I'm glad that Roy confirms something I've been saying for years.  Not that I'm speaking for him here.

1) Many horn shapes have radiation patterns which narrow as frequency goes up.  Some of this is due to the geometry of the interior of the horn and some of it is mouth size.  Mouth size or just a diaphragm does this simply because at high frequencies they are bigger antenna in terms of wavelength.

2) Olson's diagrams showed the narrowing but the diagrams seem to have been normalize to show on axis response to be constant.  I think in reality there is gain on axis as frequency goes up.

3) Drivers used on horns have decreasing output as frequency goes up, just as Keele says.  You see this when driver response on a plane wave tube is shown (about 3 kHz.  But very often the response is shown on a horn with that increasing on axis gain..

 4) You put 2 and 3 together and you get a type of equalization.

5) Constant directivity horns do not have as much change, as shown by Keele.  Note that he had to equalize the driver to a constant output, and then the horn output was constant.

6) Horns like the K-400 keep the horizontal pattern fairly constant and the vertical pattern narrows.  PWK called it controlled directivity.  So it too has some gain on axis.

7) You see the 9 kHz "glitch" in the K-55.  It may be diaphragm break up.  It is about, say 15 dB down without the horn.  But the horn has so much on axis gain it is can be heard.  It may be that something similar is happening in the Jubilee.  So it is necessary to eventually have a steep electrical filter.

8) The K-Horn is probably lossy at 400 Hz and above due to the second section where the path flips from horizontal to the vertical.  However the real problem may be that final sections face away from the front axis and there is little gain at 400 Hz and above.  If you look at the AES paper on the Jubilee, the design evolution is to have the final flares face more forward.  Hence rising gain.

9) Keele calculates an increasing rate of roll off in the driver and you see this in midrange drivers on plane wave tubes.  While a horn has some increasing gain, it can't keep up with it quite as much as the driver is rolling off.  So at some point the overall response drops very quickly.

Gil

2) The drivers

Posted by djk ( M ) on May 29, 2003 at 18:05:52

In Reply to: LS autotransformer posted by mr.x on May 29, 2003 at 09:31:48:

This is the part # from Eminence, the makers of the speaker. Note that the K33E has a longer x-max than a JBL2226, and as a direct radiator is 101dB/2.83V/1M

SPEC 15162
PART # K-33

RE OHMS 3.39 FS HZ 34.46
LE MH .96 MMS GMS 78.59
QM 7.39 CMS mm/N .2714
QE .410 RMS NS/M 2.3037
QT .390 VAS LTRS 301.66
XMAX MM 8.20 SD SCM 889.59
BL TM 11.88 EBP 84.4
EFF % 2.91 SPL dB 96.6

Autoformer: Universal Transformer #3619
P.O. Box 472
Farmersville, TX 75442
972 784-7700 voice
972 782-7000 fax

djk:

FS HZ 34.46
QT .390

djk:

This is the part # from Eminence, the makers of the speaker. Note that the K33E has a longer x-max than a JBL2226, and as a direct radiator is 101dB/2.83V/1M

Hi Mike and Hi Gil,

Wow!  Gil, what you said is correct but there are some other things going on in addition to what you have said.

3) Drivers used on horns have decreasing output as frequency goes up, just as Keele says.  You see this when driver response on a plane wave tube is shown (about 3 kHz.  But very often the response is shown on a horn with that increasing on axis gain..

You can see this as well in the nearfield.  the bottom end is limited by an inefficient magnetic gap at low freqs.  the high end is limited by decoupling of the suspension and diap breakup.  and throw in a rising impedance response and you have the makings of a headache in trying to keep a driver "flat" in the far field!

7) You see the 9 kHz "glitch" in the K-55.  It may be diaphragm break up.  It is about, say 15 dB down without the horn.  But the horn has so much on axis gain it is can be heard.  It may be that something similar is happening in the Jubilee.  So it is necessary to eventually have a steep electrical filter.

in the case of the 55, it the phase plug.  it has path length differences that cause a notch at 9kHz.  but in most cases diap breakup contributes to peaks and dips at the top end.  it's just that phase plugs are "little horns" in and of themselves.

8) The K-Horn is probably lossy at 400 Hz and above due to the second section where the path flips from horizontal to the vertical.  However the real problem may be that final sections face away from the front axis and there is little gain at 400 Hz and above.  If you look at the AES paper on the Jubilee, the design evolution is to have the final flares face more forward.  Hence rising gain.

in the case of the khorn, it is mostly due to path length differences that cause the horn to roll off at 400.  another engineer at klipsch, took on the job of taking a khorn lf and making a straight horn (without any bends) of it.  he measured the freq and it was surprising.  it goes up pretty high (800-900 hz if memory serves).  in the jub, we paid alot of attention to the folds trying to make sure that the path length differences were above 1kHz.  in addition, we kept the splay angle between the two mouths small, in a sense arraying both mouths to keep the overlap spl up.

ultimately, keeping up with all these gremlins is what makes this fun and challenging. 

bodcaw boy:

7) You see the 9 kHz "glitch" in the K-55.  It may be diaphragm break up.  It is about, say 15 dB down without the horn.  But the horn has so much on axis gain it is can be heard.  It may be that something similar is happening in the Jubilee.  So it is necessary to eventually have a steep electrical filter.

in the case of the 55, it the phase plug.  it has path length differences that cause a notch at 9kHz.  but in most cases diap breakup contributes to peaks and dips at the top end.  it's just that phase plugs are "little horns" in and of themselves.

If it's the phase plug, then why does it show up in some drivers and not in others? Or is that a limitation of the measuring? I also believe the Cornwall I that Trey measured in the chamber had the problem at 8kHz? I don't keep up with all the crazy variations in drivers so maybe that's the reason?

Hey Mike

See the test I did on the various K55 drivers I did with ETF and you will see with some varition due to Phase Plug Differences and Contruction that all of them do have the dip and then rise in response with some variation in frequencies involved but it is in the design limitations of these drivers. I believe it is common to all the K55 drivers by design!

 http://forums.klipsch.com/forums/2/733256/ShowThread.aspx#733256

DR WHO I was glad to meet you in Hope. I'm the Mike from Tennessee if you remember.

mike

Hey Miketen and Dr. Mike,

All the different 55's had some variation in the phaseplugs but as to the chamber causing something at 8kHz, that is very unlikely.  It likely due to equipment problems rather than the chamber.

By the way Miketen,

did chuck ever call you about the jub parts?

Roy

bodcaw boy:

Hey Miketen and Dr. Mike,

All the different 55's had some variation in the phaseplugs but as to the chamber causing something at 8kHz, that is very unlikely.  It likely due to equipment problems rather than the chamber.

 

By the way Miketen,

 

did chuck ever call you about the jub parts?

 

Roy

Hey Roy

No, I've not received anymore e-mails or a call from him although I guess he is pretty busy.

I did send you a copy of my only contact with him so far. Anyway hopefully soon we can get this project going!!!

mike

Hi Roy,

I heard a couple of persons in Hope mention that you were working on a passive crossover workable for Jubilee.

Was wondering about your take on that so far.

hey miketen,

i do know that he just got back from an install in southern california.  i will get him on the ball on monday and i will send you the passive schematic on monday as well.

roy