5Khz Smith Horn - need tips on design

[Fred Ireson]

might have to cheat a bit on driver and try APT50 phenolic as may have more HF power/rattling than my 806 & 807. maybe a hacked KSN1016 with stepup transformer would work too (or Leson)

anyhow - got majorly confused looking at Johnny's 5Khz horn drawing due to conical ponted piece and what appeared perhaps to be throat/plug assembly.
(a few more pix including one of driver mount would have helped)

for my intent just seems like need the "lips" front - anything beyond this concept.

are dividers helpful in mini-Smith-horn? - or just create interference lobing?

luv sound of these waveguide - and 1" version wasn't shabby with big-gap Karlson coupler loaded w, 18" haha.

I think DSH critics may not have response tailored correctly

if can ever figure out & get 5Khz built and smooth then will ask about diy slant-plate rules.

gimme hints on dims, etc which work well.

[mikebake]

Hi Fred, oh keeper of Karlson flame; nice to "see" you here. Hopefully someone with the requisite knowledge will help you on your questions; some of the more knowledgable guys here are often pretty busy.......

[yggdrasil]

Is APT50 an 1" driver, like the 806 & 807?

Does that mean that you are planning a UHF Smith horn useable from 5KHz on a 1" exit driver?

[yggdrasil]

anyhow - got majorly confused looking at Johnny's 5Khz horn drawing due to conical ponted piece and what appeared perhaps to be throat/plug assembly.
(a few more pix including one of driver mount would have helped)

When you remove the horns from the JBL ring radiators you will have to add this phase plug. You then get a driver with 12mm exit.

[Fred Ireson]

Ah Johnny - so that explains the adaptor.

I only have original article 1" format Smith horns and wish to try 5Khz version with them, conical, whatever horns laying around.

does your 5Khz Smith-horn's localization cues "disappear" as with larger versions? - if so then this might be "the" ideal HF waveguide !

did you make any other versions of the 5Khz size?

will an 1" exit driver work? - if so what needs to be done to have it acceptable and not create a narrow "neck" feed to horn?

Freddy

[yggdrasil]

The small Smith's have been lying around just waiting for some free time to appear.... Rather busy on other project.


If you are going to use a 1" exit driver you will have to increase the horn height. 2cm works on the 1200Hz.

I have used 3875Hz as f0 in my calculations. Given 2 cm/0,79 inch horn height gives a radius of 2,62cm / 1,03 inch. Approximately 5cm/2 inches wide.

That is a tiny horn...

[Fred Ireson]

Hi Johnny - 2" width is small. Are you sure to work in "real world" these rules must be followed to produce this type waveguide? - can the rules be bent with good results? - say 2CM high horn with 130/whatever degree sidewall angle?

are there modern 1/2" format HF drivers to try?

btw here's the Smith-Selsted article

http://img455.imageshack.us/img455/6...elsted13kv.jpg

http://img455.imageshack.us/img455/3...elsted20fw.jpg

http://img503.imageshack.us/img503/6...elsted32ca.jpg

Freddy

[Zilch]

Are you all considering using the 1" ring diaphragm re-entrant compression drivers like 2407H and BMS 4540nd? Seems they're the new generation of UHF drivers with UHF out to 30 kHz, 1/2 octave higher than the old standbys, if that matters. JBL has apparently embraced the technology in their latest offerings:

http://audioheritage.org/vbulletin/s...age=54&p=83669

http://www.bmspro.info/index.php?sho...10278&id=54358

In my testing on constant directivity horns, they only require standard CD compensation to flatten the response....

[Fred Ireson]

yeah - little BMS would be on list - maybe cheaper ferrite version - if works as well as neo

[John W]

At one time I had a pair of 2412h drivers with 1 in exits. These looked a lot like an LE20 converted to compression driver. The screw-threaded throat looked like it started at about 1/2 in then flared up to 1 in. I thought at the time they would make a good smith-horn tweeter, like on the westlakes, but never got around to working something up.

[Zilch]

At one time I had a pair of 2412h drivers with 1 in exits. These looked a lot like an LE20 converted to compression driver. The screw-threaded throat looked like it started at about 1/2 in then flared up to 1 in. I thought at the time they would make a good smith-horn tweeter, like on the westlakes, but never got around to working something up.

Edgewound postulates it's a repackaged 035Tia, I believe.

yeah - little BMS would be on list - maybe cheaper ferrite version - if works as well as neo

Neo's only 96 bucks.

[Ceramic's $68....]

[Earl K]

Hi Fred, ( & welcome to this forum ! )

- Here's some "small" Smith Horn info

- Designing a 5000hz Smith Horn .. ?

(a) - Some other info suggests it's prudent to keep the depth of the horn sufficiently deep enough to support at least one full wave-length, ( chosen to be one octave below the desired crossover point ). This is due to the fact that the horn mouth area for this horn type is so small .
- That means for a 5000hz Smith Horn, 2500 hz needs to be fully supported inside the horn bell .

So ; 1130 divided by 2500 = .452 ft ( or a horn depth of at least, 5.424" )

(b) This derived figure is at odds with the deeper tweeter-horn used by WestLake which is almost twice the above figure / though the figure seems to be supported by the original article that you've posted .