Welcome to one of the most active flamenco sites on the Internet. Guests can read most posts but if you want to participate click here to register.
This site is dedicated to the memory of Paco de Lucía, Ron Mitchell, Guy Williams, Linda Elvira, Philip John Lee, Craig Eros, Ben Woods, David Serva and Tom Blackshear who went ahead of us.
We receive 12,200 visitors a month from 200 countries and 1.7 million page impressions a year. To advertise on this site please contact us.
I have just read in Guitarra magazine (http://www.guitarramagazine.com/Intonation) that compensation for string behaviour can also be necessary at the nut end, which may need to be brought forward to reduce sharpness in the lower notes.
Does anyone do this?
I've read what I can about saddle compensation but in this forum you only seem to be interested in the width of your nuts rather than how far they are away from your saddle (well, you started it... )
I don't think I can be too scientific about the saddle compensation due to my working environment - I'll just slap a couple of extra mm on - so I guess nut compensation will be a bridge too far. (A BRIDGE TOO FAR!!)
I have just read in Guitarra magazine (http://www.guitarramagazine.com/Intonation) that compensation for string behaviour can also be necessary at the nut end, which may need to be brought forward to reduce sharpness in the lower notes.
Does anyone do this?
It's normally only done on classic guitars. I used to make compensated nuts but it was an enormous hassle and required getting my fingers entirely too close to a spinning router bit. The benefit is mainly to the 3rd string. You don't actually shorten the 3rd string but shift the entire string toward the bridge (make the nut wider..make the saddle more narrow) so that it plays more in tune. You can test the effect yourself by sticking something under the 3rd string just in front of the nut and checking the octave harmonic.
Yes, that article was about classical guitars I think. Maybe if the nut height is greater as I guess it is with CG, it might be necessary but I'm happy to follow in your footsteps, and not do it!
Hey, There are various methods but the best I've seen is in a new book by local Ozzie Trevor Gore. It models the characteristics of the fretboard, action, relief, playing pressure and the properties of the brand and gauge of strings and runs it through an optimisation program in a spreadsheet that aims for a nut and saddle compensation that minimises the out of tuneness at each fret for each string with great results. The more sensitive your guitar, the more likely the errors at some frets will be a problem. In the book, simply chopping 0.5mm and even up to 1.5mm from the nut end of the fingerboard is a better compromise than just adding normal compensation to the saddle only. He plots the results so you can see the improvement and shows how to do your own guitar so you can keep a track of any changes you make. Also shows a jig if you still want to individually compensate each string so your fingers are nowhere near any spinning blades, which seems like a very a bad idea for any player/builder or just a fan of fingers. Cheers Dom
In the book, simply chopping 0.5mm and even up to 1.5mm from the nut end of the fingerboard is a better compromise than just adding normal compensation to the saddle only.
The problem with this method is that as soon as you fret a string the compensation for that string is lost. The only advantage of compensating the nut is when you play a fretted string next to an open string. That's why we switched to a compensated fingerboard (the fret spacing is altered).
Nothing works 100% which explains why classic guitarists frequently retune when switching keys. The guitar by nature is basically slightly out of tune which is part of it's charm in my opinion.
John, Does this mean you're using a scale that is tempered? Like as in the olden days when lutes and viols used tempered scales?
I suppose you could say that. The fret spacing is altered to compensate for string stretch when fretted. It's rather touchy until you figure out where to place the saddle. For instance the 660mm scale length is actually 657mm on flamencos and 658.5mm on classics. Rather than tempering I'd call it spreading the inharmonicity. As Miguel de la Maria has noticed it works very well.
The only advantage of compensating the nut is when you play a fretted string next to an open string
John, I tend to agree that nut compensation is not worth the trouble but the progression open G to G# (not something you play very often) benefits greatly from a compensated nut. That particular one-semitone interval is usually quite off on guitars because of the string stretch.
Flamenco guitars are very often played with a capo.. That makes nut compensation even more questionable
Almost as much as players that just throw the fingers into the frets and take no care with the ear to intonate when they play. compensated perfect low action guitars with lazer selected perfect strings only for those guys.
Rather than tempering I'd call it spreading the inharmonicity. As Miguel de la Maria has noticed it works very well.
I thought that was what an Equal tempered scale does, theoretically. Distribute the few cents of out of tuneness evenly. Tempered scales change the relationships between frets to make certain scales play truer or for stressing certain tonic relationships to give greater weight or impact to, for example, I-V or I-IV resolutions, etc. ( Mozart's time especially liked this)
but what you're saying is not any of that and that you are making the scales smaller to compensate for string stretching? Does the fingerboard play in tune in all scales?
I thought that was what an Equal tempered scale does, theoretically. Distribute the few cents of out of tuneness evenly. Tempered scales change the relationships between frets to make certain scales play truer or for stressing certain tonic relationships to give greater weight or impact to, for example, I-V or I-IV resolutions, etc. ( Mozart's time especially liked this)
but what you're saying is not any of that and that you are making the scales smaller to compensate for string stretching? Does the fingerboard play in tune in all scales?
The frets are adjusted slightly. So a 660mm scale actually measures 328 from nut to 12 and longer from 12 to saddle. We had to experiment to find the right overall measurement on flamencos and classics and arrived at what I mentioned above. It works very well in all keys but there's really nothing one can do about the basic "out-of-tuneness" of the guitar besides spreading it out over the entire board. Call it tempering if you like. Inharmonicity is that hard to describe element that makes things that are slightly out of tune sound better than things that are in perfect tune.
Thanks guys that's not how it works but it is a common mistake. Firstly, nut compensation effects intonation at all frets, not just the open string. So using a cappo does not eliminate the effect of nut compensation just as playing a bar cord does not.
This is because you tune the open string using the new nut position and each fret is in a different position relative to the uncompensated nut. Imagine that the fretboard is cut longwise into 6 segments, nut compensation on the G string of 2mm is like you have moved the G string segment and all its frets towards the nut by 2mm. Instead the nut is moved but it the same effect. And using this method accurately both nut and saddle compensation must be calculated together because changing one means the other one is wrong. The mathematical optimisation model does this for us and cycles through all the possible combinations of nut and saddle comp and finds the one that minimises the sum of the absolute values of the cents error at each fret and does it for each string. So lets look at the facts. A normally compensated steel string guitar that's been properly set up, good action including at the nut, bit of relief etc etc, will have perfect in tune open strings and be in perfect tune at the 12 fret. However, all the notes between the nut and the 12th fret are sharp to some degree with the greatest degree of inaccuracy around fret positions 1 2 and 3. The E string at the first fret has an error of over 6 cents, the A about 6 and the other strings about 4 cents. This is enough to be heard, particularly when playing some intervals that include an open string as the intonation error combines with the error built in to the equal temperament scale. Once we are past the 12th fret all notes become increasingly flat and again the E string become flat by around 7 cents at the 18th fret. Subtracting 1.5mm from the nut end of the fretboard can reduce the errors around the first few frets by almost 2/3rds. That in itself is a vast improvement.
However, fully modelling the variables for the guitar and the player including strings, action, relief and playing pressure, the intonation errors can be virtually eliminated. The error at the first fret is reduced to around 0.5 cent and at every other fret, the error is less than 0.1 cent. This is a significant improvement over standard saddle compensation and will be a delight to those with sensitive ears.
This is because you tune the open string using the new nut position and each fret is in a different position relative to the uncompensated nut. Dom
So Dom, do you layout the frets then saw a piece off the nut end of the fingerboard ? That's what I did. This required a rather wide nut that was then shaped with a router to lengthen or shorten each string.
Does the above statement imply that you use an electronic tuner to tune the open strings?
In my opinion an electronic tuner is only good for establishing pitch on one string much like a tuning fork. After that the rest of the strings should be tuned using a variety of methods all based on what your ear tells you sounds good.
John, you just make a normal fretboard and take 1.5mm off the nut end. I build my own guitars so I make how I want. But I will usually glue the FB on with the nut sitting in its spot so I don't need a wider nut. The set up Trevor uses to shape the nut uses adjustable brass squares kind of like electric guitar saddles and once this is set up for your guitar the nut is clamped above and a simple pin router accurately cut the nut compensation. In this case you would need to take more off from the FB end and then fit the compensated nut so the distance to the 12th fret is correct.
If you are doing all this work you should use some form of electronic tuner. I run a signal into my computer which can be analysed. Or into Strobosoft. Once you have it set up correctly you can tune it how you want but since the intonation is close to perfect and we have eliminated all potential errors the best possible notes to tune the open string to will be exactly E,A,D,G,b,e. Not one Hz off. Also, when I am trimming body resonances of either the box T(1,1)1, the top T(1,1)2 or the back T(1,1)3, the measurements need to be pretty accurate. The gaps in Hz between E and F is less than 5 Hz and I want to be able to place resonances between scale notes so I need to accurately place it at 84.6Hz. The best way to measure these is with an electronic tuner. So unless you are tuning the open string by ear, any other method using harmonics or 5th fret positions will be inaccurate unless you are playing on a fully intonated guitar with properly calculated nut and saddle compensation. Not sure why it is not OK to use a tuner. A good guitar with no body resonance issues should sound sweet enough and not need the player to sweeten things up with tunings. Cheers Dom
John, you just make a normal fretboard and take 1.5mm off the nut end. I build my own guitars so I make how I want.
This sounds like what I was doing.
quote:
If you are doing all this work you should use some form of electronic tuner. I run a signal into my computer which can be analysed. Dom
Susan who is now the sole proprietor of Shelton-Farretta always uses a tuner then she hands me the guitar to test and I always re-tune it so that it sounds good. I guess it depends on what one grew up with but I just don't like electronic tuners. I always test our piano after the professional tuner finishes his work and inevitably find notes that must be corrected. Surprisingly he usually agrees with me.
I have one guitar with a compensated nut; the big advantage is that a capo does not throw the guitar out of tune very much, if at all. That's seems important.
Printable Version
) 
New Messages
No New Messages
Hot Topic w/ New Messages
Hot Topic w/o New Messages
Locked w/ New Messages
Locked w/o New Messages
Post New Thread