The frequency of the horn loaded woofers (not talking about the mids) is limited to a frequency where the wavelength is less than the circumference of the conical horn's mouth. This all but eliminates any mouth diffraction because the mouth is acoustically large compared to the wavelength that it is passing. The frequencies below this frequency are handled by the bass reflex ports which are too far forward in the horn to be horn loaded. Lastly, Danley puts an extra wide angle ending mouth flare at the end of the horn for good measure.
If there is an issue with diffraction, it comes higher in frequency than the woofers... I was more focused on the effects upon the upper mids and highs.
It is the sort of "extra wide angle ending mouth flare at the end of the horn" that is of concern to me, if the upper frequencies are "riding" along the horn walls, anything like that will look like a discontinuity.
It is the sort of "extra wide angle ending mouth flare at the end of the horn" that is of concern to me, if the upper frequencies are "riding" along the horn walls, anything like that will look like a discontinuity.
I'm listening to the Geddes Summas while I type this post, and I've built a bunch of Unity horns. From the perspective of the work that Geddes does, the designs diverge in a few ways:
1) Most listeners focus a lot of attention at the throat of the horn, but I'd say the main difference is the cabinet actually. IMHO the main reason that my Summas 'disappear' is that the cabinet doesn't have a sharp edge *anywhere.* Even the back of the cabinet is designed to reduce diffraction, and this makes a difference. Compare this to the Lambda Unity horns with their rough edges.
2) The curve at the throat isn't perfect, but it's not terrible either. If you look at the polar curves of the SH-50* you'll see their as good as it gets, and I'd guess that they wouldn't be that good if there were substantial issues at the throat. Just my opinion of course.
3) IMHO, higher order modes will be generated in the throat because the midrange energy doesn't go strictly down the axis of the horn; in fact the midranges aren't even *mounted* on the center of the horn, so higher order modes are a given. Is that audible? I don't know; I haven't listened to both designs enough to offer a valid opinion. When I've listened to the designs on the same day, I've found that the Unity wins in the articulation department, and the Summa wins in the imaging department. But the Summa's imaging is partially due to the cabinet, and the Unity cabinet could easily adopt the Summa's shape.
* download the CLF viewer and the CLF files from danleysoundlab.com to see the polar plots
The high frequencies do not "ride" along the horn walls. Those wavelengths are too acoustically small to even interact with the horn wall at the mouth.
In Toole's book, he explicitly states that diffraction looks bad on paper, but does not sound as bad. In fact, any type of comb filtering is not perceived the way it looks on the measurements. Floor bounce, for example, produces a comb filtered response, but we don't hear it that way.
Speakers disappearing is primarily down to good phase overlap through the crossover, i.e., proper crossover design, and small center to center spacing of the drivers. You can put two identical speakers side by side, one with curved sides and another with sharp edges, and I bet you won't hear the difference in a blind test. That's just my opinion.
All the reports I have heard about the Synergies are unanimous in their opinion that the stereo imaging is far superior to conventional directional speakers. It's down to the smooth crossover and near-point source behavior of the speaker.
Speakers disappearing is primarily down to good phase overlap through the crossover, i.e., proper crossover design, and small center to center spacing of the drivers. You can put two identical speakers side by side, one with curved sides and another with sharp edges, and I bet you won't hear the difference in a blind test. That's just my opinion.
All the reports I have heard about the Synergies are unanimous in their opinion that the stereo imaging is far superior to conventional directional speakers. It's down to the smooth crossover and near-point source behavior of the speaker.
Imo, that would depend on a few things.
What frequencies?
Is there any diffraction or other diffraction causing discontinuity at or near the throat?
Why do the JMMC horns (even ones like 200Hz horns) show a smoother response with very large "lips" at high frequencies if there is no effect at the mouth?
_-_-
Honk
All of which is part of the field information we use for location and "sense of the space". It may not sound bad but it impacts on the sense of realism (which is why everyone recommends damping floor bounce with a carpet)
The beyond the ariel thread has a number of discussions on the role of comb filtering in creating "horn honk".
BTW There's a nice experiment with a magazine and a towel on "horn honk" somewhere on the interweb
All of which is part of the field information we use for location and "sense of the space". It may not sound bad but it impacts on the sense of realism (which is why everyone recommends damping floor bounce with a carpet)
The beyond the ariel thread has a number of discussions on the role of comb filtering in creating "horn honk".
BTW There's a nice experiment with a magazine and a towel on "horn honk" somewhere on the interweb
Yesterday I whipped up a crossover for a new build in a couple of hours...so that I could listen to music while I did it properlyThe speakers still disappear. The vertical spacing isn't too small either.
Maybe your crossover isn't that bad after all.
How do you explain the disappearance of small speakers, such as the Harbeth monitors, or take Jeff B's Continuums for that matter. There is no rounding of the edges. Besides, there have been many counter-studies that showed that the radius that is typically achieved, like the one on the Summas, has little to no impact on smoothing the abrupt change at the edge.
Now, if we are talking about the LeCleach roundover, yes, that will have a significant impact. Especially on a horn, where it will prevent mouth reflections. That is almost a complete 180 deg roundover that is perfectly... round. That is not to be confused with the puny roundover on a flat baffle.
We are much more sensitive to errors in the crossover, bumpy response, and drivers calling attention to themselves, than we are to diffraction from baffle edges.
Last edited:
Can you provide a lead on these?
I concur with your comments on small monitors, I've always presumed some mechanism was keeping Q down and htis combined with the short distance made it less noticable.
Agreed. Horn Honk is the extreme form of "the diffraction problem" which is usually audible.
I've not heard the Danley horn but all reports indicate that it does not suffer from this at all.
Last edited:
You're trying to compare apples to oranges. The JMMC horns use a single compression driver across a broad bandwidth. The Synergy horns use multiple drivers with a specific crossover that limits the acoustic image of each driver. This segregation of frequencies prevent the high frequency output from the compression driver from acoustically seeing, or riding along the horn wall once the horn circumference has reached a certain cross sectional area.
The expansion of a conical horn is very rapid close to its apex (throat) doubling its expansion in a very short distance (as an exponential horn would). As the distance from the apex is increased the *expansion rate* essentially decreases ~ it takes a longer and longer distance down the horn from any given point in the horn for the area to double. In essence, for every point along the axis of a conical horn there is a different expansion rate, and in turn, a different cut-off frequency. By utilizing this effect a conical horn can be driven at the appropriate place along its length to yield good low frequency loading of the driver. Therefore the lower frequency drivers are placed closer to the mouth of the horn, where the horn provides better loading within the frequency region (pass band) where the drivers will be used.
From Danley Sound Labs White Paper on Tapped Horn and Synergy Horn Technologies
early in the thread comments on XO seem to miss the delay compensated LR Lipshitz & Vanderkooy XOs
cited by Linkwitz Crossovers
also elsewhere I belive there was a call for other "unity/synergy" like products http://www.diyaudio.com/forums/multi-way/207440-jbls-take-unity-horn.html
how about: http://www.acoustic-line.de/G3_Datenbl_engl.pdf
do any know the "air mass reflector" patents - I couldn't find it in Seeburg uspto search
cited by Linkwitz Crossovers
also elsewhere I belive there was a call for other "unity/synergy" like products http://www.diyaudio.com/forums/multi-way/207440-jbls-take-unity-horn.html
how about: http://www.acoustic-line.de/G3_Datenbl_engl.pdf
do any know the "air mass reflector" patents - I couldn't find it in Seeburg uspto search
Last edited:
Member
Joined 2009
Paid Member
Can't help you there jcx, but I'll repeat others that there would be a lot of interest in a home version of the Synergy Horn. Maybe Tom's business partner would like to consider the home theatre market - it's large and people will spend. A DIY version would be wonderful too, just offer us a plastic waveguide with all the holes in the right places...
2 subs are not twice as loud as 1 sub. If my car has one 12 inch subwoofer and I add a second one, it will not sound twice as loud. It will sound 3db louder.
But my friends won't say "wow! turn it down." 1 sub at 130 db is booming. Add a second one and it booms a bit harder at 133db. That's all.
To be twice as loud you would need a 130 db subwoofer to add a second one and be 260 db.
Mr Viera is incorrect; 260dB isn't twice as loud as 130dB.
But Tom might be wrong too. (I can't believe I just wrote that!!!)
According to the link below, a 6dB increase in SPL raises the pressure 100%. (Measured in Pascals)
SPL sound level to dB pressure level sound pressure sound intensity unit dB Conversion of sound pressure to sound intensity sound units levels calculating spl sound level calculation air sound units level ratio decibel Pa kPa acoustic energy power co
Can someone verify?
But Tom might be wrong too. (I can't believe I just wrote that!!!)
According to the link below, a 6dB increase in SPL raises the pressure 100%. (Measured in Pascals)
SPL sound level to dB pressure level sound pressure sound intensity unit dB Conversion of sound pressure to sound intensity sound units levels calculating spl sound level calculation air sound units level ratio decibel Pa kPa acoustic energy power co
Can someone verify?
Doubling pressure (as with 2 woofers rather than two) wil give a 6dB increase (in power, pressure, whatever you want to talk about). 6dB is double pressure, 6dB is four times power (what I imagine Tom meant to say in the quoted text).
But then when you get to loudness, suddenly there's no very repeatable measure. It's how loud something 'sounds' to someone. Sometimes 6dB (twice pressure) will sound twice as loud, sometimes 10dB will, and it's a perceptive judgement call and will vary, usually in that range, from person to person and even from moment to moment. 10dB is usually assumed by convention...
But then when you get to loudness, suddenly there's no very repeatable measure. It's how loud something 'sounds' to someone. Sometimes 6dB (twice pressure) will sound twice as loud, sometimes 10dB will, and it's a perceptive judgement call and will vary, usually in that range, from person to person and even from moment to moment. 10dB is usually assumed by convention...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.