Rockstro

[Tunborough]

Oh, and one more observation. Rather than go to any physical trouble, why not first do some computer modelling of the differences between say the Rudall bore in question and one or more assumptions of what it might/should have looked like?

Tunborough, I notice you are maintaining a low profile. A healthy instinct for self preservation at work?

No, just overly-occupied with other stuff to do. Fortunately, unlike you, the fires I've been fighting have all been metaphorical. I've been keeping an eye on this thread but haven't had the time to follow it closely, or comment on it.

I also haven't been sure how to proceed with the modelling, but I have a plan now. First, model a real Rudall, with chambers, and see what the relative tuning looks like over three octaves. Then, fill in the chambers, adjust the hole geometry a little bit to correct for tuning impacts, and see what the relative tuning looks like then.

I'd need a pile of information, and there's several rabbit holes we'd need to consider. For starters: for a specific real Rudall, the geometry of the bore, the toneholes, the embouchure hole, and the head joint; and a three-octave fingering chart. Unfortunately, WIDesigner doesn't do keys yet, so we might have to do some fudging for open keyed toneholes.

I hope to have more to say tomorrow.

[paddler]

Oh, and one more observation. Rather than go to any physical trouble, why not first do some computer modelling of the differences between say the Rudall bore in question and one or more assumptions of what it might/should have looked like?

Tunborough, I notice you are maintaining a low profile. A healthy instinct for self preservation at work?

No, just overly-occupied with other stuff to do. Fortunately, unlike you, the fires I've been fighting have all been metaphorical. I've been keeping an eye on this thread but haven't had the time to follow it closely, or comment on it.

I also haven't been sure how to proceed with the modelling, but I have a plan now. First, model a real Rudall, with chambers, and see what the relative tuning looks like over three octaves. Then, fill in the chambers, adjust the hole geometry a little bit to correct for tuning impacts, and see what the relative tuning looks like then.

I'd need a pile of information, and there's several rabbit holes we'd need to consider. For starters: for a specific real Rudall, the geometry of the bore, the toneholes, the embouchure hole, and the head joint; and a three-octave fingering chart. Unfortunately, WIDesigner doesn't do keys yet, so we might have to do some fudging for open keyed toneholes.

I hope to have more to say tomorrow.

It is clear that the effort involved in trying to use computer modeling to answer this question far exceeds the effort of taking an existing Rudall flute that has this bore profile and simply playing it with and without the
tenon chambers filled. In either case we need access to an original flute. Playing the actual flute to determine the effect has the huge advantage that it perfectly models the acoustics of the flute by using the real thing
directly. Modeling it on a computer requires detailed measurements, not just of the bore, but of the tone holes, the degree to which the pads protrude into the tone holes, etc. It also requires a computer model that
knows how to accurately simulate the behavior of all this, which we currently do not have. The model is currently not accurate enough even for predicting the lower two octave behavior of a simple keyless flute. I say
this having fairly recently spent a lot of time gathering measurement data and playing data from an old Sam Murray keyless flute and working with Tunborough to try to tune the computer model to accurately model
it. Terry is well aware of this effort and was included in the information exchange that took place.

The bottom line is that the model is impressive, as far as computer models of flutes go, but it is still far from being as accurate as the real thing, and is clearly not currently up to the task we have here. Getting it there
represents a lot of difficult and time consuming work, and may even still prove to be not currently possible. Playing a real flute with and without the chambers filled represents 15 minutes of easy work by someone who
has an R&R flute and a tuner, and yields a definitive answer.

[paddler]

Woah, the bushfires in New South Wales appear to have nothing on the fires raging in a popular Irish Flute forum!

Now there's no way I can comment on a fraction of the above (I'm about to head to bed!). So I'll pick a point at a whim.

I'd said (of Rockstro):
"Fourthly, the concept of "chambering" is the sole invention of a known and proven liar. Why should we pay him any heed when all his other assertions have been thoroughly debunked? See Welch, for example."

To which Paddler responded:
"Wrong! Several people on this thread have stated that they have used chambering themselves. I have also used it myself. It is well documented in books and articles on flute making, and pipe making."

I object to your use of the words "sole" and "liar" in that statement. You posted that statement right after others here had mentioned their own uses and knowledge of chambering, that were
independent of Rockstro's statement. I chose to declare your statement incorrect, rather than to acknowledge that Rockstro, others posters here, and I myself, are proven liars.

You also have presented no evidence that Rockstro was lying when he published his statement about the use of chambering in R&R flutes, or that he was a habitual liar. To justify calling someone a liar you
would not only have to provide evidence that what they said was untrue, but you would also need to prove some level of intent to deceive. And even then, I'm not sure it is the kind of accusation or language
that is consistent with the guidelines for posting on C&F.

[paddler]

Now, I've had a quick whip through of all my other period and more recent books on flutes - not one mentions chambering in its Index. A search on-line for "flute chambering" finds nothing apposite.

So, I have to challenge Paddler's assertion: "It is well documented in books and articles on flute making". Let us flush out these references and see if they speak to us....

If you read the book "Woodwind Instruments and their History" by Anthony Baines, you will find the following statement:

"Tuning, in this sense, mainly entails such things as undercutting a hole or chambering the bore to sharpen a note (a local expansion of the bore will sharpen the note which has an antinode at that point)."

Similarly, in the book "Baroque Woodwind Instruments: A Guide to Their History, Repertoire and Basic Technique", By Paul Carroll, you can find the statement:

"By making a flute in several pieces or joints, it also became possible to tune the instrument more carefully. Apart from the reliability of boring shorter lengths of wood using the only available tools, that is a shell bit and reamer, with less risk of misalignment or splitting, local reaming, including chambering to improve general intonation, could be carried out more easily."

Does this satisfy you, or would you like me to go on?

There is lots of evidence that chambering was used by woodwind makers, there are makers today who use it, there is a clear scientific basis to explain why it works, and there are instruments in existence that have measurable chambers in their bores.

[paddler]

Here is a little more for you to chew on regarding the underlying physics and how long it has been widely known.

Take a look at Arthur Benade's book "The Fundamentals of Musical Acoustics", pages 473 onwards. Specifically, read
section 22.3 "Adjustment of Natural Frequencies by Means of Small Changes of Air-Column Shape". You will find the following
statements, and a lot more.


The interlacing of pressure nodes and velocity nodes allows us to deduce the following general principle, which was first enunciated a century ago by Lord Rayleigh:

1. A localized enlargement of the cross section of an air column (a) lowers the natural frequency of any mode having a large pressure amplitude (and therefore small flow) at the position of enlargement, and (b) raises the natural frequency of any mode having a pressure node (and therefore large flow) at the position of enlargement.

It is possible to calculate curves giving the effect of a small, localized enlargement or contraction on the frequencies of each vibrational mode of an air column. We call such curves perturbation weight function curves, or W curves for short. The mathematical techniques for putting them to use are a highly developed part of mathematical physics. These techniques are known collectively as perturbation theory ...


The book has extensive discussion of this, and even shows diagrams to illustrate the tuning effect of specific cavities in the bore of a woodwind instrument on the first three modes of its oscillation.
If you want to understand this, I suggest you read some books and do some research.

On page 250 and 251 of Cris Forster's definitive work on "Musical Methematics" you will find extensive discussion of this topic, specifically in the context of flute tuning. There is even an analysis of the impact of the tone hole cavities in an Armstrong Boehm silver flute.

This section starts out with the following statements:

"Another important fine-tuning technique consists of changing the bore diameter of woodwind instruments at the location of a pressure antinode, or at the location of a displacement antinode for a given mode of vibration. If we expand the bore by lightly sanding the inner surface of the flute tube at the location of a pressure antinode, such an increase in diameter causes a local increase in volume (V), which in turn produces a local decrease in pressure."

It then goes on to explain the impact of this in terms of the springiness of the air, the energy in the air and the frequency at which it will vibrate. I can include more of that explanation if you really want,
but I think it would be more productive for you to study the literature. Hopefully, I have given you enough starting points. There is a lot more about this in work by Nederveen who has written books on
the acoustic aspects of woodwind instruments.

[tstermitz]

I'm a mere intermediate-going-on-beginner ...

No, no - you need to go the other direction.

This is actually the point.

You start as a Newby, knowing nothing, and knowing that you know nothing.

Then you become an Intermediate and THINK you know something.

Enlightenment only comes when you realize you know very little and become a beginner again... and forever!

[Geoffrey Ellis]

Just for fun, this is a silhouette image I took of the mandrel that I use to make the bore of the optimized xiao. This is brass and is used to make the silicone molds that I use, and because of it's shiny quality this is the only way to get a good image that shows the radical chambering effects. This is a modern creation. Those undulations were the method used to balance the tuning and enrich the harmonics of the flute, and it worked like you wouldn't believe.

[Nanohedron]

I'm a mere intermediate-going-on-beginner ...

No, no - you need to go the other direction.

This is actually the point.

You start as a Newby, knowing nothing, and knowing that you know nothing.

Then you become an Intermediate and THINK you know something.

Enlightenment only comes when you realize you know very little and become a beginner again... and forever!

Ah, of course. Of course. So seldom do we wade into the philosophy of the learning process here, that I assumed I was merely dealing with bumpy English.

[PB+J]

Just for fun, this is a silhouette image I took of the mandrel that I use to make the bore of the optimized xiao. This is brass and is used to make the silicone molds that I use, and because of it's shiny quality this is the only way to get a good image that shows the radical chambering effects. This is a modern creation. Those undulations were the method used to balance the tuning and enrich the harmonics of the flute, and it worked like you wouldn't believe.

SO interesting!


But I'm not getting how that would work. So you make a silicone mold around that mandrel. The what happens?

This whole discussion has been very interesting. I gotta go practice

[Geoffrey Ellis]

Just for fun, this is a silhouette image I took of the mandrel that I use to make the bore of the optimized xiao. This is brass and is used to make the silicone molds that I use, and because of it's shiny quality this is the only way to get a good image that shows the radical chambering effects. This is a modern creation. Those undulations were the method used to balance the tuning and enrich the harmonics of the flute, and it worked like you wouldn't believe.

SO interesting!


But I'm not getting how that would work. So you make a silicone mold around that mandrel. The what happens?

This whole discussion has been very interesting. I gotta go practice

I don't want to distract too much from the main point of this thread (i.e. that chambering is a real and practiced technique among woodwind makers), but here is how it works.

Making this bore in a conventional manner is impossible. So I made a brass mandrel (machined on a CNC lathe, since it would have been near impossible to do by hand), and I used this to make a silicone mold. Once I had the silicone mold, I would cast another mandrel from wax. This wax mandrel had a hollow metal rod down the middle. I would use this mandrel inside a wooden bore and center it. Then epoxy is poured around the wax mandrel, the whole is put under pressure (to get rid of bubbles) and allowed to cure. Once cured, I would take a thin steel rod that is sized to fit inside the hollow metal tube. I heat the steel with a blow torch, insert it into the hollow rod and it melts the wax enough to remove the hollow tube. The tube was needed to strengthen the wax mandrel. Once the metal tube is removed I can pour boiling water down the recess in the wax and melt the wax out of the bore. Voila! HUGE PAIN IN THE NECK. It was a lot of fun and I had to learn a lot of new skills along the way, but it's a fussy way to make a flute and lots of things can (and did) go wrong along the way.

But it was this unconventional bore design that allowed the creation of an amazingly balanced instrument.

[paddler]

The method Geoffrey outlined has the advantage that it allows one to create a bore with chambers that are of whatever size and in whatever
location you like. But its use of modern epoxies obviously shows that it was not used in the case of Rudall and Rose or other antique wooden
flutes. The principle is the same though, in the sense that the chambers and constrictions in the bore are placed such that they have a positive
tuning effect, it's just that historically, simpler and more restrictive methods were used.

The simplest method for achieving some limited degree of chambering is to make a flute from several short sections, ream each section to some
desired, non-chambered but still irregular, shape, and then use another reamer from the opposite end of the section to open up some part of the
bore at the narrow end so that a chamber appears in the bore once the sections of the flute are assembled together. The term chamber is used to
describe a localized cavity in the bore in which there are sections of bore either side of it that have a smaller diameter. More formally, you could
describe a chamber as a local maxima in the bore profile graph. Informally speaking, it is a peak from which you descend regardless of
which direction you travel along the line.

This is what Paul Carroll is talking about in his book "Baroque Woodwind Instruments: A Guide to Their History, Repertoire and Basic Technique", when
he says: "By making a flute in several pieces or joints, it also became possible to tune the instrument more carefully. Apart from the reliability of boring
shorter lengths of wood using the only available tools, that is a shell bit and reamer, with less risk of misalignment or splitting, local reaming, including
chambering to improve general intonation, could be carried out more easily."

Note that you can not use reaming to create chambers in the bore of a flute that is made from a single section. You can only use reaming to create
chambers at the joints between sections. In the case where those joints have a tenon on the upstream/north end and a socket on the downstream/south
end, then the reaming would be done on the section of bore under the tenon (prior to the tenon being cut!). In the R&R flutes this is precisely where we
see the chambers.

[Terry McGee]

Now, before pressing on, I need to go back to:

Okay, the language here is getting pretty heated among certain parties. If we can't avoid a tone of hostility (deny all you like; that's how it looks), the thread will have to see an unfortunate end. People can rein in the drama and aggression and still make their points quite lucidly.

It's up to you guys.

I hope we can avoid you having to boot us out, Nano. I'm not bothered by robust discussion, but if you feel I'm going too far somewhere, do feel free to rein me in. I think this discussion is one that the Irish Flute world absolutely needs to have, and I'm perfectly happy if I lose a bit of skin along the way. We have a chance here to get to the bottom of an issue that I think has been holding us back. Let's grab the opportunity with both hands.

[Terry McGee]

I also haven't been sure how to proceed with the modelling, but I have a plan now. First, model a real Rudall, with chambers, and see what the relative tuning looks like over three octaves. Then, fill in the chambers, adjust the hole geometry a little bit to correct for tuning impacts, and see what the relative tuning looks like then.

OK, but I'll add to it. What you've suggested assumes that there are only two options - "as-found" and "pre-chambered". I'd expand that to three: "as-found", "pre-chambered" and "pre-distorted". Remember I think a lot of what people assume is chambering is in fact bore distortion caused by tenon thread constriction, and possibly other forces yet to be identified. We certainly have to rule damage out before we go looking for subtleties than may never have existed.

I'd need a pile of information, and there's several rabbit holes we'd need to consider. For starters: for a specific real Rudall, the geometry of the bore, the toneholes, the embouchure hole, and the head joint; and a three-octave fingering chart.

Again, I'll add to that. The "specific real Rudall" needs to be in the hands of or in near proximity to someone who can take careful measurements on demand of both bore and tuning. In my experience, you never have all the measurements you want. This is going to be an unfolding journey. We have to be able to go back for more. I think that leaves our current subject, #3801 out in the cold, unless we can track down its current whereabouts and find someone nearby who can take the measurements we need.

I wonder if anyone can identify such a flute? Paddler seems concerned that it be a Rudall & Rose; I'm not so worried. But it clearly needs to have some bore anomaly that Paddler would accept as a sign of "intentional chambering". I'd leave that decision in his hands. Although I guess if I agreed with him that there was no other possible explanation, we could skip the rest of the study and slip down the pub for a few tunes....

Unfortunately, WIDesigner doesn't do keys yet, so we might have to do some fudging for open keyed toneholes.

I don't think that will cause us any problems, but hey, if it does, we've learned that too and we'll learn how to deal with it.

So, Good Man, Tunborough for being up for the fray. If we can locate a suitable subject....

[Terry McGee]

It is clear that the effort involved in trying to use computer modeling to answer this question far exceeds the effort of taking an existing Rudall flute that has this bore profile and simply playing it with and without the
tenon chambers filled. In either case we need access to an original flute. Playing the actual flute to determine the effect has the huge advantage that it perfectly models the acoustics of the flute by using the real thing
directly. Modeling it on a computer requires detailed measurements, not just of the bore, but of the tone holes, the degree to which the pads protrude into the tone holes, etc. It also requires a computer model that
knows how to accurately simulate the behavior of all this, which we currently do not have. The model is currently not accurate enough even for predicting the lower two octave behavior of a simple keyless flute. I say
this having fairly recently spent a lot of time gathering measurement data and playing data from an old Sam Murray keyless flute and working with Tunborough to try to tune the computer model to accurately model
it. Terry is well aware of this effort and was included in the information exchange that took place.

The bottom line is that the model is impressive, as far as computer models of flutes go, but it is still far from being as accurate as the real thing, and is clearly not currently up to the task we have here. Getting it there
represents a lot of difficult and time consuming work, and may even still prove to be not currently possible. Playing a real flute with and without the chambers filled represents 15 minutes of easy work by someone who
has an R&R flute and a tuner, and yields a definitive answer.

Yes, but.....
1. You are assuming the only choice is "as-found" and "pre-chambered". I want to rule out "pre-distorted". I can't imagine the RR owner being happy for me to re-ream their precious heirloom to my guess as to what it might have looked like when it first left the bench....
2. Who would do this experiment? Supposing it were I - would you accept my word that the filled-in chambers worked better?
3. We can do both. And may need to. And maybe more. But I think that the modelling will guide us. What have we got to lose?
4. We still need a lot more data than #3801 supplies. So we're still looking for a flute that meets Paddler's bore shape criteria and my measurement detail criteria.

[Terry McGee]

Now, further up, Paddler questions my characterisation of Rockstro as a liar.

Probably Rocksto's most egregious sins against the truth have been dealt with by Christopher Welch, in his book "History of the Boehm Flute", first edition 1882, second edition 1891. Welch systematically and painstakingly dismembers Rockstro in regard to his calculated lies about Boehm. It's quite a ride. But Welch well understood that Boehm wasn't Rockstro's only victim. His preface to his second edition ends with:

"In conclusion, I can only express a hope that when Mr. Rockstro has gone to his long rest, when the pen has fallen from his clay-cold hand, and his tongue is silent for ever, his memory may meet with more tender treatment than that which he has accorded to those that have preceeded him in the path he is treading. I do not say, may he not be assailed as he has assailed Boehm, for such an attack would only recoil on him by whom it was made. But may no future improver of the flute who proceeds to tear the perfection of others to tatters, and then to set up his own, use him as a butt for sorry jests, as he has jeered at Clinton sleeping in his quiet grave, or point at him the finger of scorn, and describe him as he has described the dumb and defenceless Siccama, of whom he has not scrupled to write: "About the year 1842, he conceived the unfortunate idea that he was destined to be the inventor of a new flute that should eclipse everything that had been made or imagined. Having become imbued with this notion, he set to work with all the vibour of an energetic nature. He had little knowledge of the flute, and less inventive genius, but he determined to bring out a flute associated with his name, and he did so."

(The reality of course is that Siccama invented the remarkable 10-key flute that Pratten later revised back to an 8-key, the Pratten's Perfected. Pratten was admired by Rockstro, and gets a number of positive mentions in the book.)

My research colleagues Adrian Duncan, Andra Bohnet and I have done some work on analysing Rockstro's campaigns against Clinton and Siccama. We have toyed with the idea of a complete revision of Rockstro - setting everything to right - but recognised that it would amount to a life's work.

Rockstro didn't limit himself to blatant lies, he also employed sins of omission. Find me a reference to Radcliff in his book. Let's not even mention the name of competitors. Especially those whose flutes have more merit than his. Rockstro's reworking of the Boehm was based on a mistaken theory that holes of equal size would yield notes of equal strength. He overlooked the fact that venting support from lower holes encouraged graduated holes, which Clinton introduced, Rockstro lampooned, but history took up.

Once one gets a feeling for Rockstro's freedom with the truth, reading his book takes on a new and heavy load. Nothing this man says can be taken at face value. It might be true, it might not. That all depends entirely on whether it will advantage Rockstro.

Interesting name, Rockstro. His real name was Rackstraw, but that wouldn't cut it in the high society in he wished to associate himself. Something a little more Italian was needed. His very good friend and flute teacher Mr Carte was born Cart. He called his son D'Oyley. This was a period of shallow self-aggrandisement.