Jump to content

Carving The Fipple In A Woodwind

Andy B

Recommended Posts

I struggle with this and it is probably the one part of making woodwinds which is the most relevant to this group as it is a very small detail that has to be carved just so. Actually, there is some room for variance that will still make sound and even produce the right notes but may have "breathy" sounds or too easily slip into the next register (on a whistle/flute) or require too much breath to change octaves, etc. There is a world of difference between a really good fipple and one that just works.


A little background...


What is a fipple?



That's an internally ducted fipple. There are also externally ducted fipples used on Native American inspired flutes, but that is a completely different subject. The diagram makes it look like there are a few different pieces but if you are making simple traditional woodwinds, there are only one or two - the "tube" and the "block". In rare cases, you can make the bore stop before going all the way through the instrument and make a windway from the other end meet it.


I am now getting into parts that are my understanding and subject to correction or other opinions...


You want the airflow to pretty much be split fairly evenly when it hits the edge. One problem with that is that the edge can't be any lower than the wall of the bore, so if you just plug the bore with room for the windway at the top, air will go below the edge. So this is solved by a couple of methods, often combined. One is to cut a ramp on the block. The idea is that the floor and roof will form a slight V and since only the floor is sloped the V will point up a little and direct air that way. The other is to cut a channel in the roof. Most manufactured instruments will have a rectangular channel and flatten the top top of the block, sometimes with a slight ramp. The trouble with a rectangle is that unless you also square off the edge of the fipple blade, it will be very difficult to accurately control how the air is split.


Another option is rounding the windway to match the natural curve of the edge. There area couple of ways to to this. What I have been doing is using a slightly larger bit to increase the bore right up to the window, then using a block that fits the larger bore and cutting a flat spot on the block. My windway is sort of a half moon shape. Another way to do it is to use a gouge of exactly the right size and cut a curved channel in the roof. Then with a block that fits the bore, you get a crescent moon shaped windway. The problems I have run into trying to do that are that I am working inside of a wooden tube with a bore size of 14mm (0.55") cutting end grain from the very end down in to about 1.5" - 2" and most stop before hitting the edge or I will destroy it. Even at that short distance, it is hard to get a tool to ride the bevel inside the tube.


I haven't even said much about the edge and window (open square in front of the edge). I need a really flat smooth cut at about 30 degrees and I need to get rid of any "fuzzies" around those cuts without rounding edges.


So I am looking for help and suggestions. If you do this sort of work with a high rate of success, I would be grateful if you would share tips. If you do similar work with wood in constrained spaces and feel pretty confident you know what tool would give you the best result, please offer your thoughts. I primarily use a 1/4" gouge sharpened at a low angle and a detail carving knife.


This is a page of someone making very similar flutes:


not a lot of detail, but he seems to use the same (or very similar) tools.

Link to comment
Share on other sites

Hi Andy,


This subject is fascinating! Thank you for the link, which is also very interesting.


I have no experience with what you are doing, but as you described how you carve the half-moon shape, you seem to be describing a standard style chisel, with the bevel on the convex side of the tool. I immediately wondered of a curved chisel, to the degree you need, was reground to have the convex side be the flat side with the bevel on the concave, interior of the curved chisel. I often use the flat side of the flat chisels for certain movements, so why not use a curved chisel similarly?


There may be a need for the tool to be the same width for the whole length that matches the depth of the cut you need to make. This is only my imaginings, not real or practical experience.



Link to comment
Share on other sites



That might work. I would have to be very careful about grain orientation so that I either have zero runout (my preference) or that it runs out on the bottom side of the whistle. That isn't optimal from an aesthetic point of view; while my preference is zero runout, if there is some it looks like a sunburst if you position it directly on top. What would be really nice is a scorp with a 1/2" OD and a reasonably thin shaft welded inside the ring. Another option I am considering is cutting a 2" long slit in a 1/2" dowel, slipping in sandpaper, putting the dowel on a chuck, slipping the whistle over the dowel, power up with downward pressure on the whistle and sand the channel.


I am hopeful that Yuri will respond to this thread. I am pretty sure he knows how to do this:


(that's from his site)

Link to comment
Share on other sites

Hi Andy,


The photo helped me to understand what you are trying to accomplish. Seeing the end-on view shows me that you are not trying to carve the airway through solid wood.


I checked, and Yuri has his email address in his Profile area, in case he does not check in here as soon as you would like.


Good luck with finding a good solution. I will have this on my mind for a while.



Link to comment
Share on other sites

I sent Yuri something to let him know I was interested and that the forum seems like a good place to discuss and share if he is so inclined.


No, the plug will be out while the airway is carved, but it is still like trying to sign your name inside a toilet paper roll with a regular pen. It's difficult even at the opening and just becomes progressively harder to control as you get deeper in. 2" doesn't seem that far, but when the width of the hole is .55" at its widest, the angle gets steep pretty fast. The most important place for precise shaping is right at the edge of the window - the deepest point.

Link to comment
Share on other sites

Is there such a tool, or could one be made, that could act like a plane with built in depth/stop guides to stop cutting at the correct depth, that cuts or peels a fine curl of wood as it slides straight in? Could there be a negative space contraption that fits into the hole that only allows the groove cutting tool to go straight in, and stops it at the precise depth?

Link to comment
Share on other sites

I am thinking of regrinding the bevel on one of my cheap Asian (there are also really good grades of Asian, but the tool I speak of is not of that class) gouges. I would grind it flat and then use a Dremel sanding drum to grind a bevel inside. I would just tape a short dowel on the shaft that prevented me from sinking it more than 2".


But I am patient. I have more wood to prep and other irons in the fire. I will think over my ideas and those of others before I move forward carving into more tubes. I really need to have a better plan for windways and fipples before I sacrifice more blanks trying to figure it out.

Link to comment
Share on other sites

Well, I have sent Andy a scan of a really good article about the subject. Can't really post it here, what with copyright stuff, but I still can add a bit of a description of how i do things.

The chisel idea for carving the windway top (into the upper surface of the tube0 doesn't work. A cutting edge is simply too powerful. There is no way you can control it to perfection. What I use (and every other maker, professional or amateur or mass-produced) is grinding of some sort. In a one-man workshop there can be many solutions. Old makers used custom-made files. I use something I described elsewhere in this forum, a piece of wood shaped like the windway's upper side, with different grades of sandpaper stuck to it using double-sided tape.


It's on the left. The file in the middle is used to file the channel into th top, after which the curved one takes over, to make that elegant curve. (By the way, old (or any other) recorders never use the complete curvature of the tube. Some modern whistles do, though. And then of course there are the cheaper ones, with flat edges and windways, you don't need to bother with the curve in those.) The thingy on the right is simply a needle file sharpened to a scraping edge, that I use to excavate the worst of the curve before doing the sanding in it. It is sharpened to about 80 degrees, not any sharper. This way it will not tend to dip into the wood. This is only to speed up the process, you can do it all with the flat file and the sanding block.


This shows the contraption I made to hold the head steady. Made from some scrap timber, a steel rod and is in turn held down to the bench by a couple of clamps. I think it is self-explanatory. The upper jaw is a floating one, not attched to anything. Both jaws are lined with leather. To accomodate different diamreters I simply use hardwood shims under the cams.

A couple more observations. The flat file to make the flat channel has its sides ground smooth, so it cuts only upwards, not sideways.

The first step is to cut the oblong window, a bit undersize in all directions. Then you start cutting the channel with the file, enough for it to reach and define the channel at the window. Then you open the window sideways, until the sides of it line up with the windchannel perfectly. Next is the full height of the windchannel, then the curving of it.

The next is carving the upper side of the edge (blade, fipple. The last word is something I never use, since no-one seems to agree just what it means.)You have to bring it down to a level about 1-1.5 mm below the upper edge of the windway. Bring it down more, and it will become increasingly breathy and unpleasant. After this you use the same sandpaper-covered block to cut the bottom part of the edge away, until you have a mere 0.1mm edge left. You do need to leave a tiny flat surface on the edge, I won't go into the physics of it, though. Put it simply- if you cut it to a knife edge, it won't work for very long, and very well.

Erm, the two photos got exchanged in the process, don't know why. Just imagine them the other way round.

Link to comment
Share on other sites

I just got on my computer to send Yuri a reply, but I guess I will follow up here instead. The information he sent me happens to exactly match the bore I am using, so I didn't have to try to extrapolate anything for a first attempt just follow directions (something I have never been particularly good at, but I gave it my best shot).


I had a bored tube a little longer than required. It is mulberry, not one of the woods listed (I warned you about me and directions) but as hard as most on the list and at hand. I measured in and marked off the rectangle and then put down masking tape around it and a tight fitting dowel in the bore. The rectangle is 10mm x 3.4mm and you have to cut that about 5mm deep (yes, this post belongs on this site :D ). The author of the article drills a few holes to remove much of the waste. He obviously pays more for his drill bits than I do; after drilling only 2, any attempts at additional holes deflected into one of the two already drilled.


So I switched over to a 1/4" chisel ground at a low angle, a detail carving knife, a small craft knife (moral equivalent of a scalpel) and a dental probe (for clearing chips). I mined wood until I hit the dowel.


To cut the ramp, I whittled a piece of hickory to fit the rectangle and then pulled up tape back to where I had marked the ramp start. I used the 1/4" chisel for this, occasionally running it down the sides but mostly cutting the ramp toward the window.


After that, I moved on to the windway. The instructions say to set in corners with the chisel first so I did that, leaving my sliver of hickory in the way, now protecting the ramp/edge. Once the corners were there, I used a split dowel to sand the top.


Next I took one of the hardwood dowels I turned to fit the bore and cut a ramp. Just an approximation to see if I can blow through cleanly and maybe get a hint of a bell tone to tell me I am on the right track. I didn't get a hint of bell tone - I got a clear full loud bell tone. I blew a little harder and got the octave. I blew harder still and got the double octave. I am not sure how much time I will have the next few days as I have family visiting, but I will make a proper block soon. Even without it, it is one of the best I have made so far in terms of tones produced on first test.

Link to comment
Share on other sites

I forgot to express my deep gratitude to Yuri for sharing this information.eusa_clap.gifnotworthy.gif


There are still more steps to the process; sub-millimeter amounts to be shaved off the corners off the block and window during voicing, but I have much higher confidence than I have had in previous attempts. Most of my previous "successful" whistles had a bit of trial and error involved and when I slid the bock in, it had to be at that "one spot". This one sounds the tone if the block (a quickly made temporary one at that) is just reasonably close. The right spot is the place where it sounds best, not the only place it sounds.

Link to comment
Share on other sites

A little postscript.

I have launched a "tour" of a couple of whistles, on chiffandfipple forum, that is the place for any whistle player to meet. Erm, turns out, when it comes to whistles, there are things I am not really quite in touch with. The above remarks about the windway/edge relate to old (that is Renaissance and Baroque) recorders, mostly (even though I do make a few more kinds of instruments in the same general type). Well, whistles do have different requirements. Since I do not play them, I asked the recipients of the tour to post their opinions, and two of them so far have. One thinks they are OK, the other does not, and both are very good players. So as far as whistles go, while the technology stands as I gave it, the whole voicing question is something I myself have to learn more about yet.

Just goes to show. It's not really about the technique when it comes to music. It's about the music, mostly.

Link to comment
Share on other sites

That is a good and well respected site. I have hopes to send a whistle or two to some members for comment one day. There are several members with large collections who order pretty much anything a respected member gives a good review. I don't get the slightly negative comment about the windway shape matching the arc of the bore unless it's just a case of expectations; maybe he is so used to the sound of a less focused air channel that a well focused one sounds funny.


One thing you do run into in the whistle world is a lot of metal, PVC and Delrin, some of the finest materials for making whistles that never lived. Due to the manufacturing precision and stability I don't know that they can be beat in a purely sonic evaluation. But when subjectivity is factored in, people gravitate to wood. You pick through 10 in the same key from the same maker because they are all different and one may speak to you the loudest. This is a generalization (and all generalizations are flawed :) ) but I think people who are all about the music tend toward metal and people who are all about the whistles tend toward wood.

Link to comment
Share on other sites



i found this to be very informative.




been there-----done this.


of everything mentioned here

the hand cut files esp the handcut riffler is a very useful tip.


something magical about the irregularity of the handcut tines that makes it smooth

the wood so much faster in preparation for a fine file.


also you may want to join

2 other groups----


and the



they have a bunch of good info. pictures and measures.

Link to comment
Share on other sites

I made the riffler from an old, worn-out file, using nothing more sophisticated than a Dremel with one of those cutting discs that come with it. Could use a diamond cutting disc, too. Just by eye, nothing fancy.

Link to comment
Share on other sites

please offer your thoughts.


The problems I have run into trying to do that are that I am working inside of a wooden tube with a bore size of 14mm (0.55") cutting end grain from the very end down in to about 1.5" - 2" and most stop before hitting the edge or I will destroy it. Even at that short distance, it is hard to get a tool to ride the bevel inside the tube.


some thoughts come to mind----first you could place a sacrificial wood dowel or wood backer inside when carving.


second is that the thinning of the wood could be done with a flat 01 steel ground to where the end was ground like a scraper---they call this tool a flats scraper and it is used in planemaking after the roughing with a float. to be truly effective, you could use some water on the endgrain, cause the grain to raise by hydroscopicity, then take a fine shaving cut or extremely fine scraper cut.



I need a really flat smooth cut at about 30 degrees and I need to get rid of any "fuzzies" around those cuts without rounding edges.


you can undercut the lip of the 30 degree slightly. but there are limits (1mm or so not more)


scraper would indeed get rid of the fuzzies.




that I am working inside of a wooden tube with a bore size of 14mm (0.55")------

-change the head shape of the recorder so its wider----whereby giving you more room to control the cuts.


the older makers did this by cheating using off axis turning around the bore.

a bore which is reamed from the end which makes it wider.


the bore is done first then the turning done so as to accomodate the bore.



the head stock shape can be significantly different than the bore shape and usually is,

the pleasing wood turned shape addresses this issue making it look straight when its not.


the air chamber is opened with very fine chisels.

Link to comment
Share on other sites

On some of my earliest whistles, I did take a slightly larger bit and enlarge the bore of the head so that I could use a dowel with a flat spot as the fipple. I did it by hand, literally holding the bit in one hand the whistle in the other. That made a whistling noise at the proper pitch pretty reliably and drilling the holes, they play recognizable songs. My kids and friends' kids love those whistles. But they are very "breathy" sounding and overblow easily and inconsistently. I tried a lot of variation on the window size and ramp angle and found lots of ways to make them unplayable, but not any to reliably improve them. There probably is one, but I lacked the specific instructions and didn't stumble on it.


Following the instructions for the method I assume Yuri uses (since he sent me the article) I got a clear tone (not breathy) that is consistent for overblowing for the next octave (easy to keep it from accidentally happening; reasonably easy to do it on purpose) on the first try. So while I am still monitoring this thread and open to some other suggestions, right now I am planning to make a few more following the explicit instructions with the hope that this problem has a solution and I just need to get better at it. Even though the first one worked, it is imperfect. The curve at the top of the ramp isn't centered perfectly; the sides of the ramp are not consistently 90 degrees, the windway roof isn't curved quite right at the beak, etc. Nit picky stuff that I will get better at by doing the same technique over and over, paying more attention to the parts that are rough in this one.


One of my favorite thoughts about building musical instruments well is from Samuel Zygmuntowicz, a renowned violin maker, who said as you get better at it over the years you begin to care more and more about less and less.

Link to comment
Share on other sites

I'm afraid you need to express your point a bit clearer.



one point is its easier to cut a ramp shape when you back it from the inside------the

backer minimizes breakout.


another point is scrapers can make a surface---even an endgrain------free of fuzzies.

scrapers work for this----wood that has the grain raised can be easier to scrape

and shape flat after its been raised---raise the grain first---let it dry then shape it.


big point: if you turn a fipple for a recorder you have a choice to use

the same axis as the bore-----or you could use a 2nd axis-----

that is one axis for the bore and another axis for the outside shaping on a lathe

--- this can place the bore itself closer or further from the players embouchure----and nearer or further from

the edge of the turning.


This is an artistic choice guided by physics. Wood turning on second

axis can also significantly change the relationship between the air passage and the vibrating part of the fipple.



what it does give you more room to cut the air passage and the plug which fits it.

this is why the mouth end of the recorder is often tapered and shaped

after the bore is cut.



(pic of a recorder fipple also)







post-2226-0-29714800-1303426967.png (on this recorder the bore is not in the center----off axis turned after initial woodturning)





Link to comment
Share on other sites

Andy, the whole windway/edge thing, about the top of the block being more-or-less aligned with the bottom of the edge is certainly right for recorders. Only very cheap school models have a windway pointing towards the edge in a way where the edge is about halfway up (or down). Whistles, however seem to have a preferred style that is exactly equivalent to the cheapest recorders' way. This obviously have to do with quite different playing techniques. That's one point where my experimental whistles have gone wrong.


Mibeck, what makes you think that the recorder plug in the photo is off-centre? Looks fairly centered to me. The thickness of the wood around the block is always unequal, but that's down to the taper on the outside. The idea itself is perfectly correct, though.


And just as an aside, I make my blocks/end of tube tapered, a slight cone.

Link to comment
Share on other sites

Mibeck, what makes you think that the recorder plug in the photo is off-centre?


because i measured it in a precise way!


i agree with the slight cone idea.



im sure its different for every instrument------especially those made by hand.


again---as always its comes down to artistic choice guided by physics.


There are probably an equal number of recorders that do it a different way.

ive seen some nice silver and brass fipple recorders that avoid the whole issue with

ring over a turned piece.

Link to comment
Share on other sites

Thanks for the additional info. I have seen some whistles with fairly steep ramps on the fipple block "aiming" the airflow at the edge. I knew it was done, but did not know it was preferred. The odd thing is that metal whistles don't seem to do that, largely out of practicality (IMO; I don't know that absolutely). On the metal ones, there is a consistent diameter and the blown edge is inside the bore. With wood, it is at the wall of the bore or even slightly outside it with a tiny bit of undercutting. So both approaches - dropping the edge into the airflow or aiming the airflow at the edge - seem to work well enough. What you (Yuri) ran into at C&F is the specific expectations of people who play ITM at more than a casual level.


I think that for the next few whistles I make, I will follow the instructions fairly explicitly and make sure that I am consistently getting playable instruments before I worry too much about getting the edge in just the right spot to please the elite players. I need to make some that impress a few people who are old enough to be allowed in the pub at all before I ask the better players to try them... B)

Link to comment
Share on other sites

BTW, Tyrone Head makes whistles that are sort of a hybrid - a silver fipple with a wooden tube and wooden blown edge. He details his process here (warning - if you haven't been there before, be prepared to waste an afternoon :) ) :


He uses a jig to form the ramp down the the edge and is really just using the small carving tools to clean up. His method is surgically precise, but I want to carve mine completely with hand tools.

Link to comment
Share on other sites

  • 4 months later...

Hi Andy,


My apologies if this reply is way out of date for where you are up to in your whistle making but I would like to add a few comments,


You cant cheat the physics behind musical instruments and the basics of achieving different tones has been well documented by Dr Guido. I assume you have read this paper http://www.ggwhistles.com/howto/index.html MIMF also has good resources. On the point on creating a stable change between the registers the subject that is least covered is the shaping of the upper and lower chamfers. Once you have cutting the fipple sorted it is in my experience the most important part of voicing. The reason is that it controls not only the direction of the airstream but also the turbulence of the stream as it hits the fipple. You can create all of the characteristics you want by controlling the lower chamfer. As an experiment try making a bunch of different blocks with no through to 1.5mm chamfers and you will see the difference that it makes. In my whistles I have no upper chamfer and about 0.7mm lower chamfer. My window is 11mm x 3mm with a 12mm bore for a D whistle. I make mostly from aluminium at the moment as I cant drill long straight holes :-(


Nice looking instrument on your website BTW. Maple? So little exists on voicing pre renaissance instruments that we are all guessing. I personally think its hard to go wrong.

Link to comment
Share on other sites

Nice looking instrument on your website BTW. Maple? So little exists on voicing pre renaissance instruments that we are all guessing. I personally think its hard to go wrong.


Thanks. It hasn't been updated in a while, but those are old in progress maple whistles.


Sometimes I wonder if we are doing worse that guessing, treating the few surviving instruments as examples of proper form. It could be the Tartu recorder didn't end up in a latrine by accident... :lol:

Link to comment
Share on other sites


This topic is now archived and is closed to further replies.

  • Create New...