Specifications

How to Draw a Plan for a Fanned Fret Guitar

Site Map
I am very pleased with the design of the 2016 9-String Fanned Fret Harp Guitar. I am completely confident recommending these dimensions if anyone else wants to build one. However, there are many variations possible, and you may want to draw a series of variations, as I did, so I am including a description of my procedure for making the drawings. I don't have any concrete advice for luthiers, as I am not a luthier, but if you make a drawing such as I will explain below, a luthier can turn it into a guitar. The drawing in the picture farther down this page is not the drawing I actually used (I drew many), but it is similar enough to demonstrate the procedure.

Although I have considered several other possible tuning schemes (the pros and cons of which I will discuss further below), the tuning that I use is this:


A4  600mm  .022" / .56mm rectified nylon
E4  615mm  .028" / .71mm rectified nylon
B3  630mm  .033" / .83mm rectified nylon
G3  645mm  .042" / 1.07mm rectified nylon
D3  660mm  .028" / .71mm silverplated copper wound
A2  675mm  .032" / .813mm silverplated copper wound
E2  690mm  .041" / 1.04mm silverplated copper wound
B1  705mm  .052" / 1.32mm unplated copper wound
F#1 720mm  .074" / 1.88mm unplated copper wound

This retains the core tuning of the six-string guitar, allowing legacy repertory to be retained without change, and adds three more strings which are all tuned in fourths, so that the fingering patterns are extended across the fingerboard without irregularities. Essentially, this combines three instruments in one:
A bass guitar F#1 - B1 - E2 - A2, tuned a whole step above a standard four-string bass;
A six-string guitar with the normal six-string tuning;
A treble instrument comparable to the Mexican Requinto, which is tuned a fourth above the usual six-string guitar, but I am using the tuning of the Renaissance lute instead of the requinto tuning in order to keep the lower tuning of the conventional six-string guitar intact, an important detail which minimized the necessary learning curve.

The differing lengths between the strings are designed on the same principle as a harp or a the pipes of an organ. High-pitched strings must be short and very thin, while low-pitched strings must be longer and fatter. Usually this is accomplished with separate instruments of different sizes.

The four unwound strings are all D'Addario or La Bella rectified nylon. The wound strings are a motley assortment of D'Addario, La Bella, and Savarez. I ordered many strings to test from Strings By Mail, which has a "Specialty Singles" page for each of these brands offering arrays of strings of different diameters. The diameters listed above are somewhat less than I had first considered optimal for tone, because I found that the cumulative tension created difficulties in barring - 50% more than a six string - so I have reduced the overall tension from what I started with.

Some important questions in fanned fret design are: What is the maximum length of the highest-pitched string? What is the minimum length of the lowest pitched string? How wide should the fingerboard be? Where should the "right angle fret" be placed for the most playable fingerboard?

The proven maximum for an A4 is about 63 centimeters, this with the thinnest string you can get, about .020" in nylon. At 65 centimeters long, almost any string will break tuned up to A4; you might make it to G4 with a thin string. 61.5 centimeters is the standard length for the A4 string of the 8-string Brahms Guitar type. My 2016 nine-string fanned fret harp guitar has its A4 at 60 c. Frances's seven-string of 2012 is at 58 centimeters. On the nine-string prototype of 2013 the A4 was at 56 centimeters, an experiment which gave unacceptable results. It is desirable to maximize this length in order to minimize the fan. I feel that 60 centimeters is a very good compromise and gives a more playable string than a longer length, and better intonation than a shorter length. Intonation is a complex issue, and careful saddle compensation is a must.

As amply demonstrated by the Yepes-design straight-fret 10-string classical guitar, the low bass strings D2, C2, B1, A1, are not very functional except as open strings when they are as short as 65 centimeters. The 9-string prototype of 2013 had its longest string, tuned F#1, at 68 centimeters, which was not long enough to make a string of that length and pitch behave well. It is, however, long enough for a very nice-sounding B1.

Demonstrating or testing string lengths in these low ranges is not easy. The fault of such low bass strings, when they are too short, is not in the sound of the open string (which is easily tested on some kind of benchtop string stretcher), but in its performance when rapidly fretted. A too-short bass string will make clicking sounds on the frets, and will fret sharp; this was the case with the 2013 prototype.

The successful 2016 nine-string has the F#1 at 72 centimeters. This is acceptable and performs well, but begs the question: other considerations aside, would not 75 centimeters for the low F# sound even better? Because, obviously, there are bass instruments with much longer string lengths than this, such as acoustic upright basses. The "other considerations", however, are whether your hands are big and strong enough, or if you could adapt your playing style to such an instrument. This is an open question which has to be answered individually. My own opinion, based now on a year and a half playing this instrument, is that there is no way I would increase the length any more, because of the difficulty of the technique.

The Specs for my 9-string as built:

  • A4 string: length nut to saddle: 600mm. Diameter: .0559mm /.022" rectified nylon.
  • F#1 string: length nut to saddle: 720mm. Diameter: 1.905mm /.075" silverplated copper wound on nylon.
  • Fingerboard width at nut: 80mm. The fingerboard will be wider near the soundhole, this is determined empirically after drawing the strings.
  • Right Angle Fret: #3
  • String Spacings: 9mm at nut, 11.5mm at saddle
At right: the lattice bracing which luthier Castillo actually used instead of the fan bracing shown in the drawing above.

About the Soundhole placement:

The unusual sound-hole placement is one of the important details. The 2013 9-String Prototype was built on a conventional body plan with the soundhole in the usual location. However, the angled bridge did not have a large enough field of vibration around it. The point at which the A4 string attached to the bridge was only a couple of inches away from the transverse bar. As a result, the A4 never developed any punch. So, in the new design, the soundhole is tucked up in the corner, and the transverse bar is farther from the bridge, and the bridge is in the middle of a larger vibrating field, and the A4 sounds "much better".

This "much better" might be better stated "relatively much better" as in my experience a high A4 string on any type of guitar is probably not going to develop the kind of punch that we have come to expect from the E4 string on a fine classical guitar. The A4 will always be a little weaker because it is necessarily shorter and thinner. As a player of such an instrument I have had to accept this. Some players use very high tension (nylon) strings for their A4s, for instance .029" instead of the .022" or .021" that I use. (I tried very high tension fluorocarbon strings also and rejected them in the end because, although loud, the sound was not sweet and vibrato was difficult.) The .021" has a much more musical sound though it is not as loud. Later note: After a year and a half, the sound of the spruce top has developed quite a bit, and between this and having adapted my playing style, the weakness of the A4 string is not an issue.

There is one more interesting weakness that has come to light with this design. During my experiments to determine optimum string diameters, at one point I put heavier strings on all of the trebles. The top developed a rather serious dimple right near the point where the A4 string is attached to the bridge. The increased pressure on the top from the angular tilt of the bridge is all focused at that spot. This was an unexpected result. Normal right-angle bridges will force the top down across the width of the bridge when overloaded, making a belly in the top between bridge and sound-hole, and a bulge behind. I was afraid that the point of the angled bridge might even punch through the top, and I removed the heavier strings. For this and other reasons I settled on lighter gauge trebles, the other reasons being for sweeter sound, and for less effort in barring. This last reason was the most critical one, because I did hurt my left hand in the summer of 2016 because of this, and it took a year to fully recover. I actually needed to develop more muscle mass in my arms and shoulders, which was unexpected and surprising; although I had known that strength might theoretically be a factor, I had discounted it.

The low F# might be longer...

There is a possibility that a 9 string fanned fret guitar similar to mine might be built with the fan from 75c to 60c. (This would be a 5:4 ratio fan, with a 15c spread, as opposed to the 6:5 ratio fan with a 12c spread of my actual build of 2016.) I think that, for some young person with big hands, that this would sound great, but it would certainly be harder to play! I have had some physical difficulties with my left hand barre muscle as it is, and so I am actually quite content with the smaller measurement 72c.

...or might it be shorter??

On the other hand, I am not sure that I can recommend going any shorter than 72c for a low F#1 string. There is some possibility it might work at 70c, but it requires another experimental build to find out, and that is, frankly, an expensive experiment that I am not likely to do. The 2013 Prototype, with the F#1 at 68c, just didn't cut it. That was a carefully calculated estimate based, so I thought, on the best information I could come by: it was about equal to the distance from the bridge to the second fret of a short scale (30" or 76.2c) electric bass. However, this was to be a round-wound string on a classical-type guitar, and there was no way to know whether the result would be the same as with flat-wound electric bass strings, and probably it was overly optimistic to expect it, and I should have built in a margin of error. (That wasn't the only design error I made in the Prototype, in any case.) Anyway, 68c turned out to be too short, and so I had to guess at what it would take to make it better; would it be 70c or 72c or 75c? So I decided that 70c was too iffy, based on the performance of the Prototype at 68c, and that 72c would be a surer bet, given that I only would be able to afford to have it built once.

(These three lengths 70c, 72c, 75c, make Pythagorean super-particular ratios to the length of the 60c A4 string, 7:6, 6:5, 5:4. Using Pythagorean ratios in this context may be voodoo, but if you have some better way to organize your thinking about it, go for it.)

Big hands? Small hands?

The thing to note if you are building a 7- or 8-string, with a low B1, is that a string length of 65 centimeters gives very poor results for bass strings lower than E2, and the builders of 8-string Brahms-guitar type fanned fret instruments would do very well to extend their lowest string, B1, out to 70c or at least to 68c. This also goes for 10-string Yepes-type instruments. The Bartolex company is now building a fanned-fret Yepes-style 10-string, but they have not yet taken advantage of the possibility of longer basses, an error in my own opinion. Although you might think that an instrument with longer basses would be very difficult to play, I assure you that it won't be if you have big hands. If you have small hands, you might think about it twice. Perhaps the best extended range design for those with smaller hands might be the so-called "alto guitar" with a short scale about 58c, straight frets, and many unfretted diatonic bass strings, but I myself don't like that design for three reasons: (1) too much sympathetic vibration from the basses, (2) non-idiomatic tuning of the basses for actually playing bass, and (3) too much technical disconnect between the bass and treble ranges; in this case the two instruments combined into one do not flow smoothly one into the other at the point of junction, whereas with my nine-string guitar there is no break in the flow across the strings.

Frances's hands are only a little smaller than mine, but she was never comfortable with a 65 centimeter scale (she even had a guitar with a 66 centimeter scale once, a beautiful guitar but which she found quite difficult). She made the radical choice to go for a much shorter scale, 58 centimeters, and she has been very happy with this as it has allowed her to develop a fluid and comfortable technique. I think there may be many guitarists who would do much better with short scale instruments. The 20th century classical guitar design after Ramirez and Hauser was optimized for use in 4000-seat concert halls, with a large body and a scale length longer than was generally customary in the 19th century, when guitars were parlor instruments. As we don't play in such large halls, and usually play amplified, we have a different set of requirements and possibilities.

In the case of my nine-string fanned fret harp guitar, an interesting thing to experience is that the topmost four strings are all shorter than 65 c., and this means that playing on those strings in the upper registers has all the ease of playing a requinto, and many useful chords are available that on a conventional six string would be above the 12th fret and so far up the neck as to play out of tune.

Does it play in tune?

Speaking of playing in tune, people often ask if the fanned fret guitar plays in tune. Yes it does: no better and no worse than a regular six-string. It shares with the six string the common problem that the G string frets a little sharp on the first and second frets, and that careful saddle compensation is necessary. The 2013 prototype, with its much shorter treble strings, had some sour chords starting on around the 10th fret. However, I never got around to addressing the issue of saddle compensation, which could improve the situation. On Frances's 58c short-scale 7-string, I have improved the intonation considerably by compensation of both the nut and saddle.

The 2016 build played in much better tune from the get-go. However, after a year and a half, when I was sure that I had the action and string gauges correct, I finally got around to compensating the saddle. I made a series of intonation maps, measuring every fret of every string with a Seiko Model SAT800 tuner which reads cents, and averaged the results (because there is quite a bit of fluctuation between one measurement and another). The G3 string, the worst, was about 12 cents sharp at the 12th fret, even though the F#1 and A4 strings required no compensation. Some rough calculation showed that this would require about 2 millimeters of extension to the string length at the saddle, but the saddle installed by Castillo was less than 2 mm wide. So, I widened the saddle slot to 4mm (a long job with a razor blade, file, and chisel) and built a new bone saddle (cut from scratch on the table saw from a large cow femur), with compensations for 7 out of 9 strings, and improved the intonation immensely. (The new saddle is roughly 15 c x 4 mm.) The operation produced a side benefit: with the fatter saddle, the bass response immediately improved dramatically. This is another point which the builders of Brahms guitars and Yepes-style 10-strings could well take note of also: if you want good basses, put in a fat saddle 4 mm wide.

I am still not happy with the intonation above the 12th fret, which runs quite sharp. This appears to be a non-linear behavior due possibly to the extra stretch required to press the strings down. A future project is to carefully quantify the intonation error, and then rip out the top frets and replace them in adjusted locations after filling the old slots. It appears that they need to be pulled back toward the nut slightly from the calculated locations. This is a rather daunting project, however, even more so than widening the saddle slot, and I am not in a rush. After the successful compensations to tune the 12th frets, the upper range above the 12th fret is really no worse than any other classical guitar, running some 6 or 8 cents or so sharp on the central strings around the 18th fret.

Isn't it strange and weird to play on fanned frets?

No, not really. If you shut your eyes and feel with your fingers, it's hardly different at all. That's not the whole story, though. The real challenge with learning an extended range guitar is to learn the musical grammar of the new strings. I have also had some physical challenges with my left hand and arm, which you might well expect. The fanned frets actually encourage the development of many new positions which can splay the left wrist and elbow out in both directions, and require some stretching out. This is an ongoing process. However, I went through the same process with the six string many years ago.

Now, back to the drawing board:

Drawing Tools:

  • Four-foot straight edge. Three feet or one meter will probably do. I have a four-foot carpenter's level and also a piece of angle iron that I use. (These measurements are easier for me in metric. Others may use hundredths of inches. I imagine that it is possible in 32nds or 64ths of an inch, but this would probably be a pain unless you are the happy owner of a Construction Master Calculator, which I used to have many years ago.)
  • Pocket Calculator
  • Small straight edge 20mm or so, or 6 to 12 inches.
  • Large sheets of paper 4 feet by 2 feet, or 120 centimeters by 60 centimeters, or thereabouts, or bigger. Since I have never found sheets this big, I have been accustomed to taping two sheets together. Put the tape on the back so it doesn't interfere with your drawing. I used plain paper at first but later I switched to graph paper.
  • Pencils
  • Eraser
  • Medium Point Felt Tip Pen
  • Note Pad
  • Compass
  • Tape Measure
  • Square(s). I use a smallish transparent plastic triangle AND a large steel carpenter's square at times.
  • Drafting tape or masking tape may be handy if you want to secure the drawing to the work table.

Procedure:

This is only for a plan view of the fretboard and body shape. Sections and elevations are un-necessary at this phase of the design process, and if needed at all they are the responsibility of the luthier who will actually build it. The choice of a bracing scheme and other interior details will depend on his or her own habits and preferences. A particularly tricky detail will be the compound angle of the headstock. How this is done I have no idea - I can only look at it and marvel.
  1. Establish a center line extending the length of the paper, bisecting the width. This line can be fairly lightly drawn. At the end, you will black in important lines with the sharpie. Until then, draw everything as lightly as possible to make it easy to erase mistakes and unwanted line extensions later.

  2. Mark the midpoint of the center line, which is the center of the paper. This will be, somewhat arbitrarily, the 12th-fret point of the center string, if there are an odd number of strings. If there are an even number, then the length of the center line splits the difference of your longest and shortest strings. In the case of my 9-string, the D string is on the center line, and it measures 660 mm from nut to saddle. Therefore, the distance from the centerpoint of the paper to the nut and saddle at the D string is 330 mm. Mark these points.

  3. Draw right-angle reference lines lightly with a pencil at the endpoints of the center string or centerline just marked. These reference lines will be 660 mm apart, or each 330 mm from the center point of the paper (or adjusting this value for your own specs) and they are are for establishing the approximate endpoints of the other strings, whose actual lengths will be somewhat different, in order to draw their lines before establishing the exact lengths. The angled nut and saddle, to be drawn later, will cross the centerline at these precise points, but at a different angle of course. It is not necessary to define the nut and saddle angles in degrees. You may measure them later after they have been drawn, if you want that info, but it is not necessary.

  4. At the nut end, make pencil tics at 9mm intervals on the temporary nut line, to establish the approximate nut ends of the strings, however many there will be. If an odd number, the center string is on the centerline. If an even number, start by measuring 4.5 mm to each side of the centerline. If you want a narrower string spacing, do it. However, bear in mind that basses have wider string spacings, and this guitar is really more like an extended range bass in a way. It's a very personal choice. Whatever you are used to may be best.

    I made the string spacing very narrow on my 2013 prototype, thinking that the reduction in the width of the neck would be worth it (69mm on the 2013 prototype, 80mm on the successful 2016 build) - my luthier, Castillo, rolled his eyes about this one, and it was indeed a mistake. I had measured a friend's Gibson arch-top which had 7.5mm spacing between the strings at the nut, and I said to myself, if he can play on that spacing, then I can. But I couldn't. I could never adjust to the narrow spacing after a lifetime of classical guitar spacing at 8.5 to 9 mm. A helpful person on the 7-string forum recommended going "with the string spacing you are already comfortable with" and this has proven to be the best course. The fact is, this is partly a bass, and the lowest bass strings are quite thick and need to be spaced farther apart than the thinner treble strings have to be. A graduated spacing is a possibility, but I have not tried it and don't know that it would be more playable. In the end, reducing the string spacing in order to reduce the width of the neck (on my 2013 prototype) resulted in more difficulty in playing cleanly than the advantage was worth. Also, consider this: if the neck is wider than necessary, it is possible to reduce the string spacing by cutting a new nut and by putting in a notched saddle, but if the neck is narrower than necessary, it is impossible to increase the string spacing unless by eliminating a string. In any case, on a wide neck like this, you will never be able to put the LH thumb over the bass side of the neck to fret the lowest string, forget it.

  5. At the saddle end, establish the string spacings at 11.5mm (or your own desired spacing) on the light right-angle pencil line which represents the temporary saddle line.

  6. Draw all of the strings by connecting the spaced endpoints with light pencil lines, running them well past the right-angle nut and saddle reference lines. The excess can be erased later.

  7. Now the right-angle fret needs to be established. On my 9-string, this is the third fret. On less extreme fans, the right-angle fret may be higher on the fingerboard. It's a personal choice. I do not recommend putting the right-angle fret at #12, because the nut will have a very extreme angle and playing in first position will likely hurt your wrist when you try to barre. The calculation for the third fret is as follows. Continue the iterations to get a higher fret.

    Iterative Calculation Procedure for the Rule of 18: (using a pocket calculator)

    1. Total String Length of center string 660mm: Enter this value in the calculator (or your own center string measurement value).
    2. Hit Memory Plus: put the value 660 in the memory.
    3. Without clearing the memory or the screen, divide 660 by 17.817152. This is the adjusted digital value for the "Rule Of 18".
    4. This give the value 37.04..., which is the distance in millimeters from nut to first fret on the centerline. Hit Memory Minus and subtract this value, in the memory, from 660.
    5. Hit Memory Recall. The value is 622.957... This is the distance from the saddle to the first fret.
    6. Divide this value by 17.817152. The new value is 34.963... This is the distance between first and second frets on the centerline.
    7. Hit Memory Minus: subtract this value from the memory.
    8. Hit Memory Recall. The value is 587.993... This is the distance from the saddle to the second fret.
    9. Divide this value by 17.817152. The new value is 33.001... This is the distance between second and third frets on the centerline.
    10. Hit Memory Minus and subtract this value from the memory.
    11. Hit Memory Recall. The value is 554.991... This is the distance from the saddle to the third fret on the centerline. Measure and mark this distance. 555mm is close enough. It is impossible to make pencil marks finer than a tenth of a millimeter, if that, but I keep the calculator working at its maximum number of decimals and only round off on the drawing itself.


  8. Having marked the right angle fret distance on the centerline, draw the right angle fret with a square, extending it lightly a little beyond the edges of the fingerboard.

  9. Now you will need to calculate all of the fret placements for the longest and shortest strings, using the recursive rule of 18. The procedure is the same as above. My longest and shortest string lengths are 600mm and 720mmm. The recursive calculations should be completed up to the highest fret you want: 19, 21, 24, whatever. Write in the notebook the complete schedule of fret placements for the two outer strings, carefully writing the fret number with each value. Even though the lowest string will have fewer frets than the highest string, you need those measurements in order to place the highest frets at their correct angles. Be careful and don't screw up, or your frets won't come out in the right places. Don't forget, you are not writing down the small values (37mm, etc) which represent the difference between frets. Instead write down the larger values, (660, 662.957, 587.993, etc, or whatever they may be), which represent the distances from the saddle to each fret. If you write them down to three decimals, you will have far better numbers than you can actually use, but keep these values for reference and round them off to the nearest .1mm when you mark those positions on the string line. Your pencil lines may only be good to the nearest .5 mm; it takes a pretty sharp pencil to get better. Do the best you can. (These outer string lines are NOT the outer edges of the fingerboard, which will be 8mm wider than the total string width, which is continually varying since the string widths are wider at the saddle than at the nut. The exact fingerboard dimensions will be drawn in later after you have drawn the soundhole.)

  10. Place your tape measure or meter-stick so that the number for the distance from the saddle to the third fret of the highest string is at the intersection of that string with the right-angle pencil line which represents the third fret. If your right-angle fret is at some other fret, line up the appropriate number. Now you can mark all the rest of the frets as accurately as you can with pencil tics. Also mark the saddle point on that string, which will be at "zero" on your tape measure or meter stick. (If you are familiar with the carpenter's procedure called "burn one", use it, but be careful to add one at the other end! Lots of screwups come from that technique.)

  11. Line up the numbers of the lowest string with the right-angle fret and mark all of the frets with pencil tics.

  12. Now connect the nut points and saddle points on the outer strings with pencil lines. These lines replace the lighter lines drawn earlier which had the points for establishing the string lines.

  13. It should be very easy at this point to draw all of the frets by connecting the fret points of the outer strings. Draw the frets around the soundhole with very light pencil lines for the moment, as some will need to be erased after you draw the soundhole and establish the shape of the end of the fingerboard.

  14. Now you will draw the body of the guitar. My procedure is just to take any guitar with the right body size and shape (a full-size classical, not for instance a Torres copy), place it carefully on the drawing aligned perfectly with the center line, and trace the body outline with the pencil. The critical decision here is, where do you want the neck to meet the body? As the frets are angled where the neck meets the body, there is no one fret to choose. On my 2016 9-string, the neck meets the body at about the 11th fret of the lowest bass string. On the treble side I have a cutaway. In theory, the body could meet the neck at any point which permits the bridge to be somewhere on the top.

    There are two variables to be traded off here. One concerns playability, the other quality of sound. (This trade-off of variables is where the design process becomes guesswork.) On the one hand what we want to avoid (I believe) is having the neck - which will be heavier than that of a six-string - too far off-center to the left-hand end, resulting in fatigue to the left elbow and shoulder. On the other hand, if the bridge is placed too far toward the butt end of the lower bout, it will be out of the central area of the top where the biggest field of vibration will be. I had an idea, at one point in the design process, that the bridge might work if placed down deep in the lower bout, because lute bridges are often placed so, and because I though that perhaps the lowest bass strings would be strong enough that their attachment point would not need to be in the center of the top; this would have brought the left-hand extension of the neck closer to center.

    However, the luthier I was working with, Castillo, did not like this idea, as he believed the entire bridge needed to be more towards the center of the vibrating field of the top, so we moved the whole body to the right and the neck-string-bridge array to the left a couple of inches or so in the final design. This is one of those details which could only be refined with many trials, which means building MANY guitars. So far, there have only been two 9 string fanned fret classicals built (both mine), and so there is a huge field of possible experimentation yet to be done.

    (For the record, there has been to my knowledge one fanned-fret ten-string classical built, and many eight-strings of the "Brahms Guitar" type, a more conservative design all the way around. That 10-string has a full bass register with fourth-tuned basses and a wide fan similar to mine. By contrast, the Bartolex company is now building their 10-string fanned fret with a conservatively narrow fan, a guitar which is intended to be tuned as is the Yepes-style 10 string, with diatonic basses, a distinctly different concept than mine. I am not a fan of diatonic basses, because I like to actually "play bass" on my 2016 9-string, in a manner which is idiomatic on fourth-tuned basses but not on diatonically-tuned basses. Diatonically tuned basses are fine for long sustained bass notes under unchanging harmonies, but worthless for dynamic bass lines.)

    Since every guitar represents a very major personal expense, I cannot afford to do the kind of systematic testing and development that would reveal the effects of the many possible design variables, and it is unlikely that any individual luthier could afford to do so either. (It appears, however, that the Bartolex company is indeed doing such experiments, and they would probably build one like mine if there were any demand, as they have built affordable models of several similar designs and appear to be the leaders in classical-style ERGs.) It was an expensive luxury to have the 2013 prototype built. However, I could see no other way forward. In the spring of 2013, I said to Frances, "I can't go any farther with this until I get one of these designs built and find out what works and what doesn't". So we had one built. It was a shot in the dark and had many errors, but every error was a question answered about the design variables.

  15. Now you may draw the soundhole, draw the lines of the outer edges of the fingerboard (I make the fingerboard extend 4mm to each side of the outermost strings), and draw the end of the fingerboard, erasing the excess fret lines or leaving them in very light pencil.

  16. The important dimensions are now established. You may draw the headstock and rosette however you please.

  17. Now, using your large square, draw reference lines to all important measuring points and mark the dimensions clearly along a line near one edge of the drawing. You might also copy the complete schedules of the fret measurements for the outer strings out of your notebook onto an un-used corner of the drawing.

  18. Using the fine-point sharpie, black in all the important finished outlines of the frets, fingerboard and body.

  19. Take photographs of your drawing and scan it as well, if possible.

  20. Duplicate the drawing however you can. If this means making another drawing, then do it. The process is valuable anyway. Cut the duplicate drawing out and glue it to a corrugated cardboard outline. Tape this onto the face of a real guitar, and go play air guitar for a while on it to see if you like how it feels, as best you can.

  21. Make a drawing of another variation. Spend a year or so drawing variations and playing air guitar on the cut-out drawings before you put your money on the barrelhead.

*    *    *

Library