Group 1 2012 Gate 5
In this gate, we finalized our findings and reviewed our process and presented them in a technical report. In addition to a written report about the project, we also presented a proposed revision discussed in Gate 4 to the class and instructors.
Our product was a 1962 upright spinet piano made by Starr Pianos. Typical of most upright pianos, it has 88 keys and two pedals, which are the only interfaces available for user interaction. Disassembling the piano reveals that it has three major components, the action, the frame, and the housing that holds the action and the frame. The disassembly and reassembly of the piano was actually much easier than initially expected, as almost all of the piano is held together with screws or glue, and we opted not to break apart any of the adhered components. We found that the housing is a simple construct of screw fastened and glued together wood, has no moving parts, and is relatively mechanically uninteresting. The frame is composed of cast iron and holds the metal “strings” of the piano at high enough tensions for them to oscillate and produce tones. While the frame contains no moving parts of the traditional sense, the harmonic nature of the strings provides interesting analytic opportunities. We found that action is the only major subsystem that has a significant number of moving parts. The component that performs its basic function is a series of levers that simulate the motion of a four bar linkage in such a way that translates the depression of a key into the motion of a hammer that strikes the strings. The action in its entirety is an array of eighty eight of these base components on a specially carved wooden bar. Over the course of the project as we researched the functionality of the piano, we discovered that pianos have a rather strict set of performance requirements that virtually necessitate the existing hardware implementation. As a result of this, the majority of our proposed revisions deal with improving upon materials, or adding functionality that does not previously exist.
Product Analysis Summary
The analysis process for the piano primarily focused on a dissection and reverse-engineering of components and systems. The following is an overview of the main stages of analysis we performed.
Dissection and Reconstruction
The parameters of the analysis project required us to disassemble the piano down to its base components, and later reassemble the piano back to its original working functionality. In general, both disassembly and reassembly were completed quickly and without major setbacks. We were somewhat surprised to see that virtually all of the piano was held together with Phillips head screws and glue. This made the disassembly and reassembly considerably easier than initially anticipated. Due to the extreme weight and relative fragility (it is made of cast iron) of the frame, we opted not to remove it from the housing of the piano. Due to safety concerns over high tension wires snapping and the fear of breaking any of the strings, we also opted not to remove any from the frame. These decisions greatly sped up the disassembly and reassembly processes.
The following is an overview of the disassembly process. For an in-depth review, see Gate 2: Product Dissection. ever cannot be removed from the action without first removing a select few mutes from the action to expose its fastening screws.
- Remove screw-fastened pieces of the housing(fig 1+2).
- Remove pedal assemblies, pedal bars, and linking dowels(Fig 3).
- Detach keys from action pull rods and then remove(fig 4).
- Remove action from housing.
- Remove action end brackets from action.
- Remove hammer springs, hammer rest, and mute lever from action(fig 5).
- Remove mutes and action linkage/hammer sub assemblies.
When dissecting the action, we found it extremely helpful to rest it on an elevated pedestal to allow access from as many angles of the action at once as possible. It is also important to note that several of these general steps can or must occur concurrently. For instance, the pedals are connected to the action by two easily removed dowels, their removal has almost no impact on the removal of the action. Conversely, the mute lever cannot be removed without first removing several select mutes to allow access to its fastening screws.
The reassembly was a relatively simple process, and proceeded almost exactly in reverse of the deconstruction process. Since group members already understood the subsystem layout and interaction, the rebuild went smoothly and quickly. Only two group members were needed for the reassembly. The reconstruction is covered more thoroughly in Gate 4.
The first significant deviation from the disassembly process was in the insertion of the action into the housing. Having not previously put it back into the housing, we were surprised at the amount of force and shimmying it took to properly seat the action back in the housing. Care also had to be taken to ensure that the pull rod for each hammer was attached to the correct key, and that the pedals were correctly adjusted so as not to move the mutes or hammers too close to the piano.
The following is a time lapse of the reassembly. File:PianoReassembly.wmv
This summarizes various parts of the analysis process that were covered in previous gates. The analysis took the forms of different models.
From a conceptual standpoint, we analyzed subsystem interaction and its affect on user interaction. For the detailed analysis, see Gate 2: Subsystems.
Another form of analysis was a component analysis. Several components were solid-modeled and analyzed for manufacturing techniques and form. The strings were then analyzed for ways in which sound quality could be improved by material choice, dimensions, and tension. See Gate 3 for solid modeling and Engineering Analysis.
The final form of analysis was mathematical. The modified four-bar linkage in the action follows known mathematical models that govern input to output motion. See Gate 4: Mechanisms.
Our analysis revealed that the upright piano can be generalized as consisting of three main components: the cast iron frame that holds the strings in place,the action that translates key depression into hammer movement, and the exterior wooden housing that supports and protects the other components.
For a more in-depth analysis of the components of the piano, see Gate 3: Component Summary.
The action is largely an array of levers and and a four bar linkage that translates the vertical motion of the keys into the largely horizontal motion of the hammers, while yielding the user a high degree of control over the duration and speed of the hammer strike. The typical piano has 88 keys, so the action contains 88 identical sets of pull rod/action linkage/hammer sub-assemblies.
The action also contains mechanisms for modifying the tonality of every key strike: depressing the two pedals results in the raising and lowering of two bars that respectively increase the volume of the piano, and change its tonality and make the piano quieter by shifting all of the hammers closer to the strings.(Fig C.)
The frame, as previously mentioned, is a very large and heavy piece of cast iron whose primary function is to provide support to the strings. Its considerable weight and thickness is to ensure that the strings can be placed under high enough tension to produce the correct sound without warping.(Fig A)
The housing is the wooden structure that supports the action and frame. It consists of glued and/or screwed together wooden planks and provides the shape of the piano. It also has a sliding cover to protect the keys(Fig B).
As part of our analysis, we also proposed several subsystem and system level revisions to the design of the piano, though due to the rather specific nature of the piano's function, we found it difficult to think of many revisions that did not in some way reduce the functionality of the piano. As a result, we typically proposed the simple replacement of parts of the piano with higher quality materials, or the addition of entirely new functions. Below are two example from each category that we found most interesting. For a complete list of our proposed subsystem revisions, see Gate 3: Design Revisions. For a complete list of our proposed system level revisions, see Gate 4: Design Revisions.
Injection Molded Action Linkages and Hammer Bodies
Save for a small number of metal connectors, a spring, and felt, the entirety of each action linkage and hammer is composed of several meticulously cut and adhered pieces of wood. We propose the replacement of all such wooden structures with injection molded plastic. After initial development costs, this would dramatically reduce the cost and creation time of all such pieces.
Chord Playing Keyboard
As previously mentioned, there are no apparent system level changes that can be made to the design of the piano without negatively impacting its functioning or desired output. Instead, we propose the addition of a second keyboard that is capable of playing chords instead of single notes. As discussed in our presentation, a considerable body of research would have to be conducted on how best to implement the extra functionality.