I need to reread terry’s excellent post–and thanks to all for this thread–and the question
remains:
Flute smyths today are capable of making simple system flutes well enough in tune. The Old Guys were great
flutesmyths and there was certainly a lot of development going into flutes, competing
with the Boehm,
and I suppose one advantage of the Boehm design was better tuning.
So, why didn’t they do what contemporary SSF-smyths have done?
This has got nothing to do with it … but why are you spelling “smyths” with a “y”? Is it an in-joke I’m not getting?
no, just like the spelling.
Weirdos–we’re people too!
I think for (at least) four reasons.
Firstly, they were dealing with a massive impediment we are quite free of. There were at least two reigning pitches that had to be accommodated (430 and 455), possibly three if we include the compromise Society of Arts Pitch (445). Very hard to optimise a flute design over such a wide range. I admit to wondering why they didn’t just resort to corps de rechange (alternate LH sections).
Secondly, we have time on our side. Flute design being iterative up to now, accuracy should get better with elapsed time.
Thirdly, we probably have a far more critical musical environment than they had. Our customers routinely listen to the world’s best flute players playing the flutes that suit them most. That sets the bar high.
Fourthly, we have technology on our side. So far that’s been in the form of tuners that, properly used, allow us a level of precision, convenience and reliability that they didn’t have. Recently we added Real Time Tuning Analysis that for the first time lets us know what we are actually doing, rather than what we think we are doing, when we play.
Soon I hope we’ll be adding computer-aided flute design (we’re getting close, but we’re not there yet - these flutes are considerably sneakier than I suspect anyone appreciates!). That should open some doors that have never even been tested before.
It’s very interesting to compare flutes from the start of the 19th century to the middle and to the end - there really were some big improvements made. But there were some losses too - the earlier flutes were prettier, more romantic, more atmospheric. I’d say the same about our modern flutes. Is it absolutely necessary that improvements in performance have to result in losses to the atmospherics?
(Note to self: must build an atmospherometer.)
Terry
Terry, I wonder if you could help with one other thing:
It’s fascintating stuff you’ve put up here. I’ve also read Carte’s article as linked on your website. But - and I may have missed the answer in all the detail - why is the bottom D somewhat flat on a lot of the old flutes? I get the stuff about the E, and the A, and the F# … but what about that flat foot? There’s got to be a reason in there somewhere …
I think this is considerably harder to be categorical about, but let me float a theory and see if it seems to make sense.
But first, we need to note that in these flat footed flutes, it’s not just the D that is flat. Low Eb is usually worse, and low C# worse again. Normally the C isn’t bad. So we have to look for a reason for that pattern.
By the mid 18th century, they were getting pretty good at making flutes. Four keys were common, with the flute ending at D. Makers at the time had found that if they continued the taper to the bottom the low D was distinctly flat. My modelling suggests that this is because if you taper the bore enough to get all the other notes in tune with their octaves, the cumulative effect will be too great by the time you get to the bottom. Fortunately, they also determined that if they flared the bottom of the bore, by the simple expedient of reversing the same reamer and coming in from the bottom, they could fix that.
But then some bright spark got the idea that the flute should go down to C. That required them to do away with the flare, and extend the taper down past where D was now a side hole, past C#, before a small flare could be introduced. That small flare was enough to get C back in tune, but low D, Eb and C# were terribly flat.
I suspect it wasn’t helped when Nicholson “improved” the flute by enlarging particularly holes 2 and 5. That would make the body notes sharper. Nobody mentioned shortening the foot to bring those notes up too.
I suspect another contributing confusion was that of pitch. At the start of the century, the pitch was circa 410. We commemorate that with modern “baroque pitch” at 415Hz. But it was on the way up, and soon settled at 430Hz, before the young bucks in the Philharmonic movement pushed it up to 455. So flutes should have been shortening, without which the foot notes were going to tend flat anyway. They were getting shorter, but not at the rate the pitches would seem to demand. Complicated by the fact that the same flute had to do for both pitches!
So I think we probably had a number of things going on, all contributing to flat feet. From about Nicholson on, things started to get better. I have yet to confirm how this happened, and at whose hands, and by what methods, but I think we can assume that it related to increasing the bore in the lower RH and foot area. Essentially, the taper had been overdone, which probably became more and more obvious as pitch rose.
I think Boehm’s 1832 conical would have come as a kick in the bum to other flute makers, and we certainly see Siccama sorting stuff out around 1847. Then of course Boehm returns with the very well tuned cylindrical bore - another kick in the bum! Pratten responds with his Perfected (really a Siccama restored to 8 keys) and the problem is just about fixed. Significantly, Siccama and Pratten have reduced the length and the taper angle so that the foot bore is not so small.
Interestingly, Rudalls pretty much carried on with the old model, but then the majority of their work was in selling cylinder flutes, so they can probably be excused. They did snap to around 1895 when high pitch was abandoned in favour of something close to modern pitch. But that meant that they had missed the whole high pitch thing! But only in their conical flutes - their cylinder flutes were definitely aimed at high pitch.
Irish players in the 60’s onwards worked out ways to play to get around this flatness, and it’s probable that similar approaches were used back in the period. Carte and Ward (both on my web site) both comment on tuning as a real problem. Eg Ward:
Akin to this mystification, is the course pursued in many instruction books. We have not seen one which candidly informs the learner of the manoeuvres required to play the ordinary flute in tune. They give a vast variety of modes of fingering the same notes; with the object, one might suppose, of perplexing the scholar, or of showing the author’s ingenuity. Their silence on the one hand, and their profuse loquacity on the other, furnish strong evidence of their having (in the old flute) a bad case to deal with, requiring a large amount of special pleading, and a studious concealment of the important truths.
The authors of these books do not tell the student that the intonation of the instrument rests entirely with the performer, and depends upon certain zig-zag manoeuvrings of his lip, and other subterfuges. They do not point out which notes require the flattening or sharpening, the forcing or tempering process. Oh, no! the flutes they recommend are, doubtless, well in tune: the student is to blame if he cannot play in tune; and he must take expensive lessons to learn, not how to tutor his own ear, but to correct the false intonation of the instrument.
Ah, lovely stuff!
Terry
Brilliant stuff, Terry!
… which leads me on to yet another question: it seems to me from the above that there is a problem in the foot area which is inherent in having an 8-key wooden conical bore flute. So, how have modern makers, like you, for instance, Terry 
, overcome this? Or have I misunderstood the original problem? Did the 19c makers simply not get around to addressing the flat foot issues? And could they have, if they’d tried?
I’m still wondering if the whole flat foot thing may have to do with a keyless fingering for F natural.
On my French flute the F# is quite flattened so that xxx xox yields something near F, while embouchure change leads to a correct F#. Alternatively, a flat foot could also help to create a valid cross fingered F. The upside of this approach would be that the embouchure change is gradually applied going down the flute instead of for just one note.
I could imagine people being taught this as a standard, which might be why it was a feature for so long after keys and large holes have made these fingerings invalid.
Well, the above is just a guess which I hope is a little educated
There might be a bore-taper design residue from that, but I think on English flutes the forked F fingering had been pretty conclusively abandoned post Nicholson, if not before.
They slowly got on top of it - as I mentioned above it was pretty well sorted out by Siccama and Pratten, and by increasing the lower end bore. There are vestiges left in their flutes, but nothing like the 40 or so cents we see in early 19th c flutes.
Compare the Prowse-Nicholson taper with the later Pratten taper, and you’ll see what I mean:
I came up with my “Long D foot” to give me even more control of the low D pitch while at the same time reducing the acoustic and aerodynamic losses of the narrow C-foot bore. I understand that at least some other modern makers have since taken the same approach.
Terry
The graph’s fascinating, Terry. I presume the y axis is bore width. Is the x axis tube length? And from where? L1? If I’m reading the thing right, are those spikes along the way areas where the bore has been made deliberately wider in order to help with particular notes?
[By the way, all this talk of the mechanics of the thing must have inspired my flute, which has really been singing out tonight. ]
Yeah, but I think it was already on the skids. Potter doesn’t give it. Gunn (late 18th c) gives it for one-key flutes but not for keyed flutes.
Interestingly, Lindsay gives it (in 1828, so well after the introduction of Improved flutes), as his second option (of 2). Short F is his other option. Seems he wasn’t a Long F person (my T.Lindsay’s Improved has no Long F).
Nicholson, writing in 1836, still gives it, but only as his last option (of 3). He gives Short F as option 1, Long F as option 2, but then comments that he never uses it. He concedes that the D-F slur is difficult (read: “Could kill an ordinary man”), but not insurmountable.
Siccama gives Short F, Long F and then Both F keys as his three options on his 10 key flute. Neither he nor Pratten give a cross-fingered option.
So I guess it’s fair to say it started petering out with the introduction of the short F key, and finally petered out about 20 years following the introduction of the large holes.
Of course, it bounced back with a vengeance when fully keyed flutes were introduced post Boehm. Carte, Clinton and Radcliff flutes had xxx xox F. But that was contrived mechanically - a hole opposite the Short F hole was opened by some simple mechanical logic along the lines:
IF .NOT. R2 .AND. .NOT. R3 .AND. NOT Long_F
Terry
Correct on the axes, x measured from top of conical section (ie top of LH section).
The spikes are controversial - some argue that they are of intended acoustical significance, I’m not so convinced. They occur suspiciously at section junctions - LH to RH or RH to foot. Keep in mind they are pretty small - about 0.3mm on diameter (the Y axis is very magnified). I wonder if they are not intended to minimise aerodynamic losses where mismatches might occur. I intend to do some modelling to test their possible acoustical implications. Anyone like to go down on record first with a prediction?
Oh, I should though mention the one at 17mm diameter on the orange curve, kindly supplied by Dave Copley. It appears far more intentional, or the result of a blunder in drilling! I’ve not investigated that. And the various horizontal ledges up at the top of the bores - they are important and intentional, but I’ll need to do more modelling to be able to characterise them adequately.
Terry
Incidentally, I’m painfully aware that we’ve totally hijacked the original post topic. I’d have no problem if the moderators, Hoovorff or Messrs Byrne or Gallagher thought it good to split this up and rename the parts! Apologies, Hoovorff!
Terry
As a jointly and probably primarily guilty party, I too have had that in mind and second Terry’s post!
Bloody good this, though, innit? I’ve learned a huge amount on this thread. Thanks in particular to Terry. Now all I need is about 15 lathes … and some talent …
Nooooooo!!! Not until some-one invents an un-clumsy pill. 
Ask Micky! (sorry, below the belt that, but…)
Besides, you need all your fingers - keep 'em for what they’re already good at!
That’s fine, Terry! I’ve enjoyed the discussions.