Building a Xylophone

Xylophone Index Page

This project page is about building a xylophone. Here is a picture of the xylophone project that is described in gory detail in the posts below.

The 3.5 octave Honduras Rosewood xylophone that my son Jack and I built and tuned.
The 3.5 octave Honduras Rosewood xylophone that my son, Jack, and I built and tuned.

I made this main xylophone page to augment the many blog posts on this site, because I sometimes find that it is easy to get lost on blog sites when trying to step through a subject that is linear, and therefore might be better described by chapters. So the list below will hopefully help you to navigate through the xylophone project. The idea is that you can look at any “chapter” and then use the browser back button to return to this index. The first post, “Overview,” contains some photos and sound files.

May 2017

Lily just finished a paper and video describing the xylophone for her high school physics project. This is a nice introduction that you might enjoy. Check it out here.

The Index

  1. Overview
  2. Bar shapes and vibrating things
  3. Frequency ratios and some terminology
  4. Research on commercial instruments
  5. Analysis of xylophone recordings from the web
  6. Spectral analysis and performance plots
  7. Sonic results for commercial instruments
  8. Determining the bar shape
  9. Putting the math to work
  10. Cutting first wood
  11. Verifying the tuning curves
  12. Making a friend
  13. Rosewood
  14. A bit more math
  15. Shaping 44 bars
  16. Rough tuning 44 bars
  17. Drilling the holes
  18. Fine tuning
  19. Designing the frame
  20. Making the frame (part 1)
  21. Making the frame (part 2)
  22. Tuning the tubes
  23. Fabricating and mounting the tubes
  24. Making the legs – and wrapping it up
  25. Epilog – The Code

Again, these are just a series of links to the discrete posts, but you can think of them as chapters that may help you to navigate our xylophone adventure!

(Note: Jack made the animated gif at the top of the page using Blender, an open-source photo-realistic rendering application that he has been messing around with lately. Pretty cool!)

21 thoughts on “Building a Xylophone”

  1. This is a wonderful instrument and a great achievement! Thank you so much for having given all the details of the building, including your “ingenieur oriented” approach, with very clear and pedagogical explanations.

    I am a woodworker, I build telescopes as a living, I built also wooden clocks, a weaving loom, etc, and I want to build also a xylophone. Your explanations will be extremely useful for me. I have already downloaded the file for design of the archs!

    I will put the adress of you site on my blog.

    Again, thank you.

    Pierre
    France

    1. Hey Pierre – So glad that you found the information useful. Documentation isn’t my first love, so it’s good to hear that you have benefited from it! BTW, I checked out the video of your loom – super cool seeing that in action! Those telescopes are cool too – gives me some ideas :-).

      Good luck with your building efforts, and please feel free to contact me if you need further clarification.

  2. Hello Rich,

    I just finished my own xylophone!

    It is not so nice, big and professional as yours. I have not followed exactly all your advices, but it works well! The padouk bars give a wonderful sonority.

    The esthetic is also quite special and unseen, but I like the unusual vintage look…

    I have to say so many thanks to you, as you gave me the understanding that it was possible to make oneself such an instrument in its workshop, without knowing anything on it before…

    I am already thinking on my next one…

    http://dobsonfactory.blogspot.fr/2016/11/xylophone.html

    My very best regards,

    Pierre Desvaux
    France

    1. Hi Pierre –

      Fantastic! Your instrument is beautiful and unique. I really like the aesthetic. Glad my site was useful to you. We very much enjoy the xylophone that Jack and I made – he plays it frequently, and recently suggested that we write down the long-term “care and feeding” of it so that he can maintain it long after I am gone. I’m sure you will get many years of pleasure from yours too.

      Best, Rich

  3. Hello Rich,
    Hello Rich,
    We try to use Zhao code for metalophone (12 Aluminium bars) with my students.
    Do you have the missing “Drilled_Hole_Predictions”?

    We are building a robotic metallophone for a robotic orchestra.
    Look at the youtube link to see last years projects.

    Thank you

    1. Hello –

      It turns out that the Drilled_Hole_Predictions function did not exist. Dr Entwistle looked into that it it appeared to be a place-holder for a future function that was never written.

      In any case, I ended up using the “salt method” of determining the optimum hole locations for my bars, so you might do the same.

      BTW, I did a youtube search for “robotic metallophone” and got quite a few hits. Can you send a link to yours? I’d like to check it out!

      Best, Rich

  4. Hi rich,

    There is a Website field in leave a reply. I thought you can see it but maybe not.
    Here is our youtube channel:
    https://www.youtube.com/user/INSAgcp
    You can’t see any xylophone or orchestra yet on that channel because the project is not finished but last years projects. We are at the first step (mechanical design)of the xylophone. It will be finished (and published ) in June.
    We update the code for aluminium 50×10 mm² and automatic calculation for 18 bars. The raw bars are 14 semitones upper the required frequency.
    Drilled_Hole_Predictions was interesting for us because the bar manufacturing will be on a CNC milling machine. It could be done at the same time as the shape milling.
    Do you know if Drilled_Hole_Predictions was written by Dr Entwistle or Zhao?
    Do you have an idea of how to calculate the hole position?

    1. Hi Ben –

      Thanks for the link. I am eager to check out the finished project! Unfortunately, the Dr Entwistle suspects that the function never existed (he suspected that it was stubbed out to be populated later).

      Code could probably be written for the longer bars to compute the hole positions, but it would likely not work for the shorter bars, since they are not well modeled by the Timoshenko or Euler methods (at least that was my experience). If I were you, I would just use the “salt method” to find the nodes for the slightly sharp bars. When you do the final tuning, the nodes locations will change somewhat, but it is likely negligible. The salt method is fun and very accurate. I used it for all of my bars.

      Best of luck – please drop me a note if you get a chance when the project is cooked!

      Best, Rich

    1. Hey Ben –

      I checked out the video and blog – very cool! The bars really do sound fantastic.

      So did you have to do any fine tuning after milling the bars to the computed profile? I guess it would be a function of how accurate the milling process was.

      In any case, well done! (I’m sure I’d fit in well with your group of budding engineers…)

      Best, Rich

  5. What a cool project! I love the blend of math and woodworking. Unfortunately for me, I’m better at math than the woodworking part. I’ve been trying to talk my high school junior (and expert percussionist ) into building a marimba. We bought him a xylophone a couple of years ago, but thought it would be cool to mimic the instrument he uses at school without just writing a check. My idea has a slight twist that makes it a bit more challenging: I want to experiment with sandwiching two materials together to form a marimba bar. I’m hoping to use wood of a fixed thickness for the top layer, with the thicker ends and arched middle composed of a 3d printed material. The speed of sound will be different in the two materials just to name the most obvious issue, but it should be fun to play with. In my most optimistic dreams, I’ll find a solution that still sounds good and derive a mathematical model from the experimental data. I’ll then use that in OpenScad (a programmable 3d modeling program) to parametize marimba bars based on the desired frequency. Wish me luck!

    1. Hey Byron – You should definitely go for it! At worse, you will spend a bunch of time with your son and you will both learn a bunch. Interesting idea to use two materials, my gut is that you will have issues with “receptance” mismatch between the two materials, but it will be fun to see what you learn. Please keep me in the loop as you progress!

      Rich

  6. Hi Rich,

    I have been researching the physics behind marimba bars for a project. In my research I have learnt about the topic but I thought that if I want the best results I should contact a professional/expert. Your pages on Building a Xylophone increased my understanding of the marimba but also generated more questions too. If you have the time (and If this page is still active), I would very much appreciate if you could answer the following questions:

    I read about the tuning of the first 3 transverse modes of a marimba bar and that they are tuned to the fundamental, fourth harmonic and the tenth harmonic. Do the resonators amplify the 2nd or 3rd transverse mode’s tones?

    Can marimbas bars be made with other hardwoods (or any other wood at all) than rosewood? (I have read that it is possible but would “sound like a toy”(Rich Wickstrom, SuperMediocre))

    Rosewood has special musical qualities, what makes it so special in regards with physics?

    While researching, I’ve ran into multiple equations for the length of bars, which one do you use? (I would attach a photo with them but I think I am unable)

    When curving a marimba resonator to a J shape, how do you calculate the length regarding the curve?

    Thanks for your time
    Grant Williams

    1. Hi Grant – I haven’t logged in to this site in years so missed your questions. Not sure if you still care, but here are a few quick thoughts:
      1) From a physics standpoint, I seem to recall that the resonator tube will boost some of the harmonics, but to my ear I predominantly hear the fundamental being boosted. In any case, it definitely changes the timbre.
      2) Other woods can be used, but the characteristics of Honduras Rosewood are unique in providing the desired timbre and sustain.
      3) I seem to recall that the bar lengths were close to those of a commercial instrument (a Yamaha?) in order to keep the overall instrument size in family with common practice.
      4) I might think a starting point for J-shaped resonators would be to just be the “un-rolled” length (i.e., if you ran a string through the ‘J’ and then measured it). However, as noted in the blog, the physics are complex and the simple equations only provide a starting point. I might think this would be even more true for curved shapes.

      Best, Rich

  7. Hello Rich,

    First comment from me, but I’ve been a fan of your build for two or three years now. Your final product is awesome, and thank you so much for sharing all the details along the way!

    I’m a percussionist turned marimba/xylo acoustics nerd. You’d be suprised (or maybe you know by now) how many percussionists, even marimbists, don’t know much about the science of their own instrument.

    This year I’ve been researching treated bamboo as an alternative bar material. I can make rectangular bar blanks from bamboo, but I don’t have the tools to make finshed, tuned bars. If I were to send you some bar blanks, would you be interested in tuning them up to compare them to your rosewood keys? I can also send you my research and some journal articles if you’re interested in why I think this will work.

    Shoot me an email at elijahmize@gmail.com if you’re interested. Thanks!

    1. Sorry I missed this Elijah. I am curious about how your bamboo bars turned out. Did you build and measure some?

      Thanks, Rich

    1. Thanks for the comment Demetria! I’ve not don’t any instrument making in several years, but have build some other stuff that I intend to blog about.

      Best, Rich

  8. Congratulations on producing such a well researched and comprehensive DIY project. Many thanks to you for sharing the knowledge you gained in your project.
    Setting out to make a xylophone seems to be a rather daunting prospect for us lesser mortals.
    Once more, thank you for the lessons.

    1. Very kind of you Keith. Thanks for the complement! Kind of crazy what a regular person can do with a suitable dose of obsession!

      Best, Rich

  9. Good day. I can’t believe I came across your site. This is simply incredible information. My name is Anton, I come from Ukraine, my son is learning to play the xylophone and I decided to make him one. Thank you for everything you have done!!! Please help, I can’t find the program you use to adjust the sound, please give some link or name of this program.

Leave a Reply

Your email address will not be published.