Giant Custom Horn Systems - How they sound and issues with sonics

I definitely dream of one day building the ultimate system and room. I see what some of these crazy guys do with truly giant custom horn installations and they make me drool, but (like most of us) I've never had an opportunity to hear one of these crazy systems (see pics).

But I'm guessing at least a few of you have and others probably possess some theoretical engineering knowledge around the issues these giant horn systems might create, but I've never seen a thread that discusses the sonic pro's and con's of these systems.

Would love to hear some thoughts (even if it is just some pontificating on theory) regarding these types of installations. Let's discuss.

 

Hello Okan?

Hello Swen and welcome to the forum! You guessed right, I guess not many multi-channel horn owners out there...

Yes, 1601 would work with flh, I am not saying anything against it. I just said TAD engineers designed it with bass reflex boxes in mind and I use that to be able to align my bass boxes to the rest of the horn system without introducing delay.

Many companies use them with FLH, including Cessaro.

 

What's that? I can't see the horn.

Impressive work Swen, both on the bass horns and your Satos!


This throat adapter was designed to optimize horn-compression driver coupling, in order to improve on- and off-axis response and minimize interferences at the horn throat (scattering, diffraction > higher order modes).

A few more images:

 

Here's the adapter with the actual horn and compression driver:






The aim of speaker design is, of course, the speaker (horn loaded in this case) should provide as natural sound as possible. Specifically, the goal is to have even frequency response not only when sitting directly in front of the horn but also when sitting at an angle of 30 ° or 45 ° relative to the sides or 15 ° above or below the horn. In addition to a large and regular spread, low harmonic distortion and a good impulse are desirable. The acoustic properties of this horn were analyzed with CLIO. The result can be seen below, where the frequency in Hz is shown on the 1st axis and the volume in dB on the 2nd axis.

The red curve is the frequency response at 0 °, sitting right in front of the horn, the blue curve is the response at an angle of 30 ° to the left or right of the horn and the dark green curve is frequency range at an angle of 45 °. The light green curve is the frequency response at 0 ° horizontally, but at an angle of 15 ° above or below. Finally, the orange curve is an additional measurement at 0 ° and is at best identical to the red curve. No Equalization was used!
The 2nd harmonic distortion at the displayed sound pressure (105-110 dB) is below 0.5% at all frequencies above 500 Hz and the 3rd harmonic under the same conditions is below 0.07%.






Note: this is not my own work, but I consider it useful and it may help forum members/visitors to better understand some aspects of horn technology.

 

For which frequency you want to use it? At least you need 4 ways. And 600 Hz with a horn of what? 15 cm depth? Is it already working with full hornload at 600Hz?

 

I believe this horn is somewhat deeper than 15cm and it's more akin to a waveguide, with an immediate expansion/shorter horn path from the throat to the mouth. Such designs have some advantages, i.e. these are less prone to distortion and beaming.
As mentioned earlier in this thread by Karel and others, multiple horns operating in their optimal range (2-3 octaves) is another, possibly even better, way to avoid such issues.

 

..As mentioned earlier in this thread by Karel and others, multiple horns operating in their optimal range (2-3 octaves) is another, possibly even better, way to avoid such issues.

Hi.
I (re-)deal my (final) experiences:
I limit the operating Fr range per driver between 1 1/2 oct to max 2 octaves/driver
Such condition(-s) for all drives give the best tech possibility to reproduce with max efficiency the "musical-signal with all his harmonics".
Music-signal is full of harmonics ..... Fourrier is omni present .
Music is plenty "impuls" (& Fourrier harmonics) ......
With such condition 1 1/2 oct/driver .... The musical scène start with a minimum 4-way (horn) & better 5-way horn (or higher) ...

It's music ... I like it
(analog&horn).

Allez, salukes.

Karel

One musical note:
3-way (horn) concept could be nice ... but not like the (3+n)way horn

 

Karel, your system appears to be very close to a no-compromise solution.
In a similar way David Wilson builds his arrayed direct radiator towers, but I am quite sure these can't touch your "horn magic".

I would (also) like to hear this custom horn system of B. Kolbrek (he is one of the few people with a PhD in Horn Physics):

 

Is this one of those Norwegian horn systems? I think because of the long winter there those guys sit home and build giant horns

 

Yes, he's from Norway.

This is perhaps his most well-known article and this is his homepage.

He is currently working as a consultant for Celestion.
I believe he uses the amazing Axi2050 driver for the big mid horn, which he designed and built with a friend.
This driver can be used from 100 to over 10.000Hz, which is quite an achievement and its sensitivity is an astonishing 140dB.
If you settle for "only" 110dB you could practically use this driver full range, with a bass solution below 100Hz.

 

Duttilleux: If you settle for "only" 110dB you could practically use this driver full range, with a bass solution below 100Hz.

Yes, if someone loves beaming

 

With such condition 1 1/2 oct/driver .... The musical scène start with a minimum 4-way (horn) & better 5-way horn (or higher) ...

I absolutely agree. I could get away with not having my smaller 1400hz JMLC horn and the 1" driver behind it. I could drive my 200hz tractrix up to where it can blend to my fostex t500amkII. It was still measuring very flat, and no crossover, no other driver, simpler right!. Simpler does not always have to be the best solution, this is a bias too.

Adding the 1400hz JMLC with 1" beryllium driver, I gained two very important things. It works between 4500-9000hz roughly. This is where we call presence, if this is dialled in right, you have the clarity. Too high, harshness, too low, laid back sound. Just right and an audio reward for my ears The second thing the 1400hz JMLC brought is the being free of the beaming of the 200hz tractrix for this band. This is why I have 4 horn channels from 120hz to 20khz. I have 6 channels in total counting the tapped horn for sub 40hz duties and the br boxes for 40-120hz. This was the least I could do with this topology!

 

Life starts with a compromise, life ends with a compromise. Every speaker is a compromise. The goal is, to minimize the compromisees.....

 

Very true, and to always adapt with them, every day.

 

Another wonderful Horn system. https://m.youtube.com/watch?v=f2SFY9zt6Xw
We've nearly copied the design from the legendary Tamaru System in Japan. But we changed the Midrange to implement our Horns

 

Another wonderful Horn system. https://m.youtube.com/watch?v=f2SFY9zt6Xw
We've nearly copied the design from the legendary Tamaru System in Japan. But we changed the Midrange to implement our Horns

Hi Swenn (&.o.)
Power-full idea & initiative & build knowledge ....
Some more info (French) from JH:
http://www.asrr.org/biblioteca/Revue Audiophile/FICHIERS/NDRS/TAMARU/TAMARU.html
Bonne chance.
Karel

 

Hello Karel, the picture I've posted is taken from the page you've linked. One of the best Pages in the www
My knowledge about Horns is very small besides the knowledge of my master and teacher....

Best Regards Swen

 

Duttilleux: If you settle for "only" 110dB you could practically use this driver full range, with a bass solution below 100Hz.

Yes, if someone loves beaming

Modern R&D and production technologies have come a long way, this driver is a product of these techniques.
Beaming occurs when the aperture diameter becomes less than the wavelengths that are to be produced, which would be the diameter of diaphragms, cones and domes in loudspeakers.

This Celestion driver features an annular diaphragm, so for breakup the relevant dimension is not the 6" dia, but the distance between the annulus ID and OD, which appears to be about 1 1/2", effectively it's even less than that because the VC is not at one of the edges.
Moreover, the phase plug design for this driver is completely different from those in regular large format compression drivers. This driver also differs from the coaxial BMS and JBL drivers.
As to horn/waveguide design: I haven't heard or read comments about beaming being an issue with the JBL M2.
If the guys at GP Group could develop a horn or waveguide that will accommodate this 6" compression driver, which holds pattern from 200hz to 20,000hz....
It's worthwhile to mention that besides Mr. Kolbrek, Mark Dodd and Jack Oclee-Brown, the engineers behind KEF’s tangerine waveguide and blade technology, reside under the same roof.

However, point taken Swen.
I think its interesting to see if advanced technologies can bring something new and refreshing to the table at which Rice and Kellogg have been sitting for almost a century

 

Duttilleux : The Celestion driver was used on the AutoTech horn at the ETF 2017. My Friend designed the matching carbon fiber throat Adapter to connect it. So we know it. You can read what JC have written about the complete setup There you can also read which driver was used to have real fun with the AutoTechs. Not the Celestion

http://hifiheroin.blogspot.de/2017/11/etf-2017-attempt-at-documenting-main.html?m=1

 

Life starts with a compromise, life ends with a compromise. Every speaker is a compromise. The goal is, to minimize the compromisees.....

I think the key is to understand what set of compromises best achieves the goal. Defining priorities and sonic preferences drives design decisions.

For me, I hope one day to have the space, time and funds for a FULL horn system. I read somewhere that having a full DIY horn system is like keeping an exotic pet... not everyone is suited!

 

Dave : Live your dreams. 40sqm and 30K are already enough. That's not even a commercial middle class speaker. With the rest of your post I agree 100%