Again on Jubilee...

[ZAKO]

That chart is not accurate.

[Deang]

Not for you it isn't.

[DrWho]

The range of human voice is about 70Hz to 1100Hz.

Hmmm...that's not been my experience. Care to share where you got those numbers from?

[DrWho]

That chart is not accurate.

I agree...

[DrWho]

Well I would say cranking a 100Hz shelf filter on a female vocal
will usually make almost no difference. Even cranking a ~250Hz PEQ makes little
difference on most females. However, ~250Hz can be the difference
between throaty and hollow sounding for most male vocals. Even as low
as a 60Hz shelf filter will dramatically affect the boominess. A 10kHz
shelf filter can easily be heard on both male and female singers and is
probably at the top end of sibilance. Sibilance can be affected down to
as low as 6kHz or so (depends on the person). The
forwardness/shrillness usually comes in around 2-4kHz and the grays or
cardboard (for lack of a better word) comes in around 600-800Hz. So I
guess I would say about 60Hz to 15kHz or so? You certainly couldn't
lowpass at 1kHz and expect the person's voice to sound the same.

I would argue that most of what affects intellgibility
of the human voice is in the 800Hz to 8kHz decade...and that can
probably be narrowed down to like 1kHz to 4kHz if you really wanna make
it tighter (which is where I think telephones tend to sit). Because of
this, you can often get away with dramatically cutting everything below
about 600Hz on a vocal channel while other instruments are
playing...in fact, it often cleans up the mix quite a bit because
guitars and keys are often competing in the lower mids too. I'm not
proposing that our playback systems do this, but the reasons for why we
can get away with this are the same reasons why I'd argue that we're
not as sensitive to the accuracy of playback in that region. Also, the
lower mids and lows of the human voice are generated differently than
the higher frequencies, so it's already natural that the timbre is
slightly different...so the person will probably sound a bit different
on the recording than in real life, but if you've never met them in
real life then it's less of an issue.
In fact, it's not too hard to get a person sounding better on a
recording than they do in real life, but that's a totally different
topic.

I'd say the instrument most impacted by an 800Hz xover is the piano, but it's bandwidth is so wide that it's impossible for a single speaker to reproduce it's entire spectrum. In fact, it is extremely difficult to even get a single mic to capture the total sound of a piano and it's even more difficult to get two microphones to work well together. A full-sized grand piano is easily going to cover nearly the full range of human hearing.

[jwc]

4-way, in real life, when done carefully, and intelligently, with measuring gear, is not as detrimental as all the talk here makes it out to be. Since we are not torquing Parametric and Shelving EQ's around by plus/minus 12 db, and jamming frequencies into a driver that doesn't naturally want to do it, there are other benefits there that 2-way Quasi-Religious pratctioners don't discuss because they have never tried 4-way

I use 4-ways all the time...the angle problem is pretty much just solved with distance. I think the MCM (as advertised by Klipsch) gels at around 40ft or so? A general rule of thumb I've heard thrown around is about 10x the interdriver spacing (as measured from the center of each driver), which has jived with my experiences too. You're looking at like 15-20ft for the Lascala...

Now that's not to say that things can sound awesome when you sit closer (which is why I reference the Lascala), but things sounded more clear when you hit the convergance point. It's really quite dramatic and probably most easily heard outdoors where one can actually get far enough away and have the room to hear the difference.

I suppose 2-way Jubs don't "converge" in most rooms either...and I really think the issue of convergance is the biggest attraction to 1-way, but of course there's other problems with that too. I might argue that we probably want our rooms to be larger anyway as that makes the room's acoustics a whole heck of lot easier to deal with too (which means sitting further back and not having to worry as much about the angles).

Assuming that one has all the processing in the world to get good acoustic summation from a 4-way, you still have the issue of getting all the polars to line up (not impossible, but can't be fixed after the fact), and probably more importantly...getting the timbre of distortion to be the same. I think this is where choosing specific xover frequencies can become interesting because sometimes you can make it such that single instruments don't walk around the xover very often...or you can pick frequencies where the timbre of the instrument is divided across the xover - I dont' think I'm describing that very well. Anyways, I would suggest that the most natural xover frequency for most music is going to be in the 400-800Hz octave (and I don't say this because it's where the Jub crosses over). It's just that I can think of a lot of instruments that are either solely above that frequency range, or that the various components of their timbre are on opposite sides of that region (for instance, the snap and thud of a kick straddle it pretty well...or the twang and girth from a bass guitar).

But the second you throw a xover in the 3kHz octave....yeouch. And as you go higher, you run into problems with the inter-driver spacing being so large that it forces the polars to beam in the overlap region of the xover. I think it'd be interesting to see how the acoustical centers shift around too...I've been doing some measurements lately that seem to indicate that it is a lot bigger of a problem than I originally thought (in reference to multi-way systems in PA settings). The off-axis probably isn't as crazy important in the home setting since the reflection/absorption coefficients of your walls could have just as large of an impact.

Again, I don't mean to imply that any of these things will make any system sound like a bose wave radio, but I also think they are clearly audible and should be avoided whenever possible.

As far as all the implications that the K402/K69 isn't meant to play higher frequencies....why then is the sensitivity at 18kHz higher than the K77? I'm going to leave it at that... (however, I will mention that I'm not saying it's the best tweeter in the world, but that has nothing to do with the skewed perspectives).

Bentz.

What is your experience with hearing MWM's in the past. Was it outside...live venues? or in a club?

The reason I ask is that the MWM shoved into a corner or close to a corner (in a small house room) may provide a lower end that the published specs. Also these guys are putting in K33 or Crites woofers. What I heard at Mark1101's house was some very tight low end....he certainly doesn't need a sub. Curves in his room were down to 30Hz. Of course some modern day techno or low rider rap music could use a little more on the low end.

Why couldn't we argue that the MWM (bass cabinet only) "converges" better than the jubilee or the Khorn bass bins?

The "Jubilee version" of dual 15" woofers can be done. The cabinet width goes from 42" to 45" with the same type path as the jubilee. But the "splay angle" is a little more of a problem. Also.....K33's wouldn't work with that simple mod. need more expensive woofers.

I do agree with the straight horn idea for the horn bass bin. then use a sub. nothing wrong with that.

jc

[Deang]

http://www.tnt-audio.com/topics/frequency_e.html

http://www.soundinstitute.com/article_detail.cfm/ID/106

Mark's numbers look on the money.

All the charts I found seem to coincide with the one posted in this thread.

[Don Richard]

Yes, harmonics exist well above that range. Yes, they will also affect the sound. But the harmonics have not a single thing to do with my comment about the vocal range.

The concept of vocal range defined as the fundamental frequency of a sound that is being produced by a human voice is a curious one. Without the harmonics it would be difficult to understand what was said or who was saying it. Unless one communicates somehow via Gregorian Chant[]. The early telephones had a range of around 300Hz to 2500 Hz for intelligibility reasons.

Certain musical instruments (violins, saxophone) create more harmonics than fundamentals. Without them it would be difficult to impossible to discern what instrument was playing. This is why we have sound systems capable of wide range performance.

[Don Richard]

The concept of vocal range defined as the fundamental frequency of a sound that is being produced by a human voice is a curious one.

I don't find it curious at all. If I were composing a song for a baritone, I wouldn't mark down a C6 on the score because he can't sing it. He can't "hit that note" as they would say in musical terms. It is, one could say, "out of his vocal range." I admit, that's a musical term, but it's not actually all that uncommon for people familiar with music.

Honestly guys, have you ever heard of the expression, "over thinking it?"

I thought that we were discussing vocals as applied to loudspeaker design/usage. Earlier in this thread you said:

"The range of human voice is about 70Hz to 1100Hz. The sound of the human voice is the most familiar sound in our internal library. Dividing that range between two dissimilar pistons with a crossover certainly has to be one of the most controversial trade-offs in all of loudspeaker design, no? "

As explained earlier, 70-1100 Hz is not sufficient to accurately portray the human voice. Community Loudspeakers in the 1970s developed a midrange horn/driver specifically for covering vocal frequencies with no xover for use in voice only high-output sound systems. It was also used in 3 and 4 way high output sound reinforcement work. It went from 200 Hz-4000 Hz with upper octave EQ. Here is a link to a modern version of this product:

http://www.loudspeakers.net/files/specs/new/m4.pdf

Getting the midrange right is critical for proper sound reproduction. It seems to me that the Jubilee midrange would easily do that.

[mark1101]

Some speaker systems even break the noted range into 3 drivers by use of a mid-bass module, or 4 with a subwoofer.

But what if you minimized the crossover points with just a 2-way system using a single crossover point and had a crossover that prevented the overlap between drivers? Then what?

[Don Richard]

For example, when a 15" driver and a 4" driver have to both produce a 500Hz fundamental tone, the character of each will be different. The singer is creating one fundamental tone that is later being produced in two very different ways, and then combined. The ultimate range of all the harmonics has nothing to do with that specific point. If you haven't performed that little experiment on your own, maybe it's not wise to comment on speaker design? If you have performed it, and you think the two pistons of radically different size sound "the same" when producing the same tone - than we simply disagree don't we?

As a matter of fact, I am presently working on a mid horn for my Khorns that goes from 200 Hz and extends to 2000 Hz. It uses a 10 inch driver and when I first tested the prototype (sine wave) I was struck by how similar it sounded to the corner bass horn at 200 Hz. Standing 20 feet away there is no difference. At 2000 Hz there was an audible difference between it and my HF, which I attributed to beaming from the midhorn. At 1600 Hz there was no difference, so when I build the other mid horn that's where I will set the crossover. Maybe the size difference isn't radical enough? Maybe that's why I am using a 10 inch driver.

[mark1101]

Just my opinion but one of the best attributes of all the Heitage products.......which are 3-way.......is the vocal reproduction. Still to this day I don't think a lascala has any serious issues when it comes to the vocals. In fact you could make a case that it is almost everything else that you can criticize ahead of the vocals.

[Don Richard]

Just my opinion but one of the best attributes of all the Heitage products.......which are 3-way.......is the vocal reproduction. Still to this day I don't think a lascala has any serious issues when it comes to the vocals. In fact you could make a case that it is almost everything else that you can criticize ahead of the vocals.

[Y]

[Islander]

Just to add a factor that hasn't been mentioned: could it be that most male vocals would fall under 500Hz, while a significant part of female vocals might be over 500Hz, so two different drivers would mostly be reproducing two different voices?

Just wondering, but it may make splitting the vocal range in two less of a problem.

[Guest " "]

Just to add a factor that hasn't been mentioned: could it be that most male vocals would fall under 500Hz, while a significant part of female vocals might be over 500Hz, so two different drivers would mostly be reproducing two different voices?

Just wondering, but it may make splitting the vocal range in two less of a problem.

wiki says...

The voiced speech of a typical adult male will have a fundamental frequency of from 85 to 155 Hz, and that of a typical adult female from 165 to 255 Hz

but, it's obvious wiki is refering to speaking voices and not singing voices.........so is this thread refering to speaking voices or singing voices...the comments seem to be refering to singing voices...but the numbers seem to be refering to speaking voices.

[Islander]

Yeah, that had occurred to me, too. It looks like all the speaking (not singing) fundamentals would be under the 500Hz crossover point.

Then, of course, there's the whistle register, that extreme high-end of the singing range (from Wiki):

The whistle register (also called the flageolet register or whistle tone) is the highest register of the human voice lying above the modal register and falsetto register. This register has a specific physiological production that is different from the other registers, and is so called because the timbre of the notes that are produced from this register are similar to that of a whistle.

In some sopranos the modal register vocal production may extend into what is usually thought of as the whistle register.^[1] Women of all voice types can use the whistle register. With proper vocal training, it is possible for most women to develop this part of the voice. Children can also phonate in the whistle register and men can as well in very rare instances.^[1]

The whistle register is the highest phonational register that in most singers begins above the soprano "high C" (C6 or 1,046.5 Hz) and extends to two Ds above (D7 or 2349.3 Hz). However, the whistle register does not begin at the same point within every voice, and there are rare voices which can extend the whistle register much further than the range listed above. For example, some opera singers can sing up to the "high" F above "high" C without entering into the whistle register.^[1] Also, rare singers like Georgia Brown from Brazil, who sang up to a G10 (25087Hz), can sing much higher in the whistle register.^[2]

[Islander]

There was actually a "Battle of Whistle Register Queens", if you can believe it, and Mariah Carey placed 3rd. It seems to be a compilation of video clips, rather than a face-to-face event, but you can see it here:

http://www.recordcup.com/item/909/nikki.html

If you mouse over "1-7 of 15", you can see and hear some of the other "contestants". This page shows Nikki, the 2nd place finisher, who seems to be a better actual singer than I Wai-Lam, the singer in 1st place.

[DrWho]

You cannot ignore the behavior of the harmonic content when trying to reproduce any source.

The
interesting thing about the voice is that the harmonic structure is
very different between the high and low notes...likely because the
acoustical mechanics of a singer's voice change at the extremes.

Crossing
over lower changes where the harmonics are divided up...and it's been
my experience that the harmonics have an equal if not greater influence
on the perception of vocals than the fundamentals. And I've got the hearing type that concentrates more on fundamentals and doesn't fill it in when it's missing.

All I can say is that with most all of the music I listen to, the vocals seem the least compromised with a xover in the 600-800Hz region. Now the piano is a totally different story...nothing works for it.

[jwc]

All I can say is that with most all of the music I listen to, the vocals seem the least compromised with a xover in the 600-800Hz region.

That's because you listen to music like the Alvin and Chipmunks soundtrack.[]

jc

[DrWho]

Doesn't everybody? []