Hello Panomaniac,
The Radian 950 is an excellent driver for the money and it can be used at quite low frequency.
The compression driver used on the small round horn is the TAD TD2001, unfortunately the match with the KT88 SE amplifier was not so good (also the small round horn appeared to not be a Le Cléac'h horn but having an empirical profile...).
The Bryston was just used at the very end of the afternoon on sunday and replaced a Gemincore class D amplifier.
Best regards from Paris, France
Jean-Michel Le Cléac'h
The Radian 950 is an excellent driver for the money and it can be used at quite low frequency.
The compression driver used on the small round horn is the TAD TD2001, unfortunately the match with the KT88 SE amplifier was not so good (also the small round horn appeared to not be a Le Cléac'h horn but having an empirical profile...).
The Bryston was just used at the very end of the afternoon on sunday and replaced a Gemincore class D amplifier.
Best regards from Paris, France
Jean-Michel Le Cléac'h
panomaniac said:Thanks JM,
nice to see the Onken W cassions in action. Ca me fait de bons souvenirs.
Those horns are, umm.... not small! They seem to have been well liked. So how are those big Radian drivers? They did well? I ask, because I just bought a pair of used Radain 1.4" drivers (at a good price). Hope they sound good.....
What was the driver on the treble horn?
From reading the posts on the Melaudia forum, it seems like you guys need a better treble amp. =) And I didn't know the Bryston was class-d.
Quite a system.
Hello Angelo,
I disagree, this horn possess an acoustical cut-off of 170Hz and as the group delay curve derived from pulse mesurement of many Le Cléac'h horns demonstrated that the group delay becomes negligeible above a frequency 2 times larger than the acoustical cut-off of the horn, we should use that horn above 340Hz.
On sunday 316Hz was used.
Best regards from Paris, France
Jean-Michel Le Cléac'h
I disagree, this horn possess an acoustical cut-off of 170Hz and as the group delay curve derived from pulse mesurement of many Le Cléac'h horns demonstrated that the group delay becomes negligeible above a frequency 2 times larger than the acoustical cut-off of the horn, we should use that horn above 340Hz.
On sunday 316Hz was used.
Best regards from Paris, France
Jean-Michel Le Cléac'h
angeloitacare said:
What if you load the horn with cone driver not a compression driver? Buyers of the Azura 204 /Oris 150, 200 may not be happy to hear that their Lowthers
work decently from up 400Hz to approx. 1K the rest of the band is just a mud (below) and harmonics above (according to dr. Edgar ) ;0). Personally I heard Azura 204 "honking" terribly while doing "fullrange" duty in a big room on certain record. It was only once but evident. I think when one is approaching midbass band compromise is inevitable .
Regards, L
work decently from up 400Hz to approx. 1K the rest of the band is just a mud (below) and harmonics above (according to dr. Edgar ) ;0). Personally I heard Azura 204 "honking" terribly while doing "fullrange" duty in a big room on certain record. It was only once but evident. I think when one is approaching midbass band compromise is inevitable .
Regards, L
Hello Limono,
Being myself very sensible to phase distortion I recommand everytime to use the horns in the frequency range inside which their acoustical impedance is the most resistive and inside which the group delay is negligeible.
I am not responsible of the way people use those horns out of my recommandations.
There was not honk when listening the large horn at Melaudia last sunday with Fc = 316Hz - 18dB/octave.
http://www.diyaudio.com/forums/attachment.php?s=&postid=1791579&stamp=1238665746
When using the same frequency for the high pass srossover but with a slope of 6dB only, the low- mid lacked some presence IMHO.
On saturday they tried lower cut-off frequecny for the high pass crossover but they found that the minimum acceptable frequency was 316Hz which is quite near of 2 times the acoustical cut-off of the horn.
Best regards from Paris, France.
Jean-Michel Le Cléac'h
Being myself very sensible to phase distortion I recommand everytime to use the horns in the frequency range inside which their acoustical impedance is the most resistive and inside which the group delay is negligeible.
I am not responsible of the way people use those horns out of my recommandations.
There was not honk when listening the large horn at Melaudia last sunday with Fc = 316Hz - 18dB/octave.
http://www.diyaudio.com/forums/attachment.php?s=&postid=1791579&stamp=1238665746
When using the same frequency for the high pass srossover but with a slope of 6dB only, the low- mid lacked some presence IMHO.
On saturday they tried lower cut-off frequecny for the high pass crossover but they found that the minimum acceptable frequency was 316Hz which is quite near of 2 times the acoustical cut-off of the horn.
Best regards from Paris, France.
Jean-Michel Le Cléac'h
limono said:
Just to follow up...
The maximum width dimension of the mouth on those horns is?
Did those horns follow your design? If so do you think they were optimal (what you would build or suggest) or different?
And which parameter values do you suggest as optimal?
My casual observation of the picture is that the horns may have been in the wrong physical location if a clean impulse response at the listening position was desired. I would have moved the bass bin back, and the little horn back, essentially making the acoustic centers even in the vertical plane. I am wondering if this was the source of the problems you heard.
Regards,
_-_-bear
PS. are we to take Dr. Geddes statement regarding using the OS throat and the Le Cléac'h expansion the rest of the way as the way to a potentially definitive design??
The maximum width dimension of the mouth on those horns is?
Did those horns follow your design? If so do you think they were optimal (what you would build or suggest) or different?
And which parameter values do you suggest as optimal?
My casual observation of the picture is that the horns may have been in the wrong physical location if a clean impulse response at the listening position was desired. I would have moved the bass bin back, and the little horn back, essentially making the acoustic centers even in the vertical plane. I am wondering if this was the source of the problems you heard.
Regards,
_-_-bear
PS. are we to take Dr. Geddes statement regarding using the OS throat and the Le Cléac'h expansion the rest of the way as the way to a potentially definitive design??
Hello Bear,
The maximum width dimension of the mouth on those horns is?
Width of the low-mid horn has been chosen to fit with the Onken W bas enclosure so it is 113centimeters (44,5 inches).
Did those horns follow your design? If so do you think they were optimal (what you would build or suggest) or different?
The profile of that square horn has been calculated using my spreadsheet.
(the spreadsheet is downloadable for free on my friend Nicolas Davidenko's website: http://ndaviden.club.fr/outils/axial.zip )
Theorically a Le Cléac'h "square mouth" horn should look more as a kind of biradial (both 4 walls will be curved in 2 directions) but as it is very difficult to build by a DIYer, square horns are generally build using curved plywood sheets... (Having listen to several of them I can say that they don't differ a lot in sound from Le Cléac'h round horns.)
And which parameter values do you suggest as optimal?
It depends on what we want as a low frequency limit of use.
If we use a 18dB/octave T ranging from 0,7 to 0,8 give good results but we have to use a high-pass crossover frequency at Fc around 2 times the acoustical cut-off of the horn.
If we can use a very steep slope crossover then using T = 0 (this means a longer horn but the same mouth dimension) allows to use a cutoff frequency of the low-pass crossover very near of the acoustical cut-off of the horn. By example a 170Hz like the one listened at Mélaudia could be used with a steep slope crossover at 250Hz if we designed it with T = 0.
My casual observation of the picture is that the horns may have been in the wrong physical location if a clean impulse response at the listening position was desired. I would have moved the bass bin back, and the little horn back, essentially making the acoustic centers even in the vertical plane. I am wondering if this was the source of the problems you heard.
Time alignment of the loudspeakers was performed (on saturday and not by me) on pulse measurements and the delays were introduced in a Behringer DCX2496.
PS. are we to take Dr. Geddes statement regarding using the OS throat and the Le Cléac'h expansion the rest of the way as the way to a potentially definitive design??
Soongsc is perfoming studies (both simulations and horn buiding + measurements) in order to keep most of the constant directivity characteristics of the OS waveguide, with less effects of diffraction. He used a step by step method , adjusting the profile semiempirically in order to reach progressively that goal. I used to give a look to the profile he published for his best design at a time. I found that for most part of the profile it was nearly a Le Cleac'h profile (except at throat where it behaves more like an OS waveguide or a Quadratic waveguide.). At the moment, for what I know, Soongsc is trying different combination between quasi OS throat and quasi Le Cléac'h mouth. I guess he'll reach his goal and the hybrid waveguide-horn will provide both a constant directivity and smooth soundfield, smooth polar, pure pulse response... and it'll allow an easy CD equalisation.
You have to know that in the last 4 years few audiophiles asked me to calculate a Le Cléac'h mouth for OS waveguides. For what it seems they are happy with the result.
Best regards from Paris, France
Jean-Michel Le Cléac'h
The maximum width dimension of the mouth on those horns is?
Width of the low-mid horn has been chosen to fit with the Onken W bas enclosure so it is 113centimeters (44,5 inches).
Did those horns follow your design? If so do you think they were optimal (what you would build or suggest) or different?
The profile of that square horn has been calculated using my spreadsheet.
(the spreadsheet is downloadable for free on my friend Nicolas Davidenko's website: http://ndaviden.club.fr/outils/axial.zip )
Theorically a Le Cléac'h "square mouth" horn should look more as a kind of biradial (both 4 walls will be curved in 2 directions) but as it is very difficult to build by a DIYer, square horns are generally build using curved plywood sheets... (Having listen to several of them I can say that they don't differ a lot in sound from Le Cléac'h round horns.)
And which parameter values do you suggest as optimal?
It depends on what we want as a low frequency limit of use.
If we use a 18dB/octave T ranging from 0,7 to 0,8 give good results but we have to use a high-pass crossover frequency at Fc around 2 times the acoustical cut-off of the horn.
If we can use a very steep slope crossover then using T = 0 (this means a longer horn but the same mouth dimension) allows to use a cutoff frequency of the low-pass crossover very near of the acoustical cut-off of the horn. By example a 170Hz like the one listened at Mélaudia could be used with a steep slope crossover at 250Hz if we designed it with T = 0.
My casual observation of the picture is that the horns may have been in the wrong physical location if a clean impulse response at the listening position was desired. I would have moved the bass bin back, and the little horn back, essentially making the acoustic centers even in the vertical plane. I am wondering if this was the source of the problems you heard.
Time alignment of the loudspeakers was performed (on saturday and not by me) on pulse measurements and the delays were introduced in a Behringer DCX2496.
PS. are we to take Dr. Geddes statement regarding using the OS throat and the Le Cléac'h expansion the rest of the way as the way to a potentially definitive design??
Soongsc is perfoming studies (both simulations and horn buiding + measurements) in order to keep most of the constant directivity characteristics of the OS waveguide, with less effects of diffraction. He used a step by step method , adjusting the profile semiempirically in order to reach progressively that goal. I used to give a look to the profile he published for his best design at a time. I found that for most part of the profile it was nearly a Le Cleac'h profile (except at throat where it behaves more like an OS waveguide or a Quadratic waveguide.). At the moment, for what I know, Soongsc is trying different combination between quasi OS throat and quasi Le Cléac'h mouth. I guess he'll reach his goal and the hybrid waveguide-horn will provide both a constant directivity and smooth soundfield, smooth polar, pure pulse response... and it'll allow an easy CD equalisation.
You have to know that in the last 4 years few audiophiles asked me to calculate a Le Cléac'h mouth for OS waveguides. For what it seems they are happy with the result.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Jmmlc said:
I made them using rather soft cardboard to get basic shape and than reinforcing it with fiberglass. Gluing cardboard edges gets you the shape so no need for fancy jigs etc. - simple. It turned out pretty nice and was quite easy. And with Altec 288 it sounds wonderful.
2way Altec/Le Cléac'h
J-M,
What happens acoustically and electrically if you use a 1st order on a horn with a T ~0.8 range??
Subjectively, does the phase shift/group delay effect the imaging? If so, how so?
Let us assume the driver's ability to do this is not an issue.
Regarding the Behringer - imho this is akin to running a Ferrari using tires filled with concrete! (nice image, eh?) While introducing the delay in digital is something that I have aspired to for a long time, I think I would prefer some means that is far far closer to have a sonic neutrality.
What I would like is a DAC with three DACs per channel of non-DSP delay, adustable delay for 2 of the 3 channels. Not being a digitally savvy designer, I have not found a good way to do this yet. Boy would that solve a ton of problems for us!
Anyhow, I am looking forward to seeing what Soongc comes up with. It looks very promising.
I agree that the RD (Rectangular Dispersion?) of Dr. Geddes design is not always useful or desireable. I don't want to call that CD, since it is an ambiguous term. Actually, I am thinking that it ought to be called RD with a number = degrees of useable dispersion. So the Geddes OS might be RD30 (I think it was 30 degrees) for example. (...you heard it here first!!)
In my room a 30 degree HF dispersion is not an advantage.
It is a disadvantage.
If I was trying to cover in a wide and deep room for many people, then it might be a benefit to have wide HF dispersion.
Also, I prefer having my speakers cross at or in front of my listening position. I find it yields a superior stereo image and depth compared to the "straight ahead" positioning.
--------------
vuki, nice work!
Pretty - nice copper color too!
After you did the cardboard layup, what did you do to thicken/stiffen the entire assembly other than a coating of epoxy resin?? Did you lay up multiple layers of cardboard?
How did you create the throat area & horn mounting??
_-_-bear
What happens acoustically and electrically if you use a 1st order on a horn with a T ~0.8 range??
Subjectively, does the phase shift/group delay effect the imaging? If so, how so?
Let us assume the driver's ability to do this is not an issue.
Regarding the Behringer - imho this is akin to running a Ferrari using tires filled with concrete! (nice image, eh?) While introducing the delay in digital is something that I have aspired to for a long time, I think I would prefer some means that is far far closer to have a sonic neutrality.
What I would like is a DAC with three DACs per channel of non-DSP delay, adustable delay for 2 of the 3 channels. Not being a digitally savvy designer, I have not found a good way to do this yet. Boy would that solve a ton of problems for us!
Anyhow, I am looking forward to seeing what Soongc comes up with. It looks very promising.
I agree that the RD (Rectangular Dispersion?) of Dr. Geddes design is not always useful or desireable. I don't want to call that CD, since it is an ambiguous term. Actually, I am thinking that it ought to be called RD with a number = degrees of useable dispersion. So the Geddes OS might be RD30 (I think it was 30 degrees) for example. (...you heard it here first!!)
In my room a 30 degree HF dispersion is not an advantage.
It is a disadvantage.
If I was trying to cover in a wide and deep room for many people, then it might be a benefit to have wide HF dispersion.
Also, I prefer having my speakers cross at or in front of my listening position. I find it yields a superior stereo image and depth compared to the "straight ahead" positioning.
--------------
vuki, nice work!
Pretty - nice copper color too!
After you did the cardboard layup, what did you do to thicken/stiffen the entire assembly other than a coating of epoxy resin?? Did you lay up multiple layers of cardboard?
How did you create the throat area & horn mounting??
_-_-bear
Hello Wuki,
Beautiful set-up, that gold color is very elegant!
You'll see in attached file another cardboard Le Cléac'h horn, "gramophone style" with 14 facets, calculated with my spreadsheet and built buy a French audiophile. (After the assembly the rear of the horn was rigidified using fiber glass + gelcoat).
Best regards from Paris, France
Jean-Michel Le Cléac'h
Beautiful set-up, that gold color is very elegant!
You'll see in attached file another cardboard Le Cléac'h horn, "gramophone style" with 14 facets, calculated with my spreadsheet and built buy a French audiophile. (After the assembly the rear of the horn was rigidified using fiber glass + gelcoat).
Best regards from Paris, France
Jean-Michel Le Cléac'h
vuki said:
Attachments
Thank you! 
The cardboard horns were reinforced by layers of fiberglass and polyester resin on the back side. Finish is "old gold" - schlag metal gold gilded with polyurethan (with a bit of brown stain) coating.
Throat area (square to round transition) is done with a bit of polyester putty and the driver mount is done with a plywood plate bonded to horn with fiberglass and resin.
The cardboard horns were reinforced by layers of fiberglass and polyester resin on the back side. Finish is "old gold" - schlag metal gold gilded with polyurethan (with a bit of brown stain) coating.
Throat area (square to round transition) is done with a bit of polyester putty and the driver mount is done with a plywood plate bonded to horn with fiberglass and resin.
bear said:
Hi Bear
Is this what you have in mind ?
http://hifiakademie.de/?id=3&si=MTIzOTExMjk2NS41MjUyfDgwLjE2NS4xNTcuMTA5fCA
Its not cheap though, if you want a full blown 3way active preamp
And some have even opposed that its poor quality
Looks good to me though
Hello Bear,
When the frequency decreases toward the acoustical cut-off of the horn, the phase rotation accelerates and the group delay increases.
This has to be taken in account because a high pass crossover is generally used with a horn. If we do'nt take in account the phase behavior of the horn then the theorical target function for the crossover will not be reached.
Consider the acoustical phase curve of Le Cléac'h horns having the same cut-off (e.g. 160Hz) but different T values (see attached graph).
As you can see for a given "limit phase" of 20° the lowest usable frequency is 320Hz for a T = 0.8 Le Cléac'h horn and a bit lower than 200Hz fro T = 0.
I guess I could agree about the DCX2496 but in fact good results may be obtain with a DCX2496 if we feed its input with the correct level signal. Remember that the input of the DCX2496 clips with 10Volts RMS sine signals when most CD drivers delivers only 2Volts RMS max at their output. Using 4:1 transformers at the inputs lead to a better sound.
At Mélaudia the room is quite large as you can see on the pictures here (hope you can see them).
http://picasaweb.google.com/sralbumsphotos/MelaudiaReuilMars2009?feat=directlink#
And using the large Le Cléac'h horn made by Frank Delbauve, nobody complaint about directivity and it was the very first time the room was not a very limiting factor.
Best regards from Paris, France
Jean-Michel Le Cléac'h
When the frequency decreases toward the acoustical cut-off of the horn, the phase rotation accelerates and the group delay increases.
This has to be taken in account because a high pass crossover is generally used with a horn. If we do'nt take in account the phase behavior of the horn then the theorical target function for the crossover will not be reached.
Consider the acoustical phase curve of Le Cléac'h horns having the same cut-off (e.g. 160Hz) but different T values (see attached graph).
As you can see for a given "limit phase" of 20° the lowest usable frequency is 320Hz for a T = 0.8 Le Cléac'h horn and a bit lower than 200Hz fro T = 0.
I guess I could agree about the DCX2496 but in fact good results may be obtain with a DCX2496 if we feed its input with the correct level signal. Remember that the input of the DCX2496 clips with 10Volts RMS sine signals when most CD drivers delivers only 2Volts RMS max at their output. Using 4:1 transformers at the inputs lead to a better sound.
At Mélaudia the room is quite large as you can see on the pictures here (hope you can see them).
http://picasaweb.google.com/sralbumsphotos/MelaudiaReuilMars2009?feat=directlink#
And using the large Le Cléac'h horn made by Frank Delbauve, nobody complaint about directivity and it was the very first time the room was not a very limiting factor.
Best regards from Paris, France
Jean-Michel Le Cléac'h
bear said:
Attachments
panomaniac said:
Sorry, you'll have a very very difficult time convincing me that anything Behringer at this point in time is going to be truly high quality and transparent when listened to with an otherwise low distortion and high resolution (for lack of better terminology) system.
Perhaps if one rebuilt the thing it might improve... I don't know.
I do know it is all teeny tiny surface mount, and we don't have a schematic.
_-_-bear
If I had the opportunity to audition a modified one, I'd give much more credence to the idea that a modified Behringer DSP engine could sound good enough to meet my personal standards. I'm not saying it can't or it doesn't.
So far I have yet to hear a DSP that I liked... doesn't mean that it won't happen.
For a long time I had not heard a single horn that I liked - until I heard one, and figured out what the problem had been. I'll keep that to myself for now.
Now I be use horns mee too.
I'm supposed to hear a new ESS Sabre DAC in the next few weeks, that's very DSP based inside, but ur stuck with their everything since it is an all in one chip type of solution. If that blows me away, I'll start to think that there is hope yet for the DSP way.
The biggest problem is that the delay is required on the HF, if you add in that last bit of >10kHz via a super tweeter, that is where you really need the delay, and that is hyper-critical in terms of fidelity, imho. So, it is the worst place to have to use a DSP that is less than pristine in all regards. Otoh, a delay on the bass (below 300Hz or lower) I can probably live with a Behringer there.
The system in my LR is pretty darn good at highlighting subtle differences between almost anything and everything that is plugged in the signal path, so nothing that is actually problematic is going to be glossed over here. That makes me very cautious about making a leap of faith on things like a Behringer product.
Anyhow I hope that makes some sense to someone...
_-_-bear
So far I have yet to hear a DSP that I liked... doesn't mean that it won't happen.
For a long time I had not heard a single horn that I liked - until I heard one, and figured out what the problem had been. I'll keep that to myself for now.
Now I be use horns mee too.
I'm supposed to hear a new ESS Sabre DAC in the next few weeks, that's very DSP based inside, but ur stuck with their everything since it is an all in one chip type of solution. If that blows me away, I'll start to think that there is hope yet for the DSP way.
The biggest problem is that the delay is required on the HF, if you add in that last bit of >10kHz via a super tweeter, that is where you really need the delay, and that is hyper-critical in terms of fidelity, imho. So, it is the worst place to have to use a DSP that is less than pristine in all regards. Otoh, a delay on the bass (below 300Hz or lower) I can probably live with a Behringer there.
The system in my LR is pretty darn good at highlighting subtle differences between almost anything and everything that is plugged in the signal path, so nothing that is actually problematic is going to be glossed over here. That makes me very cautious about making a leap of faith on things like a Behringer product.
Anyhow I hope that makes some sense to someone...
_-_-bear