How to dissect polar patterns and frequency responses for horns.

[ClaudeJ1]

1 minute ago, Droogne said:

:s , I have been trying to not trip and fall (aka making stupid remarks) but I'm talking about things I know literally nothing about  let me rephrase! What Chris mean with:

 

"If the bass bin polars match the horn/compression driver polars in each axis (horizontally, vertically), then the loudspeaker will sound very good.  If there is a mismatch in polars at the crossover frequency then it will sound not as good in-room."

It appears you talking about the Jubilee or other "bass bin" when I was assuming you were discussing Danley Synergy horns (which I own) and Chris MEH versions which use the K-402. I was referring to the latter 2, not the Jubilee.

[Droogne]

Just now, ClaudeJ1 said:

It appears you talking about the Jubilee or other "bass bin" when I was assuming you were discussing Danley Synergy horns (which I own) and Chris MEH versions which use the K-402. I was referring to the latter 2, not the Jubilee.

Oh no problem, I wasn't very clear I admit  PS those danley speakers look dang awesome! Perfect size with awesome performance (so I read). Perfect center seize too. Bit pricey for me at this point (but oh well, the K-402 is that too)

[Chris A]

The polars on the Jub bass bin (KPT-KHJ-LF) in the horizontal direction get pretty narrow up around 400 Hz (anything above about 210 Hz)--much narrower than the K-402's ~110 degree coverage angle in that frequency band.  That's what I was referring to.  It's the main thing that the K-402-MEH prototype showed me in terms of listening differences between it and the Jubs, as well as coherent point source in that same frequency band.  The listening difference: clarity/speech recognition, and general midbass presence.  It's clearly different, in fact.

 

Chris

[jazzmessengers]

4 hours ago, Chris A said:

The polars on the Jub bass bin (KPT-KHJ-LF) in the horizontal direction get pretty narrow up around 400 Hz (anything above about 210 Hz)--much narrower than the K-402's ~110 degree coverage angle in that frequency band.  That's what I was referring to.  It's the main thing that the K-402-MEH prototype showed me in terms of listening differences between it and the Jubs, as well as coherent point source in that same frequency band.  The listening difference: clarity/speech recognition, and general midbass presence.  It's clearly different, in fact.

 

Chris

 

Wouldn't it be pretty difficult to design a bass horn that extends as low as the Jubilee bass bin that would have wide pattern control at the lower limit of where the K-402 is crossed over?

[Chris A]

6 hours ago, jazzmessengers said:

Wouldn't it be pretty difficult to design a bass horn that extends as low as the Jubilee bass bin that would have wide pattern control at the lower limit of where the K-402 is crossed over?

 

The short answer is "no", but such a bass bin design would probably look different--more like a University Classic design with one horn mouth instead of bifurcated into two horns with mouths side by side, or perhaps rather with the two bifurcated mouths fully rejoined together with a pointed nose like a La Scala bass bin.

 

Chris

[Droogne]

11 hours ago, Chris A said:

The polars on the Jub bass bin (KPT-KHJ-LF) in the horizontal direction get pretty narrow up around 400 Hz (anything above about 210 Hz)--much narrower than the K-402's ~110 degree coverage angle in that frequency band.  That's what I was referring to.  It's the main thing that the K-402-MEH prototype showed me in terms of listening differences between it and the Jubs, as well as coherent point source in that same frequency band.  The listening difference: clarity/speech recognition, and general midbass presence.  It's clearly different, in fact.

 

Chris

So essentially you made a great design, better? Love ittttt. Is that the reason youre gonna be making more of them? At the price theyre selling the K-402 in the US they "only" 1,75 times my LaScalas. 

[Droogne]

19 hours ago, jazzmessengers said:

In my opinion after doing a lot of searching there isn't anything commercially manufactured that is the equivalent or has similar performance to the K-402.

@Chris A@ClaudeJ1

 

For arguments sake, would this horn (if the information is correct) be a "match" for the K-402? I already looked up the price, and it's way too much (4500 dollars) so I wont be buying them.(the shipping prices (960dollars from Australia to Belgium, which is more than 2 times the distance from Kentucky, and that for a 33pound horn each) made me realise that the K-402 shipping prices cant be that bad, which is a good thing. yeay!). I got some interesting information though: 

 

"Hi Maarten,

 

The horn has a 60 x 80 degree pattern and the great thing about this design is that it not only extends low, but even the vertical is very well behaved, with no sharp changes and nothing in the directivity that gives away crossover points. The vertical maintains 60 degrees down to 600 Hz, below which it begins to widen, increasing to 100 degrees below 300 Hz. The crossover is 350 Hz and below this point the beamwidth widens as any direct radiator will have a wider pattern. Horizontally the 80 degree beamwidth is maintained well below the crossover point, holding up even at 200 Hz."

×  Pasted as rich text.   Paste as plain text instead

• Drag files here to attach, or choose files... 
  Max total size 2MB
• Insert other media 

Uploaded Images

•  
•  

 

23897891_1832883560089191_1257489635_n.jpg

81 kb · Done

•  
•  

[Chris A]

6 hours ago, Droogne said:

So essentially you made a great design, better? Love ittttt. Is that the reason you're gonna be making more of them? At the price they're selling the K-402 in the US they're "only" 1,75 times my La Scalas. 

Well, the K-402-MEH prototype needs no bass bin.  It's full range as-is.  But that's described in the linked MEH thread, above.  I'm assuming that you're interested in the K-402 vs. other like-sized horns, and perhaps mating that with something like a Jub bass bin or perhaps something like a La Scala style bin.  

 

I could use 5 K-402-MEHs in my setup, instead of the one that I have for a center channel.   Getting all the centerlines of the surround channels at about the same height above the floor is a big deal in terms of sound performance.  That's what I'm intending to build--four more. 

 

A Jub bass bin (KPT-KHJ-LF) can be used with the above mentioned K-402-MEH but crossed an octave lower (~200 Hz) than the K-402 in order to avoid the lower midrange crossover point polar mismatch and still get the improved midbass performance of the MEH.  Much smaller woofers can be used on the MEH instead if using a separate horn-loaded bass bin.

 

To redesign the Jub bass bin to do what I referred to above to avoid polar mismatch at the top of its range with the K-402 is perhaps straightforward, but it consumes time and resources to do a proper job and I haven't found the need to explore that path to date.  The idea of adding a nose extension to the Jub bass bin has been knocked around for a few years by some Jubilee owners--and discussed in passing by Roy, but until someone gets serious and does a prototype or two, I'm not sure it's going to occur anytime soon.  It takes up more floor space than the current bass bin.   I find the Jubilees' K-402s are a little too high off the floor sitting on top of the bass bins--for my room. I'd like them lower, and the full-range MEH design makes that easy to achieve.

 

6 hours ago, Droogne said:

For arguments sake, would this horn [the PSE-144] (if the information is correct) be a "match" for the K-402?

I haven't heard this horn with attached drivers and bass bin (direct radiator woofers are typically shown underneath the PSE-144 MEH).  I also haven't seen a good set of polar plots down to below the crossover frequency for this horn/driver assembly.   The only set of polar plots that I've seen to were less than spectacular when compared with the K-402 and a good compression driver, but that's not the final word on the PSE-144 until extended FR plots are produced. 

 

The cost of the PSE-144 is way out of line as you've indicated--about 2x to 3x too high.  And when you think about it, why use four drivers on an MEH to do what the K-402 and a good 2" compression driver can already do better?  It doesn't make sense when compared.  If you want to have a 3-way, I'd instead recommend the BMS 4592ND bi-amped on a K-402 instead of the PH-144 by a large margin, including cost and sonic performance. 

 

YMMV.

 

Chris

[Droogne]

1 hour ago, Chris A said:

Well, the K-402-MEH prototype needs no bass bin.  It's full range as-is.  But that's described in the linked MEH thread, above.  I'm assuming that you're interested in the K-402 vs. other like-sized horns, and perhaps mating that with something like a Jub bass bin or perhaps something like a La Scala style bin.  

I'm having 6 LS bins build, so that would be its "mate". Not against building a jub bin in the future though. And yes I'm reading through the MEH thread today. Interesting stuff!

1 hour ago, Chris A said:

 

I could use 5 K-402-MEHs in my setup, instead of the one that I have for a center channel.   Getting all the centerlines of the surround channels at about the same height above the floor is a big deal in terms of sound performance.  That's what I'm intending to build--four more. 

That's an interesting take! The height is one of the reasons why I would mate it with a LaScala. Would you place the fronts and rears on the same height as the center? Isn't that too low? 

1 hour ago, Chris A said:

 

A Jub bass bin (KPT-KHJ-LF) can be used with the above mentioned K-402-MEH but crossed an octave lower (~200 Hz) than the K-402 in order to avoid the lower midrange crossover point polar mismatch and still get the improved midbass performance of the MEH.  Much smaller woofers can be used on the MEH instead if using a separate horn-loaded bass bin.

Would this also work with a LaScala (not sure how high the LaScala holds pattern)? Even more interesting design! 

1 hour ago, Chris A said:

 

To redesign the Jub bass bin to do what I referred to above to avoid polar mismatch at the top of its range with the K-402 is perhaps straightforward, but it consumes time and resources to do a proper job and I haven't found the need to explore that path to date. 

I'm definitely gonna remember that once I get my hands on my K-402s.

1 hour ago, Chris A said:

 

The idea of adding a nose extension to the Jub bass bin has been knocked around for a few years by some Jubilee owners--and discussed in passing by Roy, but until someone gets serious and does a prototype or two, I'm not sure it's going to occur anytime soon.  It takes up more floor space than the current bass bin.   I find the Jubilees' K-402s are a little too high off the floor sitting on top of the bass bins--for my room. I'd like them lower, and the full-range MEH design makes that easy to achieve.

Again.. height is indeed one the things I did not really like about the Jubilee. The difference with the center would be huge. (when using a tv).

1 hour ago, Chris A said:

 

I haven't heard this horn with attached drivers and bass bin (direct radiator woofers are typically shown underneath the PSE-144 MEH).  I also haven't seen a good set of polar plots down to below the crossover frequency for this horn/driver assembly.   The only set of polar plots that I've seen to were less than spectacular when compared with the K-402 and a good compression driver, but that's not the final word on the PSE-144 until extended FR plots are produced. 

 

The cost of the PSE-144 is way out of line as you've indicated--about 2x to 3x too high.  And when you think about it, why user four drivers on an MEH to do what the K-402 and a good 2" compression driver can already do better?  It doesn't make sense when compared.  If you want to have a 3-way, I'd instead recommend the BMS 4592ND bi-amped on a K-402 instead of the PH-144 by a large margin, including cost and sonic performance. 

 

YMMV.

 

Chris

Yeah I saw it mentioned in some thread about the K-402, but nothing really justified the cost and amount of drivers. Looks rad though. And about the 3-way, main reason why I was inquiring into the 3-way MEH was because of the fact that the rest of my 11 surround will consist of K77/K55 drivers, and I wanted a close timbre match. But that's quite a moot point, as I'll probably modify them all in the end to be 2-ways with the same (line of) drivers as what I'll be using in the K-402 LCR. But first the K-402. Then I'll start inquiring into K-402 drivers (the faital HF200 is recommended and is very reasonably priced, which makes it possible to put them in all 11 speakers). Then I'll worry about the horns (as the ones I'd be using in the height channels will be very size restricted) to timbre match (as close as possible) the K-402. They would all be actively crossed (the LCR with a Xilica normally, and the heights with something cheaper like a miniDSP HD), and I've read some nice things about timbre matching and EQ in one of your replies about this subject. Maybe I'll try out the MEH horn with the Bassbin along the way, who knows. 

 

All in good time. First those damn K-402 horns  I would be able to get them pretty soon if I find a European dealer, if I need to buy from the US I'll probably have to save some more first. I'll keep you posted, as I'll probably have a boatload of question when I get them. 

 

 

[Chris A]

I've found that timbre match more or less equals matching frequency response curves on axis.  I can only demonstrate this in person.  Some want to believe that this isn't true, and that it has something to do with the material of the diaphragms.  I can assure you that it's correlated to on-axis frequency response.

 

The apparent source width/height is a function of the horn-driver coverage angles.  I mentioned above, if those off-axis polars have uneven frequency response with respect to the on-axis response, then you have a timbre mismatch (that you're stuck with unless you use a lot of absorption to eat those off-axis polars in room).  However, the aural effects of uneven polars on timbre is usually minimal if the horn/driver combination are pretty good in terms of smooth polars off-axis.  This is apparently the case for Klipsch Heritage midranges and tweeters, as I found first-hand using a stock Belle center (tri-amped, time aligned with flattened frequency response) between the Jubs.  Using progressively better horns and tweeters on the Belle bass bin decreased the timbre and apparent source width differences, until I arrived at the K-402-MEH, where the center was at least as good as the Jubs in timbre and in apparent source width.  The clarity and speech recognition of the K-402-MEH is startling.  (I believe that a lot of people don't like center channel loudspeakers as specified in the 5.1 ITU-R BS.775–3 arrangement, shown schematically below) because they never have used center loudspeakers that are good enough for the role that they're placed in.  It easily needs to be the best loudspeaker that you own in order to integrate properly, I've found subjectively via experiment.)

 

 

I've found subjectively via experiment that in my room, having all channels approximately at the same height has strong advantages sonically.  The placement around the listening position needs to be at least ear height while sitting at the main listening position (LP), so I'm assuming that this is at least a metre off the floor (ITU recommends 1.2 metre, above).  The surrounds can be slightly higher (about half a metre) based on distance to the LP, as shown.  But for full-surround channel recordings (like this one, for instance), having the surrounds at the same height all around is a definite advantage.

 

Chris

[jazzmessengers]

5 hours ago, Chris A said:

 

The short answer is "no", but such a bass bin design would probably look different--more like a University Classic design with one horn mouth instead of bifurcated into two horns with mouths side by side, or perhaps rather with the two bifurcated mouths fully rejoined together with a pointed nose like a La Scala bass bin.

 

Chris

 

Why would a design like this allow the bass bin to have wide coverage at say 450 Hz (using your steep crossover to K-402) and have the deep bass extension of the Jubilee bin? Maybe this is a discussion for another thread, hope the OP doesn't mind!

 

Actually I might start a thread on this in the technical discussion forum this evening.

[Droogne]

10 minutes ago, Chris A said:

I've found that timbre match more or less equals matching frequency response curves on axis.  I can only demonstrate this in person.  Some want to believe that this isn't true, and that it has something to do with the material of the diaphragms.  I can assure you that it's correlated to on-axis frequency response.

 

The apparent source width/height is a function of the horn-driver coverage angles.  I mentioned above, if those off-axis polars have uneven frequency response with respect to the on-axis response, then you have a timbre mismatch (that you're stuck with unless you use a lot of absorption to eat those off-axis polars in room).  However, the aural effects of uneven polars on timbre is usually minimal if the horn/driver combination are pretty good in terms of smooth polars off-axis. 

Ok, so gonna try to rephrase this to see if I get it. With uneven you mean there is a relevant difference between the on and off axis frequency response. With smooth off-axis polar you just mean a good frequency response, not compared to the on-axis response? 

10 minutes ago, Chris A said:

This is apparently the case for Klipsch Heritage midranges and tweeters, as I found first-hand using a stock Belle center (tri-amped, time aligned with flattened frequency response) between the Jubs.  Using progressively better horns and tweeters on the Belle bass bin decreased the timbre and apparent source width differences, until I arrived at the K-402-MEH, where the center was at least as good as the Jubs in timbre and in apparent source width.  The clarity and speech recognition of the K-402-MEH is startling.  (I believe that a lot of people don't like center channel loudspeakers (the 5.1 ITU-R BS.775–2 arrangement, shown schematically below) because they never have used center loudspeakers that are good enough for the role that they're placed in.  It easily needs to be the best loudspeaker that you own in order to integrate properly, I've found subjectively via experiment.)

My center is easily my best speaker (and fore sure the most expensive at the moment) I have not neglected it  

10 minutes ago, Chris A said:

I've found subjectively via experiment that in my room, having all channels approximately at the same height has strong advantages sonically. 

The placement around the listening position needs to be at least ear height while sitting at the main listening position (LP), so I'm assuming that this is at least a metre off the floor (ITU recommends 1.2 metre, above). 

Normally this applies to tweeter height right? I was thinking about how I was gonna this with my LaScalas, I can easily raise the bins too but would it make a difference if I only place the MF/HF higher (by like 20-30cm?).

10 minutes ago, Chris A said:

 

The surrounds can be slightly higher (about half a metre) based on distance to the LP, as shown.  But for full-surround channel recordings (like this one, for instance), having the surrounds at the same height all around is a definite advantage.

 

Chris

 

Also, how would you try to timbre match your height channels in your setup (if you were to do that hypothetically)

[Droogne]

8 minutes ago, jazzmessengers said:

 

Why would a design like this allow the bass bin to have wide coverage at say 450 Hz (using your steep crossover to K-402) and have the deep bass extension of the Jubilee bin? Maybe this is a discussion for another thread, hope the OP doesn't mind!

 

Actually I might start a thread on this in the technical discussion forum this evening.

I dont mind derailing this thread, but I dont mind a more appropriate seperate thread either ofcourse. Makes it clearer for everyone! 

[jazzmessengers]

1 minute ago, Droogne said:

I dont mind derailing this thread, but I dont mind a more appropriate seperate thread either ofcourse. Makes it clearer for everyone! 

 

It's fine, I'll start the thread if not tonight then this weekend. It's a different enough topic to warrant it. I see @Chris A has already touched on the Jub bass bin sitting a bit too high and I thought the same thing when I heard the full Jub/K-402 system.

[Droogne]

1 minute ago, jazzmessengers said:

 

It's fine, I'll start the thread if not tonight then this weekend. It's a different enough topic to warrant it. I see @Chris A has already touched on the Jub bass bin sitting a bit too high and I thought the same thing when I heard the full Jub/K-402 system.

It might also take my focus away from this forum  spending way too much time here haha! At this tempo I'll still be distracted when my exams start in january, cant have that  

[Chris A]

55 minutes ago, jazzmessengers said:

Why would a design like this allow the bass bin to have wide coverage at say 450 Hz (using your step crossover to K-402) and have the deep bass extension of the Jubilee bin? Maybe this is a discussion for another thread, hope the OP doesn't mind!

 

I can answer the question above on bass bins here briefly.  (If the discussion continues on this sub-subject, I recommend another thread be started on that.) The problem is lobing due to the dual separated horn mouths that are not rejoined smoothly (like the La Scala is rejoined).  See figure below:

 

 

47 minutes ago, Droogne said:

Normally this applies to tweeter height right?

I'd pick the midrange centerline instead.  If the tweeter is separated by more than an inch or so, you'll get lobing, like in the figure shown above, which is a function of vertical separation distance and the wavelengths at the crossover frequency interference band.

 

47 minutes ago, Droogne said:

Also, how would you try to timbre match your height channels in your setup (if you were to do that hypothetically)

First, you would need to select height loudspeakers having the same frequency response range as your other loudspeakers, then you'd EQ their frequency response flat like your other loudspeakers using a DSP crossover or similar device upstream that does parametric equalization.  I found this out using a tri-amped, time aligned, dialed-in Belle center between the Jubs--its timbre isn't the same due to the fact that its frequency response lacks the bottom octave or so of the Jub bass bins. Once the K-402-MEH hit the scene, I had essentially perfect timbre matching.  The surrounds also need to be timbre matched (mine are, presently).  If you use narrow coverage front loudspeakers (i.e., L, C, R) and wide coverage bipole or dipole surrounds, you will never timbre match them all in room.

 

One reason why I've not been interested in Atmos is because the timbre match between elevation and ground-plane loudspeakers in the array that manufacturers are selling can't do this.  It's apparently a problem with the Dolby specification for home theaters and their "certification"--if there is such a thing, such as what THX provides in terms of QA of loudspeaker designs/implementation.  Like 3D movies, it looks to me like the Atmos thing is basically going to go by the wayside over time.  It's a sound effect thing thus far--not music like 5.1 music recordings (multichannel SACD and DVD-A, 5.1 downloads) have been doing for 10-15 years now.   YMMV. 

 

Chris

[Droogne]

2 minutes ago, Chris A said:

 

I can answer the question above on bass bins here briefly.  (If the discussion continues on this sub-subject, I recommend another thread be started on that.) The problem is lobing due to the dual separated horn mouths that are not rejoined smoothly (like the La Scala is rejoined).  See figure below:

Does this mean the LaScala is better in pattern control? 

2 minutes ago, Chris A said:

I'd pick the midrange centerline instead.  If the tweeter is separated by more than an inch or so, you'll get lobing, like in the figure shown above, which is a function of separation distance and the wavelengths at the crossover frequency interference band.

So when the MF and HF are coupled I could get away with a separation of 30cm/1foot from the bass bin (that's 50% of the 500hz crossover point to the bassbin)?

2 minutes ago, Chris A said:

First, you would need to select height loudspeakers having the same frequency response range as your other loudspeakers, then you'd EQ their frequency response flat like your other loudspeakers using a DSP crossover or similar device upstream that does parametric equalization.  I found this out using a tri-amped, time aligned, dialed-in Belle center between the Jubs--it's timbre isn't the same due to the fact that it's frequency response lacks the bottom octave or so of the Jub bass bins. Once the K-402-MEH hit the scene, I had essentially perfect timbre matching.  The surrounds also need to be timbre matched.  If you use narrow coverage front loudspeakers (i.e., L, C, R) and wide coverage bipole or dipole surrounds, you will never timbre match them all in room.

So using height channels with horns with a similar dispersion pattern, same drivers and appropriate EQ should get me pretty close right? And I would cross all speakers at a point which I think I can reach with all 11 speakers. Ofcourse it's not perfect, but mounting  full blown LaScalas (or K402 MEHs is both expensive and ridiculously difficult)

2 minutes ago, Chris A said:

 

One reason why I've not been interested in Atmos is because the timbre match between elevation and ground-plane loudspeakers in the array that manufacturers are selling can't do this.  It's apparently a problem with the Dolby specification for home theaters and their "certification"--if there is such a thing, such as what THX provides in terms of QA of loudspeaker designs/implementation.  Like 3D movies, it looks to me like the Atmos thing is basically going to go by the wayside over time.  It's a sound effect thing thus far--not music like 5.1 music recordings (multichannel SACD and DVD-A, 5.1 downloads) have been doing for 10-15 years now.   YMMV. 

 

Chris

Well I'd be using a setup which is pretty versatile, and not so specifically or exclusively for Atmos alone. Because I would be using regular height channels (not modules) I think I can evolve with future advances in the field. I am aware of the superiority of the 5.1 (in the sense that it makes the most sense and has the most native sources) so I'll be giving the 5.1 the priority for now. Using the heresies as front height channels for now, which is pretty good match with the LaScalas. 

[Chris A]

36 minutes ago, Droogne said:

Does this mean the LaScala is better in pattern control? 

Yes--in terms of suppressing lobing at higher frequencies.

 

36 minutes ago, Droogne said:

So when the MF and HF are coupled I could get away with a separation of 30cm/1foot from the bass bin (that's 50% of the 500hz crossover point to the bass bin)?

You need less than 1/4 wavelength separation distance to avoid lobing. 

 

36 minutes ago, Droogne said:

So using height channels with horns with a similar dispersion pattern, same drivers and appropriate EQ should get me pretty close right? And I would cross all speakers at a point which I think I can reach with all 11 speakers. Of course it's not perfect, but mounting  full blown La Scalas (or K402 MEHs is both expensive and ridiculously difficult)

Yes.  Having "auxiliary loudspeakers" in the elevation channels will never really be timbre matched to the rest of the array, in practice.  (In commercial cinemas, they have the real estate to do a more proper job in terms of timbre matching all channels.)  They're more like "sound effects channels" than real loudspeaker channels.  YMMV.

 

36 minutes ago, Droogne said:

Using the heresies as front height channels for now, which is pretty good match with the La Scalas. 

This is probably the best that you can achieve.  The way that I check for timbre matching is to play pink noise through each channel, like the AVRs/AVPs provide in order to set channel gains using a hand-held SPL meter.  Any difference in timbre between channels is maximized using pink noise (something that Floyd Toole pointed out in his 3rd edition book). 

[Droogne]

10 minutes ago, Chris A said:

Yes--in terms of suppressing lobing at higher frequencies.

Would a LaScala stack ameliorate this even further?

10 minutes ago, Chris A said:

 

You need less than 1/4 wavelength separation distance to avoid lobing.  This is the same principle used the multiple entry horns that Danley provides.

Oh my mistake, so it's okay if I place the horns 15cm higher. Separation is between the mouth height or the borders of the horn? 

 

Again thank you! I dont think I got any questions left for now  no I just gotta wait on some good K-402 news! 

 

[Chris A]

"Separation" is centerline to centerline of the horns/drivers.