Belle listening fatigue

[Honeybadger]

16 hours ago, Deang said:

I use Fastrac horns in my LaScalas. I like them, but the sound isn't as dynamic as the K-400s at the volumes I listen at - which is pretty low - I'm a late night listener. 

    I had a very different experience, to me the tractrix had far better detail at lower levels.

   

    

[Deang]

"Detail" (resolution) and "dynamics" don't mean the same thing to me. 

[Schu]

6 hours ago, Deang said:

"Detail" (resolution) and "dynamics" don't mean the same thing to me. 

agree...

[Honeybadger]

In my opinion its all about the detail at low levels.

 

Not trying to start an argument Dean, don't jump on the offensive.

[Marvel]

I don't the he was on the offensive at all, just stating how he hears things. I agree with everyone though, it's the detail in low level listening that makes it. I never was one to listen to really loud music, and even less as I get older.

 

Bruce

[Deang]

Not the kind of thing that sets me off. :-)

[mikebse2a3]

6 hours ago, Marvel said:

I don't the he was on the offensive at all, just stating how he hears things. I agree with everyone though, it's the detail in low level listening that makes it. I never was one to listen to really loud music, and even less as I get older.

 

Bruce

Loud doesn't require dynamics.

 

Dynamic resolution is one of several important factors in the formation/perception of detail resolution at any listening level.

 

miketn 

[Chris A]

On 9/9/2016 at 4:58 PM, hcnelly said:

Marvel and Chris, I'm confused about the digital crossover? Is this difficult to setup, I think this would be too complicated for me....I'd like to keep the Belles looking the same, keeping any horns or upgrades in the cabinets...

It takes an active crossover between your preamp and your amplifiers, three XLR or RCA-XLR cables (available at Guitar City, etc. for a reasonable cost--they're typically sold as "microphone cables"), and two more zip-strip speaker cables per loudspeaker, or you can use 6-conductor cable instead. 

 

Everything else is simply dialing in the settings that I referenced above--about 20-30 minutes for complete novices to complete.  Then you have complete control over the sound that you find most pleasing. 

 

Chris

[Deang]

... And don't forget the microphone, REW, book, and instruction video. :-)

 

Need capacitors between HF drivers and amplifiers channels driving them. 

 

Expect some noise if mixing consumer and commercial gear (gain structure). 

 

Its not insanely difficult stuff, but I don't think it's something a novice can do. 

[Chris A]

No, no.  Now you're going off.  If you're using unreliable amplifiers that fail regularly with DC bias because you haven't done preventative maintenance, then some sort of DC protection might be prudent. The crossover itself is supremely more capable if anything in the preamp or upstream of it drives too hard or creates a killer transient - the crossover itself has fast limiting built in.

 

You actually don't need REW and a microphone if you're just doing "plug and play".  That's why I created those two postings--for those that don't want the measurement part.

 

If you get a good quality crossover, noise isn't an issue.

 

Be nice, Dean.  I'll be nice in return.

 

Chris

[Deang]

Tongue in cheek Chris, I was just playing. 

 

Honestly though, I've never read anything related to the active route that didn't involve a recommendation to add the capacitors between the amp and HF drivers. Good info, thanks.  

[hcnelly]

Well, I'll let you guys know how my speakers sound once I knock the midrange down 3b...Dean's helping me out with that...looking forward to hearing the difference...thanks for the help guys! Very much appreciated! I hope it's just what I need...

[Schu]

Whatever you do... make sure to spend some time listening to them for a good while before making another change.

[Travis In Austin]

On 9/10/2016 at 4:41 PM, mikebse2a3 said:

Loud doesn't require dynamics.

 

Dynamic resolution is one of several important factors in the formation/perception of detail resolution at any listening level.

 

miketn 

Not to sidetrack this thread, but the "crew" is in here already, Mike, Dean, Chris, et al, so I thought I would ask.

 

Is it known what causes "listening fatigue."  I have experienced it myself and know what it is, however,  I have found it isn't necessarily tied to a particular set of speakers.

 

Doesn't the source and type of music have to be factored in (Chris)?  Does lack of DR in the recording contribute to "listening fatigue."

 

What about the room it's acoustical properties  (Mike)?

 

What about amp topology, SS v. Valve (Dean, Mike)?  Frequency Response?

 

Some days I can sit and read a book for 20 minutes and I'm done. Other days, 6 hours straight.  It depends on a lot of factors like light, fatigue, posture, size and type of font, and obviously the content.

 

Is "listening fatigue" pretty much understood to be the result of the speaker, or do you also have to look upstream?

[Chris A]

With the Belle, there are a few specific factors that I think contribute to a particular sound that can be very fatiguing on the listener:

 

1) When you align the drivers so that the tweeter and midrange are either physically aligned or aligned using delay in an active crossover, you get a very large timbre shift that will sound almost indistinguishable from that of Jubilees, i.e., extremely neutral sounding.  This is probably the biggest single adjustment in terms of controlling timbre and the overall sound.  Aligning the bass bin to the midrange will also shift the vocal timbre noticeably, curtailing that nasal/midrangey timbre of the stock Belle.

 

2) There are also big peaks in the frequency response of a Belle using stock crossovers in the 90-200 Hz band.  Smoothing these out balances the the entire loudspeaker considerably.  The midrange actually is quite flat in frequency response, so the only crossover adjustment necessary is the relative level of the midrange to woofer and tweeter. The tweeter has peaks at about 5 and 12.5 kHz that need to be smoothed out.  Once these frequency response irregularities are smoothed out (bass bin and tweeter), the whole presentation is much smoother and much less fatiguing.

 

3) The K-500 midrange horn has its well-known vertical axis collapsing polars from its high pass point (below 500 Hz) to about 1.8 kHz, and it is this extra acoustic energy that needs to be absorbed in the near field, and any near field reflective objects near the midrange horn mouths moved away so that there is a 1 yard/metre clear area all the way around the speaker, and there also needs to be "salt-and-pepper EQ" applied in-room to balance the early reflections from the floor and ceiling with the direct sound.  The best way to balance the sound of the Belle is to have carpet on the floor and a high ceiling, with nothing around the loudspeaker, especially between the Belles, which will disrupt the imaging performance.

 

4) The tweeter passive crossover in the Belle overlaps the midrange frequencies over an extended band, and this causes many issues in timbre, polars, and clarity in the critical 2.5-6 kHz band.  This also leads to listener fatigue. Using an active crossover facilitates the use of steeper slope filters and moving the crossover point down a bit.  The K-500 horn doesn't sound very good up high above 4 kHz.

 

5) Both the midrange horn and the tweeter are very rough sounding relative to a K-510/K-69-A midrange and a Beyma CP25 tweeter, crossed at 8 kHz using 24 dB/octave Linkwitz-Riley filters.  Also, the frequency irregularities of the CP25 can be addressed using the PEQ filters in the crossover, at ~5 kHz and at 10.5 kHz.

 

All of the above (with the exception of the tweeter and midrange roughness and midrange horn collapsing polars) can be adjusted using a good digital crossover.  The difference in sound quality is really quite startling.  I found that I had no fatigue at all after fixing these issues.

 

Chris

[Chris A]

As far as the music dynamic range and mastering EQ is concerned, the more you listen to music originally produced after 1991, the more that the "loudness war" will be apparent.  These popular genres of music produced from 1999-present can be extremely fatiguing.  This is actually a fairly involved discussion, and is fairly specific to the exact music tracks that you're listening to.  Every album and musical number on the album is usually mastered differently in EQ, and even clipping/compression from the other tracks on the same album. 

 

I'll leave it at that for the time being.  I can supply specifics for many popular albums. 

 

Chris

[Travis In Austin]

17 hours ago, Chris A said:

As far as the music dynamic range and mastering EQ is concerned, the more you listen to music originally produced after 1991, the more that the "loudness war" will be apparent.  These popular genres of music produced from 1999-present can be extremely fatiguing.  This is actually a fairly involved discussion, and is fairly specific to the exact music tracks that you're listening to.  Every album and musical number on the album is usually mastered differently in EQ, and even clipping/compression from the other tracks on the same album. 

 

I'll leave it at that for the time being.  I can supply specifics for many popular albums. 

 

Chris

I meant in general.  I don't listen to much after '91.

 

Travis

[Chris A]

In general, music produced will have various "spectral shaping" techniques applied to the tracks, in the form of rolled-off bass below some point (depending on the album, generally below 80-100 Hz, but it could be below 50 Hz, or as high as 800 Hz).  This is done to make the apparent loudness of the tracks even louder than the original recordings.  Music produced after 1991, bass will not only be attenuated, but clipping and compression will be much more apparent, sometimes approaching 10 dB of overall boost from the original mixdown tracks.  This is especially true for any music that has been marked as "remastered". 

 

Additionally, mastering engineers usually apply some sort of boost to higher frequencies, depending on the genre type and the exact instrumentation.  This boost usually starts at 1 kHz and continues up to at least 10 kHz, sometimes continuing up the highest audible frequencies (20 kHz).  This boost usually is in the 6-12 dB range, and is usually boosted at its maximum at 6-8 kHz. 

 

Here is a fairly typical unmastering curve for a rock and jazz fusion track which is used to return the music tracks back to a mixdown track equalization state (i.e., as recorded in the studio).  Note the 9+ db de-emphasis from 2-6 kHz, and the 15+ dB bass boost at 30 Hz:

 

 

What this means in terms of listener's fatigue is that if listening to these musical tracks without unmastering, and if the loudspeakers have any distortion in the 2-6 kHz band (like I mentioned above for the stock Belle), there will be more more severe listening fatigue based on the equal loudness curves of human hearing, shown below:

 

 

Frequencies that are boosted on musical tracks anywhere in the 2-6 kHz range will sound strident due to the sensitivity of the tuned-tube eardrum/ear canal response at those frequencies.  The human hearing system is exceptionally sensitive to any frequency response peaks or dips, distortions (harmonic, modulation, compression, etc.) or relative time delays in crossing HF drivers in this region. 

 

Chris

[jimjimbo]

Hey, if I stick my Lava Lamps in the middle of the room, can I observe those Spectral Shapes?  That would assuredly be very cool....

[Deang]

Great analysis Chris - man, you love to get into the weeds. I would have just said, "garbage in, garbage out." :-)

 

The narrow throats and rearward reflections off the mouth of the horn don't lend themselves well to live level listening - especially with complex material.