JBL 2397 with 2405 time aligned

[Dr.db]

Hey,


I´d like to built myself a pair of the 2397(e.g. Westlakes...) and want to mate them with a 2405 time aligned on top.

I actually have no equipment to measure the impulse-response which could help me find the correct placement of the drivers.
I think vertically matching voicecoils should be aprox. right, but not 100% correct.
Has anyone of you allready find out, what´s the best position for the tweeter on top ?

To avaid reflections of the Tweeter on the surface of the 2397-horn I would have to place the 2405 something like 10cm above the horn, due to the 25° dispersion. Is that true !?

If anyone has a detailed drawing of the vanes, I would really appreciate that :)



Thanks a lot & have a great week,
Olaf

[hjames]

Not trying to be snarky in any sense, but the 2397 horns are JBL horns - and are not Westlakes ...
http://www.jblpro.com/pub/obsolete/2397.pdf (http://www.jblpro.com/pub/obsolete/2397.pdf)
I believe the main difference is the curve of the "lips" and the thickness of the top and bottom plate ...
I use a pair of 2405 slots in my L200-custom speakers and for a while had a pair of the (consumer version) 077 drivers
mounted on a small walnut base externally positioned on top of the smith horns.
Sounded good - good luck with your project!


Hey, I´d like to built myself a pair of the 2397(e.g. Westlakes...) and want to mate them with a 2405 time aligned on top. I actually have no equipment to measure the impulse-response which could help me find the correct placement of the drivers. I think vertically matching voice-coils should be approx. right, but not 100% correct. Has anyone of you already find out, what´s the best position for the tweeter on top ? To avoid reflections of the Tweeter on the surface of the 2397-horn I would have to place the 2405 something like 10cm above the horn, due to the 25° dispersion. Is that true !? If anyone has a detailed drawing of the vanes, I would really appreciate that :) Thanks a lot & have a great week, Olaf


64048

[Dr.db]

This is an example, but I´m not shure if this is right....

[Dr.db]

I know the 2397 is a JBL, but I propably build the westlake version which is nearly similar.

[hjames]

Lee in Montreal had a really neat pair of Aluminum brackets for mounting 2405 slots and the CD drivers for smith horns ...

Seems that if you mounted some wood to the top screws, then mounted the 2405 to the back of that,
you could offset the 2405 back from the aluminum ears and, using shims, shift the offset to time align the diaphrams as desired.

(not my drivers ...!)

64049

[Mr. Widget]

I´d like to built myself a pair of the 2397(e.g. Westlakes...) and want to mate them with a 2405 time aligned on top.

I actually have no equipment to measure the impulse-response which could help me find the correct placement of the drivers.I would suggest you mock it up testing various heights and depths to determine what sounds best at your listening position.

Regarding Westlake vs. the 2397 and the exact contours etc., years ago I posted drawings of both as a comparison as well as an aid to allow others to build their own. Quite a few people have successfully used the drawings as these are fairly easy to build.

I am out of town and typing on my phone, but you should be able to search for the drawings... or perhaps someone can help you find them.


Widget

[pos]

I am out of town and typing on my phone, but you should be able to search for the drawings... or perhaps someone can help you find them.

That would be these:
http://www.audioheritage.org/vbulletin/showthread.php?3813-2397-Smithhorn-help&p=35121&viewfull=1#post35121
http://www.audioheritage.org/vbulletin/showthread.php?5425-Westlake-HF-Horns&p=49874&viewfull=1#post49874
http://www.audioheritage.org/vbulletin/showthread.php?28146-Westlake-style-horn-drawings&p=283617&viewfull=1#post283617

[Horn Fanatic]

Hey,


I´d like to built myself a pair of the 2397(e.g. Westlakes...) and want to mate them with a 2405 time aligned on top.

I actually have no equipment to measure the impulse-response which could help me find the correct placement of the drivers.
I think vertically matching voicecoils should be aprox. right, but not 100% correct.
Has anyone of you allready find out, what´s the best position for the tweeter on top ?

To avaid reflections of the Tweeter on the surface of the 2397-horn I would have to place the 2405 something like 10cm above the horn, due to the 25° dispersion. Is that true !?

If anyone has a detailed drawing of the vanes, I would really appreciate that :)



Thanks a lot & have a great week,
Olaf

Greetings Dr.db -

Have you considered building a pair of real Smith horns, but with the Westlake look? The math for the design is quit simple, and can be found in Bob Smith's 1949 article. Using his math would allow you more flexibility, and could easily be modified to accommodate the 2328 adapter. His original horn was made of cheap 1/4" thick plywood, but you could design your horn using thick hardwood as Westlake did. There is no substitute for mass!

The only real part of the 2397 that resembles a Smith horn is about the first nine inches from the throat, with the dividers making it kind of a quasi-Smith horn. The Smith horn is a true exponential horn from throat to mouth, whereas the 2397 and Westlake horns are not. As a test ten years ago, I built a pair of the original Smith horns following his plans using six millimeter Baltic birch. I have to say they sound interesting. Smith gave his horn the proper designation it certainly lives up to, DSH. Distributed Source Horn.

One of these days I will purchase a pair of Visaton throat adapters for a 1" driver, and design new Smith horns for my ALTEC 902 drivers after the suggestions I gave you. One thing I would do, is incorporate a slight conical taper to the horn panels to avoid standing waves. The 2397 and Westlake horns are subject to a standing wave of approximately 5K. The 2405 needs to be crossed over higher than that.


Good luck,

H.F.

BTW - Bob Smith co-wrote a subsequent article with Walter Selstead on their design of a horn tweeter. You might be interested in reading that one also. I have thoughts of building that tweeter horn for a ring radiator using the 2404 throat cavity and phase plug. Some day I will build a pair, since I bought the parts from the service center about 5 years ago.

[Dr.db]

Thanks a lot for your assistant response, I really appreciate that!! :)


@ James: Lee´s Aluminum brackets do like nice indeed! But I´m doing an all wood-job :bouncy:
Your walnut 2397 look great, too! Pretty neat enclosures beneath too.



@ Mr. Widget: Your drawings are a very big help, thanks so much for sharing! :applaud:


@ Pos: Thanks for searching the links and posting them :)


@ Horn Fanatic: Thanks for bringing up your ideas! What is the advantage of the smith horn over the JBL 2397 in your opinion ? What´s the advantage of being a true exponential horn ?

I like your idea of the slight conical taper of the superior and bottom horn panels. How does the standing wave affect the sound ? Is there a peak at about 5khz ? If so, how much db`s am I dealing with ? Would a simple EQ help ?

Why did JBL and Westlake keep building their horns with this fault?

[Maron Horonzakz]

I detect no standing waves in my measurments.. (smith horn)

[ivica]

I detect no standing waves in my measurements.. (smith horn)

Hi

A so called "waterfall" presentation or Wavelet Analysis of Time Domain Behavior,
of the measurements would show a kind of "echo" (prolonged response) on some frequency regions.

It wold be nice if You can show us your data results.

Regards
ivica

[Horn Fanatic]

@ Horn Fanatic: Thanks for bringing up your ideas! What is the advantage of the smith horn over the JBL 2397 in your opinion ? What´s the advantage of being a true exponential horn ?

I like your idea of the slight conical taper of the superior and bottom horn panels. How does the standing wave affect the sound ? Is there a peak at about 5khz ? If so, how much db`s am I dealing with ? Would a simple EQ help ?

Why did JBL and Westlake keep building their horns with this fault?

Greeting Dr.db -

The first thing I noticed about the Smith and Westlake horns, is the hot spot in the middle, and the increasing off axis attenuation. With the Smith horn, what you hear in the middle cells is the same that can be heard in the outer cells. Consider the Smith horn a single row multi-cell radial horn, with only two flared sides for each cell. As Arthur Gordon Webster demonstrated in his 1919 article, the exponential profile is the most superior, however, as it provides more design flexibility I use Vincent Salmon's Hyperbolic equation for my calculations. It allows one to maintain the horn length while being able to adjust the mouth area within reasonable parameters. In some cases I design my horns using Harry Olson's Manifold Section approach, even for low-frequency horns. It all depends what the horn is being used for, and where it is deployed. Not ever having one in hand I can only speculate that the TAD horn is a manifold section horn. The initial section that of a mini 2397, and the rest of the horn following an exponential flare.

As for standing waves, they are an acoustic detriment to any sound source. Yes, there would be a resonant peak at a standing wave. An EQ will not help, as by attenuating a resonant frequency the adjacent frequencies above and below would be attenuated as well. A comb filter is not the answer either. If it were that easy, the resonant frequencies found in compression drivers could be eliminated. Not going to happen. That problem can only be addressed mechanically.

IMHO, the best course of action would be to a build Smith horn in the same manner as the Westlake, for the density. Should you wish to build the horn with parallel panels it would be prudent to cross it over below the standing wave frequency. I believe employing a slight conical taper to the Smith horn would be an improvement. As I suggested before, by using Bob Smith's calculations you could design a custom horn to fit your application, not only by cut-off frequency, but by physical size and throat adapter dimensions. Bob Smith's method of providing a square to circular transition was by means of using putty. The transitions on my two horns were made from the rear plate used on the 604 Mantaray horns I received from Bill Hanuschak. Not the best means, as by cutting up the rear plate the horn is rendered useless. It's a remedy I would not repeat in any numbers. Ideally, if an aluminum die cast throat adapter is not an option, a wooden adapter can be made on a CNC machine, which is what I have had done for other types of horns

"Why did JBL and Westlake keep building their horns with this fault?"

Beats the Hell out of me!

H.F.

[Horn Fanatic]

I detect no standing waves in my measurments.. (smith horn)


Are you measuring a real Smith horn, and not a 2397? There should be resonance at approximately 6.7K for the Smith horn, and approximately 4.9K for the 2397. Both frequencies of resonance as a result of parallel panels with 1" separation for the Smith horn, and 1.375" separation for the 2397.

[Dr.db]

Thanks a lot for your detailed answer!

The smith horn was the first contender and decades later JBL and westlake copied and modified this type of horn. What made them change these good properties ? They must have accomplished at least some improvements!?

[Horn Fanatic]

Thanks a lot for your detailed answer!

The smith horn was the first contender and decades later JBL and westlake copied and modified this type of horn. What made them change these good properties ? They must have accomplished at least some improvements!?

As far as I know, the 2397 was introduced by Bart Locanthi, rest his soul. The Westlake horn is nothing more than a pretty 2397 with thicker chamfered panels. No doubt the chamfer has a legitimate function. IMO, the Westlake horn sounds better for the density of the material. I had the opportunity to hang out in a Westlake designed studio in the middle 70's that had the big Westlake control room monitors soffet mounted. I can only speculate why Locanthi designed his horn in that manner. I'm sure he had a good reason.

I built a couple of pair of 2397's for Hal Cox some years ago, and he told me that by using MDF my horns sounded better than the JBL horns. What I was using at the time was called, 'Ranger Board', which was manufactured in Canada. JBL used what they called in the day, "Timblend" for many of their wood products. A fancy name for particle board slightly better than the material used for sub-flooring.

A hot dog by any other name...

[Dr.db]

Thanks a lot for your guidance!!


Would you have any suggestions for the placement of the 2405 ontop of a 2397 or smith-horn ?

[Horn Fanatic]

Thanks a lot for your guidance!!


Would you have any suggestions for the placement of the 2405 ontop of a 2397 or smith-horn ?

The photo of the bracket used by a chap in Canada HJames posted on the 7th looks very cool. As for time alignment? By the looks of the bullet placement it doesn't look like it. I noticed two chaps on Face Book discussing the time alignment issue recently. One of them is of the school that the voice coils aren't always require to be on the same plane. His reasoning appeared to be sound, but that is just one opinion. If I can find the discussion I'll post it here so that others may review it.

However, I would use some type of damping material on the top of the wooden horn you end up building, to eliminate reflections from the top of the horn should you mount the 2405 within proximity of the horn throat. Perhaps some felt, it's cheap!

Good luck. Post pics of your horns when you make them. :)

[Dr.db]

That would be nice, looking forward to that discussion!


Actually I´d like to avoid any kind of damping material on top of the wooden horns. Simply because of the looks. :bouncy:
So I would try to mount the tweeter high enough so that it will beam completly above the midrange-horn.
Could this height be determined by simply using the diffraction-characteristics of the 2405 ? It has 25° vertical diffraction, so 12,5° up and 12,5° down from the horn-mouth. This is about 1inch down per 4"inch lenght.... If the 2397 is 13,5inch long plus the 2328 adapter with 4inch of length this would be a total of 17,5inch which the 2405 is mounted in behind.
17,5inches divided by 4inches = 4,375inches
So the horn mouth of the 2405 has to start at least 4,4inches above the midrange horn to avoid reflections, right ??

I hope this wasn`t too confusing.

[Mr. Widget]

That would be nice, looking forward to that discussion!


Actually I´d like to avoid any kind of damping material on top of the wooden horns. Simply because of the looks... So the horn mouth of the 2405 has to start at least 4,4inches above the midrange horn to avoid reflections, right ??I agree that a pile of absorbant material on the horn would be hard to pull off and make it look good.

Realize that the dispersion spec is at -XdB, it isn't absolute. I think you will need greater separation to pull this off without absorbant material.

Also on the voice coil alignment discussion be aware that time alignment must take into account listening position relative to each driver's acoustic center... Meaning if you stand up or move back you will need a different alignment than a lower or closer listening position so simply lining up voicecoils will almost never be correct.


Widget

[ivica]

I agree that a pile of absorbant material on the horn would be hard to pull off and make it look good.

Realize that the dispersion spec is at -XdB, it isn't absolute. I think you will need greater separation to pull this off without absorbant material.

Also on the voice coil alignment discussion be aware that time alignment must take into account listening position relative to each driver's acoustic center... Meaning if you stand up or move back you will need a different alignment than a lower or closer listening position so simply lining up voicecoils will almost never be correct.


Widget

Hi Mr.Widget,

Not to mention the influence of the (usually applied networks). As comb effect (dye to the time delay mismatch) can be seen on the FR measurements, an experimental work would be easier way, I think.
I have here one question for You ( as a man of great experience and knowledge): using "brick-wall filters" (in VHF/UHF sections) would it help or not?

Reagrds
ivica

[Dr.db]

That´s correct; time alignment always depends on listening position.
I think the height of sitting in a couch is allways about the same, just the distance varies considerably.
But it has to be considered, you´re right!


I`m going to use an active-crossover, but this will still affect the time alignment, won`t it ?


With which equipment could I actually determine the correct position in a field-test ? I´ve nothing for measuring so far and I don`t want to spend a lot money either...

[Fort Knox]

I've already been though all this horn alignment business ...
and your better off w/2 way system and some serious eq..FK

[Mr. Widget]

...using "brick-wall filters" (in VHF/UHF sections) would it help or not?
With regards to comb filtering absolutely, regarding improving basic impulse response (the goal of time alignment) I'm not sure. Since there are several other variables, I guess it would depend on the application.


Widget

[ivica]

With regards to comb filtering absolutely, regarding improving basic impulse response (the goal of time alignment) I'm not sure. Since there are several other variables, I guess it would depend on the application.


Widget

I agree that impulse response would be the best measure, but for sure, in such situation , a kind of time-delay, compensation has to be applied.

regards
ivica