Some Holdbacks

Ideally, I'd like to have the final testing results here today. However when gluing the bridge, it's come to our attention that it is in fact warped and that only it's ends have been attached to the guitar. This was discovered when attempting to restring it, where upon tuning and playing the guitar the bridge abruptly snapped off the guitar.

I haven't any photos to document this as we immediately went about re-gluing the bridge, however, the final results should be up either tomorrow or Saturday.

Reconstruction of the Guitar

Everything's cut and ready to go. Time to glue. 

There are three things here that have to go back together. First, the soundboard needs to be glued on, then the bridge and fret board segment. The soundboard presents a problem because the body of the guitar without a soundboard attached has warped, and needs to have its sides pressed back in. 

For the soundboard we considered a few options. The first was using several C-clamps to line the guitar. However, due to the irregular shape of the guitar, this seemed to present significant room for error. The adhesive used was LocTite epoxy, which sets in about five minutes, so setting clamps in place was a race against the clock and would likely require a few sets of hands. Bar clamps were another consideration, but similar perceived difficulties. Garret decided that the best thing to do would be to create a jig, a setting shaped like the guitar which the pieces would fit into. 

The edges of the soundboard were scored, or roughed up with sand paper to enable better adherence of the two pieces. The soundboard was then set in the jig, followed by the guitar, and weight was placed on top of the guitar to ensure a good seal. This method was also preferable to the clamp method for one other reason. Our sandwich board has wooden honeycomb inside, which can collapse under too much pressure. A clamp applies acute pressure, which has the potential to puncture the soundboard. By distributing weight over the whole guitar, the board is less likely to be crushed. 

Taping a perimeter to score the edges

Scored Soundboard

Soundboard in jig with guitar and weight on top

Soundboard glued to body

After this the placement of the bridge and fret board were determined, outlined on the face of the guitar, and glued in to place. The final piece without strings is shown below. 

Construction of the Soundboard

Well, the guitar's been taken apart, so the next step is to take the composite plate we've received from ACP Composites (http://imgur.com/qXf5QPn). The plate is of wonderful quality and as mentioned in the FAQ rings like a drum when struck.

We sketched the outline of the guitar by tracing the bottom plate of the body (The top is slightly warped without the soundboard attached). A bandsaw that's built to cut steel was used to cut the shape. As much excess as possible was cut to leave as close to the actual shape as possible, then the edges were covered in masking tape, the outline was redrawn, and detailed curvature was removed little by little. We had to use a pneumatic sander to do the tighter curves on the side. I'm afraid we haven't many pictures of this process as I was assisting Garret by removing the carbon fiber dust with a vacuum, as it's toxic and itchy if it makes contact with your skin. Most of the process is rather intuitive though.

Taping the rough cut soundboard 

The second task in cutting the soundboard to size was removing the sound hole. A cardboard template was made from the original sound hole and used to trace the new hole. The center was a identified and a hole saw was used to remove the sound hole. 

Soundboard with hole under the hole saw

The other task completed while in the machine shop was trimming down the bridge to fit the new soundboard (the composite is slightly thicker than the wood). The pneumatic sander was used again as well as a rotary sander. A mill was considered for a more precise cut, but the irregular shape of the fret board made this a difficult task. However, as the bottom was sanded away, the frets were reached, and the fretboard had to be left slightly arched as too much was taken away. This may pose a problem if the height of the strings off the sound board, or the action as it's called, is not high enough.

Garret Sanding a rough edge on the fret board

Deconstruction of the Guitar Sound Board

      For the C40 Yamaha classical guitar model the sound board of the guitar is attached to the body with a meltable adhesive. The bridge and portion of the fret board which are attached to the soundboard are also sealed in place with the same adhesive.

     

In order to remove the sound board the bridge must first be removed from its face. A heat gun and a razor blade are ample tools to do this procedure. As the heat gun melts the adhesive a razorblade can be wedged under the bridge to lift it off the guitar. Once it was removed it was sanded and polished to remove any exess adhesive or debris.

Removing the bridge with a Dremel

Close up of bridge

Polished and Sanded bottom of bridge

Following the bridge, the portion of the fret board which extends on to the soundboard must also be removed. A backsaw can be used to cut behind the twelfth fret to separate the inner portion of the fret board from the neck. It can then be removed with heat and a razor blade, similarly to the bridge.

Cutting the fret board at the twelfth fret

The final task is to remove the soundboard itself. In order to obtain the best results you first want to remove as much of the center soundboard as possible with drummel while leaving a good half inch perimeter of the remain soundboard around the outer edge of the guitar. This perimeter can then be segmented using a backsaw or drummel and removed using a heat gun piece by piece. If the board isn't segmented, you are faced with the difficulty of heating the entire perimeter of the guitar at once and removing it as a whole, leaving a lot of room for error. 

Removed Soundboard

Guitar without soundboard

Control Testing of Guitar Sound and Sustain

In order to analyze the differences between the spruce and composite material's effects on the soundboard a qualitative measure of sustain is necessary. However the sustain may vary depending on the force applied to the string. Playing the string manually would not then be an appropriate means of measuring the sustain. A simple mechanical device was made to ensure that the force of each pluck of the string was the same. The device is a hollow wooden box that is placed vertically lengthwise, with two slits on the opposing horizontal sides which act as a track for a weighted center piece which slides up and down and holds a guitar pick. The device can be seen in the photos below. 

Strumming Device

 This instrument allowed for the playing of a string at a constant force. The next step involved finding software which would record a sound and display a graph of its amplitude over time. (Frequency is not needed as the guitar strings are always tuned to the same frequencies). The freeware program Audacity was a prime choice for this task as it features the graph and allows it to be zoomed and manipulated. A sample of this graph is shown below.

First test of the high E String

By plotting the maximum value of each oscillation as a function of time, a linear regression model can be applied to the points to find where the amplitude reaches approximately zero. The time at which the amplitude is zero will be used to measure the sustain. That value will be averaged over each test of a given string and compared to the value found for the other soundboard to indicate the changes in sustain. 

Initial Considerations

The objective set for the final composite model are that the guitar has better resonance ( a longer tone) and a more mellow timbre (tone quality). In order to complete the first of these two criteria a high modulus material is desired that is as thin as possible without sacrificing the structural integrity of the soundboard as whole. In other words it is necessary that it can resist warping from the tension of the strings as well as under conditions such as changes in temperature. The soundboard is attached with a glue adhesive, thus the resin must have a melting point higher than the glue used to attach the board so that it can be removed and replaced.

The motivation for using a high modulus material arises from the fact that the greatest interruption to the acoustic resonance of the sound emitted from the guitar comes from the wooden ribbing that holds a convectional wooden soundboard in to place. The material itself does much less to effect the guitars sustain, although it can influence its timbre, or tonal quality (i.e. mellow, metallic, poignant).

The guitar being used is a Yamaha C-40 classical acoustic guitar. It features a traditional spruce soundboard with inner ribbing.

A carbon fiber-epoxy resin composite is the material of choice for this task.