Phase is the issue. The audibility is not to be underestimated. Either do it perfectly or leave it alone (forget the super tweeter). And I believe there are situations where it is just cannot be done. May be similar with adding another woofer (e.g. subwoofer) to an already working 2-way.
Duelund: It is believed, that one can divide another place in the frequency band and use the same type of filter once more to create a 3-way system. So it is done in the textbooks, but it is very wrong. Every single unit must "know" the existence of the others.
I tend to agree with this but I imagine the medium format drivers they use with this 425 horn in this project probably lacks sufficient HF energy off axis so it probably lacks the last bit of realism and leaves the listener a bit disenchanted. The solution could be a treble horn but as I said before good luck tuning it in for acceptable power response with the little midrange (actually large physically do to the round over!) horn. With that small of a horn a 1" throat driver makes better sense where the midrange can be padded down with possibly better HF respose - yet that still doesn't fix the power response 😕
No, not yet. Still a work in progress 😎. Not trying to hide anything, but I hate to post information that becomes outdated shortly afterwards. I will be happy to share more when I'm done.
Marco,
Yes if you looked the phase alignment was obviously not correct, but I couldn't hear anything that bothered me. I will say I heard things in the music that I had never been able to identify before with even the TAD 2001 driver. The resolution of things like a cymbal crash was astounding. I don't remember what I was listening to but a tune that I had heard many times before finally came into focus. I could actually hear that I was listening to a small triangle being hit, it was not in fact a cymbal splash, it was a revelation to say the least. I think obviously it is nice to get all the time alignment correct and the phase angle at crossover perfect, but I have a real question that we can really hear all of this, perhaps with very simple music with a couple of instruments but once we go to more dense music with many things happening I don't really think it is anything to worry so much about.
Now since this was now about 20 years ago can I even hear that high anymore, I don't know, I haven't checked my high frequency hearing for about 7 years now. I still could back then.
Yes if you looked the phase alignment was obviously not correct, but I couldn't hear anything that bothered me. I will say I heard things in the music that I had never been able to identify before with even the TAD 2001 driver. The resolution of things like a cymbal crash was astounding. I don't remember what I was listening to but a tune that I had heard many times before finally came into focus. I could actually hear that I was listening to a small triangle being hit, it was not in fact a cymbal splash, it was a revelation to say the least. I think obviously it is nice to get all the time alignment correct and the phase angle at crossover perfect, but I have a real question that we can really hear all of this, perhaps with very simple music with a couple of instruments but once we go to more dense music with many things happening I don't really think it is anything to worry so much about.
Now since this was now about 20 years ago can I even hear that high anymore, I don't know, I haven't checked my high frequency hearing for about 7 years now. I still could back then.
While doing a hearing test using individual frequencies this is true, but when you take away while listening to sounds we are familiar with, it is quite easy to detect. But then there are also design trade offs to consider when looking at the total speaker design. For horns that operate below 7KHz, I would imagine that some equalization would be a better choice, but that also depends on the type of horn. Difficult decisions.
It's a matter of preference. I was never able to get as much musical realism with a small format driver.
I believe that an AH425 could just as easily be made to fit a 1" driver; it would just be a bit longer. Would its HF dispersion really be all that much greater?
Gary Dahl
I believe that an AH425 could just as easily be made to fit a 1" driver; it would just be a bit longer. Would its HF dispersion really be all that much greater?
Gary Dahl
If some of use older guys still have upper frequency hearing then in my eyes the 1.4" and 1.5" and definitely the 2" drivers are just not going to cut it. This larger driver thinking has been going on for to long. If you are not doing a PA system I would never use those large format drivers. The 1" drivers can just as easily get down low enough with the limited output you are asking of them in a home environment and then you have much greater high frequency response. It has always been this way, I just don't understand why people are still chasing this poor selection of compression drivers. If you are truly using a compression driver down to 700 hz any good 1" driver should do that in a home system. I'd much rather have the extra high frequency response of the smaller driver than a driver that can get down to 300hz, there is no real reason to do that. We surely aren't still building A7's are we?
Gary you could fairly easily make an adapter to fit the smaller driver. The only real difference is going to be the increased high frequency response. You are never going to ger a 1.4 or 1.5" driver to go much above 12Khz without some serious eq and cutting the bottom in a passive xo and you will never get clean sound up to 16khz in my opinion.
Yup, seen them on Ebay and they are nothing short of beautiful!
What bothers me though is the rough throat, which makes me think that they are more a thing of visual beauty than musical enjoyment.
Ceramic Multicell Horns for 1" HF Driver Altec JBL and Others | eBay
/Anders
What bothers me though is the rough throat, which makes me think that they are more a thing of visual beauty than musical enjoyment.
Ceramic Multicell Horns for 1" HF Driver Altec JBL and Others | eBay
/Anders
Phase is actually widely reported to be virtually inaudible beyond ~5kHz.
In other words, what matters at mid frequencies is different from what matters at very high frequencies.
To superTw or not to superTw?
The fact is that no solution is perfect, there are always trade-offs.
1) Pushing a compression driver into service for the full high frequency spectrum, including VHFs (> 7 kHz or so), does away with the crossover issues (time alignment, relative phase, comb filtering); however, the VHFs are produced by the drivers's break-up modes, and they 'beam'. The onset of break-up can be pushed higher in frequency by using a small format (1") driver and/or a Be diaphragm, and the 'beaming' issue may be ameliorated by using an OS waveguide, but neither issue can be completely 'solved'.
2) OTOH, using a good supertweeter results in less distortion and more resolution at VHFs, and can largely solve the 'beaming' issue. The flip side of the coin is that the necessary crossover produces the known issues mentioned above.
The question, as I see it, is which of the two options results in audibly (as opposed to visually, on a graph) better sound?
Admittedly, this may depend at least in part on subjective preference.
For instance, it seems as if Gary Dahl and GedLee are happier to settle for compromise 1), while Kindhornann and myself (and maybe also Lynn Olson?) prefer compromise 2).
Which option is preferred may also depend on the specific type of horn/waveguide employed.
For instance, largish axially symmetric horns like the AH-425 used by Gary Dahl and the OS waveguide used by Dr. Geddes force one to place a supertweeter at a larger distance from the acoustic centre of the compression driver, and this may make the integration more difficult (i.e. less audibly transparent).
Interesting stuff!
Marco
The fact is that no solution is perfect, there are always trade-offs.
1) Pushing a compression driver into service for the full high frequency spectrum, including VHFs (> 7 kHz or so), does away with the crossover issues (time alignment, relative phase, comb filtering); however, the VHFs are produced by the drivers's break-up modes, and they 'beam'. The onset of break-up can be pushed higher in frequency by using a small format (1") driver and/or a Be diaphragm, and the 'beaming' issue may be ameliorated by using an OS waveguide, but neither issue can be completely 'solved'.
2) OTOH, using a good supertweeter results in less distortion and more resolution at VHFs, and can largely solve the 'beaming' issue. The flip side of the coin is that the necessary crossover produces the known issues mentioned above.
The question, as I see it, is which of the two options results in audibly (as opposed to visually, on a graph) better sound?
Admittedly, this may depend at least in part on subjective preference.
For instance, it seems as if Gary Dahl and GedLee are happier to settle for compromise 1), while Kindhornann and myself (and maybe also Lynn Olson?) prefer compromise 2).
Which option is preferred may also depend on the specific type of horn/waveguide employed.
For instance, largish axially symmetric horns like the AH-425 used by Gary Dahl and the OS waveguide used by Dr. Geddes force one to place a supertweeter at a larger distance from the acoustic centre of the compression driver, and this may make the integration more difficult (i.e. less audibly transparent).
Interesting stuff!
Marco
Last edited:
The book didn't say anything about "phase inaudibility beyond 5kHz".
The book talked about "phase locking", e method how ears (of living things) decodes sound frequency. Experiments were mostly with non-human and this is just an hypothesis.
Even without knowing about hair-cell of the nuchlea anatomy, we can still understand from real-world phenomena that the higher the frequency, the harder we can decode the information. This works for human and animals. But "where" and "how" the threshold is, is UNCLEAR.
Here is a perfect statement from the book (which is my belief): It remains possible that perceptual degradation at HF reflects lack of experience with such frequencies, or their relative unimportance for typical behavioral judgments, rather than a physiological limitation.
So our body (and brain) has been designed such that it will develop itself the way we develop them, intentionally or forced by circumstances.
In the ancient world, no circumstances is available for human to develop their phase acuity. But in a modern world, we have complex music where timing is important. An orchestral conductor for example, might have been forced by his activity to train his ears to hear phase issues.
Many of us here, as audio DIYers who have searched for audio perfection for years, have also trained our ears such that I believe no human who has no access to music has the same ability to decode phase information from a sound.
Exactly the key point!
I have mentioned about subwoofer as analogy to the super tweeter. Many of us here will be happy with any commercial REL or Sonus Faber or even Sony subwoofers. But a few select others (and you should be able to differentiate this two) have never been able to accept subwoofers.
Subwoofer is very important because without sufficient LF, we have lost too many musical information from a recording. Super tweeter as a comparison? Not many critical musical information up there!
GedLee has stated his position. Perfect implementation of super tweeter is impossible so forget it, and make sure the HF horn driver can reach as high as possible frequency.
Many of us here are perfectionists, but it can happen only if we can hear the difference.
All homes ain't the same. Of course the larger drivers have advantages, using the small format drivers below around 1.5K compared to using a good medium or large format driver is just as compromised as expecting uncompromised playback from the larger drivers in the upper treble. 🙄
Gary.
For you in particular unless you personally are unhappy with the selections you have already made and want to chase every change possible I say forget about making more changes for change sake. You already have sorted the speakers and the electronics and are more than happy it seems.
For others who are not there yet I would say that I prefer the 1" drivers over the larger format drivers for the reasons I have given. There are many 1" drivers in two basic styles, the long throat style that has been around for ever from Altec, JBL, WE, ElectroVoice, Yamaha and others. The other options is the pancake style with much shorter throat and greater exit angle allowing wider dispersion. Radian and others such as Emilar and Rankus Heinz were early adopters of this style. What I don't like about some of these, actually all of the earlier designs is the use of low quality ceramic magnets, there is no reason except for price that Neo or AlNiCo magnets aren't used. I'm not sure that Radian has a Be diaphragm for these smaller drivers or if they ever will have those or better magnets. Perhaps if asked Materion will make a replacement diaphragm? There are just so many drivers to choose from, many made in Europe also such as B&C, RCF and others, the list goes on.
Bottom line it is just much easier to have a wider dispersion angle and high frequency response while meeting the lower frequency demands if not pushed to an absurd lower frequency. I don't understand why anyone would want a compression driver to go as low as 300 cycles, that just seems to be a mis-use of the devices.
On the question of phase alignment and its brother time alignment, really the same thing, I say today that is not an issue if you are willing to use modern electronics. FIR and IIR filters using modern DSP and DAC solutions are here, are viable and take care of many issues that in the past had no easy answers. I don't understand those using horns in the raw, by that I mean not in an enclosure that helps to smooth the response if we are using more than one horn lens. Stacking round horns one upon the other with the higher horn set back and looking at the outside of the larger horn is not a smart move, you have solved no problems this way, you have made things worse. Align the front edges to eliminate these issues and electronically correct the phase response, that is the wiser solution.
I just do not want to throw out the baby with the bathwater and say who cares about the upper octave or more of the output in the FR. Until you hear the extreme hgh frequencies and realize the real loss of those frequencies all we have are the studies that say we don't need those frequencies.Bose would tell you that you don't notice what is missing, I think we know better than that. I personally don't believe those results and have heard the difference, we hear in a range that is larger than is commonly accepted. And it may be that we do this with sensory perception that is beyond our understanding of our ears and brains. Don't forget that some blind people can seem to use echo location to find there way around.
On those very interesting looking multi-cell horns I have read some of the threads where people are attempting to reproduce the old Altec horns. I think there are so many flaws in their design that I don't recommend them, the physical mismatch in the throat itself would make me just say NO! Now if someone wants to take the time and actually study how this could be done. I have, you could redesign these ideas and make a much better device, I won't go into that here but I have had ideas about that for 40 years that would take some very advanced tooling techniques to correct the very obvious compromises that were originally overlooked in-order to be able to produce these horns from the sheet metal used so long ago.
ps. POOH, I myself don't use a compression driver lower than 1.5Khz, I don't know if we have come to the same conclusions for the same reason but that is my choice. I just happen to like a cone driver horn for those frequencies lower than that, I just can't get past whatever it is that makes a compression driver below that just sound like a PA to me, a very distinctive sound. I have used cones from 10" to 6 1/2" cones to cover this range. I was the first that I know of in 1976 in Pro Audio to use a 10" mid horn rather than what everyone else was trying to do with 12" drivers, the beaming was not desirable with the larger cones. 10" was the upper size limit in my eyes for satisfactory loading and dispersion. The outlier was the attempt by Community Light and Sound, their original name, when they tried to make a very large compression driver device, it didn't really work all that well but was an interesting exercise.
For you in particular unless you personally are unhappy with the selections you have already made and want to chase every change possible I say forget about making more changes for change sake. You already have sorted the speakers and the electronics and are more than happy it seems.
For others who are not there yet I would say that I prefer the 1" drivers over the larger format drivers for the reasons I have given. There are many 1" drivers in two basic styles, the long throat style that has been around for ever from Altec, JBL, WE, ElectroVoice, Yamaha and others. The other options is the pancake style with much shorter throat and greater exit angle allowing wider dispersion. Radian and others such as Emilar and Rankus Heinz were early adopters of this style. What I don't like about some of these, actually all of the earlier designs is the use of low quality ceramic magnets, there is no reason except for price that Neo or AlNiCo magnets aren't used. I'm not sure that Radian has a Be diaphragm for these smaller drivers or if they ever will have those or better magnets. Perhaps if asked Materion will make a replacement diaphragm? There are just so many drivers to choose from, many made in Europe also such as B&C, RCF and others, the list goes on.
Bottom line it is just much easier to have a wider dispersion angle and high frequency response while meeting the lower frequency demands if not pushed to an absurd lower frequency. I don't understand why anyone would want a compression driver to go as low as 300 cycles, that just seems to be a mis-use of the devices.
On the question of phase alignment and its brother time alignment, really the same thing, I say today that is not an issue if you are willing to use modern electronics. FIR and IIR filters using modern DSP and DAC solutions are here, are viable and take care of many issues that in the past had no easy answers. I don't understand those using horns in the raw, by that I mean not in an enclosure that helps to smooth the response if we are using more than one horn lens. Stacking round horns one upon the other with the higher horn set back and looking at the outside of the larger horn is not a smart move, you have solved no problems this way, you have made things worse. Align the front edges to eliminate these issues and electronically correct the phase response, that is the wiser solution.
I just do not want to throw out the baby with the bathwater and say who cares about the upper octave or more of the output in the FR. Until you hear the extreme hgh frequencies and realize the real loss of those frequencies all we have are the studies that say we don't need those frequencies.Bose would tell you that you don't notice what is missing, I think we know better than that. I personally don't believe those results and have heard the difference, we hear in a range that is larger than is commonly accepted. And it may be that we do this with sensory perception that is beyond our understanding of our ears and brains. Don't forget that some blind people can seem to use echo location to find there way around.
On those very interesting looking multi-cell horns I have read some of the threads where people are attempting to reproduce the old Altec horns. I think there are so many flaws in their design that I don't recommend them, the physical mismatch in the throat itself would make me just say NO! Now if someone wants to take the time and actually study how this could be done. I have, you could redesign these ideas and make a much better device, I won't go into that here but I have had ideas about that for 40 years that would take some very advanced tooling techniques to correct the very obvious compromises that were originally overlooked in-order to be able to produce these horns from the sheet metal used so long ago.
ps. POOH, I myself don't use a compression driver lower than 1.5Khz, I don't know if we have come to the same conclusions for the same reason but that is my choice. I just happen to like a cone driver horn for those frequencies lower than that, I just can't get past whatever it is that makes a compression driver below that just sound like a PA to me, a very distinctive sound. I have used cones from 10" to 6 1/2" cones to cover this range. I was the first that I know of in 1976 in Pro Audio to use a 10" mid horn rather than what everyone else was trying to do with 12" drivers, the beaming was not desirable with the larger cones. 10" was the upper size limit in my eyes for satisfactory loading and dispersion. The outlier was the attempt by Community Light and Sound, their original name, when they tried to make a very large compression driver device, it didn't really work all that well but was an interesting exercise.
Last edited:
I kinda like the JBL CMCD (cone compression driver) and the medium size Community M200 for midrange. All others sound thin and smeared (strained?) in the lower midrange to me compared to a horn loaded cone- I switched to a front horn loaded Tannoy 12" dual concentric a couple of months ago and am pretty satisfied with that for now. 😛