Rockstro

The Chiff & Fipple Irish Flute on-line community. Sideblown for your protection.
Post Reply
User avatar
paddler
Posts: 752
Joined: Sat Nov 03, 2007 7:19 pm
antispam: No
Please enter the next number in sequence: 8
Location: Hood River, Oregon, USA
Contact:

Rockstro

Post by paddler »

I recently stumbled across this online copy of Rockstro's "Treatise on the Construction the History and the Practice of The Flute" from 1890 (revised edition 1928). I know this is old news for some of you, but for those who haven't read any of it before, it is a very entertaining read, not least because of how bold and unapologetically opinionated he is on various issues.

I was just reading chapter XI, which I recommend as an entertaining read on flute design, materials etc, when I came across this interesting comment regarding chambering, which we have discussed several times here when questioning whether the bore profile variations that we observe in antique flutes are intentional or simply a result of shrinkage etc. The quote below (from page 160) seems to be definitive evidence that some of them at least, are very much intentional:

"Much was, however, done by actual experiment to improve the defective notes of the older instruments by "chambering," or slightly enlarging, the bore at certain places. Some of the flutes made by the old firm of Rudall and Rose were marvels of ingenuity in this respect."
User avatar
Geoffrey Ellis
Posts: 585
Joined: Thu Sep 13, 2012 1:15 pm
antispam: No
Please enter the next number in sequence: 8
Tell us something.: Crafting fine quality folk flutes from around the world since 1997, my goal is to create beautiful instruments that have the best possible voice, tuning and response by mixing modern methods with traditional designs.
Contact:

Re: Rockstro

Post by Geoffrey Ellis »

This had been added to my reading list!
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: Rockstro

Post by Terry McGee »

Unfortunately, I think this rather unclear statement has caused a lot of makers to confuse subsequent bore damage due to tenon thread wrap strangulation with intentional "chambering".

For example the bulges in the bore of the Richard Potter flute shown below are clearly caused by tenon crushing, not "chambering". See where the compressions lie, in and around the tenons (indicated by the vertical dotted lines). You couldn't produce such a shape with a reamer if you wanted to. It's hard to imagine what sort of tool could make such a shape.

Image

Note that Rockstro defines chambering as "slight enlargement" of the bore at certain places. I think he's talking about minor deviations from a straight taper probably produced by feeding in shorter fatter reamers.

I'd be interested in seeing measurements of any Rudall & Rose flutes that seem to fit the "marvels of ingenuity" promise.

An interesting question arises. Rudall & Rose get going in 1821. Rockstro is writing in 1889. An early Rudall and Rose flute could have up to 68 years to deform before Rockstro gets to see it. Plenty of time for serial strangulation to have screwed up its bore. Marvel of ingenuity, or damaged goods?

Interestingly, Rockstro and Ellis actually measured the intonation of a Rudall & Rose flute from about 1827, and got a result not dramatically different to many I've measured. Note the sharp A and B notes, and the flat foot.

Image

(from: http://www.mcgee-flutes.com/RockstroAnalysesRRflute.htm )
User avatar
paddler
Posts: 752
Joined: Sat Nov 03, 2007 7:19 pm
antispam: No
Please enter the next number in sequence: 8
Location: Hood River, Oregon, USA
Contact:

Re: Rockstro

Post by paddler »

I think the statement is pretty clear Terry. I've seen a few Rudall bore profiles that seem to be chambered near
the joints, and I have always assumed this was most likely the result of deliberate back-reaming, since it seems
to occur in a regular and repeatable way. I have yet to hear any other plausible explanation for it. I'm not aware of
anyone who confuses the bulges that you identify in your Potter flute as an example of chambering. These may well
be the result of various deterioration processes at the tenon, including compression. When you have chambers that
widen the bore, not just relative to an adjacent bulge, but also relative to a straight line average for a long section of the
bore, and perhaps even the whole bore, it is difficult to explain such chambers as being a result of compression,
especially when many flutes with compressed tenons don't exhibit them, and multiple Rudall and Rose flutes do.

Isn't it also true that many people like the way Rudall and Rose flutes, with their irregular bore profiles, play?
I doubt that re-reaming them with a straight reamer would improve them.

I think it is also worth noting that a tuning analysis of first octave notes does not tell the whole story. The Rockstro
article talks about impact on third octave tuning as being a major motivator in making various tuning tradeoffs.
We tend not to pay much attention to the third octave tuning because our music doesn't tend to go there, but
these flutes were originally designed to excel across their full range.
User avatar
paddler
Posts: 752
Joined: Sat Nov 03, 2007 7:19 pm
antispam: No
Please enter the next number in sequence: 8
Location: Hood River, Oregon, USA
Contact:

Re: Rockstro

Post by paddler »

Here are a few pictures and links to illustrate what I'm talking about in the previous post.
The evidence of chambering I'm looking at are the short sections of the bore that rise
up above the general trend line. On R&R flutes these look like small peaks that coincide
with the location of the joints between sections (hence creating an opportunity to use short
reamers to produce the chambering in a precisely repeatable way). On the graph below,
look at the line for R&R 5501 at the approximate x-axis locations of 200 and 320. You'll
see that it rises up above the general trend line in localized peaks.

The specific shape of these peaks is also important. An unaddressed mismatch between bore
size either side of a joint would appear on the graph as a vertical line, as in the line for the
Clinton flute at location 320. But the intentional peaks I am referring to have sloping sides,
indicating that the bore is either parallel sided or flaring as it approaches the join, from both
sides.

So looking at this bore profile chart from Terry's site. It shows that R&R 5501, and
to a lesser extent RC 7174, have clear evidence of this type of chambering, but flutes from
Prowse-Nicholson, Boehm-Laube, Clinton, Prattens, and Hawkes, do not.

Image

Perhaps this is just coincidence. But now take a look at charts that Casey Burns posted here some
years ago. These shows that "Typical R&R" and "Later Rudall and Carte" flutes both exhibit this same
chambering profile:

Image

Similarly, Terry's Rudall bore study here shows that R&R flute numbers 5047, 655, 5501, 7174, 519, 7120, 3522, and R&R flutes from Norman
and Hedges, all exhibit this exact same pattern. Scroll down to the bottom of the page for the bore profiles.

Now, you might wonder, well maybe this is just distortion due to tenon compression, but this graph on Terry's site shows that RR 5384 has this chambering pattern, but none of the other flutes do.

I have profiled a lot of bores myself, and have yet to see this intentional chambered profile for non R&R flutes, regardless of
whether their tenons are compressed or not. And I have yet to see a bore profile from an R&R flute that does not have this chambering
profile at the joints between the two body sections and at the foot joint. I also have yet to see evidence of this kind of chambering profile
evident at the head tenon on any flute, despite many old flutes exhibiting tenon compression symptoms. It is for all these reasons that
I assumed that R&R deliberately modified their bore profile at these points (and perhaps others) to create chambers. This assumption is
consistent with, although admittedly not proven by, Rockstro's statements that R&R flute bores exhibited chambering.

I have not done any careful analysis to try to understand what effect they were pursuing via chambers at these locations. To understand
this one would have to construct a map of the bore and locate the node locations for all notes in all three octaves of the flute's range,
and see which notes or harmonics had nodes that fell in the region of these chambers. These would be the notes affected.
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: Rockstro

Post by Terry McGee »

Nah! To convince me that a narrowing under a tenon was intentional, you'd first need to be able to demonstrate the acoustical benefits of said compression, and then convince me that the period maker was aware of said benefits and how to achieve them! If they were that clever, why did they make flutes that were so out-of-tune, and even more out-of-tune at high pitch than when extended?

As you note, the three-part body flutes show the distortion at all three tenons, but the two-part body flutes show it only at the two ends of the long body. This is not intentional reaming. This is later damage.

We have to remember why the wood under the tenon is so at risk of compression, compared to the wood anywhere else along the flute. Firstly, it starts life only half the thickness of the wall as the thickness is shared between socket and tenon.
Secondly it then has the thread trough cut in it, so the bit under the thread trough is now about a 1/3rd or less of wall thickness.
Then it has the thread wound fairly tightly around it so it is under constant load, 24/7.
Finally, when the wood expands due to moisture, the thin wood is constrained by the thread wrap and has its cells crushed. When it then dries out, it shrinks away from the thread pack. We know that happens because we so often meet flutes that can't be pulled apart for that very reason - the thread pack has come away from the bottom of the trough and adhered to the inside of the socket. Looking at the Potter graph I showed earlier, we see the result. It might look like reaming, but it's not.

And we have several indicators to confirm that:
- when you take the thread off, you can see the compression at the bottom of the trough. The outside of the trough is concave.
- when you steam the wood without the thread in place, the wood returns to its original shape, both inside and out.
- the steamed flute now plays much better than when it was compressed under the tenons
- period flutes from the same maker and even with close serial numbers tend to show quite different degrees of distortion, probably because of different playing histories.

Now, with the modelling now available to us, we should be able to determine what the acoustical effects of that bore compression/alleged reaming is. Our excuses are starting to run out!
User avatar
paddler
Posts: 752
Joined: Sat Nov 03, 2007 7:19 pm
antispam: No
Please enter the next number in sequence: 8
Location: Hood River, Oregon, USA
Contact:

Re: Rockstro

Post by paddler »

Terry, you really need to read my posts more carefully. Nobody, except you, is arguing that a narrowing under a tenon is chambering.
I have explicitly stated that I do not consider this to be chambering. Chambering is a localized expansion of the bore, not a narrowing.
I also explicitly stated that the chambering effect that I a describing does not take place at the head tenon, so your statements about what
I noted regarding the upper tenon are misrepresenting what I have said.

This thread is not about tenon compression. That is not the focus of my initial post and it is not what Rockstro discussed. Rockstro's
statements are very clear and unambiguous. The focus is on chambering, and specifically, chambering in Rudall and Rose flutes. Rockstro's
statements about this, and all of the data I presented are very specific in this regard. Your continued focus on a tenon-compressed Potter flute,
that exhibits none of these chambering patterns, is simply not relevant.

If you want to make a separate argument that Rudall and Rose flutes are uniquely affected by thread compression in such a way that they
consistently exhibit a pattern of distortion that matches the shape of chambering by back-reaming at the joints, and that no other maker's
flutes distort in this way, and that this affects them such that they are especially out of tune and hence undesirable compared to other flutes,
then I think the onus is on you to provide some evidence and supporting reasoning for why you think this is the case.
User avatar
Geoffrey Ellis
Posts: 585
Joined: Thu Sep 13, 2012 1:15 pm
antispam: No
Please enter the next number in sequence: 8
Tell us something.: Crafting fine quality folk flutes from around the world since 1997, my goal is to create beautiful instruments that have the best possible voice, tuning and response by mixing modern methods with traditional designs.
Contact:

Re: Rockstro

Post by Geoffrey Ellis »

Chambering is definitely it's own thing, unrelated to tenon strangulation. I've talked with several Baroque flute makers who have acknowledged the use of the this technique on past and present (replica) flutes of that era. I didn't dig deep into the subject with any of them, but merely asked if they knew about it and they all said, "Oh, yeah." So I've been trying to learn more about it ever since.

And forgive me for trotting out my shop-worn example (but I have limited experiences along these lines to use as illustrations), but that was one of the things that struck me about that optimized xiao created by Yang Lan from the University of B.C. In order to create superior tuning balance, he "chambered" the bore radically and repeatedly, creating an undulating profile throughout with spectacular results. So it wouldn't surprise me if our flute making predecessors did their own optimization within the limits of what they could achieve with their tooling. In the case of the xiao I reference, there was no effective way to back-ream or sand the bore to achieve this so he opted for a split-bore approach and a CNC mill. I used resin casting to get the same result.
User avatar
Mr.Gumby
Posts: 6603
Joined: Mon Jun 07, 2010 11:31 am
antispam: No
Please enter the next number in sequence: 8
Location: the Back of Beyond

Re: Rockstro

Post by Mr.Gumby »

Chambering is definitely it's own thing, unrelated to tenon strangulation.
It's also found in chanters from the classic Uilleann/Union pipemakers of the 19th century. They were that clever.
My brain hurts

Image
User avatar
gorjuswrex
Posts: 437
Joined: Thu Oct 03, 2002 6:00 pm
Please enter the next number in sequence: 1
Location: England

Re: Rockstro

Post by gorjuswrex »

I don't how much they knew about nodes and waves , theory and practical; measuring technology, etc , or much else.

Very interesting debate of experts on bores.

The whole approach of the 19thC makers may have been largely practical to overcome theory and technology limitations.
Geoffrey Ellis wrote: In order to create superior tuning balance, he "chambered" the bore radically and repeatedly, creating an undulating profile throughout with spectacular results. .
So maybe to find the effect these chambers had they could have produced test bore profiles that ''overdid'' the chambers so the exact effect became clear rather than a subtle effect that could be masked a little by the way the player blew a note ;without meaning to for the test.

or does the below mean they also would not be able to ''overdo'' the chambers? (for a test piece)
Geoffrey Ellis wrote: In the case of the xiao I reference, there was no effective way to back-ream or sand the bore to achieve this so he opted for a split-bore approach and a CNC mill. I used resin casting to get the same result.
Is it possible they had a way?
User avatar
paddler
Posts: 752
Joined: Sat Nov 03, 2007 7:19 pm
antispam: No
Please enter the next number in sequence: 8
Location: Hood River, Oregon, USA
Contact:

Re: Rockstro

Post by paddler »

I think gorjuswrex is probably right that early flute makers likely approached this practically, using
trial and error experiments to develop a knowledge base about how to do it systematically. I have read
reports of recorder makers, baroque flute makers, and shakuhachi makers all experimenting
with bore profile modifications to influence the tuning of notes and harmonics.

It is fairly easy to experimentally test the impact of bore restrictions by attaching a small lump of clay or
beeswax to the end of a slim wire and then inserting it in different positions in the bore to see what impact
it has on the tuning of various notes. You can repeatedly set or modify the location and test the
tuning in order to fairly quickly get a ball-park estimate of where a bore restriction should be located.
This is a practical method that could be used to produce a map of node locations along a flute bore,
without having to precisely calculate it.

Chambering is a bit more difficult because it is hard to reverse once done. Methods for creating
chambers include the use of a rod with a set of sand-paper flaps on the end. The rod is inserted
a set depth into the bore, and then the rod is rotated (or the flute section is rotated and the rod
held stationary) in order to sand out a chamber. The locations for such chambers would be based
on a bore map that is either calculated, produced experimentally using a method such as that described
above, or produced experimentally using destructive chambering. The latter is less likely, at least in
the early phases, because you would go through a lot of flute sections.

The sanding approach to chambering could certainly be "overdone", and in fact is quite difficult to do in a
precisely repeatable way. Using reamers is much more controllable and repeatable, but reamers can only
be used to create chambers at the intersection of two flute sections. The precise location of such intersections
along the bore is restricted, and may not coincide with the node locations you are interested in. But I think
it is worth realizing that the location of joints is not totally fixed, since the flute maker has some latitude in
where to divide the flute, how long to make the tenons and sockets, and which way around to do the tenon/socket
joint, etc. Tone holes also create chambers in the bore, and represent opportunities for adding chambering
using undercutting etc. Similarly with keyed tone holes. There are actually quite a lot of opportunities for
influencing chambers in the bore of a flute.
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: Rockstro

Post by Terry McGee »

Now the problem with this conversation, and we've been here before, is that we're talking in useless vagaries. And this goes right back to Rockstro. What sort of starting point is "Some of the flutes made by the old firm of Rudall and Rose were marvels of ingenuity in this respect."?

To talk seriously about "chambering" we need to see detailed bore profiles, and we need to see a proposed datum line for the same bore without the "chambering". Otherwise, how do we know whether it's really been "chambered" over this region, or compressed over an upper region, resulting in an apparent chamber lower down?

So, here's the challenge. Someone, anyone, find us a bore that they feel is reasonable evidence of "chambering". Overlay a graph of said bore with a dotted line that you reckon would represent the same bore if it were not "chambered". And let us test alternative explanations for that bore and see where it leads. If we can't demonstrate such evidence, the claim falls victim to Hitchen's Razor:

Hitchens's razor is an epistemological razor expressed by writer Christopher Hitchens, asserting that the burden of proof regarding the truthfulness of a claim lies with the one who makes the claim; if this burden is not met, then the claim is unfounded, and its opponents need not argue further in order to dismiss it.

Hitchens has phrased the razor in writing as "What can be asserted without evidence can also be dismissed without evidence."


This graph illustrates the approach I'm talking about. The measured diameters are shown by the thick navy blue trace. The thin lines indicate the various possible original bore scenarios I want the reader to consider.

Image
User avatar
Conical bore
Posts: 505
Joined: Sat Aug 04, 2012 7:12 pm
antispam: No
Please enter the next number in sequence: 8
Location: Pacific Northwest USA

Re: Rockstro

Post by Conical bore »

What this thread needs, for those of us following along, is some nice X-Ray images of the flutes under discussion. Doesn't anyone with their hands on a bunch of antique flutes know a friendly (or bribe-able) radiologist? I would love to see a cross section image of what's being discussed here, in the actual flute.
:)
User avatar
paddler
Posts: 752
Joined: Sat Nov 03, 2007 7:19 pm
antispam: No
Please enter the next number in sequence: 8
Location: Hood River, Oregon, USA
Contact:

Re: Rockstro

Post by paddler »

Terry McGee wrote:Now the problem with this conversation, and we've been here before, is that we're talking in useless vagaries. And this goes right back to Rockstro. What sort of starting point is "Some of the flutes made by the old firm of Rudall and Rose were marvels of ingenuity in this respect."?

To talk seriously about "chambering" we need to see detailed bore profiles, and we need to see a proposed datum line for the same bore without the "chambering". Otherwise, how do we know whether it's really been "chambered" over this region, or compressed over an upper region, resulting in an apparent chamber lower down?

So, here's the challenge. Someone, anyone, find us a bore that they feel is reasonable evidence of "chambering". Overlay a graph of said bore with a dotted line that you reckon would represent the same bore if it were not "chambered". And let us test alternative explanations for that bore and see where it leads.
I agree, but I believe I already addressed this point quite clearly when presenting graphs that illustrate the specific kind of chambering that occurs in Rudall and Rose bore profiles.
The specific regions I identified rise above pretty much any datum line you can reasonably imagine. They are localized peaks, not just relative to the bore profile of the tenon immediately next
to them, but relative to any reasonable trend line that you care to draw along the entire length of the bore. Unless, of course, you assume that the original bore was entirely different and much
larger across its entire length (maybe cylindrical, and it only looks conical now because of damage?) and has shrunken in a non-uniform way such that these small cavities are left in the same
specific location on all Rudall and Rose flute and not on any others. But this kind of argument stretches credibility beyond its limits, and certainly violates Hitchen's razor.

Your argument that the bore cavities I identified may be caused by tenon compression is completely blown out of the water by the following observations about the data:

1. They occur only at two of the three tenons in the Rudall & Rose bores profiled, and never at the head tenon. Why would the head tenons not exhibit these chambers, especially
when the head tenons are exposed to more wetting and drying action than the lower tenons, and when they often quite clearly exhibit tenon compression effects?

2. They occur on all of the 10 or so Rudall and Rose flutes profiled, and on none of the other flutes profiled, regardless of how much tenon compression the others seem to exhibit.

3. The cavities extend across the joint, with matching bore diameters on two separate sections of the flute, one of which has no tenon. You can verify this by the lack of vertical
lines in the graphs, showing that these Rudall flutes do not have discontinuities in the bore profile across joints, unlike the Potter flute you keep showing.


In answer to conical bore's observation, it may be nice to have X-Ray images of the bores, but the bore profile graphs really present the same information in a clearer way. You can view them
as an image of one side of the bore, but with the bore variations exaggerated (depending on how the axis are labeled) to make them more visible. When the graph line climbs up (i.e. the
values get larger as you go from a point on the left to one to its right) you have a section of bore that is expanding (not just contracting slower than the datum line suggests that it should!).
If you look carefully at the graphs you will see very few points where the bore actually expands when traveling from left to right. Of these, most have a discontinuity in the graph line,
indicating that we have moved from one flute section to the next, across a joint, and that the bore diameters do not match. These are NOT the cases I'm looking at. The ones that interest
me are those that show an expansion and do not show a discontinuity. These seems to occur predominantly (perhaps exclusively?) in Rudall and Rose flutes.

So, I encourage everyone to take a close look at the graphs and try to identify sections of bore that exhibit all of the following features: a localized expansion (i.e., the line climbs up), a set of
values that exceeds those of any reasonable datum/trend line you care to come up with (this would be a straight line you draw that approximates some non-chambered bore profile you care
to hypothesize), and no discontinuity (no completely vertical section in the graph line). Then see how many of the examples you identified are Rudall and Rose flutes. Make sure that
you look at the following graph which shows 9 Rudall bore profiles and comes from Terry's website. I linked to this graph earlier, but it did not appear directly in the thread. I'm hoping that
the localized bore expansions that rise above any reasonable trend line are pretty obvious. I think it is also worth noting that many flutes show bore expansions at the very bottom of the foot.
I contend that these too are intentional, and that they are likely formed by back-reaming.

Image
User avatar
Geoffrey Ellis
Posts: 585
Joined: Thu Sep 13, 2012 1:15 pm
antispam: No
Please enter the next number in sequence: 8
Tell us something.: Crafting fine quality folk flutes from around the world since 1997, my goal is to create beautiful instruments that have the best possible voice, tuning and response by mixing modern methods with traditional designs.
Contact:

Re: Rockstro

Post by Geoffrey Ellis »

And back-reaming at the foot would not be at all surprising--it gets used on other types of flutes as well, though using it with vent holes does probably have a different effect than using it as more of a bell effect on a flute without vent holes (such as a shakuhachi).

But those peaks on the graph are quite high. I made a few "datum lines" by the simple expedient of holding a straight edge up against the computer monitor :-) It took some squinting on my part because the colored lines are like a tangle of thread, but any line I created that could accommodate the entire bore, did not rise near the peaks. And as you point out, there are no vertical lines in any of the bores represented, which supports the assertion that the spikes don't come from mismatched sections (where one is accidentally smaller or larger due to inconsistent reaming of the sections). And the fact that this effect repeats so consistently does support the notion of a deliberate piece of engineering.
Post Reply