(this is the text of a talk given to the Lute Society in 2007)
It’s ironic that I should be talking about lute tuning, because I have always found it very difficult. But my early struggles have made me very aware of the problems and determined to make everything as easy as possible. Most of the things will be known by most of the people in this room, but hopefully we can draw together all the issues and there may be little wrinkles that you haven’t spotted.
There are 3 broad areas to cover:
• The lute and its strings
• Perceptual things
• Tuning meters: how they can help and the different kinds that are available
The lute and its strings.
The way the strings are arranged on the pegs
The height of the action
The knot on the bridge
If you can iron out all the niggling practical problems, you can spend a lot more time concentrating on the really crucial issue of whether the instrument is actually in tune. If every time you feel happy with a string’s pitch, a peg slips or the string’s tension causes it to “click” over the nut, you’re going to have a lot trouble.
The strings. After the initial settling-in, new strings are much easier to tune. They produce a much richer sound with more harmonics, which are more accurate multiples of the string’s fundamental pitch, and this makes it far easier to hear when a note is tune. Even people like me with fairly poor pitch perception will find it easier to hear unisons and fourths & fifths. My feeling is that this is also true of tuning meters and that they also hear new strings more clearly.
I think having new strings is even more important for the upper courses, particularly the first course, and we should make sure we change them more regularly. I would suggest every 3 months or so for the first course, but there will be people who will want to change much more often. With bass strings, it is often less important and some people like the rather dead sound of old wound strings. Gut basses also seem to have a much longer life.
Certainly old strings can be difficult to tune because they are more often false so that their harmonics are more likely not to be properly in tune with their fundamental note and fretted notes will not be correct. (BTW tuning by harmonics is also bad news for two reasons: firstly harmonics can be false and secondly that lute tuning is always tempered in some way and thus tuning a perfect fifth or fourth will disagree with the tempered intervals required, particularly in equal temperament where all the fifths are tempered.)
I find modern strings like Nylgut are very rarely false, but new gut strings often are. The old texts tell you how to check before you put the strings on the lute. You hold the string fairly tight and observe the appearance as you pluck it, if the wave form is nice and clear, it’s good, but if the string seems to wobble and show unevenness it’s probably false. I find this difficult to do, but that may be because I rarely meet a new string that’s false.
It seems to be easier to see this effect on the lute with old strings. Particularly if you hold the lute under an electric light, false strings show the uneven wobbling quite clearly. If you’re having difficulties with a string, this check will help you to confirm if it’s false.
So go home tonight and fit a new set of strings, it will be the best thing you can do to improve your lute’s tuning and sound.
The nut. The nut must be the right height and the slots must be in-line with string’s natural course over it; if the string goes through a little “Z” bend or chicane, it will not slide easily and tuning will be difficult. These are basic issues and if your lute suffers, get a maker to fix it for you.
Even good lutes usually have a “sticky” nut, where the strings do not slide easily over it. As a result, when you first tune, the tension of the string will be different on the sounding length and on the length between the nut and peg. The string’s pitch will change as this difference gradually equalises.
This is always a problem to some degree and thus one needs to be consistent and tune either by dropping the pitch below the required note and pulling up to it or raising the pitch and coming down. Using both techniques will result in some strings tending to sharpen and others flatten with a bad effect on the relative tuning of the lute. I think pulling up to pitch is generally preferable, but know people who do the opposite with good results. I guess this is also a problem when playing in ensembles if players use opposite techniques, but I must say I haven’t noticed it.
There are number of things you can do to improve a sticky nut:
• Polish the grooves. Take the nut off the lute and clamp it in a padded vice. Polish the grooves by pulling a rough string back and forwards in the groove - old gut string is ideal. Coat the string with a fairly mild abrasive like metal polish or “T Cut” (a polish for “restoring” the paint finish on cars). You could also try these abrasives with an old electric toothbrush.
• Lubricate the grooves with graphite using a sharp HB pencil to lay graphite in the groove.
• Play loudly when you tune so the vibrations encourage the string to slide in the groove. This also has the benefit that you hear the note better.
• You can also pull on the sounding length of the string to lower its pitch or press on it above the nut to raise it. This can be a bit difficult to do without over-shooting the note you want.
• Keep the pegs moving smoothly and then sticking securely by coating them with peg paste or French chalk. They both seem to work fine, but don’t mix them. Use one or the other. It seems to be good to re-do the application each time you change strings. Like changing strings this is something we tend to do too rarely, so try to get into the habit.
• Some pegs and also their holes go oval with time. This doesn’t seem to be a major problem, but it’s good to recognise it’s happened and if it becomes a serious problem, go and see a maker.
• A tuning wrench can be a help, particularly if you have a multi-course lute or lots of sticky pegs, but they do have be used with great care – very great care. You can easily break a peg that’s locked fast. The tuning wrench I use is very nicely made, to my specification, by Bruce Brooks, but if I had another one made, the diameter would be substantially smaller (say 25mm rather than 50mm) to reduce this risk. Tuning wenches are also a great help when restringing and make it much easier to get the strings on the peg as discussed below.
The way the strings are arranged on the pegs. This is really important and doing it right is essential. The points to consider are:
• The string must be wound on to the peg from the hole toward the head of the peg. This means the string leaves the peg as close to a right angle to the peg’s axis as possible and minimises the sideways force which tends to pull the peg out of the taper and cause the peg to slip.
• The string should be arranged so that when up to pitch it lightly “kisses” the cheek of the pegbox. This helps to stop the peg slipping by preventing the sideways force mentioned above pulling it out of its taper. It takes a bit of trial and error particularly with new strings that are still stretching and may require some adjustment in the first days after fitting new strings. It’s important not to over-do it as it can make the peg jam too tightly and make further increases in pitch impossible.
• The string should be wrapped just one layer thick around the peg at the point it leaves the peg toward the nut. More than one layer changes the effective diameter and leverage, and thus the sensitivity of tuning. The pitch will change more rapidly for a given rotation of the peg making fine tuning more difficult.
The frets need to be in good condition and correctly positioned. This has to reflect the temperament you use and is usually done by a combination of measuring and listening using your ear or a meter. It’s quite difficult and something it’s worth getting help from your teacher or an experienced player. Charts giving the measurements are available on the LSA website downloads section and in various back-issues of our journal.
If the action is too high and particularly if it is significantly higher on the bass side, placing frets correctly will be difficult. The fretting charts allow for an increase in tension and thus pitch when a string pressed down to the fret. If the heights are too different it will be very difficult to get the fret in tune for both the bass and treble. If yours suffers, go and see a maker.
The knot on the bridge needs to be carefully tied so it doesn’t slip – this is usually only a problem on the first course. Well made lutes have a slight undercut on the rear of the bridge and if you can tuck the string under it and trap the end in the knot that works well. It you can’t do that, tie a simple knot in the end of the string and when that gets pulled up to the knot the slippage will stop; not very elegant, but effective.
The previous section was I hope was fairly uncontentious, but now for more difficult topics.
Just how do you judge tuning? What is accurate tuning and how accurate does tuning need to be for a lute to sound nice?
Tuning is measured in cents; there are 1200 cents in an octave and thus 100 cents in each of 12 equal tempered semitones. Lutes are often tuned in unequal temperaments, but 100 cents is still a fairly useful approximation of a semitone.
So how accurate do you have to be? The Harvard Dictionary of Music states “A difference of six cents is discernable to very sensitive ears” (Second edition: Comma, shisma), implying to me that less than that is not discernable. It doesn’t make clear what the “sensitive ear” is hearing, but I think it means that if two notes are played separately anything less than 6 cents will appear to be the same. Certainly, if notes are sounded simultaneously one can hear much smaller imperfections, one cent or less, because the absence or presence of beats and their speed helps us to just more accurately.
When unisons, octaves and fifths are sounded, we can judge quite accurately if they are true, because these intervals require the second, third and fourth harmonics, which are quite loud in relation to the first harmonic or fundamental note, to be in tune with one another. The next harmonic, the fifth, produces the interval of a major third and by this time the harmonics are much weaker and the ear will accept a wider variation in the tuning of thirds.
Most people’s ears are used to quite a lot of out-of-tuneness; for example: all fifths in equal temperament are 2 cents narrower than true, major thirds 14 cents wider and minor thirds 16 narrower. Meantone temperaments also have a lot of out-of-tuneness that we accept, and the intervals that we do criticise, the so-called “wolf” fifths are around 20 cents out. So it looks like 2 cents is nothing much to worry about, 10 cents or even more is often OK and 20 cents is not. As an aside, opinions can seem to differ about the accuracy of tuning of false strings if people are listening different harmonics; the first or fundamental harmonic may out of tune, but if someone is listening to say the second harmonic they may judge the string to be in tune. It would seem to be best always to listen to the fundamental.
On a lute unisons and octaves must be as good as absolutely possible – within +/- 1 cent – and fifths, no more tempered than necessary, are also nice to have. Thirds will sound good even when they are more significantly out of tune. This means that the double octave between the first and sixth courses must be a good as you can possibly make it, the g on the fourth course second fret must be good, and the unison between the two strings of the second course must also be good. The other unisons between courses are also important, but it’s the one on the second course that is more of a problem if it’s wrong.
After that I think the tuning becomes a bit less critical. It’s remarkable just what the ear will accept; when you first pick up a lute it may well sound out of tune, but most ears rapidly adjust to a degree of out-of-tuneness in the same way as they adapt to different temperaments. In an ideal world every string should be perfectly in tune with a chosen temperament, but if, as usually happens, strings won’t settle on the desired note (due to the “clicking” over the nut discussed earlier) there are compromises which most ears will accept.
In temperaments from equal tempered through to sixth comma meantone and perhaps beyond, the second & third courses can be flatter and the fourth & fifths sharper than the required note. This is because movement in these directions takes the lute towards the purer fifths and nice thirds of meantone temperaments. In theory the degree of flatness and sharpness is quite small - -/+2 cents for the second & fifth courses and -/+4 cents for the third & fourth takes an equal tempered tuning down to sixth comma meantone, but in practice one often finds bigger variations are acceptable.
What is definitely not acceptable is to make compromises in the opposite directions, so the second & third courses must never be sharp and the fourth & fifths flat. That will sound bad.
In all of this, I wouldn’t want you to think that I’m advocating playing out tune, I’m not, but I am trying to highlight what is important and what can be to some extent disregarded.
Stress seems to affect our ability to hear pitch differences and it does it in very unfortunate ways. I find that any stress makes it much harder for me to tune, but sadly much more likely to hear imperfections. This can make you unhappy with the tuning of your lute and that uncertainness affect you concentration and cause lapses. I haven’t yet found a way round this, but it helps to remember that your audience is probably unaware of tuning errors that seem very obvious to you.
I’m a great fan of tuning meters and think they have revolutionised amateur lute music. They didn’t exist when I started and my tuning and the tuning of many amateurs was awful. Playing with other people was even more difficult because no ones lute was tuned to precisely the same standard and often a whole lute would require retuning to get to a common pitch.
It took hours. Duets were terrible, trios and quartets almost impossible. Those instruments that habitually played in ensembles, particularly viols, wasted hours before they played any music. You might think this an exaggeration, but viol ensembles really did often spend an hour before they started playing.
Tuning meters changed all that. People adopted pitch standards, often a=440, and instruments were kept fairly well intune at those pitches. At first the meters were all in equal temperament, but more sophisticated meters came with unequal temperaments and amateurs started to use various meantone temperaments. I remember the very first time, with Chris Goodwin actually, that we got a pair of different sizes lutes in a fair approximation of sixth comma meantone; it was a revelation.
At first sight there is a confusing range of meters available and I had thought of doing a really complicated analysis showing all the various models and comparing their features, but in the end it’s not that hard because most of the models are not really suitable for amateur lutenists. However let me summarise some the features and their implications:
Meters can be chromatic, meaning they cover every note of the octave, or dedicated to a particular instrument like a guitar and only playing the notes of their open strings. Clearly lute players need a chromatic tuner.
The cheaper meters are equal tempered like a modern piano or guitar. Even these cheap meters now automatically determine the closest note to the one you play, tell you what it is and how far out you are, usually in cents. The meter will show 50 cents flat and sharp of the note. After that it goes to the next note in the scale.
They’re usually accurate to +/- 1 cent, which I think is perfectly acceptable, but the cheapest may not “hear” as well as the more expensive. That said you can get fairly acceptable ones for as little as £10. Their choice of +/- 1 cent accuracy is interesting, because it shows, I think, the level of tuning accuracy necessary for an instrument to sound good. And when you compare that with the very small 2-4 cent adjustments for meantone it shows how accommodating the ear can be.
Many meters will have a ¼ inch line-in jack for electric guitars and this can be used with a clip-on mike for tuning in noisy places or next to other instruments. It would nice if plugging in an external mike switched off the internal one, but this often doesn’t happen. Perhaps they will start to do that.
These cheaper meters will be capable of a small range of pitch standard adjustment, say 435-446 Hz, which is OK for people only ever play in equal temperament and only have lutes in g.
You can use these cheap meters to tune unequal temperaments by tuning the -/+2 and -/+4 cents for the second/fifth and third/fourth courses discussed earlier. That will allow you tune in sixth comma meantone, but it’s fiddly, the meters seem to hear better at the exact centre of their range and it definitely does not work with lutes of different sizes. Those lutes can sound really nasty if you try to use an equal tempered meter to get meantone.
Unequal temperaments and lutes in pitches other than g are much easier if you get a more expensive meter. For most amateurs there’s really only one manufacturing that makes affordable meters and that’s Korg.
Their meters have a number of unequal temperaments built-in, but are not programmable for any additional temperaments. Renaissance lutenists tend to want sixth comma meantone, and although this is not one of the pre-programmed ones, Vallotti and Young is and this is the same as sixth comma for the white notes only. This means that a g & d lutes are fine because all their open strings are white notes, but most other lutes are not. There is a work round, because these meters have a wide range of pitch adjustment from 349Hz to 499Hz and setting the pitch to 392 for f, 370 for e, and 494 for a and then pretending they’re g lutes works fine for equal lutes, but again not for unequal ones.
After that, I think you’re into much more expense and probably these meters are more suitable for professionals. It’s interesting that I should say that they’re for professionals, because a few years ago professionals wouldn’t dream of admitting they used meters, but now often use them in public when the occasion demands. These meters are like the VioLab Pitchman and Peterson Strobe Tuner reviewed in the Lute issue No. 74 and if you want more detail I suggest you read those reviews.
One point of interest here is that the accuracy of the Peterson is rated as +/- 0.1 cents. I find it quite difficult to get to within the +/- 1 cent rating on the Korg meters and don’t find that a problem for my ears.
When you look at the cheap to medium priced meters that suit amateurs there are a few names we’ve never heard of doing really cheap ones then Korg, Boss and Seiko. When you get into the detail the latter two don’t seem to me to be really suitable because they don’t offer any unequal temperaments or the wider range of pitches that we need. Their products seem to be mainly aimed at guitarists.
So that leaves as Korg and when you exclude their ones for guitarists there are just a few options.
If you want a simple meter, the Korg CA-30 or CA-40, the newer model, is probably fine, but you may want to buy a Korg CM100 or Seiko STM230 clip on mike as well. You might also consider the simple clip on tuner/mike the Korg AW-1 which avoids the trailing wires of a separate mike.
If you play in unequal temperaments and/or pitches other than g, get a Korg OM12 or the new OM120, both of which are supplied with a clip on mike (BTW I don’t think you need to be tempted to get a mains unit for these. The batteries seem to last 6 months or more, as long as you avoid using the light.)
So that’s it, simple really:
• Make sure your lute and strings work well.
• Compromise on the tuning of the inner courses if you have to and don’t worry if you think you’re not in tune when you’re under stress.
• Get a meter of your choice and don’t be embarrassed to use it.