Jump to content

Subconscious Auditory Effects of Quasi-Linear Phase Loudspeakers


Chris A

Recommended Posts

I recently recalled a quote from Joseph Campbell from his "The Power of Myth" series from the mid 1980s--just before Campbell's passing:

 

Quote

"My friend Heinrich Zimmer used to say the best things can't be told because they transcend thought.

The second best are misunderstood because those are the thoughts that are supposed to refer to that which can't be thought about, and one gets stuck in the thoughts.

The third best are what we talk about."

So try as we might: the second best things are misunderstood because we're trying to talk about those things that we can't fully hold in our minds as we're writing/talking about them.  On the subject of harshness elimination, as soon as I start to write about it, I lose probably 90% of what I was intending to say--that's important to say. 

 

As I write this, I remember first listening to a DVD-A from 2007 that I acquired used last year: Fear of a Blank Planet by Porcupine Tree.  As I put it into the Oppo player before I had ever heard it before, I found that I was consciously steeling myself for the sound I knew was going to emerge from the setup. (I wasn't wrong.)

 

I say "steeled" because it's a common experience nowadays.  We have audio recording technology that, in my childhood I couldn't even dream about its quality of capturing the recorded signal (which is now considered "old hat" technology)...but somehow whose audio fidelity on this disc has actually been compromised to the point that my hearing can't brave the resulting sound without first having to distance myself from its sound (and also forgetting about its lyrics). 

 

Today, I'm listening to it again and it's a totally different story.  While it's clear that the harshness that I experienced last year is still underlying the sound of this DVD-A, I'm now able to listen without having to steel my hearing. 

 

So what changed? 

 

Flat-phase loudspeaker performance. 

 

It's the "why" that's so much more difficult to talk about.  I'll give you a clue:

 

When music or sound in general is played back on typical loudspeakers having non-flat SPL and phase performance, it's altered in terms of the time that all the harmonics of each instrument/voice arrive at our ears.  It turns out that our hearing systems can actually perceive those relative arrival times of harmonics--how they sum and cancel each other in the resulting sound received by our hearing system.  If those in-room harmonics are lined up like the microphone recorded them, we perceive a pleasant and realistic sound.  If however those higher harmonics are shifted in time (usually forward) relative to other frequencies, suddenly our hearing system must deal with a different resulting waveform with the peaks and dips at different times and at different levels than the original recorded signal waveform.  If the original sound that's been recorded has been heavily processed via audio mastering techniques--mostly to increase its perceived loudness vs. time via limiting and compression--but also the musicians' electrical/electronic instrumentation in the recordings altered to sound more "aggressive", we often perceive the resulting altered music (produced via the loudspeakers) as much more unpleasant and harsh sounding.  When that same music is played back using flat amplitude/phase response loudspeakers, the perception of the sound experienced can be much different.

 

So does this have anything to do with the difference in sound between direct radiating loudspeakers and horn-loaded ones?  Well, it is reasonably well known that the phase shifts in many horn-loaded loudspeakers can typically be much greater than in some for-purpose direct radiating "monitors on a stick" that are designed to minimize their phase distortions. 

 

Can horn-loaded loudspeakers be made to eliminate most or all of those phase distortions? 

 

Yes.

 

Chris

 

[EDIT 22 June 2020:  After I was able to rip the multichannel DTS 5.1 tracks from disc, I found that there was about 3 dB of clipping that was applied to them during the mastering of them--that I was able to reverse using "Clip Fix" to eliminate the harsh artificially introduced odd-order harmonics from the clipping of the tracks.  After this, I reduced the loudness of the tracks further via Audacity such that the overall loudness of the album tracks averages 0 (zero) dB on the ReplyGain scale used by the foobar2000 ReplayGain component plugin.  Here is the old vs. new DR (dynamic Range) database log files, the first before de-clipping and leveling to zero dB:

 

--------------------------------------------------------------------------------
Analyzed: Porcupine Tree / Fear of a Blank Planet
--------------------------------------------------------------------------------

DR         Peak         RMS     Duration Track
--------------------------------------------------------------------------------
DR10      -1.46 dB   -14.71 dB      7:30 01-Fear of a Blank Planet
DR12      -1.56 dB   -17.66 dB      5:06 02-My Ashes
DR11       0.00 dB   -15.58 dB     17:42 03-Anesthetize
DR12      -1.82 dB   -18.18 dB      5:27 04-Sentimental
DR11       0.00 dB   -16.57 dB      7:38 05-Way Out of Here
DR9       -1.57 dB   -15.66 dB      7:33 06-Sleep Together
--------------------------------------------------------------------------------

Number of tracks:  6
Official DR value: DR11

Samplerate:        48000 Hz
Channels:          6
Bits per sample:   24
Bitrate:           3938 kbps
Codec:             FLAC
================================================================================

 

...and the second Dynamic Range Meter log file after de-clipping and leveling the output (now including the four bonus tracks from the disc):

 

foobar2000 1.3.19 / Dynamic Range Meter 1.1.1
log date: 2020-03-25 02:47:35

--------------------------------------------------------------------------------
Analyzed: Porcupine Tree / Fear Of A Blank Planet (DTS)
--------------------------------------------------------------------------------

DR         Peak         RMS     Duration Track
--------------------------------------------------------------------------------
DR12      -5.65 dB   -22.89 dB      7:30 01-Fear of a Blank Planet
DR12      -5.12 dB   -24.41 dB      5:06 02-My Ashes
DR12      -4.48 dB   -22.33 dB     17:42 03-Anesthetize
DR12      -7.31 dB   -24.93 dB      5:27 04-Sentimental
DR12      -6.39 dB   -23.33 dB      7:38 05-Way Out of Here
DR10      -5.84 dB   -22.42 dB      7:33 06-Sleep Together
DR11      -4.68 dB   -22.40 dB      6:14 07-Nil Recurring
DR12      -6.65 dB   -24.49 dB      7:10 08-Normal
DR12      -6.57 dB   -24.22 dB      7:10 09-Cheating the Polygraph
DR11      -6.77 dB   -23.20 dB      8:28 10-What Happens Now
--------------------------------------------------------------------------------

Number of tracks:  10
Official DR value: DR12

Samplerate:        48000 Hz
Channels:          6
Bits per sample:   16
Bitrate:           1897 kbps
Codec:             FLAC
================================================================================

 

The real story is in how these sound on the Jubilees...which is a dramatic change from the as-produced tracks.  Now I can listen to the entire disc without clinching my jaw or wanting to turn it off at several points.   It's now "music" in the most complete sense--without the added audible clipping that was placed there during mastering.  How anyone could prefer the prior sound to the de-clipped (or un-clipped tracks if they happen to have access to the original mixdown tracks before mastering) is beyond me.  There just is no comparison.]

 

Chris

  • Like 8
Link to comment
Share on other sites

The best things I have personally recorded are the ones where I didn't jack the eq around and do complex compression or other effects. Mostly I recorded that way because I didn't have the tools with which to accomplish that. So now I go back and listen to those tapes I made 50 years ago and find them easy to listen to. Not because it was me/friends, but the recordings sound more natural.

 

You post above makes perfect sense to me, as has your long thread on de-mastering. I am learning to appreciate you explanations more and more as time goes by (partly as I understand more and more).

 

Bruce

Link to comment
Share on other sites

21 hours ago, Chris A said:

I recently recalled a quote from Joseph Campbell from his "The Power of Myth" series from the mid 1980s--just before Campbell's passing:

 

So try as we might: the second best things are misunderstood because we're trying to talk about those things that we can't fully hold in our minds as we're writing/talking about them.  On the subject of harshness elimination, as soon as I start to write about it, I lose probably 90% of what I was intending to say--that's important to say. 

 

As I write this, I remember first listening to a DVD-A from 2007 that I acquired used last year: Fear of a Blank Planet by Porcupine Tree.  As I put it into the Oppo player before I had ever heard it before, I found that I was consciously steeling myself for the sound I knew was going to emerge from the setup. (I wasn't wrong.)

 

I say "steeled" because it's a common experience nowadays.  We have audio recording technology that, in my childhood I couldn't even dream about its quality of capturing the recorded signal (which is now considered "old hat" technology)...but somehow whose audio fidelity on this disc has actually been compromised to the point that my hearing can't brave the resulting sound without first having to distance myself from its sound (and also forgetting about its lyrics). 

 

Today, I'm listening to it again and it's a totally different story.  While it's clear that the harshness that I experienced last year is still underlying the sound of this DVD-A, I'm now able to listen without having to steel my hearing. 

 

So what changed? 

 

Flat-phase loudspeaker performance. 

 

It's the "why" that's so much more difficult to talk about.  I'll give you a clue:

 

When music or sound in general is played back on typical loudspeakers having non-flat SPL and phase performance, it's altered in terms of the time that all the harmonics of each instrument/voice arrive at our ears.  It turns out that our hearing systems can actually perceive those relative arrival times of harmonics--how they sum and cancel each other in the resulting sound received by our hearing system.  If those in-room harmonics are lined up like the microphone recorded them, we perceive a pleasant and realistic sound.  If however those higher harmonics are shifted in time (usually forward) relative to other frequencies, suddenly our hearing system must deal with a different resulting waveform with the peaks and dips at different times and at different levels than the original recorded signal waveform.  If the original sound that's been recorded has been heavily processed via audio mastering techniques--mostly to increase its perceived loudness vs. time via limiting and compression--but also the musicians' electrical/electronic instrumentation in the recordings altered to sound more "aggressive", we often perceive the resulting altered music (produced via the loudspeakers) as much more unpleasant and harsh sounding.  When that same music is played back using flat amplitude/phase response loudspeakers, the perception of the sound experienced can be much different.

 

So does this have anything to do with the difference in sound between direct radiating loudspeakers and horn-loaded ones?  Well, it is reasonably well known that the phase shifts in many horn-loaded loudspeakers can typically be much greater than in some for-purpose direct radiating "monitors on a stick" that are designed to minimize their phase distortions. 

 

Can horn-loaded loudspeakers be made to eliminate most or all of those phase distortions? 

 

Yes.

 

Chris

 

On the Campbell quote, maybe I need to read a bit more of his writing to understand his viewpoint, but, I immediately find myself wondering if for Campbell and his friend "the best things" ARE the best BECAUSE they transcend thought.  That "things" (experiences) which challenge our ability to think about them, which can't easily be encoded in thought or word  are by that definition "better" or "best" BECAUSE they challenge us to expand our thinking about them and expand our communicating about them.

 

I put on my copy of Fear of A Blank Planet DVD-A and immediately thought I heard the "harshness" you speak of.  A question is how much harshness, aggressiveness, and distortion is intentional, especially if the recording is being mixed and mastered on playback systems that aren't flat amplitude/phase response?  This is particularly interesting in this case where you have Steven Wilson having so much creative control from writing to playing to mixing to mastering of FoABP and his many other stereo and surround creations. More that most other musicians, SW may have a more fully controlled and realized intent for the listener.  Are we straying from his intent for our listening experience by eliminating distortions that were present during the mixing/mastering? 

 

Perhaps I'm being a bit insincere on that line of argument.  Ultimately, we can't know exactly the creator's intention for our listening experience.  Also, one person's unpleasant harshness is another's exhilarating aggressiveness.  Personally I love PT and Steven Wilson's music and want to enjoy listening not have to "steel" myself against it.  So, I'll continue to follow you down the road of flat amplitude/phase response horn loaded systems since at bottom - it sounds better to me.  What can be said of "subconscious auditory effects"?  Not much unless we invoke words about the inadequacy of words - such as "ineffable" or notice the indirect evidence as you have observed - I find myself listening more and with better attention and enjoyment. 

 

Link to comment
Share on other sites

2 hours ago, Delicious2 said:

Are we straying from his intent for our listening experience by eliminating distortions that were present during the mixing/mastering? 

If that's an issue, I'm certainly in the wrong profession :wink: (or listening to the wrong albums...🤔).

 

2 hours ago, Delicious2 said:

Ultimately, we can't know exactly the creator's intention for our listening experience.

This is due to the RIAA industry refusing to set standards for sound reproduction and standing by them by disassociating from those that don't adhere to them.  The MPAA has sound reproduction standards, and that's why movies usually have much more consistent and higher fidelity historically.

 

2 hours ago, Delicious2 said:

Also, one person's unpleasant harshness is another's exhilarating aggressiveness. 

This is something that both Floyd Toole and Sean Olive have been proving to be largely not true (at least if the listener has average hearing that hasn't been significantly damaged).

 

2 hours ago, Delicious2 said:

So, I'll continue to follow you down the road of flat amplitude/phase response horn loaded systems since at bottom - it sounds better to me.

Thanks for that...

 

2 hours ago, Delicious2 said:

What can be said of "subconscious auditory effects"?  Not much unless we invoke words about the inadequacy of words - such as "ineffable" or notice the indirect evidence as you have observed - I find myself listening more and with better attention and enjoyment. 

That's the bottom line here, too.  I'm having much more trouble pulling myself away from the setup than I used to--even 6 months ago.  I think it's also been reflected in my album demastering (which mostly now consist of updates to my initial demastering attempts--which now sound much better than when I started on that path 4 1/2 years ago).  I've learned a lot--even more than I've learned about loudspeakers, rooms, and how to use DSP crossovers.

 

Chris

  • Like 1
Link to comment
Share on other sites

  • 2 months later...

Now that some time has passed since I started this thread, I'm now in a position to report back a bit more quantitatively on the effects of paying attention to phase flattening of loudspeakers which in my case includes all channels in a 5.1 array: left-right Jubilees w/TAD TD-4002 drivers, center K-402-MEH with BMS 4592ND dual diaphragm driver, and surround ESS AMT-1s on top of Belle bass bins.  All of these loudspeakers have been carefully bi-amped/tri-amped, and crossovers reworked to yield zero phase shifts through the crossover interference bands.  Note that all of these loudspeakers have matching timbre due to their full-range directivity (100 Hz--20 kHz) and dialed-in frequency responses. 

 

All of the above comments about "smoothness", "listener engagement", and "enhanced transparency" apply.  What's missing from the above discussion is how flat the measured phase and group delay now is across the entire 5.1 array to achieve the above subjective listening levels.  That performance to achieve that is shown below:

  • 180 degrees or less phase growth from 20 kHz down to 200 Hz:

948063150_PhaseResponseof5.1Array.thumb.jpg.f24774b0a6c91fb845ad7868259586b3.jpg

  • less than 1 ms of group delay above 500 Hz and less than 2 ms GD from 200 to 500 Hz:

1082800665_GroupDelayResponseof5.1Array.thumb.jpg.4730128ad6fbf9d3835dde5320b671d2.jpg

  • outstanding step response, rivaling the best studio monitors:

1021542237_TypicalStepResponse.thumb.jpg.4d9c94aadde3cadb66a2884699a6ecf9.jpg

 

In order to do this, time and attention to detail is required, as well as a little practice paying attention to the elements that can affect in-room measurement and getting the settings correct.  Using relatively good DSP crossovers is also required.  (No FIR filters were used to achieve the above results.)

 

The latest piece of the phase flattening effort came with the compensation of the high frequency diaphragm in the BMS 4592ND in the center channel K-402-MEH.  The addition of 0.145 ms (145 microseconds) of delay on the HF diaphragm channel produced not only an amazingly flat loudspeaker phase response (the blue trace, above), it also produced the last bit of "magic": apparent depth and seamless soundstage with multichannel music sources (5.1) without any hint of harshness.  In fact, it opened up some orchestral (classical) recordings to critical listening in that some 5.1 recordings the violins sound very subdued, while other recordings, the violins still sound very sharp (and these recordings have some level of mastering EQ applied, which also shifts their inherent phase responses).

 

Additionally, before I managed to get the center loudspeaker phase flattened to the level it is now, I listened to a lot of stereo recordings, namely the ripped music library of my CDs and DVD-As.  I initially found (a few months ago) that the flattened phase response of the Jubilees (purple trace, above) completely changed the sound of the demastered albums to the point that I have been systematically re-doing the demastering of these albums from the original discs off the shelf. 

 

So, bottom line...the changes have been pretty dramatic when taken as a whole.  The effect of flat phase and group delay response, when coupled with full-range directivity of fully horn-loaded loudspeakers, yields significant changes in the sound of the recordings. And the better the recordings (in terms of fidelity), the more dramatic change in the resulting sound.

 

This is probably the most important and surprising personal discovery of the effects of hi-fi setup/dial-in that I've written about on this forum.

 

...But I expect little reaction to this news because so many people already "know" that phase/group delay doesn't matter... :wink:

 

Chris

  • Like 2
  • Thanks 2
Link to comment
Share on other sites

I now have a different understanding now of the attraction of "full-range drivers" to some hi-fi listeners, even though I have never managed to enjoy the same level of enthusiasm in their music reproduction overall effect.  For many years, I had assumed that the reason for their popularity was due to their point source capabilities, but now I see this in a different light. 

 

I'm beginning to believe that the point source effect isn't as much the root of the listening experience, but rather their flat phase/group delay presentation.  (Full-range drivers all suffer from the same issues of polar directivity loss at lower frequencies due to the limited diameter of the moving diaphragm, and the effects of modulation and compression distortion due to their direct-radiating nature, especially at lower frequencies.)

 

119gdreams.passsystema.jpg

 

Fully horn-loaded loudspeakers, such as the Jubilee (suitably dialed in) and multiple entry horns (MEHs), like the K-402-MEH and the Danley Synergy series of loudspeakers all exhibit very low modulation and compression distortion, as well as very flat phase/group delay response--flatter than full-range drivers in fact--due to the horn loading of their moving diaphragms which reduces phase growth across the audible spectrum.  These fully horn-loaded designs can have very flat phase and group delay curves.

 

p5030811_1.jpg

 

198125039_Screenshot-2017-11-29-3TheHardpop.thumb.jpg.8e22eafb7fd360beed2c75dc38c69a19.jpg

 

Chris

  • Like 2
Link to comment
Share on other sites

19 hours ago, Chris A said:

(No FIR filters were used to achieve the above results.)

 

Do you have any notion whether such good overall phase results would be obtainable along with steeper slopes going FIR, and if so would it generally be worth the money / time / trouble?

Link to comment
Share on other sites

Certainly, if you have the compute horsepower in your DSP crossover or dedicated PC, FIR filtering will be able to pull the phase/SPL back to zero/flat SPL on axis with steeper slopes, but currently I've achieved 12-24 dB octave effective slopes at the crossover interference bands, but without phase penalty in the crossovers.

 

I generally have viewed FIR filtering as something that you can use but if you can bring the SPL and phase back to minimum phase (i.e., no excess phase), I doubt that more correction will be audible. 

 

I've tried to set my threshold of phase audibility to within 90 degrees of zero, and anything beyond that as probably not worth my time--but I could be wrong.  When I acquire a dedicated PC with associated 5.1 correction capability (probably JRiver on a higher horsepower PC running with a multichannel AES card), I'll let you know if that last few degrees of phase and 0.5--1 dB of on-axis SPL correction is audible, and whether or not effective phase correction below 100-200 Hz is achievable and audible.

 

Chris

  • Like 2
Link to comment
Share on other sites

  • 4 weeks later...
On 6/6/2019 at 2:41 PM, Chris A said:

As I write this, I remember first listening to a DVD-A from 2007 that I acquired used last year: Fear of a Blank Planet by Porcupine Tree.  As I put it into the Oppo player before I had ever heard it before, I found that I was consciously steeling myself for the sound I knew was going to emerge from the setup...

 

Today, I'm listening to it again and it's a totally different story.  While it's clear that the harshness that I experienced last year is still underlying the sound of this DVD-A, I'm now able to listen without having to steel my hearing...

Today, I'm listening to this DVD-A (multichannel) disc and I have to say that it actually sounds fairly dull much of the time on the surround setup.  The setup has been further dialed in to minimize phase/group delay growth. 

 

I no longer wonder if most of the comments on this forum about "brightness" and other forums about horn-loaded loudspeakers being "harsh" are related solely to the phase/group delay growth of the loudspeakers.  Remember that cone/dome type loudspeakers usually have lower amounts of phase growth than horn-loaded...unless someone has decided to use steep-slope crossover filters in the cone/dome direct radiating loudspeaker.

 

Group Delay Response of 5.1 Array.jpg

Violet trace = Jubilees in front corners

Blue trace = center K-402-MEH

Peach trace = surround AMT-1/Belle bass bin

  • Like 1
Link to comment
Share on other sites

I read the Greenfield article.  I'm also a follower of Nelson Pass and particularly his description of effects of adjusting the strength and phase of the second harmonic.

 

My observation is that Greenfield et al.'s subjective test is done with stereo sound as is Pass's.  I don't know about just mono.  Perhaps that is in other research.  Soundstage is a major issue.

 

The descriptions  of subjective effect in Greenfield and in Pass (and maybe in Chris A's) seem a bit similar.

 

I think PWK's favor for first order crossovers came from the days of mono.  FWIW.

 

WMcD

 

 

  • Like 1
Link to comment
Share on other sites

Ah, yes, but did you know that virtually all the listening in Hope was done on a mono loudspeaker (usually some form of a Jubilee or other Cinema configuration)?  In Toole's latest book edition, he devotes a whole chapter section to the subject of monophonic listening trials over stereo or multichannel to find differences between loudspeaker sound quality.

 

The subject of "desirable harmonic distortion" (a subject by the way that I don't believe in) in amplifiers is most assuredly wrapped up in pushing around the higher frequency sideband distortion--otherwise known as modulation distortion.  This is an interesting diversion--but not the same as the phase and group delay flattening, which I'm reporting on above. 

______________________________________________________________

 

I really wish that I could demonstrate the difference to more people.  A couple of the "remotely dialed in" Jubilees that I've done since 1 May of this year--when I re-created the effect in my setup from comments made by Danley--have been this flat-phase crossover approach that actually achieved flat phase like I've experience in my setup.  I just mentioned at dinner time to my wife that my tastes in music have been changing since this discovery.  I said to her that my taste in 19th and early 20th century music has dramatically increased since 1 May (including now some opera classics).  This is about as dramatic a statement as I can make about what has occurred--since I've never really preferred 19th century music beforehand. It's the sound of the massed string orchestras that has me spellbound (among other effects).

 

I'm now of the opinion that the subtle sound qualities of our sound systems are more determinants of what we prefer to hear than I had allowed myself to realize.  I'm (as they say across the pond)...gobsmacked.  If anyone is through the D/FW area anytime, please feel free to drop me a PM beforehand.

 

SilverShockedHochstettersfrog-size_restr

 

Chris

 

 

  • Like 2
Link to comment
Share on other sites

CHRIS,

 How ya doin', brotha ? !

I am almost finished with my travels for 2019; ...11 international trips this year, has been crazy !

  ( ...on Nov 26th , I return from a 1/2 month in CHILE !)

      ...I'm off December. 

            (My vacation from VACATIONING !)

 

I'd love to come stop by,  and pay ya a visit and experience your 'Danley-style' tuning.

And, even attempt to get my 5.1 'low-rider' Jubes, dialed with ya., soon.

 

 

Hope all is well.

 

Craig LeMay

281-469-1111

Link to comment
Share on other sites

Sorry for the cryptic response just above.  It's been a little disjointed for the past 6 weeks or so--including taking on a 5-month-old Samoyed puppy in July who likes to visit our neighbor's yard and pool unannounced, etc.  Everyone here is in full health now.  We're looking forward to the cooler weather expected on Monday.  Tex (the Samoyed puppy) is especially looking forward to it...

 

Chris

 

  • Like 1
Link to comment
Share on other sites

  • 4 weeks later...

One interesting discussion that I read over at diyAudio was the thread on "beryllium vs the best soft domes" strangely enough.  While I don't post to that forum anymore, it was, I believe, yet another discussion of the subconscious auditory effect of quasi-linear phase/full-range directivity loudspeakers--although the participants in that discussion didn't actually identify the root cause. They didn't address the problem as a phase issue, but the effect has been clearly shown in my listening room, and quite dramatically so even though the explanation for the experienced effect is not very easy to put into words.

 

In any case, of the systems that I've helped others dial-in with their Xilica crossovers, virtually all have strongly preferred the phase-coherent, very low group delay crossovers described above.

 

Chris

  • Like 2
Link to comment
Share on other sites

  • 2 weeks later...

Chris,

Just finished reading this thread and am looking forward to getting more into this area as soon as my Xilica arrives in a few days.  I still am not too clear on how your make a crossover without actually using a "named" crossover on a DSP, but I'm sure you will shed some light on this.  So far, everything you have taught me has proven to be an improvement!  I am very curious as to how this method will work on something as different as full range driver in an Oris horn coupled with a very large ported bass bin.

Link to comment
Share on other sites

Okay, so if you think about the function of crossover filters in multiway loudspeakers, they have some requirements.  The following requirements apply to both active and passive crossover filters:

 

1) They shall attenuate the SPL from lower frequency drivers/horns beginning at some settable breakpoint frequency and continue upward in frequency until reaching the upper limit of natural SPL response output of the lower frequency drivers/horns (i.e., a low pass filter), without also affecting the SPL/frequency output of the drivers/horns within their passband.

 

2) They shall attenuate the SPL from higher frequency (crossing) drivers beginning at some settable breakpoint frequency and continue downward in frequency until reaching the lower limit of natural SPL response output of the lower frequency drivers/horns (i.e.,a  high pass filter), without also affecting the SPL/frequency output of the drivers/horns within their passband.

 

615px-Linkwitz_vs_Butterworth.svg.png

 

3) They shall keep the intervening "interference band" between the lower frequency driver and higher frequency driver relatively flat in SPL, summed across both the lower frequency drivers/horns and higher frequency drivers/horns.

 

n160fig1b.png

 

4) The crossover filters shall keep the width of the interference band between the drivers/horns relatively narrow--so that the contribution of the interference band where both sets of drivers/horns are both audibly playing is minimized to the point of not being audible (and this not as narrow a frequency band as many would like to believe because of the way the human hearing system works).

 

<<<<<< Most people want to stop at this point and ignore the following requirements (but they suffer from poorer overall sound quality if they do) >>>>>>

 

5) The crossover filters shall not affect the phase curve of the loudspeaker's otherwise natural output, i.e., they must have a continuous phase across the crossover interference band that does not increase measurably over the natural phase shifts of the drivers/horns alone taken separately, and avoid phase discontinuities such crossing non-time-aligned drivers/horns resulting in multiple cycles of phase (multiples of 360 degrees) across the interference band.

 

d4oF2.png

 

6) The crossover filters shall not affect the group delay curve of the loudspeaker's natural output, i.e., the must have a continuous (no "poles" in group delay) response across the interference band that does not increase group delay growth substantially.

 

7) The crossover filters shall control the phase of the higher frequency and lower frequency drivers/horns at off-axis angles so that three-dimensional lobing in front of the loudspeaker does not result or materially affect the off-axis SPL vs. frequency of the loudspeakers, resulting in a net uneven SPL/frequency curve across the crossover interference band.

 

n160fig1a.png

 

<<<<<< The following requirements generally apply only to passive crossover filters >>>>>>

 

8 ) The crossover filters shall control their input impedance swings to make sure that the input impedance doesn't fall below a lower limit of impedance or a higher level of impedance within their pass bands (the limits of which are subject to opinion, but the nominal limits are 2 ohms and ~30-40 ohms).

 

MCM1900 Input Impedance Plot.JPG

 

9) The crossover filters shall control their reactance vs. frequency to avoid presenting too complex a load to the amplifier(s) output(s), or present a reactance vs. frequency that may cause a driving amplifier to begin oscillating at some frequency and output level (subject to heating of passive crossover components that change their electrical characteristics while they are hot).

 

Richard C Heyser KHorn Review impedance plot fig 3.jpg

 

10) The crossover filters shall present some form of control through their network in order for the amplifier feedback circuitry/components to damp and directly control the reactance of the drivers coming back to the amplifier's output. Virtually all passive crossover filters fail in this requirement. [But all active crossover filters that directly couple to the amplifier output to the drivers pass this requirement.]

 

afDhR.png

 

10) The crossover filters shall be able to absorb large amounts of electrical power--up to the maximum rated power of the loudspeaker without changing their electrical filter characteristics, or destroying/permanently changing via heating any resistive/reactive components in the filter networks.

 

11) The sum total of passive crossover networks shall not materially reduce the overall efficiency of the loudspeaker above the lowest-efficiency "way" of that loudspeaker.

 

 

As you can see, there are many requirements--the above requirements not being exhaustive.

 

Chris

  • Like 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...