Page 1 of 4

Clapper height tonal variation

PostPosted: Sat Mar 30, 2013 1:54 am
by TerryMcGee
I thought I should open a new topic so we don't "contaminate" other discussions. I'll do the clapper height effect analysis first, as the difference is clearly audible, and therefore should show up clearly in analysis. I'll probably use Bill Hibbert's decay analysis tool, as it shows both the amplitude (loudness) and the decay of the various significant partials. It will look something like this:

Image

But I have a question first. You can see Bill used modern bell partial names in the example above (see legend on right). Which nomenclature do you think works best for describing partials in these analyses? Here are some options:

    Traditional (organ mutation stop) names: hum, prime, tierce, nominal, superquint etc
    Modern names: hum, fundamental, minor third, fifth, octave, major third,......, upper fourth, etc
    actual pitches: F#2, F#3, A3, F#4, C5, etc
    actual frequencies: 729, 1462, 1743.5, 2919, 4300.5Hz, etc
    semitones and cents relative to Prime: -12.05, 0, 3.05, 7.03, 11.96, 12.52,
    a combination of the above: Tierce, 3.05; Nominal, 11.96; Nominal 12.52; etc
    or other, please specify?

The semitones and cents one is interesting in that it also tells us the tuning of the partial. Note the 12.52 value in the list above. That means the Nominal is a smidge over half a semitone (52 cents) sharper than the octave (12 semitones) we would expect. The combination line tells us that we seem to have two nominals, one at 4 cents flat (11.96), the other at 52 cents sharp (12.52).

Terry

Re: Clapper height tonal variation

PostPosted: Sat Mar 30, 2013 5:34 am
by JohnGouwens
Well, for the note terminology, I favor the following, which is what I used in my just-finished Campanology book:

Double Octave
Tweltfh
Two Elevenths
Major Tenth
Nominal
Quint
Minor Tierce
Prime
Hum Tone

If you're reporting deviations (not relevant for this particular exercise), it's best to use cents. (A cent is 1/100th of a semitone. The difference in frequency number is of course greater the higher you go.)

Is this graph from one of my sound samples or some other bell? If it's mine, was this the one with the raised clapper?

Re: Clapper height tonal variation

PostPosted: Sat Mar 30, 2013 6:07 am
by TerryMcGee
JohnGouwens wrote:Well, for the note terminology, I favor the following, which is what I used in my just-finished Campanology book:

Double Octave
Tweltfh
Two Elevenths
Major Tenth
Nominal
Quint
Minor Tierce
Prime
Hum Tone


Hmm, so a bit of both modern and trad. OK, I'll see if I can run with that. Pick me up if I get it wrong!

If you're reporting deviations (not relevant for this particular exercise), it's best to use cents. (A cent is 1/100th of a semitone. The difference in frequency number is of course greater the higher you go.)


It might be worth looking at deviations as I am seeing some interesting stuff, eg the number 12.52 I mentioned above. That seems to be a second nominal, spaced about a 1/4 tone above the real nominal. I might go with the format "Nominal, 12.52" as an experiment to see if that works. So that would mean that it's in the Nominal range, but at twelve and a half semitones (or more accurately twelve semitones and fifty-two cents) above the Prime.

Is this graph from one of my sound samples or some other bell? If it's mine, was this the one with the raised clapper?


No, this is one that Gideon posted, analysed by Bill Hibbert who wrote the software. I just grabbed it as an example. I'm just playing with the different sounds to see how best to present this stuff.

Terry

Re: Clapper height tonal variation

PostPosted: Sat Mar 30, 2013 6:12 am
by JohnGouwens
I HOPED I didn't have a nominal that far out of whack!!

Re: Clapper height tonal variation

PostPosted: Fri Apr 12, 2013 1:46 am
by TerryMcGee
OK, at long last (sorry!), I've done a page on how the bell sound varies between a clapper at the right height and one artificially raised.

http://www.mcgee-flutes.com/HighClapper.htm

Have a look and a listen, and tell me if it makes sense!

Terry

Re: Clapper height tonal variation

PostPosted: Fri Apr 12, 2013 4:51 pm
by JohnGouwens
OK, very interesting to see this. I would correct the part about the "sweet spot." The real "sweet spot" is the point on the circumference of the bell that produces the best, most in-tune sound. Try as they do to achieve it, bells are never completely round, and there are irregularities which can at least be somewhat de-emphasized by striking the correct point. I was confronted with this issue directly when inspecting bells at Paccard for Ball State University. I was insistent that the minor tierces had to be in tune with the corresponding bells (so the minor tierce of c had to be in tune with the prime of e-flat). We ended up changing several markings of sweet spots in the process. The experiment you had me do was more related to my statement that you can indeed bring out (emphasize) certain partials by tapping the bell at a different point. Gideon, you will note, denied that could be done. We have now proven that it can and does happen, and this clapper wasn't that much higher, actually. So, changing the clapper height by only 7% of its length made a very obvious difference - it brought out the quint (which I expected, as I hear that whenever I demonstrate this principle on tours people take in the tower) as well as a false nominal (in every sense of the word - it really shouldn't be there!). It is interesting that the "tierce octave" (or tenth, as I usually call it) is in this case a minor tenth, and a rather accurate one at that. None of my bells are perfect in their tuning (though I love the rich, characterful sound of them!), and of course what *should* be there is a major tenth. The emphasized quint isn't a particularly pleasant sound, though the false nominal is likely the biggest problem.

I also sent demonstrations of the slightly worn clapper as well as a fresh strike spot, and also using different touches at the keyboard. Whatever differences they showed are likely to be subtler - a whole lot subtler than this part of the study! I hope you'll share what you were able to find about those! Thanks!

Re: Clapper height tonal variation

PostPosted: Thu Apr 18, 2013 8:51 am
by TerryMcGee
JohnGouwens wrote:OK, very interesting to see this. I would correct the part about the "sweet spot." The real "sweet spot" is the point on the circumference of the bell that produces the best, most in-tune sound. Try as they do to achieve it, bells are never completely round, and there are irregularities which can at least be somewhat de-emphasized by striking the correct point.


I had only thought in those terms too, until I noticed our new bells from Taylors had been marked for both radial location and height, so that the installer would get the clapper height right. Normally of course we are only concerned with rotation. So I think it's fair to speak of the sweet spot as a two dimensional location, even if the vertical dimension is not normally under our control. Of course it is for you. With power comes responsibility!

I was confronted with this issue directly when inspecting bells at Paccard for Ball State University. I was insistent that the minor tierces had to be in tune with the corresponding bells (so the minor tierce of c had to be in tune with the prime of e-flat). We ended up changing several markings of sweet spots in the process. The experiment you had me do was more related to my statement that you can indeed bring out (emphasize) certain partials by tapping the bell at a different point. Gideon, you will note, denied that could be done. We have now proven that it can and does happen, and this clapper wasn't that much higher, actually. So, changing the clapper height by only 7% of its length made a very obvious difference - it brought out the quint (which I expected, as I hear that whenever I demonstrate this principle on tours people take in the tower) as well as a false nominal (in every sense of the word - it really shouldn't be there!). It is interesting that the "tierce octave" (or tenth, as I usually call it) is in this case a minor tenth, and a rather accurate one at that. None of my bells are perfect in their tuning (though I love the rich, characterful sound of them!), and of course what *should* be there is a major tenth. The emphasized quint isn't a particularly pleasant sound, though the false nominal is likely the biggest problem.


But fortunately a problem you can tune out. Perhaps many carillon owners/managers wouldn't relish the additional power!

I also sent demonstrations of the slightly worn clapper as well as a fresh strike spot, and also using different touches at the keyboard. Whatever differences they showed are likely to be subtler - a whole lot subtler than this part of the study! I hope you'll share what you were able to find about those! Thanks!


Yep, working up to it. It's good having this richness of material to analyse.

Terry

Re: Clapper height tonal variation

PostPosted: Thu Apr 18, 2013 3:51 pm
by JohnGouwens
Power vs responsibilty - simple, really, in practice - trial and error - raise and lower the clapper slightly to get the most agreeable sound. I'm very glad I have that "power" and I do require that when I consult - at least I request it. (Too many other hands got involved at Ball State.)

Re: Clapper height tonal variation

PostPosted: Thu Apr 18, 2013 3:53 pm
by JohnGouwens
As for the false nominal, note it didn't show up at all when striking the coreect spot (vertically - actually, it might be that deviating from the correct spot on the circumference would bring it out at the correct height).

Re: Clapper height tonal variation

PostPosted: Tue Apr 30, 2013 3:32 pm
by Gideon Bodden
I haven't signed in here recently, so I missed some of what has been going on.
John Gouwens wrote:
The experiment you had me do was more related to my statement that you can indeed bring out (emphasize) certain partials by tapping the bell at a different point. Gideon, you will note, denied that could be done.

Please explain, John, what exactly did I deny? Did I deny that tapping a bell at different points would make differences in sound? I have of course never said that. Then what did I say? I don't recall.