Improving the Upright Piano Action: A Mechanical Engineering Project

The STEminists
Nov 2 · 5 min read
Photograph by Andrik Langfield

During the spring semester of 2019, we were tasked with choosing a project in which we could apply the engineering knowledge we had gathered during our college years. We called ourselves the STEMinists, chose redesigning the upright piano as our project, and finished our final design back in April.

For a detailed explanation about what we are trying to achieve and the evolution that led to our final prototype, check our previous blogs:

Part 1

Part 2

Part 3

Part 4

For the last blog update, we are happy to share that the team finished building the first key of the final model back in October. Initially, the team planned on building an octave of keys based on our design, but it wasn’t feasible to do because we had to redirect our focus back to delivering the best possible design. After we finished assembling the first model, we finally had the chance of tinkering with it and run validation tests.

That’s us, except that we did for a grade and the technicians did it for free… Cartoon by Mike Baldwin

After consultations with the piano technicians at Moore’s School of Music, the team realized that we only needed a single key to run our validate our design, as the evenness is already regulated by drilling lead weights into the key. Therefore, the team has decided to create a duplet of keys to at least demonstrate that we can reliably replicate our design and results.The team has started cutting the parts of maple wood that will be used to build the second piano action and preparing them for assembly. The team has also started validation testing on the first action along and getting feedback from the technicians on the responsiveness of the key.

Upon simple inspection, it became clear that our action was significantly heavier than the standard upright action and that makes it feel sluggish. The jack might reset faster into place, but that improvement is not perceived if it takes more effort to press the key a single time. We reduced the size of all the action parts in the repetition mechanism in an attempt to fit the modified action in an actual upright piano. We used maple instead of basswood to build the device, as the former is considerably stronger and less prone to fractures; however, the density of maple (620–750 kg/m^3) can be double than that of basswood (320–590 kg/m^3). So for a slight decrease in volume of these parts, the repetition mechanism made from maple would still outweigh the one made from basswood.To compensate for this increased weight, the team removed two lead weights in the front end of the key to make the action lighter that were placed to regulate the evenness of an actual upright action. When we tested the evenness, we discovered the force required to push our key exceeded 165 grams, as opposed to the preferred 50 grams required to press an upright key. Other than the density of the materials, we also realize that we are adding friction points and another torsion spring that opposes the key movement in order to make the action reset faster.

The team ordered calibrated weights to put on the front part of the key in an attempt to make the required force to press it closer to 50 grams. Furthermore, we will also apply graphite lubricant in the friction points to make it run smoother. However, if adding the counterweights affects the functionality of our key, we might need to accept the heavier weight as an idiosyncratic characteristic of our action. If this counterweights work we will also succeed at improving the method by which these keys are regulated, as regulating the counterweight without having to drill and possibly damage the keys, should be considered an improvement.

Grand piano action re-purposed to fit an upright piano, by the STEMinists with a table band-saw

The team also struggled to insert the bushings (little pieces of felt) inside the pinholes over the last week. None of the member of the team have much experience with tools and techniques (pins, felts and fine motoring skills!!!) of this craft. The technicians did the bushings for the first action; however, we were tasked with doing them for the second one. The team has gotten better at this after a thorough demonstration by the technician and has gained some valuable insight into the process, but inserting bushings is still one of the most challenging of the processes used in action building. Maybe we are biased after a week of failed bushings, but at least we managed to properly do three bushings on our own and are capable of repairing them just in case. Moreover, they gave us permission to use their band-saw and press drill after watching the them operate this machines and watching a safety video.

A duplet of STEMinists

The plan for the work period of November 1 — November 15 is to continue performing validation testing on the device and building one more action. All the pieces for the last action have been made and are ready for assembly. At this point, the team is also simultaneously testing and adjusting the device. The piano technicians have surveyed our results and offered their advice throughout this process, and will kindly continue to do so until the culmination of this project. Despite the current and anticipated challenges, the team is still confident in finishing the overall product on time.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade