Khorns Time Alignment

[jcmusic]

Here are my final IR measurements for mid/tweeter drivers time aligned...

[Schu]

Audability?

[jcmusic]

Meaning what? One is the left side the the right side...

Edited September 14, 2014 by canyonman

[DizRotus]

Time alignment issues and the benefits thereof, are what makes a good single driver appealing. A single driver is time aligned. The cohesiveness, imaging, soundstage, etc. are excellent. Horns, female vocals, etc. sound very good. The problem is it all comes at the sacrifice of extreme high and low frequencies.

After listening to an excellent single driver, one is compelled to achieve similar results through alignment of tweeters, squawkers and woofers. That is the way to get the imaging of a single driver with the dynamics and frequency range of a 3-way system. Physically aligning a horn loaded woofer is impractical, leaving the electronic solution.

Or . . . . you could always use headphones.

Edited September 14, 2014 by DizRotus

[russ69]

Time alignment issues and the benefits thereof, are what makes a good single driver appealing. Or . . . . you could always use headphones.

Or a large electrostatic, no frequency limits. King Sound King, 28hz-35khz.

[ClaudeJ1]

Time alignment issues and the benefits thereof, are what makes a good single driver appealing. A single driver is time aligned. The cohesiveness, imaging, soundstage, etc. are excellent. Horns, female vocals, etc. sound very good. The problem is it all comes at the sacrifice of extreme high and low frequencies.

After listening to an excellent single driver, one is compelled to achieve similar results through alignment of tweeters, squawkers and woofers. That is the way to get the imaging of a single driver with the dynamics and frequency range of a 3-way system. Physically aligning a horn loaded woofer is impractical, leaving the electronic solution.

Or . . . . you could always use headphones.

Or go with Synergy Horns from Danley. Seriously.

Edited September 14, 2014 by ClaudeJ1

[Quiet_Hollow]

Or a large electrostatic, no frequency limits.

All the same problems as a line array, without any benefit of the work-arounds. Very limited dynamic capability, dipole radiation, and all the the IMD of a direct radiator. 

[jcmusic]

The IR I posted is of near perfect alignment of the tweeter/mid drivers in my room, what it has done is cleared up any smearing that was present and opened the sound stage tremendously!!! Everything now is so clear it's spooky, it has even made the bass sound clearer.Well worth the effort for me, I think the only thing better would going active to get the woofers aligned with the other two drivers!!!

Edited September 14, 2014 by canyonman

[Schu]

you don't know what audible means?

 

 

is this specific difference from here:

 

 

 

  to here:

 

 

 

 

actually significantly audible?

 

or is it more of a measured difference?

Edited September 14, 2014 by Schu

[erik2A3]

Good single, full-range drivers DO, as mentioned above, sound very good because of being single-point-source. I have Lowther PM5As (they are not inexpensive drivers, however) on open baffle panels and they they lightening fast, electrostatic-like. The other benefit of this type of driver is the fact that they do not use the sort of necessary evil of passive crossover components. The most expensive capacitors and chokes in the world would do far more to mess up an otherwise very clean signal than running a driver like this full-range. However, some do use passive networks, particularly series networks (also my preference in many cases) in similar open baffle arrangements with a higher crossover point.

The only low frequency (say from about 200 cycles and below) I have used that is able to keep up with the Lowther driver has been the bass bin of either Klipschorns (as Forum member Cut-Throat is now doing with his Oris Horns) or as in my case, La Scalas.

[russ69]

 

Or a large electrostatic, no frequency limits.

All the same problems as a line array, without any benefit of the work-arounds. Very limited dynamic capability, dipole radiation, and all the the IMD of a direct radiator. 

 

"Limited dynamic capability". They are for home use, not for filling stadiums. I still have some hearing left, I don't need 115 db inside my house.

"dipole radiation". Some of us do not look at that as a negative.

"Same problems as a line array". Bring it on, most of my favorite speakers are line arrays.

"IMD of a direct radiator". Well if IMD is a detriment to detail and precision, I can live with that, electrostatics are known for their detail and accuracy.  

[jcmusic]

you don't know what audible means?

 

 

is this specific difference from here:

 

 

 

  to here:

 

 

 

 

actually significantly audible?

 

or is it more of a measured difference?

Of course I know what audible means, I just didn't know what you were asking. Please be more specific, if you are asking do I hear and audible difference I answered that in post #48...

[artto]

The only problem is that even though you can get the drivers time-aligned, you still won't have a phase coherent (linear phase) system. Time alignment helps with better focus but the rasp & harshness will still be there albeit to a lesser extent. In the end I finally decided to can the idea. In the old days this probably didn't matter so much because other forms of distortion and distortion in other components masked it. This is one area where Klipsch has really missed the boat and unfortunately Paul Klipsch never quite got it either. However one has to keep in mind the era/generation PWK is from and what was known and available. I'm 62 and PWK graduated college years before my parents were even born. He worked at GE when Thomas Edison was CEO.

 

I guess what I'm saying is the time alignment exercise with Klipschorns is interesting, and while not without benefit, is of limited usefulness in that it still doesn't solve the problem. Not to name names, today there are other products which have. And I'll bet my booty that PWK or Roy Delgado wished that they had thought of it first.

[mark1101]

Arto,

 

Interesting comments.  I don't know why you say you can't phase align.  You sure can and it's not that hard to do.

 

I would say that once aligned any left over harshness is either the old exponential horns or something outside the speaker.  Because with a good processor you should be able to dial out the smear (misaligned overlap) which I always thought was the source of the "Klispch harshness".

[Chris A]

Phase linearization of loudspeakers is not only possible, it's fairly cheap to try out:

 

http://www.diyaudio.com/forums/multi-way/221434-rephase-loudspeaker-phase-linearization-eq-fir-filtering-tool.html

 

However, all the instances that I've read say that it's basically inaudible.  This coincides with all the data I've seen on audibility of phase in human hearing.  Phase misalignments in the crossover regions is another matter, however.

 

Chris

Edited September 18, 2014 by Chris A

[Chris A]

Because with a good processor you should be able to dial out the smear (misaligned overlap) which I always thought was the source of the "Klipsch harshness".

I'm beginning to believe that the harshness you speak of is more related to the K77 tweeter (ElectroVoice T35A) and the upper reaches of the K-400/K-55 series of midrange horns-drivers. 

 

Once you swap tweeters and lower the crossover frequency of the midrange-tweeter crossover, things seem to get much smoother.  Using the K-77 time- and phase-aligned in the crossover to the midrange horn was still harsh, IMHO.  But using a Beyma CP25 tweeter and lowering the crossover frequency to 2.5 KHz had a real effect on the elimination of harshness.

Edited September 18, 2014 by Chris A

[Fjd]

Phase misalignments in the crossover regions is another matter, however.

 

Chris, are you aware of any suggested reading for this issue?  I started to think about this issue during the summer and look for sources on this topic as I've read many debates on various forums regarding choosing a crossover point between drivers with what seems to be very little regard to driver performance, but life issues have gotten in the way and put my reading on hold for a little while and I would like to catch up at some point. 

 

I'm not sure if I'm thinking about it correctly; however, I keep thinking that if I start out with an odd-order crossover (i.e. 18 dB slope) and the phase between drivers starts out at 90 degrees off, it seems that any driver response anomalies around the crossover region could push the phase into a greater degree off (i.e. into that 120 - 180 degree range) and create more cancellations potentially resulting in dis-jointed sound.

 

 

 

[Chris A]

Chris, are you aware of any suggested reading for this issue?

I'd start with Linkwitz: http://www.linkwitzlab.com/crossovers.htm and http://www.rane.com/note160.html and http://en.wikipedia.org/wiki/Linkwitz%E2%80%93Riley_filter

 

Note that I'm using an ElectroVoice Dx38 digital crossover for my Jubs and another one for my center Belle, and in particular, the Linkwitz-Riley 24dB/octave digital filter options for the crossovers filters, so my experience using passive filters of any order isn't recent.  I'm reporting on my own listening tests using these hardware items that I listed above (tweeters, crossover, and stock Belle midrange/driver), not a third-party opinion.

 

EDIT: If you look at the CP25 tweeter data sheet you'll see that the FR at 2.5 Khz is good to even as low as an octave below, i.e., 1.25 KHz and at least -24 dB from its 2.5 KHz level if using a 24 dB L-R crossover filter

.

 

Chris

Edited September 18, 2014 by Chris A

[artto]

Sorry guys. You are quite wrong.

 

While time alignment and phase are related (as is frequency), there are differences.

 

There is absolutely no way to make a large horn system like a Khorn or Jubilee phase coherent.

 

An example of a linear phase fully horn loaded speaker system is the Danley SH50. Some of the Frazier models are also linear phase. Even a single driver speaker system is not necessarily phase coherent, nor behave as a single point source.

 

http://www.danleysoundlabs.com/danley/wp-content/uploads/2012/01/line-array-paper.pdf

 

There is a difference. And it is audible. How audible depends on the program material, room acoustics and associated equipment used, AND your own personal experience.

 

Example: If you do what PWK did ~ take a single Heresy into a church and put it up on a stand and play a multi-miked recording to a couple hundred people who are not trained listeners & don't really care all that much about audio in the first place and none of them can hear any difference, what does that prove? It proves that a bunch of people who aren't familiar with the recording being played on a non-linear phase speaker in an extremely reverberant room (way more reflections than direct sound reaching their ears), can't hear the difference with nothing to compare it to. No Kidding. What a surprise.

 

FWIW (here comes my schpeel) I have dozens of original master recordings that I've made myself, most in the same church with the same singers and musicians several times a year over 25 years, not including the ones I've been involved with in my own bands the past 40+ years. Unless you have gone through that exercise, and done so many times, you don't (can't) have a known point of reference. Unfortunately most audio enthusiasts haven't had the same opportunity or ever been in a recording studio and heard the original master recording of anything much less have heard the original performance so that there's at least some point of reference.

 

On the other hand, if you're right, we need to see the combined radiation pattern of all drivers operating at the same time, not the impulse response(s) taken from the individual drivers one at a time as shown above. There should not be any lobbing error in the radiation pattern across the entire frequency range and especially at the crossover frequencies.

 

http://www.rane.com/note160.html

Edited September 18, 2014 by artto

[Chris A]

I've read many debates on various forums regarding choosing a crossover point between drivers with what seems to be very little regard to driver performance

Here is another thread on that issue, and one reason why I believe that the Jubilee sounds as good as it does (and there are other reasons why the Jubilee sounds so good, but that one reason will do for now), since it has one crossover point at ~425 Hz, which is just below the interaural directionality frequency--governed by the distance between the two human eardrums, corresponding to about 500 Hz. 

 

There is but one driver and one horn above that frequency in the Jubilee so its "phase coherence" (group delay) characteristics are basically unsurpassed above that point.

 

Note that the point of greatest aural sensitivity is about 3.5-4.5 KHz, found also by looking at the now proverbial Fletcher-Munson (equal loudness) curves.

 

 

 

Chris

Edited September 18, 2014 by Chris A