Baroque Alto Recorder in F




Introduction: Baroque Alto Recorder in F

In this first instructables I write I'm covering all the details to print, post process and assemble this fully working alto recorder. The recorder can be printed in any standard RepRap printer or commercial home printer such as ultimaker or makerbot. As long as your max build height is greater than 190 mm you are fine (most RepRap printers can print up to 200 mm in height).

As for the recorder itself, it has a fairly good quality. It can produce tones for two full octaves and a little bit of a third. The fingering is the modern English fingering, also known as baroque fingering. In this page you can find the complete fingering chart for this recorder.

Go ahead download the thing and continue reading this instructables.

Step 1: You'll Need...

  • The thing from thingiverse (I recommend to download the source zip).
  • A small flat file, around 12 mm width on the tip is ideal.
  • A pair of pliers.
  • Sandpaper.
  • A calliper (optional).
  • Acetone (optional).
  • A 3D printer (duh) and ABS plastic.

You can try and use other type of plastic, but some steps in this process use acetone to weld and polish the plastic, so I recommend to use ABS. The parts are basically thin, tall cylinders so you shouldn't have any trouble with warping.

Optionally you'll also need the following software:

This software is necessary if you want to compile the source, which is necessary to print a calibrated recorder. You should be familiar with the use of that software, as this tutorial doesn't cover the details of their use.

Step 2: Print Settings

To slice the parts, I recommend to use a thin layer height, like 0.1 mm. It's rather slow, but you'll get the most precision you can from your printer. Some parts of the recorder, such as the labium, are delicate and will greatly benefit from the thin layer height.

As for other settings, I've used "normal" print speed (around 50 mm/s) and 30% infill, although I think those things don't really matter. Just use the settings you are comfortable with.

Pay attention to the joints. You want them to be as solid as possible. Some slicers (like slic3r) tend to leave gaps in thin walls like those, so experiment with the perimeter extrusion width to get the best results.

Don't scale the recorder. If you are wondering, you can't get a soprano recorder by just scaling it down to 66%. Not all measures shrink the same amount to get it tuned.

Step 3: Calibrate Your Printer

To achieve the most accurate tone the recorder needs precision sized holes. Most printers tend to print undersized holes, so it's a good idea to calibrate the size of the holes printed by your printer before printing the whole thing. To calibrate, you'll need to compile the sources, so you'll need OpenSCAD and Python.

If you know your printer prints accurate sized holes or don't bother about tone accuracy, you can go ahead and print the STL's from thingiverse directly. Otherwise follow the following steps to calibrate.

  1. Print the file "calibration.stl" using the slicing settings you'll use in the final print.
  2. Measure the small lateral hole and the inner bore. The hole should be 6 mm in diameter, and the inner bore 16 mm. The outer diameter is 25 mm. If you get less than that, you need to compensate. For example, if you get 5.70 mm for the small hole (like in the picture) you need a compensation of 0.3 mm.
  3. Uncompress the file "" in some place on your computer and open "barroca.scad" with OpenSCAD.
  4. Change the variable "internal_compensation" to whatever value you measured (in my example, 0.3).
  5. Uncomment the line with "calibration_piece()" and compile a new calibration piece.
  6. Print using the exact same slicer settings you used before.
  7. Measure again. You should now get correct (or close to correct) measurements.

If you are wondering what the "external_compensation" variable is, it is intended for "external" features, such as the width of a cylinder. If you measure more (or less) than 25 mm in the cylinder width of the calibration piece, you should change this value as well, although I've found most printers are accurate for exterior perimeters.

Once you got the new calibration piece printed, you need to make a final adjustment. Open the file and change the variable "internal_compensation" to the value you obtained before. Save and execute the script. It should generate a new "windway.scad" file. Now you can compile and print all the pieces with their measurements calibrated. Just uncomment "full_plate()" and you are ready to go!

Step 4: Print

Now you are ready to print the recorder. In my case it took roughly 13 hours to finish, so go get some snacks while you wait.

Step 5: Remove Support Bridges

You've probably noticed that the bore of the three main pieces are obstructed by one or two layers of plastic (depending on your slicer). Those bridge layers are there as support. Remove them using the pliers. Doesn't require much effort.

Step 6: Sharpen and Polish the Labium

The labium is the sharp edge where the air stream strikes, producing the sound. Ideally, the edge should be as sharp as possible, and the surface should be smooth, but the typical printer has a 0.4 mm extrusion tip and produces rough walls. You can see the edge is not sharp at all.

Before sharpening, do a quick test of the sound. Assemble the mouthpiece and blow while you cover the bore with your hand. As you blow harder, you'll notice that it squeaks very unpleasantly (check the video). This is what we want to eliminate by sharpening the labium.

Using the file, sharpen the edge and polish the roughness of the labium. try to file equally in the outside and inside of the bore to keep the edge centred.

Go slow and gentle, it is really thin and easy to break. Notice also that the edge is slightly curved outwards. Try to keep that curved shape as you file.

Get the edge as sharp as in the picture. Then, using a cloth wet in acetone, do a final polishing. The acetone will give the surface a nice shiny finish.

Do a test again. It shouldn't squeak like before (it should sound as in the video).

Step 7: Make the Windway Bevel

After polishing the labium, you can still hear kind of a warble in the sound test. This warble will make the recorder unable to reach the higher register notes, so we want to get rid of it!

Turns out you need to make a small bevel to the exit of the windway. I don't really know what's going on here, but it works. I also failed to get the bevel directly from the print, for some reason it just doesn't want to work if not done manually.

Using the square tip of the file, make a small bevel. Use the pictures and video as reference. The bevel should be around 0.5 to 1mm deep. Make sound tests periodically until you can't hear warbling any more (just as in the last video).

Step 8: Assemble the Fittings

The fittings (the two small cylinders) need to be glued to the main body. Probably you can't quite fit them into the grooves, so file them down until they fit tightly. Then, using a little bit of "ABS juice" (ABS plastic dissolved in acetone), weld the fitting into place. Be sure to fit the piece all the way to the bottom. The ABS juice sets really quickly, so be careful!

Step 9: Sand and Polish the Fittings

The fittings are intentionally slightly thicker than the bore where they fit. This is to give room to sand and polish them to get a tight seal.

Smooth the inner bores using a cloth wet on acetone. One or two passes will do.

Sand the fittings using the sand paper until they barely fit into the bore. Then smooth them down using acetone on a cloth. Use some kind of lubricant to fit the pieces together. I had some "recorder cream" from a Yamaha recorder, but vaseline works just fine. If it doesn't quite fit, just sand a little more and polish again.

Once this step is complete, you can assemble a play the recorder for the first time!

Step 10: (Optional) Weld the Mouthpiece

Optionally you can weld the mouthpiece to the upper segment using ABS juice. I designed the mouthpiece to be welded together, but I found that the recorder plays just find without welding, and it is easier to clean if you can remove the mouthpiece, and you can experiment with modification on the labium too. But, if you feel the joint is too loose, just weld and forget about it.

Step 11: You Are Done!

Enjoy your new alto recorder! Check the video to see it working once finished!

Be the First to Share


    • Reclaimed Materials Contest

      Reclaimed Materials Contest
    • Paint Challenge

      Paint Challenge
    • Laser Challenge

      Laser Challenge



    1 year ago on Step 11

    Been spendng a few days making this. Having real trouble with mouthpiece upper segment. I printed at 0.1mm layer height. Made a few prior. Have filed the blade/labium and yes I got a sound but as soon as I applied a bevel it stopped working.


    Question 2 years ago on Step 11

    Hello Aitor,
    thank you for you data!

    From which Alto recorder maker is this recorder original (Bressan?)?

    Is it possible to make a bigger or smaller recorder with the same
    data of the Alto recorder?

    E.g. making a Sopranino recorder with scaling the Alto recorder model x 0,5 or in the
    other direction making a Basset recorder in F with scaling Alto recorder model x 2).

    If it is not possible in this simple way, which parts of the windway can't be scaled
    in this way (an how can I get these data)?

    Aitor Esteve
    Aitor Esteve

    Answer 2 years ago

    Hello ZugpferdchenD,

    I did not know about Bressan recorders before you mentioned them so, no, it's not based on them.

    The profile is based on a technical drawing from a 18 century recorder made by "Johann Pörschmann". However, the critical dimensions (bore and holes) are not from that recorder, since it's an A415 instrument and I wanted it tuned to A440. The critical dimensions were taken from a modern yamaha recorder and tweaked a bit afterwards.

    And no, the recorder cannot be simply scaled down. The hole sizes, in particular, do not scale linearly. I don't know of a method to design a recorder from scratch, they are actually pretty complex. Historically I think they were designed mainly by trial and error. I think the mouthpiece might be linearly scalable, but I haven't tested it. The fastest way to design a different size wold be to buy a working recorder and copy the critical dimensions from it.

    EDIT: I wanted to add, I by no means an expert, I just found that drawing I mention on a google search and researched a bit online about recorder physics.


    Question 3 years ago on Introduction

    Hi Aitor, I have tested to print in PLA and look amazing. But, does not play well. After the flute was printed sound was quite OK, only when You cover the whole windhead, the lowest tone wasn't there, only the high overtone. When I brushed the windhead blade, the result tone became worse. Have You any idea, how to improve the tone, the make flute playable? Thanks for any advice. Ondrej

    Aitor Esteve
    Aitor Esteve

    Answer 3 years ago

    Hello Ondřej,

    Well, that's a little difficult to tell, but here are some advices:

    1- I assume that you are familiar with recorders, right? The lowest notes are actually a bit unstable if you blow too hard, they actually overblow (they sound high). Try to blow very gently.

    2- Have you done the chamfer, or bevel, at the exit of the windway (the slot you blow through)? I found this to be *very* important, you can't get a stable sound in the lowest and highest register if you don't do this properly.

    3- The labium edge (the blade you write about). The sharper, the better.

    4- Be sure that the windway is clean. Some people seem to have trouble getting the printer to print cleanly the mouthpiece. If this is your case, some slicers offer the option to make the printer avoid crossing perimeters, this should avoid any plastics strings inside the mouthpiece.

    I've attached a few detailed pictures I've just taken showing you the labium edge, the mouthpiece chamfer and the (clean) windway.



    Reply 3 years ago

    Hi Aitor,

    in first, thanks a lot for Your quick answer. I have noticed, Your kind reply little bit later.

    About Your 1st point, yes, i played recorder in past and know, that lower tones are harder to get, in this case, at the moment, I can get the lower tone even I am as decent blowing as I can ...

    At 2nd point, I think, that I made the bevel adjustment as You mentioned, I will rather attach the photo...

    At 3rd point, I was trying really to get it as sharp as possible (again I attach the photo)

    At 4th point, yes the windway is clean, I was trying to print the flute in thin layers, to have a good reasult.

    My feeling is, that maybee (but that is only my guess), the PLA material was shrinked a little more then ABS, and whole model is little (mybee around 1 milimiter) smaller, so each printed peace is mouted together very freely (without glue or welding part together would flute not be able to stand in one peace). And in light of this, maybee there was changed the angle, where the blowing wind goes on the blade. This guess is little supported by fact, that when I take wind way parts (they are not sticked together yet) and make by hand manually by feeling little more ange, then it starts making tone, and when I do it very, very carefully, in one position I can reach also the lowest tone either. But that is position, where is angle of those two parts aroud 12degrees aproximatelly.

    Any idea I will very welcome... Thanks for Your time and effort.


    Aitor Esteve
    Aitor Esteve

    Reply 3 years ago

    Sorry for the delay, I forgot to reply :P

    About the pictures, it's difficult to tell with the white plastic, but maybe the chamfer is not as deep as mine? If you blow the recorder with the body detached while covering the lower opening with your hand (as in the video) you should get a stable low tone. If the tone is instable or there is warble, then the chamfer is not deep enough. Refer to the second video in step 7, titled "Final test".

    I don't think it is the PLA shrinking, PLA shrinks less than ABS (that's the reason it warps less). And if that were the case, the recorder would just sound lower, but should sound fine. In any case, if I understand correctly, your pieces are not glued yet and they don't fit tightly, is that correct? Then maybe there are air leaks at the joints. You could try sealing the joints temporally using some kind of tape (maybe electrical tape) and see if you have more luck.


    5 years ago

    Can I print this with PLA?

    Aitor Esteve
    Aitor Esteve

    Reply 5 years ago

    Yes, but then you'll have to put it together using glue and polish it with sandpaper, as PLA is not soluble in acetone. It should work OK, but haven't tested it.


    6 years ago on Introduction

    This is awesome! I love recorders. Thanks for sharing how you made it.

    Aitor Esteve
    Aitor Esteve

    Reply 6 years ago on Introduction

    Hey! You are welcome. Thanks for commenting. And thanks for featuring my instructable!