.  Mark has confirmed the phenomenon, but we still don't know WHY.  It would effectly prove the point about 22khz being too low a cut off point for Redbook if you did this experiment with an amp that did not pass anything beyond 22khz and one that did and were able to clearly tell the difference. 

Sure would like to understand the why of it...

The phase response of the steep filters used for anti-aliasing are probably the culprit. Frequency-dependent delays that are created by the filtering tend to smear time-related information that extends into the audible region.

Don

Honk if you love Horns

>The phase response of the steep filters used for anti-aliasing are probably the culprit. Frequency-dependent delays that are created by the filtering tend to smear time-related information that extends into the audible region.

No sir.  There are several situations we've discussed not involving those filters. 

Your thoughts on this, Mark?

Dave 

Sam Tellig reviewed the La scala for Stereophile in November 2006 ( I think ) and found that adding a pair of super tweeters (don't remember brand, am at work now) improved their sound (to him).  He thought very highly of them, and the La Scala is currently listed as a "Class A - limited low frequency extension" speaker by Stereophile. 

Mallette:

>The phase response of the steep filters used for anti-aliasing are probably the culprit. Frequency-dependent delays that are created by the filtering tend to smear time-related information that extends into the audible region.

 

No sir.  There are several situations we've discussed not involving those filters. 

Your thoughts on this, Mark?

Dave 

 

Yeah. My thoughts? Keep it analog, keep it fast, and keep the NFB to the barest minimum = "musical sound". All the rest is just mechanical BS and frankly it doesn't matter what software tricks they are playing-----they are already operating on a corpse.