 The Clarinet BBoard
|
Author: Eva F.
Date: 2008-11-16 16:22
my clarinet, which is good in every other way, has a weak, airy, flat-sounding Bb
key. no one else cares, and when i talk about, everyone says i'm over-reacting,
and that it sounds fine. i think it detracts from my playing, and the only person
who agrees with me is my mom, who played on the clarinet before me. no one
else really cares enough to do anything about it or look at it. my clarinet is a Jean
Cartier Professional, and i would greatly appreciate any help you can give me.
AND DON'T YOU DARE SAY I'M OVER-REACTING AND IT'S ALL IN MY HEAD!!!
-Eva
8th Grade Band Nerd!!!
|
|
Reply To Message
|
|
Author: mrn
Date: 2008-11-16 16:41
Are you talking about the Bb on the middle line of the staff, played with the throat A-key and register key? If so, that's a bad note on *all* clarinets, not just yours. It's "supposed" to sound bad--it has to do with the fact that in the standard Bb fingering, the register key works double duty as a "Bb key," a job it's not that well suited for.
There are two easy ways to improve the sound of this note:
1.) Don't use the register key, but instead use the 3rd side key (counting from the bottom--so this is the second side key if you're counting from the top). Obviously, you won't always be able to use this fingering because of technical demands, but you should use it whenever practical.
2.) Use a "resonance fingering" to improve the normal Bb. Basically what this means is that you add some additional fingers to the normal fingering to make it resonate a little better. There are a variety of ones you can use. My favorite one calls for adding the middle finger and ring finger of both hands to the regular Bb, like this:
o x x | o x x
And actually, you can combine techniques 1 and 2 (use the side key Bb and a resonance fingering together) for even better results.
|
|
Reply To Message
|
|
Author: Chris P
Date: 2008-11-16 16:48
"YOU'RE OVER-REACTING AND IT'S ALL IN YOUR HEAD!!!"
Which Bb do you mean?
Former oboe finisher
Howarth of London
1998 - 2010
The opinions I express are my own.
Post Edited (2008-11-16 16:48)
|
|
Reply To Message
|
|
Author: Eva F.
Date: 2008-11-16 16:53
thank you, mrn and Mark Charette! mrn, i tried your alternate fingerings and
resonance techniques, and my Bb sounds wonderful! now i know that it's not just
me or my instrument! amazing what nice sounds a clarinet can make! WOW!!!
Thanks again,
Eva
8th Grade Band Nerd!!!
|
|
Reply To Message
|
|
Author: Eva F.
Date: 2008-11-16 16:55
Chis,
I'm talking about the register key and a, the one on the third line of the staff.
-Eva
8th Grade Band Nerd!!!
|
|
Reply To Message
|
|
Author: Chris P
Date: 2008-11-16 17:54
The resonance fingerings and use of the side key (whenever possible) as explained above are the best bet to add substance and clear up the crappy throat Bb.
I've just been playing bass clarinet in a concert and the throat Bb is fine on my bass as it has better venting (double register mechanism), though I kept using the side (lower trill) key for Bb in some instances, so sometimes old habits die hard.
Former oboe finisher
Howarth of London
1998 - 2010
The opinions I express are my own.
|
|
Reply To Message
|
|
Author: Ed Palanker
Date: 2008-11-16 18:55
All good suggestions but also make sure the tube is clean. It's easy to take off the register key and clean the tube with a pipe cleaner. Also, make sure the key is opening enough, if it's not it will sound stuffy, a repair man can "bend" the key higher or if it has a thick piece of cork under it you can sand that thinner. ESP
www.peabody.jhu.edu/457 Listen to a little Mozart, live performance
|
|
Reply To Message
|
|
Author: clariknight
Date: 2008-11-16 19:07
Record yourself playing the Bb and then tell me you aren't overreacting. I used to think the same thing, that the Bb was a nearly unfixable (with the exception of those different fingerings) note that was just meant to sound bad. But, when I recorded myself and listened out for the Bbs, I found that they sounded only slightly less clean than the other notes. To me, in fact, they sounded cleaner than a low F or E.
|
|
Reply To Message
|
|
Author: Don Berger
Date: 2008-11-16 20:08
Hi Eva - Welcome, you are bringing what may be our most discussed topic !! Try a Search [above] for "pinch Bb" , "mid-staff Bb", "register key" for many "hits". There have been many mechanical and register tube improvements, mostly to the pro cls, an intriguing history. Don
Thanx, Mark, Don
|
|
Reply To Message
|
|
Author: Bassie
Date: 2008-11-16 20:13
> There have been many mechanical and register tube improvements
For example, the intriguingly named 'Hasty pad' (search!), a conical pad for the register hole.
|
|
Reply To Message
|
|
Author: Eva F.
Date: 2008-11-16 21:42
thank you for your help! And i did try recording my playing (g-g chromatic scale),
and i barely noticed the difference. i listened to it several times, and decided i
sounded fine. it sounded hardly weaker than any of my other notes. i wonder
what causes you to think it sounds so bad? i will have to look into this...
-Eva
8th Grade Band Nerd!!!
|
|
Reply To Message
|
|
Author: tictactux ★2017
Date: 2008-11-16 22:08
Eva F. wrote:
> i wonder what causes you to think it sounds so bad?
Simple - it's the note next to your ears, except maybe the A just below (which is, unlike the Bb, proper vented). This and the fact that throat notes, thanks to a very short "sound tube" length, tend to appear less full and vibrant and whatnot.
You hear it (partly because you learned to hear it), but nearly anybody not quite so close won't.
Recording oneself often sets the record straight, mostly in a sobering manner, but sometimes, as in your case, in a reassuring and calming way.
--
Ben
Post Edited (2008-11-16 22:08)
|
|
Reply To Message
|
|
Author: mrn
Date: 2008-11-17 16:50
tictactux wrote:
> Simple - it's the note next to your ears, except maybe the A
> just below (which is, unlike the Bb, proper vented). This and
> the fact that throat notes, thanks to a very short "sound tube"
> length, tend to appear less full and vibrant and whatnot.
> You hear it (partly because you learned to hear it), but nearly
> anybody not quite so close won't.
For the same reason, Jack Brymer (in his book) suggests that one avoid practicing clarinet in an acoustically-dead room. The instrument sounds different (and usually better) at a distance. For that reason I usually practice in a non-carpeted room.
|
|
Reply To Message
|
|
Author: clariknight
Date: 2008-11-17 20:17
Jonathon Cohler has also brought up a point about the throats tones fundamental being out of tune with their respective overtones, causing the stuffy sound when it is not properly amplified (i.e. when your ears hear it before it has a chance to travel the whole length of the instrument.
|
|
Reply To Message
|
|
Author: mrn
Date: 2008-11-18 01:13
clariknight wrote:
> Jonathon Cohler has also brought up a point about the throats
> tones fundamental being out of tune with their respective
> overtones, causing the stuffy sound when it is not properly
> amplified (i.e. when your ears hear it before it has a chance
> to travel the whole length of the instrument.
That's wrong. Even he (Cohler) knows that's wrong (at least now he does, thanks to Tony Pay). He has since revised his "vibrato" article so it does not make that claim anymore (I assume that's where you got that statement from). The revised article still has some errors in the physics section, including one I brought to his attention, but it is better now. If you really want to know how your instrument works, read Arthur Benade's book Fundamentals of Musical Acoustics (or one of the many other reputable references on the subject).
As long as we're talking about a periodic waveform (as a clarinet would produce), it's not physically possible (or, more precisely, not mathematically possible) for the fundamental of a note to be "out of tune" with the note's overtones. Overtones (also known as harmonics or partials) are *always* perfect integer multiples of the fundamental frequency. The sum of the fundamental and overtones of a periodic function (or signal) is called the "Fourier series" of that function. All real-world periodic signals can be broken down into corresponding Fourier series.
See http://www.e-dsp.com/8/ for a nice explanation with few mathematical details.
What IS possible (and indeed happens) is that the clarinet itself does not happen to resonate one or more of these partials strongly enough to make the note "ring true," because the resonant frequencies of the instrument don't match up with the frequencies of the partials--the partials aren't out of tune, the clarinet just won't resonate them all properly. Resonance fingerings can fix this because they change the resonance characteristics of the instument (in fact, that's what all fingerings do--change the resonance response of the instrument).
But PLEASE don't say that the fundamental and overtones are out of tune with each other. That doesn't happen with periodic waveforms!
Post Edited (2008-11-18 06:39)
|
|
Reply To Message
|
|
Author: Mark Charette
Date: 2008-11-18 02:51
mrn wrote:
> It's not physically possible (or, more precisely, not
> mathematically possible) for the fundamental of a note to be
> "out of tune" with the note's overtones.
Right. Damn pesky stiffness gets in the way in some practical situations (like strings on a piano, where there is some inharmonicity ...) but that problem is negligible when using an air column.
|
|
Reply To Message
|
|
Author: mrn
Date: 2008-11-18 07:20
Mark Charette wrote:
> mrn wrote:
>
> > It's not physically possible (or, more precisely, not
> > mathematically possible) for the fundamental of a note to be
> > "out of tune" with the note's overtones.
>
> Right. Damn pesky stiffness gets in the way in some practical
> situations (like strings on a piano, where there is some
> inharmonicity ...) but that problem is negligible when using an
> air column.
While you don't have that problem with a clarinet like you do with a piano, and we were discussing clarinets, you raised a good point, which led me to edit my previous post.
When I said the overtones of a note are always integer multiples of the fundamental, I should have qualified my remark by noting that it is the overtones of a *periodic* signal that are always integer multiples of the fundamental.
The difference between a piano note and a clarinet is that the clarinet note, unlike the piano note, can be considered to be a periodic signal, because if you maintain a constant volume, the signal should consist primarily of a sequence of virtually identical periodic cycles. A piano note, on the other hand, is most definitely not periodic, because, as a percussion-type instrument (of the struck-string variety), it has a built-in decay. In other words, with a piano note, each cycle is progressively smaller in amplitude than the previous one, so it is not periodic in the sense that a clarinet note would be.
Hence, there would not be a Fourier *series* for a piano note, and my discussion above, which would be valid for a periodic note (like on a wind instrument), would not apply to a piano. However, you could still do a Fourier *transform* on the piano note (which is what a spectrum analyzer does), in which case you would see spectral components that are somewhat inharmonic, as you describe.
The same thing would occur with a guitar, as well, since notes from plucked string instruments have a natural decay, as well.
Thanks for mentioning this. Since I am most accustomed to dealing with periodic signals (both in music and in radio), it's easy for me to forget to address the fact that I am *assuming* periodicity here (because it's appropriate to do so in the case of a clarinet), but that that assumption cannot always be made.
|
|
Reply To Message
|
|
Author: Koo Young Chung
Date: 2008-11-18 08:13
To mrn:
Why do you assume that clarinet sound is (strictly or mathematically)) periodical(harmonic)?
It is not periodical (harmonic) in the strict sense.
(Fourier transform is only an approximation here,because a musical note is not infinitely long.)
It only appears (sounds) to you (or us) that the clarinet sound is periodical,but it isn't.
Our ears cannot distinguish an absolutely harmonic(or periodical) sounds from the sounds which contains inharmonicity in them.It still sounds almost harmonic (nice or musical) to us.
And any clarinet sound has inharmonicity built in(i.e.,out of tune overtones),no matter how it "sounds" harmonic (or periodic) to us.
It's is a little bit confusing and difficult to understand but that's what happens when we play a clarinet.
|
|
Reply To Message
|
|
Author: Mark Charette
Date: 2008-11-18 10:36
Koo Young Chung wrote:
> And any clarinet sound has inharmonicity built in(i.e.,out of
> tune overtones),no matter how it "sounds" harmonic (or
> periodic) to us.
No, it doesn't ... the overtones of a clarinet are integer multiples. Do a spectrum analysis and look. I have, using the measurements of Dr. Jim Pyne at Ohio State. As far as the equipment could measure the overtones (partials) were spot on integer multiples.
String instruments are a different story due to the way they vibrate - they have appreciable mass, appreciable stiffness, and the effective length of a plucked string changes over time.
|
|
Reply To Message
|
|
Author: Bassie
Date: 2008-11-18 12:19
mrn -
If you take a short piece of a periodic signal, you just get broadening of the peaks in the Fourier transform (i.e., the pitch becomes /uncertain/). They remain harmonically related.
There's a recent thread on this, which I think came to the general conclusion that (1) the components of a clarinet tone are indeed harmonically related (2) the higher partials don't really 'want' to be harmonically related, which can lead to a change in pitch as the volume of a note is varied.
|
|
Reply To Message
|
|
Author: Koo Young Chung
Date: 2008-11-18 18:52
To MC:
When you're doing a spectrum analysis,the analyzer takes one cycle(or average of many cycles) and assumes it repeats itself perfectly.
No wonder the overtones are perfect multiples of fundamental.
In other words,the machine only computes the amplitude of each overtone
without questioning its perfect harmonicity.
When you play a note on clarinet,as we all know,each overtones are activated
and what we hear is the linear superposition of all those not-so-perfect harmonics.
***There are no internal mechanism which magically force all those overtones to align to be perfect multiples of fundamental.***
I also point out that on any decent clarinet the overtones are very close to perfect multiples ,especially middle notes.
|
|
Reply To Message
|
|
Author: mrn
Date: 2008-11-18 22:33
Koo Young Chung wrote:
> To MC:
>
> When you're doing a spectrum analysis,the analyzer takes one
> cycle(or average of many cycles) and assumes it repeats itself
> perfectly.
The Fourier transform (which is what a spectrum analyzer displays a representation of) is derived mathematically by considering the period of the signal analyzed to be *infinitely* long.
Fourier analysis (spectral analysis) is a mathematical technique (like long division). You take a function f(t), which is a function of time, and compute a new function F(w) from it that is a function of frequency. A spectrum analyzer is nothing more than a machine for computing an estimate of the Fourier transform of a signal.
You're trying to define a mathematical concept in terms of a machine used to compute it. That's like trying to define division as being "what the calculator does." That won't work. Fourier, who developed spectral analysis, let's not forget, lived in the early 1800s before anything even remotely resembling today's electrical test equipment had been invented.
> No wonder the overtones are perfect multiples of fundamental.
If it was the spectrum analyzer making the overtones appear to be harmonic because of some property of the spectrum analyzer, then we wouldn't expect to get different results from a piano than a clarinet, but in fact, the results are different.
> In other words,the machine only computes the amplitude of each
> overtone
> without questioning its perfect harmonicity.
The machine computes a function relating frequency to amplitude. You get a nice little graph on the screen where the X-axis is frequency and the Y-axis is amplitude. If overtones are inharmonic, you'd see it on the screen because the little peaks in the graph corresponding to the different overtones would show up in different places, for example at X=1, X=2.2, X=3.1 (not integer multiples of 1) instead of at X=1, X=2, and X=3 (integer multiples of 1).
The machine doesn't "question" the harmonicity of what it processes, because a spectrum analyzer doesn't question anything. It simply computes a function and graphs it on the screen. But it so happens that that function contains all you need to know about the harmonicity of the signal components.
> When you play a note on clarinet,as we all know,each overtones
> are activated
> and what we hear is the linear superposition of all those
> not-so-perfect harmonics.
>
> ***There are no internal mechanism which magically force all
> those overtones to align to be perfect multiples of
> fundamental.***
I know this is going to sound weird, but overtones don't really exist in the real world in the sense that you can measure them directly. Musical instruments (with the exception of some synthesizers) don't generate individual sine waves (overtones) and mix them to create a composite sound. The physical phenomenon you can measure, coming from a clarinet let's say, doesn't look much like a collection of sine waves if you display it on an oscilloscope.
However, if the physical waveform is periodic, then it is mathematically equivalent (i.e., equal to) a sum of harmonically-related sinusoidal functions. This is a mathematical abstraction that allows us to understand the signal. Just like you can take any non-prime number and factor it into prime factors, you can take any periodic waveform and break it down into the sinewaves you'd have to add up to obtain that periodic waveform. (and there's usually an infinite number of them, actually)
If the physical signal is not periodic, you can still pretend that it is, give it an infinitely long period, and compute the Fourier transform. If you don't have a periodic waveform, the Fourier transform will look less like a Fourier series (i.e., harmonically related sinusoids) and more like something else.
It's not magic--it's not even physics--it's purely a mathematical abstraction. That's why it seems so unbelieveably "perfect."
Post Edited (2009-02-22 20:27)
|
|
Reply To Message
|
|
Author: Koo Young Chung
Date: 2008-11-21 06:46
I found my answer in your link on another topic.
http://www.phys.unsw.edu.au/music/clarinet/C6.html
The overtone peaks for (any) clarinet notes are very close to the integral multiples of the fundamental frequency,however they are not at the exact muitiples of the fundamentals.
They(center of the peaks) are slightly higher or lower than the exact multiple of the 1st peak,which is ,by definition, anharmonicity of the clarinet
overtones. You can see this better on the higher notes.
For example, the peaks are located at 1.0000(by definition) , 3.0427.. , 4.9532 .., etc.
Not at 1.0000 , 3.0000 , 5.0000 etc. (just an example,not an actual quote)
When you play a long notes on clarinet,even though it sounds like a periodic (or perfectly even ) tone,it is not because of the anharmonicity contents of the overtones. They just sounds like one to our ears.
Any acoustical instruments have an (varing degrees of) anharmonicity built in its overtones.
If the overtones are exact mulpitles of the fundamental,the tone is monotonous and boring like (simple) computer generated sound usually
found in computer games or electronic music.
Post Edited (2008-11-21 13:43)
|
|
Reply To Message
|
|
Author: Tony Beck
Date: 2008-11-21 16:15
If you are interested in this subject, download the PC audio spectrum analyzer that Chris J posted under; http://test.woodwind.org/clarinet/BBoard/read.html?f=1&i=287643&t=287643
It shows the overtones perfectly, just like an HP or other spectrum analyzer, and it’s free. I was driving my wife nuts playing with it last night. You can easily see on the spectrum display why a pianissimo sounds differently from a forte, and different clarinets (even different mouthpieces) have different tone qualities.
A spectrum analyzer performs a “Fast Fourier Transform” to pull out the frequency components of any random input waveform. It converts a function of time f(t) into a function of frequency f(fq). Obviously, this function puts limits on the integration time, but if the time span is reasonable, the approximation errors are very small. The FFT doesn’t assume that the harmonics will run in lock step with the fundamental, it computes what’s really there. In fact, when you put in random signals, it will pull out everything, from whatever source, and display all the signal components that are above the device’s (or the microphone’s) noise floor.
A clarinet sounds like a clarinet because the straight bore produces approximately a square wave (which gets more nearly square as volume increases), with only the odd harmonics (3rd, 5th, 7th, and so on). The “brighter” the tone, the more upper harmonics are present. Oboes, Bassoons and Saxes, have even and odd harmonic components due to their conical bores, so sound quit different. Strings have a different set of characteristics all together.
With a piano it is possible to have harmonics that are more out of sequence with the fundamental because pianos have multiple strings for each note. Each string is usually tuned slightly differently. This produces mixing products. Tone A+Tone B+Tone C equals a bunch of harmonics that aren’t in either A, B or C alone. You also have an e^-t decay to factor in, which shifts all of the harmonic series.
|
|
Reply To Message
|
|
Author: Chris P
Date: 2008-11-21 16:22
And with a spectrum analyser you can get a readout of your favourite clarinettist's tonal spectrum - and then try and match it with your tone!
Former oboe finisher
Howarth of London
1998 - 2010
The opinions I express are my own.
|
|
Reply To Message
|
|
Author: Mark Charette
Date: 2008-11-21 17:23
Tony Beck wrote:
> With a piano it is possible to have harmonics that are more out
> of sequence with the fundamental because pianos have multiple
> strings for each note.
Actually that's not the reason - in the case of multiple strings you may have multiple fundamentals, differences in tension due to different pin/bridge spacing, phasing effects, and other variables introduced.
The effective length of a string with mass & stiffness is not exactly the length between the pin and bridge, and the effective length of the string changes with amplitude (which changes as a function of time since power is being removed by the bridge and there are other losses - some of the losses are actually due to the inharmonicity) and frequency (the overtones that are excited when the string is struck by a hammer end up with a slightly different effective length due to frequency vs. stiffness and other effects). That causes inharmonicity in a single's string overtones that change over time.
|
|
Reply To Message
|
|
The Clarinet Pages
|
 |
.jpg){kind=small}
.jpg){kind=small}
.jpg){kind=small}
.jpg){kind=small}
.jpg){kind=small}
.jpg){kind=small}
