Jump to content

A K-402-Based Full-Range Multiple-Entry Horn


Chris A

Recommended Posts

42 minutes ago, Delicious2 said:

...but what about the MEH with 2 channel music to anchor the image in the center?   some modern version of PWKs center channel circuit?  something derived from the 2 channels but not synthesized in the sense of adding any effects.  Maybe just another memory preset on the Xilica could create this when there is no actual center channel.

 

I listen to a lot of multichannel music so I get to hear the center MEH daily for at least 1-2 albums and in the early evenings with ~2 hours of movie and/or streaming video.  (Remember that my daily listening time is in excess of 12 hours.)  I've just found that stereo sounds better in stereo with the Jubs at that spacing--when dialed in.  My Jubs and TH subs are against the front corners in a room that's 15.5 feet wide (40' deep, 9' high).  If they were in a room that was 25+ feet wide, then the story might change.  I've found, however, that 15.5 feet is definitely enough for both stereo and multichannel at my listening distance: about 10-11 feet from the front wall. 

 

For multichannel music and stereo that has sufficient stereo separation properties (i.e., extremely wide with a hole-in-the-center effect), the center channel creates a seamless wall of sound with the Jubs on each side--that spans the entire front of the room, with a phantom soundstage image that is much larger.

 

As far as creating a center image with the Xilica: I haven't tried it yet, but that is infinitely adjustable and is easy to control with XConsole running on my PC in real time.  It's quite easy to dial in a full mix of input channels A and B to the center channel, then adjust the center output channel gain to taste.  That's effectively what PWK's center channel mixer boxes did but without the adjustment in real time. 

 

Chris

  • Like 1
Link to comment
Share on other sites

Thanks Chris.  

11 minutes ago, Chris A said:

...stereo that has sufficient stereo separation properties (i.e., extremely wide with a hole-in-the-center effect), the center channel creates a seamless wall of sound with the Jubs on each side--that spans the entire front of the room, with a phantom soundstage image that is much larger.

and how do you achieve that center channel from stereo 2 channel music?

 

thanks for the much fuller explanation with only this question lingering.

Link to comment
Share on other sites

You can set the input of the center channel (output channels 5, 6, 7 in my setup that corresponds to the woofer, midrange and tweeter channels) to be the sum of input channels 1&2 (left and right front), as shown.  THen you can adjust the three output channels to the same relative gain--up or down--from the present positions (circled in red pen under the "Gain" column):

 

1526697674_BridgingcenterchannelXilicaconfig.thumb.JPG.4a5d662c5f5e1c48e1a097dba7b1c9cc.JPG

 

Chris

  • Like 2
Link to comment
Share on other sites

1 hour ago, Delicious2 said:

...and how do you achieve that center channel from stereo 2 channel music? 

I actually use one of the presets within the AVP if I sense something is amiss with a stereo mix: so I try "All Channel Stereo", "THX Music", etc.

 

I found more than one pop music DVD-A (i.e., I can upload the tracks to my computer without having to use special means to get them off the disc--like is the case with SACDs and DSD files on them) in 5.1 multichannel with almost no center channel output at all on the tracks.  I had to add between 10 and 12 dB of gain within Audacity to bring the center channel up to anywhere near the front left and right channels, and push down the front left and right down ~3 dB each in order to match the surround channels.  The resulting sound much different than the original tracks, with a locked-in center image even if listening a bit off-center.

 

I'm not quite sure why it is/was fashionable to severely attenuate the center channel gain in 5.1 discs, but I've found several like it.  A bit irritating--to be honest--but the fix is easy to implement.

 

Chris

  • Like 1
Link to comment
Share on other sites

1 hour ago, Chris A said:

I'm not quite sure why it is/was fashionable to severely attenuate the center channel gain in 5.1 discs, but I've found several like it.  A bit irritating--to be honest--but the fix is easy to implement.

 

I read something a while back, with regard to multichannel music, that stated that vocalists don't like to be placed "naked" (only their voice, no instruments) in the center channel. So many mix engineers avoided the center channel altogether.

Link to comment
Share on other sites

The particular album that I'm thinking of...is Alan Parsons' On Air:

 

MI0000325984.jpg?partner=allrovi.com

 

Many other DVD Video discs also have this problem. 

 

Chris

  • Like 1
Link to comment
Share on other sites

3 hours ago, Chris A said:

I actually use one of the presets within the AVP if I sense something is amiss with a stereo mix: so I try "All Channel Stereo", "THX Music", etc.

 

I found more than one pop music DVD-A (i.e., I can upload the tracks to my computer without having to use special means to get them off the disc--like is the case with SACDs and DSD files on them) in 5.1 multichannel with almost no center channel output at all on the tracks.  I had to add between 10 and 12 dB of gain within Audacity to bring the center channel up to anywhere near the front left and right channels, and push down the front left and right down ~3 dB each in order to match the surround channels.  The resulting sound much different than the original tracks, with a locked-in center image even if listening a bit off-center.

 

I'm not quite sure why it is/was fashionable to severely attenuate the center channel gain in 5.1 discs, but I've found several like it.  A bit irritating--to be honest--but the fix is easy to implement.

 

Chris

Let's face it. Allowing so called, engineers, to mix 5.1 for music, gives them infinitely more ways to screw it up than 2 channels ever did, which, with only 2, most of them screw up just fine most of the time.

 

  • Like 1
  • Haha 3
Link to comment
Share on other sites

On ‎3‎/‎12‎/‎2019 at 3:40 PM, ellisr63 said:

What would have happened if the woofers had been mounted x" away from the horn sides...would this have given you the acoustic delay you needed?

 

This is an interesting question when thinking about the design. I see it has been answered, concerning the fully functional model currently being enjoyed by the folks who have them, but some of us like to daydream about building one.

  • Like 1
Link to comment
Share on other sites

Yes...adding air compliance between the woofer cones and the off-axis port in the horn is like adding an acoustic low pass filter in series with the electrical one in the DSP crossover, resulting in an undesirable loss of bandwidth on the woofer's top end. 

 

However, you can move the ports closer to the apex of the horn (throat) to reduce the time lead of the woofers if that's the concern.  The tradeoff, of course, is that it will affect the crossover point and off-axis polars if the ports are too large or crossed at too low a crossover point.I found that a simple first order filter crossover adds enough delay to the woofer channel to time align them to the throat compression driver.  With a first order crossover filter in the woofer-midrange channel, the K-402-MEH is presently time aligned. 

 

Chris

  • Like 1
Link to comment
Share on other sites

  • 1 month later...
On 1/11/2016 at 4:46 PM, Chris A said:

 

Well, the side entry ports on the conical section of the horn determine the low pass point of the woofers. I chose 475 Hz and used Hornresp to determine where that point within the horn is. I know that was a good crossover point from the last 7 years of using Jubs crossed about at that point to the bass bins.  I wanted to get as far away from the throat of the horn as I could to avoid diffraction issues with the holes because I knew that they needed to be big holes--due to using 15" woofers. 

 

I surveyed all the Unity and Synergy designs and scaled the through hole sizes for each model.  I knew that the holes needed to be in the "ditch" between the side and top/bottom walls based on this survey.  I used the maximum diameter that I dared, and elongated the holes toward the mouth along the "ditches" until I got about a 10:1 compression ratio on the 15" woofer piston area vs. the hole slot area [which is a rule-of-thumb maximum for the compression ratio used in practice]. 

 

No one puts a hole in the middle of a rectangular conical horn wall, I noticed.  I followed suit (I had no reason not to). All the designs have two holes per woofer--except the SH96, which is a special case having one very elongated hole per each of its four 15" woofers--two per side.

 

 

On 1/11/2016 at 4:46 PM, Chris A said:

 

I am trying to understand this...

1: the ditch is the area between the walls of the horn?

2: How do you determine the compression ratio for a 15" woofer? Would it be the same for all 15" woofers?

 

I  am ready to drill my holes in my k402, and want to get it right the 1st time.

 

Thanks, Chris for having the balls to cut a k402 up! 

 

Link to comment
Share on other sites

I would have a question regarding point source.

Not entirely sure that I understand it. 

Real music is not poit source, meaning when musicians are on stage, every one of them is emiting its own sound from diferent direction. 

 

What is the "point source speaker" regarding that fenomenon? Of course it has to be connected with reproduction, but how?

Poeple are talking about point source "behaviour" of speakers. Is this where the catch is, and how?

Link to comment
Share on other sites

Each instrument and each vocal is a point source though. There's no doubt that a speaker that has a driver that covers most of both the treble and midrange sounds considerably more coherent. When you have gotten used to it it's difficult to go back to a speaker with a traditional crossover in the 2-4KHz region.

 

Link to comment
Share on other sites

Each element (instrument, voice) being reproduced was a point source.  Of course there's no way to recreate that with a recording apart from each element having both its own recording track and faithful sound reproducing mechanism positioned appropriately.  If an ensemble were recorded with two faithful mikes/channels and two faithful speakers were positioned approximately in the microphone positions a pretty good illusion is obtainable (even more so with the mikes at a listener's ears and played back on faithful headphones).

 

But none of all that pertains to this discussion, where "point source" merely means that all frequencies produced are emanating from a single point in space, or at least very nearly so.  Multiple drivers each handling a portion of the range makes it difficult for a few reasons, only one of which is that they can't occupy the exact same place.  Personally, I think it matters little if the sound of a cymbal is coming from a couple feet higher than the sound of a kick drum, but the transition area between any two "adjacent" drivers' coverage is the difficult part.

 

I really like the present idea of a multiple-entry horn and some day I intend to own a pair.

Link to comment
Share on other sites

20 hours ago, parlophone1 said:

...Real music is not point source, meaning when musicians are on stage, every one of them is emitting its own sound from diferent direction.  

 

What is the "point source speaker" regarding that phenomenon? Of course it has to be connected with reproduction, but how? 

People are talking about point source "behaviour" of speakers. Is this where the catch is, and how?

I think that we need to go back to the idea of how performances are recorded.  The simplest case (nowadays) is a stereo pair of microphones spaced apart a few feet horizontally across a stage inside a performance theater near ear height of the conductor--each with an omnidirectional microphone pattern:

 

spacedomnis.jpg

 

Once the recording is made with these stereo microphones, the user wants to recreate the sound field that was recorded by replacing the microphones with loudspeakers in order to create the best facsimile of the actual music performance that's possible from the recording (i.e., from the sampling of two points in physical space).  So you place two loudspeakers with omnidirectional radiation patterns where the original microphones were located and replay the music recording back through these two loudspeakers. That's the best that you can do to recreate the soundfield that was recorded on two channels.

 

If you now realize that a home hi-fi room doesn't have the same scale in terms of distance from the walls/floor/ceiling as the original (theater) performance, then you know that you don't want omnidirectional loudspeakers because the acoustic energy that they would emit would bounce off of the nearby walls, as if you lived in a room with mirrors on all the walls, floor, and ceiling:

art_room-setup3.gif

 

If you position a light at one of the real loudspeakers, sitting at the listening position you would see the real loudspeaker, but then you'd see a reflection in the side wall, front wall, ceiling and floor of four other loudspeakers that also produce light.  In the case of acoustic energy, you're looking at pretty good reflections from the floor, ceiling and front room corners, except that the acoustic energy from those reflections aren't "white light", but rather tinted toward the red end of the spectrum (longer wavelength/lower frequency), thus producing a net total of one white light and four red lights that you're looking at, and the apparent physical separation of those reflections produces phase delays.  When two sources of emitted energy are spaced apart, they begin to form a diffraction pattern with null axes that extend out radially from the apparent center between the sources (which also happen to move as a function of frequency of the acoustic waves), thus producing the well-known diffraction effect:

 

u12l3a1.gif

 

273748176_sidelobelevelvsseparationdistance.png.39544522797c7ae3461b5680c695639f.png

 

If you're trying to reproduce the music performance of the stereo microphones in a small hi-fi room, you'd want to avoid those extra acoustic reflections in the room because their acoustic reflections would not be "in phase" with the direct arrival energy from the loudspeaker.  At 1/2 wavelength spacing between the emitting sources (1/4 wavelength at the wall interfaces), they produce nulls at those 1/2 wavelength points:

 

speaker-boundary-interference-response.j

 

So what you'd like is to point all the acoustic energy at your listening position, but none on the walls/floor/ceiling, because of the undesired reflections that are spread apart in phase and in "color" or the balance of frequencies to make the equivalent of "white light" or "white sound".  Otherwise, you've got one good source of "white light" and four bad sources of "red light" that are all spread at different phases/apparent distances from the viewer/listener.  All those extra sources are not desirable because they degrade the fidelity of the original direct arrival energy. 

 

Directivity is your new friend, except that most loudspeakers with directivity (many are horn-loaded, but also some MTM-type designs--at least have directivity in the vertical plane) have moving directivities vs. frequency--which is again not desirable.  But you want the same coverage angles vs. frequency, otherwise you'll have the equivalent of blue-colored light on-axis, and red-colored light off-axis.  You want white light being emitted into the room--not on the room's boundaries but directly into the room volume, and you don't want the "color" of the acoustic energy to change vs. angle.  So a point source of acoustic energy with very good directivity control out to about 45 degrees off the center axis of the loudspeaker (90 degrees included angle).  Additionally, you still don't want the reflected acoustic images of the walls, floor, and ceiling to be phase-delayed relative to the direct arrival energy, so you place the loudspeaker in the corner of the room, and the reflected images off three of the room's boundaries (floor, walls) are coincident--i.e., the phase of the nearby reflections are approximately the same as the direct arrivals. (Some absorption on the adjacent walls and floor will aid a great deal in reducing the highest amplitude reflections that are almost in phase, but not quite.)

 

Klipschorn_speaker_drawing_1948.png

 

Voilà!  Point source loudspeakers are your best choice:

 

aHiN5RirjWrVgYUZ1fZ0I1u1tubnCZl1MA_44482

 

Chris

  • Like 2
Link to comment
Share on other sites

I don't know whether this is the best example, but it was handy.  The same sort of interference distortions Chris speaks of are handily comprehended when looking at images which can be degraded in exactly the same way as with audio information.  A picture can be worth a thousand words (or sounds).  Your eyes can easily pick out such anomalies, and you ears can just as easily do so, especially when you're aware of their existence.

 

Increase and decrease the size of ForteEbonyPair.jpg and watch what happens.

 

  

Link to comment
Share on other sites

Chris A:

You forget that a speaker which has a good power response, reflections don't contribute negatively anymore. Just ask Toole and his followers!

 

Kidding and sorry couldn't help myself. But it's quite disturbing and sad to see that what I wrote above is considered by many as science now and isn't even ut for debate! Just read some of the posts in the discussion I had with amirm and Toole at audiosciencereview forum.

Link to comment
Share on other sites

2 spaded Omni's is how PWK made ALL of his recordings. I heard several at his house in 1985 and they were terrific. I also own both of the Klipsch Tapes DVD's, which are still available. Amazing live recordings from when I was only 3 years old in the 50's!!

 

  • Like 1
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...