simple windway cutter

[Andy Barnhart]

I came up with a simple way to cut an internal duct that is consistent and repeatable. On top of that, it actually works.
The test whistle sounded good on the very first attempt. I just drilled a hole at the end of the windway squared it up, filed the ramp, put in a plug and it worked immediately.

Anyway, I used a router with a 3/8" straight bit inside a 1/2" template bushing. I know that means it isn't truly concentric, but a little sanding could fix that. Having a straight clean roof at the right height to start with gets me past a major stumbling block. In its current form, it only works on tubes with a consistent diameter and wall thickness. Actually, changing the wall thickness would change the windway height. I am also using the router in a jig on my lathe to get consistent diameter.

Fairly self explanatory:










After taking the picture of the end of the whistle I realized I needed to clean it up after cutting the beak. The interior is smooth as glass.

[Andy Barnhart]

I just cut a few more and it occurred to me that I should mention something that is second nature if you sue router templates much but might not be known to some...

set the router in turned off, wiggle it around a little to make sure the bit is in the hollow of the bore when first turned on. Turn it on, push it forward the tiny amount it moves, pull it back, turn it off and wait until the router has stopped completely before lifting it out.

[Stephen Bacon]

Interesting. I know the relationship of radius of the roof to the edge and bore has an effect on the tone and response, but I have never worked with such a short radius.

[Andy Barnhart]

Do you mean proportionately? In the sample above, the whistle bore is 14mm, or just under 5/8" so the windway radius is over 60%. I know optimal is over 100% but it still has to have "corners" of a smaller radius.

[Stephen Bacon]

I think in time you will discover that the best windway curvature is less than the bore. Curvature matching the bore will in time lead to problems. I use a number seven gouge as a reference to what I find in better historic instruments. I think your current method will lead to moisture build up. If you added a little work with a flat file to create sides to your channel and left your block and edge a bit more in relation to the bore you will have better success with this new system.

[Andy Barnhart]

What I found attractive about the idea is that before I was starting with nothing (no windway, no sound) and trying to get a good sound. Using an approach like this, I am starting with something and possibly improving it. I say possibly because the sample whistle above actually sounds very good and I am afraid to do much to it. I did 4 more yesterday using the jig and found it is not magic; they still require some fiddling.

Your comment is interesting. I though that a concentric arc at the top of the windway was optimal (not from experience, but from reading). What I have is an arc smaller than the bore - the windway was made with a 3/8" diameter bit in the top of a (roughly) 5/8" bore. So the radius of the windway arc at the top (all the way around, actually) is about 60% of the bore arc.

What width #7 ?

[Stephen Bacon]

Any width #7 , it is an interesting system. In your case think perhaps the curve of a 3/4" dowel or larger, not less. Positive results are always encouraged , not matter how it is done. Yes, concentric is wonderful , but I noticed Fredrick Morgan put a dome in the center of the arch of just the roof with excellent results.
Your math might be correct but your use of it in this instance is in reverse, the 3/8 bit has a tighter arch( shorter radius) than the 5/8 bore. You want an ideal arch of a longer radius, than the bore.

[Andy Barnhart]

We are saying the same thing after all then - "You want an ideal arch of a longer radius, than the bore." When I referred to thinking that optimum was over 100%, that is what I meant - the roof arc having a radius at the top that is more than 100% (longer) than the bore arc. But if the roof arc is larger, then it has to have "corners" to make it end unless you mean having a different bore size in the head.

I am confused when you say any width #7. Below is a Pfiel chart. The arcs of the smaller sizes are tighter.

[Stephen Bacon]

don't worry bout the #7.

[Andy Barnhart]

Quick note in case anyone follows these directions - I have made a new one based on this that I will likely post soon. It is larger, the notched board has been replaced with a glued up V made with 2 miter cut pieces on a backing and some tweaks were made so the cut is a little better in its intended purpose. I used a piece of ply slightly thicker than the 1/2" router template bushing needs so I could leave the bottom in when I first made the template and then cut it using the 3/8" bit. Then when I put the whistle body in, I can see exactly what will be cut.