“How do I choose a good guitar?” After years of hearing this query I decided, several years ago, to write a brief outline of those things that are pertinent to the question. At the same time I decided that this guide could serve as a format for the lectures I give on this subject. I would like to share that outline with you now and then proceed to a more detailed discussion of some of the points it raises.
The four areas to look into are:
2. Action and feel
3. Condition and construction
1. The most important of these is Sound: That’s what the guitar is all about …sound.
There are several ways to check for sound:
A. Bring a good sounding guitar for comparison.
B. Bring along a friend with a good ear who can also play.
C. When testing guitars do it outdoors or if that isn’t possible, do it in the largest room you can find.
The important facets of overall sound quality are:
1. Timbre. (Quality of the individual tones.)
2. Balance. (Trebles must match bass.)
3. Separation. (The clarity with which individual tones can be distinguished in a chord.)
4. Sustain. (The rate of decay of a tone after it is struck.)
Always remember that sound can rarely be greatly improved in a guitar without tremendous expense.
2. Action and Feel.
Action is the height of the strings above the fret and fingerboard.
Feel pertains to those features that comprise the playability of a guitar other than the action,
such as: neck size and shape; string length; string spacing and location from the edges of the
fingerboard; body shape.
Actions can usually be corrected at moderate expense. Other than reducing fingerboard width and
neck size, little else can sensibly be done to change the feel.
3. Condition and Construction.
If the guitar is new, then examine it for clean construction inside the body and carefully tap
around the face and back to check for broken struts. Check for depressed or swollen face.
See if the bridge is on tightly. Check the condition of the neck and frets.
If the guitar is old, examine it for the above conditions plus cracks in the face, back, sides,
neck-to-body joint, head-to-neck joint, purfling and centerjoint of the back. Also examine the tuning
machines for worn gears or sloppy installation.
Let the buyer beware! Know the seller! Ask about a guarantee. Shop around. Remember the most costly guitar isn’t always the best. Think about re-sale.
While this outline is basic in content, it does generate questions from the audience and we often delve at some depth into the various facets of guitar construction, testing, and sound.
The most important subject we discuss is sound. Here are some of the things I discuss with them:
Loudness. If you intend to give recitals and concerts in large halls, you had better be sure that the guitar you choose projects well. The best place to test for this is outdoors. If weather deters you, the second best method is to use an auditorium, gymnasium, or a church. Lacking all of the above, use the largest room you can find. When making this test and, for that matter, all test pertaining to sound, it helps to have a proven guitar along (or several) to use as a basis for comparison and, naturally, someone to play for you and to listen while you play. If you do not plan on concertizing or if you intend to amplify electronically, loudness is not the most important factor of sound to you, but all other sound qualities will be. So at this time, with guitar in hand, let us test for them.
Timbre (pronounced tamber, tanber, or timber according to which authority you choose) is purely subjective, so that what sounds great to me may not impress you at all. However, the instrument must have a tone quality that truly satisfies you, or you will not enjoy playing it no matter what other attributes it may have.
Balance. This I prefer to think of as mostly an objective test because if either the treble or bass end is weak, it will be very noticeable heard at a distance. Be sure to test for this by barring each fret from the first to the twelfth because some guitars have weaknesses more pronounced in certain areas of the fingerboard than others.
Separation (or clarity) is, to a great degree, a quality that goes untested by most players because it is such a difficult and elusive feature to listen for. When a guitar has loudness, good timbre and balance, it is hard to remind yourself to really listen to chords to see if you can hear individual tones (like a good barbershop quartet) or only a glob of sound.
Sustain. Some guitars have an even output of sound and will appear to have good sustain, whereas a guitar with a robust or popping initial output of sound will seem to have less sustain. Therefore, when comparing guitars set a metronome at some fast tempo and count the beats from pluck (or pick) to silence. Some interesting facts will emerge by trying this with different guitars. As to the amount of sustain, all tones on the guitar should have some, with the lowest tones having more than the higher tones.
One word of advice about testing guitars: be sure to play the instrument for at least ten minutes or more before testing in order to “wake up” the soundbox. This is particularly true for spruce-faced guitars. Cedar faces are less likely to require this.
Intonation is included as a branch of sound quality because if the guitar doesn’t play in tune it sounds bad. Fortunately, you can check for fretting accuracy and saddle and nut placement. If errors are found, they can be easily corrected by any competent repair person.
There are several ways to test for tonal accuracy. Let us start with one that many of you are familiar with. Play each string at the 12th fret. Then strike the 12th fret harmonic. These should be identical in pitch. If they are, it tells us only that the maker placed the saddle correctly. If all six strings play sharp, it tells us that the saddle wasn’t set back far enough. If all strings play flat, it tells us that the saddle is set too far back. The cure for either of these conditions is to have the saddle or nut reshaped or repositioned. Again, a repair person should be consulted. Keep in mind that faulty strings can also sound either sharp or flat, but never all six in the same direction. So you should be able to rule out the occasional bad string.
An accurate way to mechanically test for saddle positioning is to determine the basic string length (see below) and add .050” to it if the guitar has a low action and as much as .080” for a high action. The reason for the saddle setback ( or compensation) is best explained thusly: If the 12th fret is equidistant between the nut and saddle, a sharping condition will exist when a string is played at the 12th fret due to the stretch of the string (we make it tighter).
Now suppose the guitar passes the above test, but still plays out of tune. We can then be suspicious of incorrect fret placement. To determine the correct fret placement you will need a ruler with .010” graduations, preferably 18” or longer.
Determine the distance from where the strings leave the nut (usually the end of the fingerboard) to the center of the 12th fret. Multiply by two. The answer is the Basic String Length for that string. Now divide the B.S.L. by the constant 17.817. The answer is the distance from the 1st fret to the 2nd fret. Continue this procedure for the remainder of the frets. Then use the ruler to see if the frets are in the proper position.
Let’s do a hypothetical guitar for practice:
1. Distance from nut to center of the 12th fret = 13.00.”
2. 13.00” times 2 = 26.00.” ( our Basic String Length)
3. 26.00” divided by 17.817 = 1.459.” This is the distance from the nut to the center of the 1st fret.
4. Subtract 1.459” from 26.00” = 25.541.”
5. 25.541 divided by 17.817 = 1.377.” This is the distance from the 1st to 2nd fret.
6. Subtract 1.377” from 25.541” = 23.163.”
7. 23.163” divided by 17.817 = 1.300” This will be the distance from the 2nd to 3rd fret.
By adding the distance from nut to fret and from fret to fret we can arrive at direct readings from the nut to all of the frets. Using the above example, 1.459” plus 1.377” plus 1.300” = 4.136”, the distance from the nut to the 3rd fret. See how easy that was? The use of a calculator with a memory makes the job quite easy.
As to the allowable inaccuracies, it varies with the listener. Some people can detect errors in the fret placement as much as 1/100th of a semi-tone. Others with .100” errors may never hear the discrepancies Keep in mind that a .030” error at the 1st fret is equal to a 2% error, while a .030” error at the 19th fret is almost a 6% error and will be detected more readily. Anyway, with the possible errors caused by faulty strings, who needs errors in frets?
Another important topic for discussion is the action of the guitars. It is possible to mechanically check to see if yours has a good one. Here is a simple and effective test.
Depress the strings one at a time at the 1st fret without sounding the note. Measure the height from the top of the 12th fret to the bottom of each string, using a steel ruler with .010” graduations.
The readings should be:
1st string, .100” to .115”
2nd string, .110” to .120”
3rd string, .120” to .130”
4th string, .125” to .135”
5th string, .130” to .140”
6th string, .135” to .145”
These readings are for classical guitars. The lower readings constitute a low action and the higher readings, a normal action. Anything higher would indicate a high action. Keep in mind that the required action height will vary for different players due to two things: one, the player’s attack style and two, how hard he /she plays the strings. Also, these readings do not take into account the fret condition or the straightness of the fingerboard.
Another test for action height (and the one most overlooked) is that of the height of the strings over the first fret. This is how it is done. Push each string down to the 2nd fret using a finger nail. Be sure you’re pressing on the very center of the fret radius. Now check the height of the string over the 1st fret. The first fret to the fifth strings should have about .003” (approximately newspaper thickness) and the sixth string should have almost double that to allow for the wearing effect caused by tuning down to low D. Also, because the sixth string has a tendency to vibrate sympathetically behind the left hand on certain barred chords, which causes a buzzing sound if it touches the frets.
The question will be raised, “how do I perform these tests on a guitar I haven’t even bought yet?” Well, there are several ways. You can buy the guitar contingent upon its passing these tests, or you might have to do the testing right at the store or the guitar maker’s shop. After all, besides the cost, you are choosing an instrument on which you will spend as many as a thousand hour or more a year playing. The care taken in selecting is priceless. You have every right to expect the guitar to perform to your expectations.
John Gilbert is a well known builder of classical guitars. John now devotes his time to the production of his line of tuners while his son William continues the Gilbert tradition of high quality concert guitars. Gilbert guitars have been used by David Russel, David Leisner, George Sakellariou, David Tanenbaum, Frederic Hand, Earl Klugh, Raphaella Smits, and many others.