Hello all.
I would like to make a copy of the western electric KS-6368 horn. Sadly this horn has very little information published about it since it is no where near as popular as the larger theater horns.
I am trying to find some sort of expansion equation I can use to govern the shape of the horn. I know I am probably overcomplicating things and I can probably use a simply exponential horn equation, but if anyone has more detailed information on how WE horns were designed, I would greatly appreciate it. Any help is welcome.
I would like to make a copy of the western electric KS-6368 horn. Sadly this horn has very little information published about it since it is no where near as popular as the larger theater horns.
I am trying to find some sort of expansion equation I can use to govern the shape of the horn. I know I am probably overcomplicating things and I can probably use a simply exponential horn equation, but if anyone has more detailed information on how WE horns were designed, I would greatly appreciate it. Any help is welcome.
You know, that brings up an interesting point. Im thinking about making this a straight horn. From what I understand, the Klein bottle shape is simply there to make the horn more compact.
I would tend to think that making the horn straight would improve performance. Am I off base?
I would tend to think that making the horn straight would improve performance. Am I off base?
Once it was understood that sound is 'round' and expanded at 1/f [exponentially], then it seemed reasonable that the horn should be too, so TTBOMK all early 'wide range' horns are, so don't let the odd folding fool you as it once did me. 🙁
Note too that many of these horns are truncated to a greater or lesser extent, so can be shortened at one or both ends, making figuring its flare frequency a bit problematic if you don't know its acoustic axial length.
GM
edit: All that said and if I used Hornresp right just now, all one need do is input the throat and mouth area in expo mode and it calculates the rest, then export the data to get a dimensioned plot to build a straight horn! 😎
Note too that many of these horns are truncated to a greater or lesser extent, so can be shortened at one or both ends, making figuring its flare frequency a bit problematic if you don't know its acoustic axial length.
GM
edit: All that said and if I used Hornresp right just now, all one need do is input the throat and mouth area in expo mode and it calculates the rest, then export the data to get a dimensioned plot to build a straight horn! 😎
Last edited:
Since we have modern computer tools to design a much better horn, why not design one from scratch with an ideal profile that you prescribe from a simulation. But format it to resemble the WE as close as possible. I don’t think anyone will notice (visually) that it’s not the same expansion or flare rate etc. it will look like the antique but sound much better.
Or is the idea to reproduce the sound of the antique WE?
Or is the idea to reproduce the sound of the antique WE?
Good question. The short answer is....it's complicated. 😀
So if I can reword the easy part of your question to ... "does it need to be a WE horn?". Then answer is no. I am not a zealot and I am definitely open to other ideas and suggestions. I also could care less about aesthetics. I know some people take great care in making an exact 1:1 replica of WE horns, but that ain't me 😀
If I can reword the more difficult part of your question to ... "what are you trying to achieve?" or "what sound are you looking for?" the answer is far more complicated.
The simplest answer is that I want to make a midrange horn (2-300hz up to 4-5K would make me very happy) whilst minimizing the horn honk/coloration/whatever it is called. I am very sensitive to that weird echoy reflection sound that a lot of horns have.
Some incredibly popular WE horns like the 15A have too much of that weird honkyness. Some very popular modern horns like the magicos or the avantagarde turn me off as well.
I do however like the WE 753C, Jensen Imperial, and even the more modern klipsch corner horn.
But out of all the horns I have come across, the WE fabric horns seem to be the most natural sounding. That is really the only reason I am pursuing it vs some other design.
So if I can reword the easy part of your question to ... "does it need to be a WE horn?". Then answer is no. I am not a zealot and I am definitely open to other ideas and suggestions. I also could care less about aesthetics. I know some people take great care in making an exact 1:1 replica of WE horns, but that ain't me 😀
If I can reword the more difficult part of your question to ... "what are you trying to achieve?" or "what sound are you looking for?" the answer is far more complicated.
The simplest answer is that I want to make a midrange horn (2-300hz up to 4-5K would make me very happy) whilst minimizing the horn honk/coloration/whatever it is called. I am very sensitive to that weird echoy reflection sound that a lot of horns have.
Some incredibly popular WE horns like the 15A have too much of that weird honkyness. Some very popular modern horns like the magicos or the avantagarde turn me off as well.
I do however like the WE 753C, Jensen Imperial, and even the more modern klipsch corner horn.
But out of all the horns I have come across, the WE fabric horns seem to be the most natural sounding. That is really the only reason I am pursuing it vs some other design.
In that case, I would take a look at the tractrix horn calculator on Volvotreter’s website for mid range horn. I have made one for a 5in high sensitivity driver and it sounds superb. Measures well too. To get the lower reach you are looking for it just needs to grow a bit.
http://volvotreter.de/downloads/tractrix_v1.3a.xls.zip
Here is a 300Hz horn:
https://volvotreter.de/downloads/300hz_d54_horn.pdf
Here is mine:
PRV 5MR450-NDY for FAST/WAW applications
Sounds great in foamcore with constrained layer damping.
http://volvotreter.de/downloads/tractrix_v1.3a.xls.zip
Here is a 300Hz horn:
https://volvotreter.de/downloads/300hz_d54_horn.pdf
Here is mine:
PRV 5MR450-NDY for FAST/WAW applications
Sounds great in foamcore with constrained layer damping.
Last edited:
Ah very cool. I need to modify it a bit for the compression driver I have though.
I have looked at a tractrix horn a few times. I will do some more research.
Foam board is interesting. I have been playing around with cut up usps boxes and have gotten remarkable results. In fact if I hadn't cobbled together a horn as a demonstration piece for a friend, I don't think I would have ever believed making horns out of cheap material was possible. 😀
I'd add the LeCleach horn, the modern interpretation of exponential that considers spherical wavefronts. Very good loading, self-terminating into free-space, low diffraction and with variations for midrange quality or for LF duty.
Tjj, we took away the instant quote function in the last post in the thread to further discourage quoting when next in line. Please play along.
Oh is that a new thing? I thought my computer was just having a hissy fit.
I had a long break from DIY audio so I must have missed something along the way.
I had a long break from DIY audio so I must have missed something along the way.
Allen, could you please elobarate on the «variations for midrange quality or for LF duty»? Is it by using different T-values when calculating the horn? Waste difference in SQ?
Br
Gisle
@Tjj226
Have you seen this thread? Klangfilm spherical wave horn calculator - from round to elliptical
More on it here: DrBA horns at sphericalhorns.net
Even though docali focussed on Klangfilm horns it might be useful to you.
Have you seen this thread? Klangfilm spherical wave horn calculator - from round to elliptical
More on it here: DrBA horns at sphericalhorns.net
Even though docali focussed on Klangfilm horns it might be useful to you.
Ummm... no they don't. Unless you don't know how to design a crossover or use a horn properly.
But that would be true of most horns.
Also one thing to think about with the WE horns is that they have a pinched waist. Pinched in the vertical axis. They start out axi-symetric round then square, transitioning to a wide aspect ratio then back out to a lower aspect ration at the mouth.
Somewhere I plotted this shape change in a spreadsheet. I'll post it if I can find it.
Somewhere I plotted this shape change in a spreadsheet. I'll post it if I can find it.
Yes the T value. In this post, Jean-Michel discusses choosing this value and shows comparison plots - https://www.diyaudio.com/forums/multi-way/140190-jean-michel-lecleach-horns-2.html#post1775427
Here, the shape is discussed - Beyond the Ariel
Lynn Olson used power factor to justify the choice of T factor, focussing on the azurahorn - Beyond the Ariel
Ummm... no they don't. Unless you don't know how to design a crossover or use a horn properly.
So that's an interesting comment. I thought the whole point of a 15a with the 555 driver was to get as much bandwidth as you or die trying.
Can you expand on the idea of using a horn "properly".
The original idea was to cover dialog in motion pictures, so somewhat more than the telephone band. And to fill large cinemas with the amplifier power available at the time. These weren't always used alone, they were also used with open baffle woofers.
These days we do not need to use a horn like this full range, it can be supported at top and bottom with cone drivers below and other horns above. Just like any other horn, these should used where they sound best. And like other big horns it may be fun to stretch the bandwidth to the breaking point, but it's not the best way to use them.
When used part of a multi-way systems, these WE style horns are magnificent.
These days we do not need to use a horn like this full range, it can be supported at top and bottom with cone drivers below and other horns above. Just like any other horn, these should used where they sound best. And like other big horns it may be fun to stretch the bandwidth to the breaking point, but it's not the best way to use them.
When used part of a multi-way systems, these WE style horns are magnificent.
I have not seen the western electric KS-6368 horn in person, but photos of it make it look like it is unusually flat sided for the last 1/3 to 1/2 its length. Is this something you want to duplicate, or would the profile of the WE 15a, Vitaphone 11, Sato style horns be better for you?
AFAIK, these are just long exponential horns that change aspect ratio twice. The KS-636 appears to start at 1:1 (of course) widening out to about 4:1 then ending up at 1:1.5 tall at the mouth.
AFAIK, these are just long exponential horns that change aspect ratio twice. The KS-636 appears to start at 1:1 (of course) widening out to about 4:1 then ending up at 1:1.5 tall at the mouth.