Effect of thread wrapping on flute tenons


Discussion rages between proponents of thread wrapping and cork lapping on tenons in flutes.  Thread wrapping has been around since flutes (and other woodwinds) started to be made in multiple parts in the baroque period.  Cork lapping seems to date from France in the mid-19th century. Thread wrapping enthusiasts quote the simplicity of the method, its stability with changing weather or flute usage patterns, the ease with which non-technical people can make any adjustments as what attracts them.  They dislike cork because of a perceived risk to the socket if the cork is put on too thick, although this applies to thread equally if not more.  Cork enthusiasts (like the writer) feel they have overcome any corking problems, enjoy the method and the results they get, and are concerned that the tightness of thread can pose problems to the flute bore, especially over time. 

Over recent years the writer has received for repair two 19th century flutes, one in boxwood, the second in cocuswood, where too-tight thread wrapping does appear to have damaged the tenons, causing significant reduction in bore diameter.  Indeed, so significant, that the direction of the normal bore taper under the tenons was reversed!  And, when the thread was removed from the tenon, the tenon was distinctly "hour-glass" shaped, not cylindrical.  The word "strangled" came immediately to mind.  I'm defining strangulation as sufficient bore compression to cause a significant bottleneck in a bore that should have had a straightforward reducing taper.

Most other 19th century flutes also show signs of less extreme bore compression.  Reporting all this on the Chiff & Fipple flute forum led to a very lively debate as to whether or not thread tension could be enough to cause such damage, although no-one seemed able to suggest any other probable cause for the distortion.  Clearly, there was a need to investigate.

A short survey

I went looking among flutes I had easy access to for evidence of bore compression and strangulation.  I hoped to find flutes showing varying levels of damage, and, guess what?  Richness beyond the dreams of avarice!  The results are shown in this graph:

I've sorted them into provisional categories, and advanced definitions for each.

Seriously strangled flutes

These are flutes where bore compression is so advanced that a significant bottleneck forms under the tenon area, constricting operation of the flute.  The bore minimum under the tenon is more than 1mm less than the rest of the bore might suggest at that point.  The acoustic and aerodynamic impact on the performance of the flute is dramatic.

  • The anonymous boxwood strangled flute, in yellow.  Note the massive compression - a reduction in diameter of over 1.5mm - in the middle of the tenon area, stretching all the way in to 50mm along the flute.  The degree of compression reduces as you go further - this makes sense - it's much easier to compress the thin wood of the tenon than the relatively thick wood of the body.  But in the boxwood case, the process has continued well into the thicker body.

  • The cocuswood strangled flute, in pink.  Similar pattern to the boxwood, but milder in terms of both diameter and length, perhaps due to the greater strength and stability of cocuswood.

  • In aqua, a 1-key flute in stained boxwood by Schuchart (presumably John Just Schuchart, flourished London 1731-c1753), from the Bate Collection in Oxford, drawn by Australian Ken Williams in 1984.  Average of vertical and horizontal measurements (both show the same level of strangulation).  Nice to have an example on the public record predating this study.  I'm not making this stuff up!  Ken records the observation: "Tenon hollowed, matching socket is straight".  We know why it's hollowed, don't we....

Moderately strangled flutes

Bore compression under the tenon forms a mild bottleneck (less than 1mm) followed by a region with reversed taper.

  • A boxwood flute by Richard Potter, circa 1790, in orange, showing substantial strangulation, distortion going back to around 50mm.

  • An extremely famous original 18th century boxwood baroque flute, shown here in cobalt blue.  It's the GA Rottenburgh owned by Bart Kuijken.  Note the even larger bore of this early flute (from 40mm onwards), and that the compression also extends well past the tenon.  This flute has a number of corps de rechange (alternative left hand pieces used for different pitches).  Although wear patterns indicate that some of the corps were rarely used, they all showed bore compression, suggesting that the combination of thread and weather was enough to cause it.  I.E. a thread-wrapped flute doesn't have to be played for the damage to set in.

  • A boxwood flute by Bilton, in very light green, with the area under the thread wrap dipping below the region to follow.

  • My Nicholson's Improved flute, cocuswood, in sky blue, "just past the cusp of strangulation".  Note the larger bore on this flute, compared to most of the other 19th century flutes. 

Flutes "on or about the cusp of strangulation"

Bore compression has been enough to flatten out the taper, but not yet cause a bottleneck.  Nonetheless, the mixture of acoustic and aerodynamic disruption is likely to weaken flute response.

  • A cocuswood flute by Camp, showing bore compression a smidge beyond "the cusp of strangulation", in brown.  You'll notice that, in the Camp, the compression is limited to the tenon area, the first 25mm from the left.  It also starts earlier than the cocuswood flute - this might be because there is less of a tip shoulder before the threaded area starts.

  • A perfect case of taper negated, the very cusp of strangulation!  (The Camp had come close, but this is perfect.)  Check out the 20 to 25mm region of the Clinton 8-key in cocuswood, in light yellow.

Flutes showing bore compression, but not strangled

Such flutes will show compression to the bore, but not to the extent that the original taper is negated or reversed.  At this level, impact on performance should be slight.

  • My late Rudall Carte, cocuswood, shown in navy, with only mild compression.  By my definition, this flute is definitely not strangled - at no point is the bore taper negated or reversed.  But you can see there is a definite dip in the area of the tenon.  Now, very interestingly, this flute has an original wrap of a thick black thread, overlaid with some additional thinner white thread.  Keep that in mind when we get to the topic of serial strangulation...

Flutes with no sign of compression

Such flutes will show a straight or otherwise plausibly intentional taper.  They serve to remind us what general form the other flutes should take.

  • An anonymous Pratten-like flute with cork lapping and no sign of compression.  In grey.  Note the bridging taper over the first 6 mm, presumably intended to make a smoother transition from the bore of the head (19mm) and the main taper, which would otherwise have come out at around 18.3mm.  This may not have been an original feature, as the taper is quite crudely cut, unlike the rest of the bore which is very nicely cut.  But note no compression to the taper in the lapping zone X = 5 to 22mm.  Unfortunately for our study, English conical flutes with original cork lappings are rare.  It's unlikely that we'll come across enough to draw much in the way of conclusions.

  • And finally, the Rosetta Stone - a threaded flute that appears to have suffered no compression at all and shows no sign of any tampering.  Shown in pea green, it's my Geo Rudall, Willis Fecit - one of the flutes made by Willis for George Rudall in 1820, before Rudall teamed up with Rose.

Time to count heads

I could go on, adding flutes to the survey above, until I run out of flutes.  But the graph is already rather cluttered, so I think time to pause and do a tally.  Ignoring the cork-lapped flute, then, we have looked at 11 thread-lapped flutes.  Of these, and rounding the numbers:

  •  10% are undamaged

  •  30% are seriously strangled

  •  60% are strangled

  •  80% are at or beyond the cusp of strangulation, and

  •  90% show some compression

Ideally of course, we would have a much bigger sample size - that could be achieved by a trip around the museums such as I had done in 2002.  But even if a large sample halved the figures above, they would still tell a dismal story, and one we can hardly ignore.

Willis' Secret

Now, what was Willis' secret?  How come his 190 year old flute has survived unscathed, while earlier and later ones have suffered compression or even strangulation?  The answer is actually pretty obvious on inspection.  Willis' tenon is thicker than those on more recent flutes.  Indeed, it's more reminiscent of a baroque flute, although that didn't seem to have helped our boxwood baroque instrument! 

Unfortunately, the solution available to Willis is no longer attractive to us.  In Willis' time, flutes were generally much thicker all along, and bores were thinner.  The outside of the barrel at the base of his socket is 29.7mm, the inside diameter 23.5mm.  The tip of the tenon is 23mm, the inside at that point 18mm.  Since then, it's been realised that a thinner head plays more adroitly, and bigger bores give more power.  As the flutes became thinner (all over, for more elegant appearance and comfort in holding), and the bores bigger, the wall strength was whittled down from both sides.  And as the strength of a piece of wood varies at approximately the square of its thickness, a small reduction in thickness brings a bigger reduction in strength. 

Willis had a second trick, and I suspect this has even more to do with his freedom from damage.  He also made the thread trough very shallow - so shallow that the current fairly thick thread has room for one layer only.   This thread is 0.4mm (estimated under the microscope) and, from its condition, is probably not original.  Less thread to cause the damage, more tenon wall to resist it.  Bravo, Mr Willis!

Indeed, here's an interesting observation.  Supposing you deepen Willis' thread trough just enough to accept a second layer of thread.  And you might have good reason to - as it stands there isn't much room there to accommodate any wood movement due to seasonal change.  The second layer of thread will double the constraining force on the tenon, and accommodating it will slightly weaken the tenon's strength to resist that force.  A third layer would treble the force, and weaken the tenon further.  The ratio of thread depth to timber left might be the critical element in Mr Willis' legacy.

This seems to be confirmed when we contrast the unscathed Willis with two of the worst affected flutes in the survey - the strangled boxwood and cocuswood flutes.  They both had very slim barrels, around 28mm, leaving little room for all that has to fit inside.  But the Schuchart from the Bate seems to have pretty juicy dimensions (34mm OD and 24.8 ID in the middle of the socket, around 22.5 and 17.2 for the tenon) and yet its tenon is dramatically crushed.  Confirming it may not be enough to consider wood strength alone, but necessary to look to the ratio of that and thread thickness.  The Schuchart's tenon is so distorted we might never be sure what thread thread depth was supposed to be.  But the current difference between the socket's 25mm ID and the tenon's 22.5mm OD suggests that a lot of thread would be needed to fill that gap.  I'd suggest serial strangulation might be the killer here, see the discussion on this topic later.

Measuring compression

I should mention that measuring compression is not altogether straightforward.  Most people measuring flutes, eg when preparing to make copies, use the regular T-gauge, in conjunction with the micrometer.  Set the T-gauge to say 18mm, and see how far down the bore it will go.  Then move on to 17.8mm, repeat etc.  But, as you can see from the graphs, once you pass the constriction, you need to start opening up the gauge again, not just go on to the next place it stops.

Follow the yellow Strangled Boxwood curve in the graph above.  Once you get down to 16.3mm, it skims along from 10 to 15mm, and then won't stop until somewhere near the middle of the section!  That's a dead giveaway that something is wrong.  Obviously, it won't be so noticeable on flutes with mild compression, but you'll still notice a bigger step than normal.  Follow the brown Camp trace, and note that there's only about a 10mm step across the cavity.  But you can still detect that.

So, the golden rule has to be, when you find a step that's bigger than any other step, it needs full investigation, not glossing over.  Open your T-gauge a step, introduce it on an angle to duck under the low overhang, straighten it up and note how far in it will go and how far back it will come.  Keep increasing the setting until you've mapped the whole chamber.

What effects can we expect?

We can expect two kinds of effects.  There will be an increase in aerodynamic losses, as the oscillating air column encounters an increased constriction in the middle of the flute.  So performance will be down on that basis.

But there will also be a marked acoustic effect which will evidence itself in two ways.  Tuning is going to be different to what the original maker expected, and, if the spacing between the octaves changes, that is likely to bring a weakening of performance, as the harmonic alignment will suffer back at the jet. 

I ran a few computer simulations to predict what we might expect to see.  The software is currently being developed by a group involving the Physics Department at the University of New South Wales, the Powerhouse Museum and myself.  It's not quite ready for use in computer modelling of flutes yet, but more than adequate for comparative analyses like this, where only small changes are involved.

You'll see three traces in the graph below:

  • The blue trace is the boxwood severely strangled flute we met at the top.  You'll see that the model predicts sharpening of the low octave, to a startling 24 cents around low A, and a flattening of the second octave, again by up to 24 cents around 2nd octave G#.  With those midrange octaves now being narrowed by up to 48 cents, we can certainly expect significant loss of performance, as well as distinctly odd tuning!

  • The yellow trace shows what a moderately strangled flute such as the Richard Potter will suffer.  Same general shape, but with a bit less than half the effect.  I'd still expect some performance loss from both acoustic and aerodynamic sources.

  • Finally, in pink, a flute that was on the very cusp of strangulation, i.e. the bore is sufficiently compressed to flatten the taper under the wrap, but not enough to actually cause a measurable bottleneck.  But enough to narrow the G# octave by about 14 cents.  That may or may not be enough to introduce acoustical misalignment losses - this would depend largely on luck.

I could model more of the flutes, but it would mostly serve to confuse the graph.  They can be expected to come out between the pink and blue traces.  A flute like my Rudall Carte that only shows mild compression would come out between the pink trace and the horizontal axis.  That's about as much uninvited change as I'd like to see.

We can draw much from these simulations:

  • Note the general shape of the tuning change is one we are already familiar with.  Most old flutes have LH notes that tend sharp in the low octave.  Some of that is attributable to other reasons, but it's certainly not something to be encouraged further!

  • There isn't that much between the Cusp and Moderately strangled flutes.  Once the bore has constricted enough to negate the taper, significant errors are already being introduced.  Which is logical enough.  The maker made the flute tapered to bring the octaves into line.  If part of the taper gets cancelled, so does that part of its beneficial effect.

  • Given that we found above that most old threaded flutes were strangled or beyond, we can now say that most old flutes no longer perform as the maker intended.

  • If we have an old flute that is at the cusp or moderately strangled, we have to ask ourselves what are we doing to prevent the problem advancing.

Just how strong is thread?

I took a length of the thread taken from the strangled cocuswood flute, supported it from a retort stand, tied the free end to a hook, and carefully added calibrated weights until the thread broke.  I got to 700 gms weight, or 1.5lb.  That's pretty strong for a thread with a diameter of about 0.1mm!

I measured the thread from the Richard Potter "somewhat strangled" flute.  This was very old thread, and no doubt weakened by time and compression.  It also broke at 700 grams weight, but unless we can find a modern equivalent, we don't know what it was capable of when young and fresh.  There were 5.4 metres of that thread on the top tenon.  It was squashed too flat to measure by normal mechanical means, but using a stage micrometer on the zoom microscope, I'd estimate it as varying between 0.3mm and 0.9mm, depending if I was looking on the flat or the side.  It had become a ribbon, rather than a thread.  If we took the average of 0.6, it's 6 times thicker than the modern sewing thread.

I also measured some "soft-looking" yellow hempy sort of thread I'd bought 30-odd years ago from a Scottish bagpipe shop, sold for the express purpose of wrapping tenons.  It measured about 0.3mm on the stage micrometer.  A single length of that broke at 2.7Kg (6lb).  Thread is strong!

Another useful observation I can make about these tests is that none of the threads seemed "elastic", in the bungee cord sense.  When the first calibrated weight went on, the thread straightened out.  As I added more weights, there was no sense of the thread getting longer, up to the point where it broke.  I can obviously measure this if needed, but I suspect the observation is enough.  A coil of a hundred or more turns of any of these threads will present a formidable barrier to expansion.

Calculating the forces

An informal attempt to calculate the possible maximum forces involved lead to numbers so great that some readers rejected them outright as preposterous!  I asked my friend Professor Neville Fletcher for a confirmation (reproduced below in italics).  Characteristically thorough, he responded with two ways of looking at it - the total force applied by the thread on the tenon, and the pressure the thread applies to the tenon.  You'll remember that pressure is force divided by area, and so takes into account the area covered by the wrap.

Total force applied to tenon

From what you write, you want to evaluate the total effective inward force acting on the tube. Suppose the string is tightened to 700 grams weight and that there are 13 layers each of 150 wraps. Now imagine that you slice along the tube on both sides and remove one half of all the threads, but that magically the thread stays in place and taut. To make this happen you have to put a weight of 700 grams on each end of each of semicircular wrap. This makes a total weight of 13 x 150 x 700 grams or 1365 kg on each end of the semicircle, making 2,730 kg altogether.

For US readers, this translates to 6020 pounds.

Pressure applied to the tenon

If the question you are answering is "What is the effective compressive pressure provided by the winding?" Then each turn has an inward force of T/R per unit length, where R is the radius of that turn, acting over a length 2.pi.R, giving a total inward force of 2.pi.T. (Note that this is independent of R.) If we have N turns in a single layer and M layers. then the total inward force is 2.pi.N.M.T. But this force is spread over a total area 4pi.N.R'.r where R' is the average radius of the winding and r is the radius of the thread, so the inwards force per unit area, or equivalently the inwards pressure, is (2.pi.N.M.T) / (2.pi.N.R'.r) = (M.T) / (R'.r). To evaluate this you need to decide whether to use SI units and get the answer in pascals. Doing it on the back of an envelope, I get about 2x10^8 Pa or about 2000 atmospheres.

For US readers, this translates to about 29,000 psi.

These are indeed frightening figures.  But keep in mind the following:

  • we based the thread tension on the breaking strain of the thread.  It's unlikely that anyone would wind that hard.  But even if they wound at one tenth that tension, one tenth of those forces is still a lot to apply permanently to a thin-walled tube of wood.

  • I'm told that this is no news to serious kite flyers.  It seems that, if you wind up the kite string onto a hollow plastic reel at the tension the kite is exerting, the reel can crack and collapse under the accumulated force!

  • The moment you apply anything like that pressure to the tenon, it will start to compress.  And will continue to compress until it takes off enough pressure that it can support the remainder of the force.  In other words, it will crush to reach equilibrium.  We will see that happen at the start of the experiment below.

  • A more realistic way to look at these figures is that they represent the maximum resistance to expansion that the threadband (even if relatively loosely wound on) could present before the thread would break.  As we'll see later, that is totally relevant.

  • The actual value of force applied to a real tenon is not really calculable, because of all the unknowns.  But it suffices us to know that it is more than enough to cause problems!

A Test Tenon

In order to test the possibility that a thread wrap could damage a tenon within a reasonably short time frame, I decided to make up an under-strength test tenon, wind it firmly and subject it to some rigorous climatic change.

It has to be noted that this is an experiment "in extremis".  The process I used is a form of artificial aging.  Such an experiment is designed to bring results in as short a timescale as possible.  It is therefore not designed to mirror reality accurately.  It will be enough for now to prove possibility, and to learn anything else we can from it.

As can be seen from the first column, Just Made, the bore of the tenon was 18mm, and the outside diameter 20mm.  There is an 18mm wide trough for thread lapping with a floor diameter 19.6mm, but of course we'll lose track of that once the tenon is thread-wrapped.  The full width of the tenon being 30mm, this left shoulders of around 6mm wide on both sides of the thread trough.

In the Just Wrapped column, we can see that a wrapping of 0.5mm depth was added, producing a wrap of 20.6mm diameter.  The thread I used was the first to come to hand, a left-over of a domestic sewing project.  The manufacturer advises: "Rasant - the functional polyester/cotton core spun thread.  Rasant is a functional sewing thread with a wide variety of uses. The perfect synthesis of polyester core and cotton covering makes Rasant outstandingly efficient, not only in the sewing process, but in the seam as well."

Because the test tenon was small and a bit fiddly to hold, the thread was wrapped on firmly, but not overly tight.  It was also not wound on as neatly as had been intended, which would have achieved a stronger wrap. 

Immediately, a small reduction of the bore under the wrap was noted.  This is to be expected, as mentioned above, as the tenon will be compressed until the force exerted by the  thread is balanced by the restorative force produced by the distorted wood.  The fact that we can measure compression immediately after the thread has gone on does however confirm that we are dealing with a significant force.  I left it in this condition overnight to make sure that balance has been reached before proceeding.

Effect of Weather

The next day, there was little change, so it was felt appropriate to start the "Effects of Weather" phase.  The tenon was subjected to an environment at 25% RH overnight.  Naturally, the timber shrunk, as can be seen in all curves.

The tenon was then subjected to a very humid atmosphere by suspending it in a closed plastic ice-cream container over, but clear of, a wet sponge.  My laboratory grade hygrometer reads an impressive 99.9% RH in there.  Note that, in After Humid1, all the wood swelled, excepting that bound by the thread, which continued to shrink slightly.

Subsequent drying and humidification cycles have presented an increasingly interesting pattern.  The bore under the wrap (pink) has continued to collapse in diameter.  But the ends of the tenon have started to increase in diameter, both outside and in.  This dramatically enhances the "hourglass" shape which was a very visible feature of the damaged flutes.  And it is consistent with a previously unexplained issue with the damaged cocus flute - all three tenons, even when their thread wrapping has been removed, jam noticeably as they enter their sockets.

I was concerned that I mightn't be giving the drying phase enough time for it to equilibrate, so, after the third drying, I gave it some extra time.  As you can see, not much more happened in the More Drying phase.  It was felt at this time, it was appropriate to let the tenon then equilibrate to local atmospheric conditions, the results of which you can see in the "After Airing" column.  Not shown there is the length, which has returned to the original 30mm.  It had increased to as much as 30.3mm in humidifying cycles, and reduced to 29.8 in drying.  The fact that it is back to normal suggests we have equilibrated.  This pattern was to be repeated throughout the tests.

Summarising the Effects of Weather phase, we can see that:

  • the diameter at the middle of the wrap on the outside (aqua) has dropped by 0.3mm

  • the outside diameter of the ends has increased by nearly 0.4mm,

  • the bore diameter under the wrap has dropped by 0.8mm,

  • the bore diameter at the ends has increased by just over 0.5mm

Explains a few bumps

Interesting now to look at the RC7174 (Rudall Carte, aqua) and R&R 5501 (Rudall & Rose, black) curves in the chart below.  Note the flares at the ends of the LH sections (at around X=210mm) and the RH sections (at around X=320mm).  Are they, as sometimes claimed, deliberate flares introduced by the maker, or are they the result of severe bore compression?  The location and amounts are consistent with bore compression.  Most of the flutes also seem to have signs of compression at the top ends (around X=15mm).

Effects of Playing phase

With clear trends in the weather established, I think it's now time to try emulating the effects of playing. 

I simulated an hour's practice, by lightly stuffing the tenon with a damp rag for an hour.  I felt it may not have any different effect from the humidifying we did in Effects of Weather, but it seemed possible that only humidifying from the inside could produce a different result. 

Certainly, the first hour "playing" illustrates that a wet bore is a much faster way of getting water into the tenon than a moist environment!  Not really surprising, I guess. 

In order to keep things moving, I popped it back in the dry environment for an hour or so, then let it air on the bench.  As you can see in the final After Airing, there hasn't been much change since the airing before playing.  Like most things, change is faster at first and slows as time progresses.  There probably would be more change if we kept this up, but I think we've reached the point where we can draw some useful conclusions.

How collapse manifests

The image below is of a slightly strangled flute, the (Richard) Potter mentioned at the top.  I've shown you one edge of the top tenon, with the rest of the LH section out to the right of image.  You'll see that the worst of the reduction in diameter occurred in the first 14mm (although the minimum in the bore was at 15mm).  But more surprising is that the width of the combing (the thread retention grooves) varies, being a regular 1.5mm, then a narrow 1mm one, followed by several at 2mm.

Note also the absence of shoulders at each end of the thread band.  They were a feature of later flutes.  Their absence might be of significance in the shape of the distortion.  Indeed, the more strangled cocuswood flute presented a little differently.  The bore trough was a more predictable hourglass shape, but there was still evidence of a change near the middle, with the land areas between the thread retention grooves normally flat-topped on the tip side, but sawtooth on the body side.

We're tempted to imagine that the tenon is made of a homogeneous stuff like plastic, that will behave in a linear, predictable sort of way. But in fact it's made of fibres, running lengthwise, made up of empty cells.  And the outer ones have been slashed across every mm or so by the thread retention grooves, not necessarily to the same depth. And with a tapered bore inside, leaving the tip end wall the thinnest.  Put a death grip on that, and why should it behave in a simple way? But short of slicing along the tenon and examining it on the scanning electron microscope, how can we find out what it is doing? 

Hmmmm, CAT scan?

Comparison of bore distortions

I thought it might be helpful now to compare the distortion we had introduced into the test tenon with the kinds of distortions I had found on the two "strangled" flutes.  Not directly comparable of course, as the flutes had tapered bores, while the test tenon had a cylindrical bore.  So I corrected the test tenon figures to introduce a taper, based on my guess as to what the cocuswood flute taper might have been originally.  So, in the graph below, we see:

  • The boxwood strangled flute, in yellow,

  • The cocuswood strangled flute, in pink, with my guess as to what it might have been originally in pink dashes,

  • The "corrected" data from the test tenon, in the purple trace.

You can see that the test tenon distortion lies between the two other well-strangled flute traces.  Even uncorrected, it pretty much followed the pink solid trace, whereas it should have been (if it hadn't been compressed) horizontal along the 18mm diameter line!

Note that, had our test tenon been part of a real flute body, the distortion from the middle of the tenon onwards (X between 15mm and 30mm) would have been less, as the more rigid body on that side would have helped resist the distortion.  Although we would have seen some compression in the body.  In order not to confuse the image, I've chosen to map only the first half of the tenon, which is analogous to the other flutes.  I would expect to see the purple trace following a path approximately halfway between the pink and yellow traces.

Not just the top tenon...

Now, I've been concentrating in this article on the top tenons of all these flutes, probably because it's that tenon that has the most profound effects.  A constriction at this point of the flute will effect all the notes, but in different directions and to differing degrees.  Constrictions in the lower tenons will mostly impact on notes there and lower.

But it needs to be remembered that most of these flutes are also constricted where the LH meets the right, and where the body meets the foot.  Compression there is actually easier to detect, as it forms a single point of inflection, rather than two.  There is also no chamber formed, until you plug in the mating section.

Serial Strangulation?

Now, let's just imagine that the test tenon on our make-believe flute had suffered the kind of distortions we've seen above.  It had started out with a thread wrap of 20.6mm, which would have been exactly what it needed for a nice snug fit in the socket.  But that wrap has now reduced to 20.25mm in diameter, 0.35mm too small for a snug wrap.  Indeed, well before this stage, the joint would be sliding about uncontrollably, and even leaking.  So what would we do about that?  Put on 0.35mm more thread of course, or perhaps even remove the old thread and start again.  Either way, we now have a tight grip on the tenon again, and an even stronger band of thread.  The game starts over!

Remember my Rudall Carte we met right at the top?  Black thread covered by some additional white thread.  Only mild compression so far, but clearly we're on the way...

Isn't it one of those cruel ironies that the more you look after your flute, lovingly changing or augmenting the lappings when it gets a bit loose, the more likely you are to kill it!  I'm reminded of Oscar Wilde's terrifying poem the Ballad of Reading Goal.  "Yet each man kills the thing he loves..."  Indeed, the very benefit touted for strung flutes, that seasonal variation can be taken up by the owner, might in fact be part of the strangulation cycle.

The Good Old Days?

Was it better in the Good Old Days?  Some have suggested that perhaps threads were softer then.  Glancing at the graph at the top though seems to dispel that hope.  The most damaged bores were flutes from the 18th and 19th centuries, and the baroque instruments in particular would not have seen much if any use until the Early Music Revival in the 1970's.  By then the Schuchart was in museum hands, yet it it one of the most damaged.  The Richard Potter is missing several keys, and has probably not been played for 100 years or more.  There was nothing soft about the thread I took off it.

Duplicated Distortion Doubled

Imagine this ghastly scenario.  A modern flute maker faithfully copies a period flute without making allowance for the thread-induced bore compression.  So the copy is also distorted.  But then the maker also uses thread to wrap the copy's tenons.  After some time, it starts to work its ugly magic, and the copy is now more distorted than the original!  I'm told it happens regularly in the early flute field.  Perhaps it happens in our field too?

Clearly, we must understand bore compression and take steps to guard against it.

What about re-reaming?

Some have suggested that a flute displaying signs of constriction should simply be re-reamed to remove the offending in-growing tumour.  A moment's thought though should ring warning bells.  When you ream off the protrusion into the bore, you're further weakening the tenon wall.  Carry on like that - putting more thread on the outside and reaming off the wood on the inside - and the two will finally meet.  Probably with a bang!

But wait, there's more....

What happens when you play a country song backwards?  According to the popular song, it reverses all the sad things that usually happen to you in country music.  Your truck engine springs back into life, your dog doesn't die, and even your wife comes back to you.  Powerful magic indeed!  If we reverse what we did to our strangled tenon, could we reverse some of the damage?  Can we use the tenon to work out possible cures for strangulation in real flutes?  We need to know, so...

I measured it once more, to ensure no further changes over the last few days.  Nothing worth reporting.  I then pulled the thread off.  For the record, it was 20 metres long.  (Easier to measure coming off than going on!).  That is substantially less than I had taken off the cocuswood strangled flute. 

I remeasured the tenon, no immediate change other than the bore under the thread increased in diameter to 17.45 as the remainder of the thread tension came off.  Originally 18mm, it's still pretty squashed. 

We can also now pick up the diameter of the bottom of the thread trough that we lost track of when it was wrapped (purple trace).  As we can now see, it had started at 19.6mm, been crushed down to 19mm.

The first test is to simply humidify it.  It seems too hopeful that it will obediently spring back into place, but let's see.  Back into the ice-cream container....

After rehumidifying overnight (humid 3 column), we can see that all the diameters have increased (as expected).  Indeed the crushed bore under the thread wrap (pink) and the crushed bottom of thread pack (purple) are now well over their original uncrushed sizes.  So, what will happen when they dry out?

Oooh, that's promising!  After the final airing, on 28 January, we can see that the bore under where the thread wrap went is back to pretty well normal, as is the outside diameter at the bottom of the thread trough.  So this does suggest that removing the thread wrap and then humidity cycling might be helpful in attempting to deal with a collapsed tenon.

Surprisingly, it's the free end(s) of the tenon that have ended up still wrong, and oversize at that.  Now who would have thought?  But I think I can see the basis of a way of dealing with that too.  If one were to dry the tenon first to make it all undersize, wrap some thread around the free end(s) to constrain them and then humidify and dry, one could probably get that back to size too.  If wrapping thread isn't the answer (might be too hard to predict the end size), then maybe installing a delrin ring which can be popped off or turned off later.

The proof of the pudding...

... is traditionally located in the eating.  So, can we really cure a real flute from the distortions of the past, by the same processes that got us there?  I resolved to find out.  I opted for the approach used by violin, guitar and harpsichord makers, suitably adapted for flutes, but there would be many approaches.  This method offers minimum water uptake and so speedy recovery. 

I turned up a brass plug following the bore shape I guessed at for the cocuswood strangled flute.  Holding the tenon in boiling water, and then pushing the heated plug into it produced steam, which infuses into and softens the wood.  Each time I did it, I could push the plug a little further, until it finally reached the point I wanted.  I let the wood cool and dry with the plug in place, to prevent it sneaking back when my back was turned.  (If you want to try this yourself on an affected flute, be careful to constrain the outside of the tip of the tenon from expanding too far, as the thin wood of the tenon is at considerable risk of splitting from the wedging action of the plug.  Paradoxically, wrapping some thread around the tip as mentioned above might be just the answer!)

You can see the result in this next graph, with the steamed cocuswood flute in green.  Comparing my previous guess (pink dashed) with what we have now, it looks like I might have overdone it at the tenon tip, but that's easily fixed if necessary.  But looking at the steeper bore taper down around the 60-70mm mark, I'm also wondering if this approach might not have reached far enough into the thicker wood just beyond the tenon, circa 30-40mm.  A longer, slower soak might do that.  But the flute has to be far better off now than with the constriction seen on the pink curve in its throat!  I decided to let its performance tell me if we're done yet.

The strangled cocuswood flute is now back together, and working very nicely.  It is impossible to say how important fixing the choke in the top tenon was, as I also found lots of other little things that warranted attention along the way.  This is normal with old flutes.  They might have one or two dead obvious things wrong with them, which is usually why they were sent in for attention.  But it's often the accumulated effect of a dozen or more little unnoticed problems that pull an old flute's performance down.  If you want to give it back its will to live, every one of them has to be detected and ruthlessly eradicated.  Bore strangulation is now a new one on the agenda.

Flute, heal thyself [Luke 4:23]

(See, it's true.  Even the devil can misquote scripture!) 

Our experience with the test tenon might bring us some hope for a self cure for strangled flutes.  Imagine you have a flute with bore compression.  You remove the thread-band tourniquet and replace it with something that won't restrict expansion, say cork.  You put the flute back in service, meaning it gets subjected to wide humidity cycling.  If our test tenon's recuperative experience is any guide, slowly the damage might be undone.  At least you know it won't be getting any worse!

Wrapping it up (get it?)

Puns aside, I don't think there's much point in attempting much more on the test tenon, as we're now getting into finer points, which is likely to depend on matters like how compressed a real tenon is, what timber it's made from, how thick it is, how long the damage was sustained over, how long since the damage was sustained, and so on.  But we can certainly thank our test tenon for confirming how this compression comes about, and offering us some hope that it can be reversed or at least ameliorated.

I'm very happy with the in-extremis test procedure adopted.  In just over a week, we were able to simulate what might have taken years under normal circumstances, and yet end up with entirely plausible results.  And discover an unexpected side effect.  And explore a possible road to recovery.  Rewarding!

And to apply our new-found knowledge successfully to a real-life patient and effect a cure. Priceless!

Conclusions so far....

To summarise, let me advance this analysis.  For tenon-wrapping (cork or thread) to work it must just firmly fill the void between the socket and the tenon.  Any less and it will leak or not hold the flute together satisfactorily.  Any more and it will put unacceptable outward pressure on the thin wood of the socket.  But the tenon wood is going to get wet during playing, and will want to expand.  If the tenon wrap is thick-enough cork, it will have that room, provided by the resiliency of the cork.  But if it is thread it will not, and the wood of the tenon will be compressed.  Even if the thread wrapping is applied loosely, it will be compacted by the same expansion-contraction wetting-drying cycle, until more thread is needed to secure and seal.  Once enough thread is applied, the pressure goes onto the wood.

This article started out titled "effects of extreme thread wrapping...", but I have taken out the word extreme.  Most of the old thread-wrapped flutes surveyed (indeed, all but one) showed considerable bore compression or strangulation.  The test tenon was wrapped in a common polyester-cotton sewing thread, with considerably fewer turns than I had taken off the strangled cocuswood flute. The only extremes involved were the humidities of 25% and close to 100%, applied in quick alternation to accelerate the passing of time.  For most of us, these are not uncommon conditions.

Perhaps it's appropriate to ask ourselves why, so long ago, makers of Boehm flutes, clarinets, and oboes shifted to cork, leaving only conical flutes string-wrapped.  The simple, if painful, answer is that conical flutes were the cheap end of the market, and string is cheap.

It is now safe to conclude that it is clearly possible, indeed almost inevitable, to damage a tenon by wrapping it in thread, and subsequently exposing it to the rigours of weather and playing.  Indeed, the question now becomes what can we do to ensure such damage doesn't happen?  Unfortunately, the rigorous and numerous experiments needed to prove what is safe using multiple layers of thread might take dozens or even hundreds of tenons, and years, tens of years or even more!  It may be misdirected energy to conduct such tests.  Indeed, maybe we should be putting our energies in identifying newer, more appropriate materials and methods than either cork or thread.  

What, me worry?

So, having read all this, and having a flute that is thread wrapped, should you worry that strangulation should happen to you?  Unfortunately, although our experience shows that flutes get strangled, and our experiment shows a mechanism by which it can happen, we can't yet predict whether and when it's going to happen to your flute. 

But here's one circumstance that would definitely ring alarm bells.  If you find, from time to time, you have to add thread, but you don't find you ever have to remove any, I'd be worried.  Unfortunately, it's probably a bit late to be worried, as it probably means serial strangulation has already set in.  But the sooner you stop making things worse, the better.

I'd be more worried about the softer and less stable boxwood than the stronger and more stable blackwood.  But having said that, it's only because I have no experience of blackwood compressing, and that's because I have no experience of threaded blackwood flutes. 

Cocus, from our survey above, is obviously at risk.  The older the flute, the more likely the risk.  The thinner the tenon, the more likely the risk.  The deeper the thread pack, the more likely the risk.

I'd keep up the oiling, as this should slow water intake.  I'd be very thorough about mopping out, as that at least terminates the water uptake period.  I'd definitely unwrap any flute with a tight binding, and look for something softer I could apply more loosely.  And while I had the wrapping off, I'd look to see if the thread trough is uniform in diameter or hourglass shape.  Holding a rule against the tenon shoulders and looking at the size of the gap along the trough is a reasonably sensitive test.  if in doubt, I'd measure and graph the bore for the first 50mm (2") or so at each tenon, and look for signs of compression. 

If you don't feel motivated to go that far, at least investigate the top few layers.  If you found more than two or three separate windings were involved on any one tenon, I'd be worried that serial strangulation might be happening.  If that were the case, I'd definitely check out the tenon below using the rule method.

A single point check

Although to really understand what's going on, you need to measure and graph the bore for at least the first 50mm, our survey at the top suggests one pretty easy "single point check".  The typical 150mm (6") vernier callipers can reach about 16mm (5/8") down the bore if you use the internal anvils (the pointy ones).  Larger callipers have longer anvils, but often only the first 16mm or so are chamfered to prevent interference, so they can only read accurately to the same depth.  Fortunately it's enough to take us to the middle of the tenon, where the shrinkage is likely to be mostly felt.  If you haven't any callipers, you probably know someone who has, even if it means a trip to your motor repair shop.  Be gentle in taking the measurement, you don't want to rough up the nice bore on your flute with the sharp pointy ends! 

So the question becomes what would we expect to see at 16mm down the bore.  That's obviously going to depend on the flute type, but we could hazard a few guesses.  A small-bore flute, if we take the Geo Rudall, Willis fecit as an example, should be around 17.75mm.  A small bored Rudall style flute more like 17.9mm.  A larger bored Rudall or Nicholson's Improved say 18.1mm, and a Prattens around 18.3mm.  If you compare your measurement with the flutes on the graph at the top, you should be able to form a good idea of what condition it's in.

Checking the lower tenons on either the LH or RH section is much easier.  Measure just inside the end of the tenon, and compare it with the measurement you get when you insert the callipers to full depth.  The end measurement should be smaller.  If the full depth measurement is similar or smaller, then you have clear evidence of bore compression.

There's one case where that might be misleading though.  If the foot on your flute flares dramatically (eg on a flute with a Short D foot), the back reaming might have been taken past the foot and into the lower end of the RH section.  In which case you might legitimately see a very small reduction in diameter when you insert the callipers fully.  But if it's a large reduction, it's bore compression.

Getting it corked

You might decide, on the evidence above, that it's time to get your pride and joy corked.  In theory, that's a straightforward operation.  Any woodwind repairer who handles clarinets (and that's probably all of them) is accustomed to replacing cork.  But there are some significant differences between old wooden flutes and clarinets.  Clarinets were made with cork in mind, so the trough depth is ready-to-go.  Clarinets have smaller bores than the top end of flutes, but similar outside diameters, so the wall thickness is much greater, lending much greater strength.  Old wooden flutes were usually designed with thread in mind, so the trough depth could be anywhere.  And if bore compression has set in, it may well be deeper in the middle than at the ends.  Putting cork on tenons made for thread is likely to put too much force on the thin socket walls, unless the tenon and/or socket dimensions are adjusted to accept the cork safely.  That's a job for a wooden flute specialist, although once done, any woodwind repairer should be able to replace worn or damaged cork in future.  An alternative process is to put on the cork, then sand it to fit the socket nicely. 

It would be a shame to compound your tenon compression issue with a split socket issue!  So, ask around and find someone specialising in wooden flutes and cork.  Or at least alert your woodwind repairer to the issue and ask them how they intend to deal with it.

Future work

Perhaps the most surprising discovery was that the free ends of the test tenon actually swelled in size, probably explaining why the three tenons on the strangled cocuswood flute jam on entry to their sockets.  This means that we may be misinterpreting what the bores of old flutes originally looked like, and a more realistic test might be warranted to help guide our interpretations.  This test would still take some time, because the thicker tenon attached to a real flute body would have much longer reaction time, but at least such a test would be achievable.  Even so, it would still have to employ artificial aging techniques, or the results may not be available within a useful timeframe!


Thanks to all those at Chiff & Fipple who took part in the lively debate, including the sceptics!  It was useful to conduct this experiment in the light of public scrutiny, as it alerted me to the issues people found hard to accept.  Hopefully I've now answered them!

As with any topic, there are the deniers who will never agree with any proposition, no matter how well proven.  Some of these are just downright argumentative by nature.  They deserve our sympathy.  Others clearly feel that their business interests are being threatened.  If they put their business interests before the interests of their customers, they cannot expect our sympathy.  Others quibble about details of the experimental approach.  They would have criticised poor Fleming - he didn't predict penicillin after all, he merely discovered it.  It is clearly possible for anyone sufficiently motivated to repeat or improve on the experiment conducted here.  I'll happily advise them.  Anyone not sufficiently motivated is not worth listening to. 

To all of them, we simply pose the question: "If the thread didn't squash these tenons, what did?"  Aliens perhaps?  Come up with a plausible alternative scenario and we'll sit up and take notice.

Thanks to Neville Fletcher for clarifying the calculation of forces involved.  Thanks also to a number of individuals from other flute email lists who have chipped in with the benefit of their specialist knowledge.  I'll decline from naming any as I may forget some, but you know who you are!

Discussion of issues raised.

Discussions on Chiff & Fipple about earlier versions of this article became so heated the moderators felt obliged to lock the original and related threads a number of times.  After discussion, we have agreed to try a "For Information Only" approach, in which I can announce any developments, but further discussion is discouraged.  We'll see how that goes.

I'm happy to hear from anyone directly, and public discussion is available on any of these Internet fora:

  • woodenflute, primarily Irish flute players
  • flutemakers, a discussion group for makers of all types of flutes
  • earlyflute, a discussion group aimed at early music flute players.


Back to McGee-flutes Index page...

  Created: 23 January 2011; last updated 20 February 2011.