You are here

Intonation: Nut compensation

For all things relating to guitars, basses, amps, pedals & accessories.

Intonation: Nut compensation

Postby Stvy » Sat Sep 04, 2004 11:39 am

Hi,

I just recently got a peterson strobe tuner, and have reintonated my guitar. I did this by adjusting the bridge saddles based on the 5th and 17th frets.

The results were amazing the guitar has never sounded so sweet from the 4th fret to the 22nd.

If I intonate based on the nut and the 12th fret the results are so so all over the neck.

So clearly the nut is causing the problems... Upto 5 cents on the open D and G strings.

So I was thinking of making one of these:

http://mimf.com/nutcomp/

Since I can't but a left handed Earvana. Any thoughts?

Cheers

Steve
User avatar
Stvy
New here
Posts: 7
Joined: Sun Mar 02, 2003 12:00 am

Re: Intonation: Nut compensation

Postby Mats G » Sun Sep 12, 2004 11:59 pm

Hi Steve. Nut compensation theory and techniques like those presented by Stephen Delft, Buzz Feiten and others have gained a lot of attention in later years. Although interesting in many ways, and despite the fact that I know people who swear by their compensated nuts, I have yet to find one theory that convinces me.

Firstly, different compensation technique developers recommend slightly different measures and have slightly different ways of calculating these measures, which indicates there is a certain amount of arbitrariness involved.

Secondly, since the compensation is basically taking place at the nut, it can only deal with fretting related to open strings; as soon as you leave open string territory, e.g. when using a capo or barré chord, the compensation is in pratice overridden. Fretting with a barré is the equivalent of moving the nut, which means that you should move the compensation as well. This in turn means that to make a difference to the overall tuning, every single fret should be compensated - a technique that has been tried by numerous luthiers over the years, but has never caught on.

As a former guitar tech I find nut-related intonation problems mainly have to do with the nut slots being to shallow. They should be deep enough to make the string height over the first fret equal to the string height over the second fret when you hold down the string against the first fret. I'd say most factory-setup guitars have too shallow slots, probably because filing them down to the optimal height is a precision work that requires time, and time is money. And if you file down too far, you either have to replace the nut or fill the slot with some superglue and start over again, which again takes too much time. Better leave a safety margin, then.

A nut slot that is ever so slightly too high, forces the finger to press the string harder near the nut than higher up the neck to reach the fret. The extra force brings extra tension to the string, which sharpens the pitch. Of course, fretting a string anywhere on the fingerboard causes string tension, but if the string-to-fret height is consistent from the nut upwards (slightly increasing on the way up), this tension is effectively compensated for "globally" by intonating the saddle(s). In fact, the optimal solution is the old (not so fasionable today) "zero-fret" construction, which uses a fret - similar in size/height to all other frets - at the nut position and just uses the nut to guide the strings and give them the correct spacing.

Another thing to consider is the nature of the tempered tuning which has been used in western music since the 17th century, and which more or less makes all intervals on instruments with fixed notes (like guitars, pianos etc) slightly out of tune. This, of course, is nothing you can compensate away from, but in combination with a nut that is a bit too high, this often means that when e.g. getting an open position C major in perfect tune, the G# in the open position E major is way too sharp. If your nut slots are optimal in height, your overall intonation correct, and you still get these sharp notes you should ask yourself if maybe you are using excessive force when fretting. Letting go slightly on the pressure could be all that is needed to solve the problem.

Just my thoughts and experience.
User avatar
Mats G
Poster
Posts: 14
Joined: Tue May 20, 2003 11:00 pm
Location: Sweden
Mats G

Re: Intonation: Nut compensation

Postby Stvy » Mon Sep 13, 2004 12:29 pm

Hi,

Well I've gone and super glued some extra delrin strips to the D G and B strings on my MIA strat. I will cut a new nut completely with these compensations once I am certain that they are spot on after a few more string changes....

I have to safe the results are better than I could have imagined. All open string chords sound MUCH better, it is a very pleasing experience just to sit and strum the guitar randomly and listen. I am very happy. I can play chords all over the neck with open strings and they sound wonderfull.

I have set up my guitar with a very slight give in the neck, and with a high enough string at the nut to play slide. I needed a great deal of compensation on the open G string. Before which the open G string was 6 cents flat ( the D was 5 cents adn the B 4 cents flat. Now it is within a half a cent (the error of the strobe tuner) as are all the open string and frets all the way up the fingerboard...

I agree with you that no one theory suits all. I can almost gaurantee to you that with my string gauge (13-17-20-36-46-56) and my higher than normall action the Earvana nut would not provide as good a solution as I now have. I don't need any compensation on the low E A and high E strings. I believe this must be because of the high tension of these strings for the guages I use is counter acting the effects of the high action, high nut slot height. My testing has proven to me at least that you have to compensate the nut for the actual guitar, guitar, player, string gauge, and action. Anything else is a compromise.

This guitar is now the sweetest sounding I have ever heard.

The only downside is that the nut looks a bit messy. Although this will improve when I get around to cutting a custom one piece version.

Cheers

Steve
User avatar
Stvy
New here
Posts: 7
Joined: Sun Mar 02, 2003 12:00 am

Re: Intonation: Nut compensation

Postby zenguitar » Tue Sep 21, 2004 1:07 am

Just a quick comment. Mat G is right to explain that most intonation problems are due to poorly cut nut slots. So your starting point has to be a professional set-up. As Mat implied, there is no such thing as a perfect set-up, just the best compromise for the individual guitarist, and a decent guitar tech will set-up the guitar to meet your needs, not some abstract ideal. And, for our sins, we love an instrument that highlights the diffences between harmonic tuning and equal temperament tuning. In plain english that means that the G string can be as much as 12% sharp compared to a harmonic G, and all other notes on the G string are equally sharp. We get used to that and we all tune our G strings slightly diffently, to find the best compromise for how we play. But some guitars have subtle errors in their fretting, and if those errors mean a guitar plays slightly sharp, the G string will be intolerable when fretted (the same applies to all the strings, whether the guitar tends towards sharp or flat, but the G string is the one that highlights the problem most). In those cases a compensated nut can pay great rewards
User avatar
zenguitar
Moderator
Posts: 9251
Joined: Thu Dec 05, 2002 12:00 am
Location: Devon
When you see a fork in the road, take it.
Yogi Berra

Re: Intonation: Nut compensation

Postby octavedoctor » Sun Sep 07, 2008 8:13 pm

The whole issue of nut compensation is nonsense and I have yet to hear a sound technical explanation of it that holds water. The Feiten System in particular is a farrago of misapprehension about equal temperament and harmony, with a number of fundamental errors of logic and science.

As Mats G rightly says, I think most problems with intonation at the nut are owing to incorrect cutting of the nut slots. Most manufacturers leave these too high and few players bother to have a proper set up done.

As most guitars these days are CNC machined I don't think that errors in fret placement contribute significantly to this problem (except for Gibson, some of whose guitars use a fret placement factor that is closer to the rule of 18 than 2root12) but there is a perceptual factor which may contribute to the myth of "first fret sharpness".

If one plays a perfectly tuned equal temperament first position e chord you will observe that the note at the first fret on the G string sounds sharp; this is because in equal temperament the major third between the A flat on the third string and the E on the fourth is slightly stretched, by approximately 13 cents. By comparison the minor third between the A flat on the third string and the B on the second string is compressed by about 15 cents. The combination of the two, to someone who is not fully cognizant of the compromises inherent in the equal temperament system (and that's most guitarists i'm afraid :frown:) leads to the misapprehension that it is a misalignment of the first fret that is responsible and so the wild goose chase for a solution starts.

If you think about it, any rationale that is applied to the nut must also be applicable to every fret as well. Nut compensation can contribute nothing globally because if you put a capo on the first fret you immediately eliminate its effect and restore the uniform constant ratio geometric progression you started with ...

I don't think that Zenguitar's point about fret inaccuracy is without merit though as i have seen plenty of, um, handmade guitars with questionable fret locations, but more of an issue for me is that of badly dressed "flat-top" frets where the string sounds from the back edge of the fret instead of the crown of the bead. I don't think this is something that nut compensation could or needs to solve; a simpler solution is to just get the job done properly.
User avatar
octavedoctor
Regular
Posts: 115
Joined: Sat Sep 06, 2008 11:00 pm
Location: Bridgend, Cardiff and Bristol

Re: Intonation: Nut compensation

Postby octavedoctor » Sun Sep 07, 2008 8:28 pm

Stvy wrote:I have set up my guitar with a very slight give in the neck, and with a high enough string at the nut to play slide.

This will be the cause of the intonation error that causes you to seek this solution. Excessive arc relief plays havoc with intonation but if that is what you need to do to pay slide then it's a legitimate compromise. The Stephen Delft compensation can only make open chords in the first three positions sound better and only certain chords at that; more unfamiliar chords will sound worse, inevitably as the changes in intervals required for the chords that sound more in tune will be entirely wrong for chords of a different shape, however there is a definite placebo effect apparent in users of these systems who will swear that their guitars sound better even though, to an objective observer, they don't.

Myself, I feel that a better solution is to educate guitarists so that they can accept equal temperament the way keyboard players accept it. The problem is that keyboard players don't have to tune their own instruments and usually have a more sound musical education than guitarists, so they understand the compromises necessary in a way that guitarists don't.
User avatar
octavedoctor
Regular
Posts: 115
Joined: Sat Sep 06, 2008 11:00 pm
Location: Bridgend, Cardiff and Bristol

Re: Intonation: Nut compensation

Postby zenguitar » Tue Sep 09, 2008 12:58 am

Hi Octave Doctor, and welcome.

You've made some very constructive posts over the last few days, and it is always good to have someone with your experience on board. So please, feel free to keep on posting.

You are right, it is increasingly rare to find guitars with the frets in the wrong position. And yes, in most cases the problem has far more to do with the profile of the fret than the location. But even with CNC, tools get worn which can result in the front edge of the nut being slightly further away from the bridge. Making the actual scale length fractionally longer. A lot depends where the factory Quality Control puts the reject threshold as to how many actually get through.

However, to some extent, that is irrelevant. A decent tech could and should be capable of spotting where the problem lies. The essence of this thread is not about that, but about the benefits, or otherwise, of sweetened tuning systems like Buzz Feiten or Earvana.

My approach is quite simple. The modern guitar is a compromise. The location of the frets is based around Equal Temperament which averages out all the harmonically derived scales to deliver an average tuning. However, that averaging out works better for some strings than others, leaving some notes more out of tune than others. What the sweetened tunings do is to redistribute the out of tuneness differently. Still a compromise, but a different compromise.

All other things being equal, my preference is for a traditional nut. Not because it is in any way 'better', but because almost all of the recorded guitars over the last 70 years, which are our common references, have traditional nuts. However, where a guitar has a problem, or a player tends to play in keys where the traditional compromise doesn't work so well, I can make a compensated nut from 1st principles.

Andy :beamup:
User avatar
zenguitar
Moderator
Posts: 9251
Joined: Thu Dec 05, 2002 12:00 am
Location: Devon
When you see a fork in the road, take it.
Yogi Berra

Re: Intonation: Nut compensation

Postby octavedoctor » Wed Sep 10, 2008 11:57 am

I take your point about that, but my feeling is that any such adjustments should be done empirically rather than built in as part of the construction.
Anything else places limits.

My issue with Earvana, the Feiten system and the Fretwave guitars is that they proceed from a standpoint of ignorance, i.e. their progenitors have identified a problem but have insufficient understanding of the problem to realise that it is nothing to do with guitar construction and everything to do with how we perceive and define harmony. The True Temperament guitar is another animal entirely; Anders Thidell clearly understands the history and theory behind harmony (although I take issue with the idea that there is any such thing as true temperament when there is no constant value for tone and semitone) but I wonder what the point of it is. When almost every other instrument in the western musician's panoply except the violin and the voice is designed for equal temperament what point is there in producing a guitar that is fixed to a key?

I'm not sure what you mean when you say that the averaging of equal temperament works better for some strings than others. Modern string manufacturers produce strings of remarkable consistency and reliability. Certainly inharmonicity will create different perceived effects but this can be compensated for without altering the guitar's construction in any way. Arc relief will also affect the pitch accuracy at different points on the fingerboard; quite dramatically, in fact, and this is something that a lot of people forget about or are never aware of. The compensation applied at the bridge represents an increasing percentage lengthening of the string's sounding length the further up the neck you travel, however with the introduction of arc relief the fingerboard plane levels out as you go higher so that from the 12th fret on there is little change in the applied strain so that the higher frets are being overcompensated. The traditional method of intonating using the 12th fret and the associated harmonic as a reference results in the higher frets sounding flat however the discrepancies can be quite easily balanced out using alternative methods of intonating and the adjustable saddles of a modern electric guitar are really all you need to achieve this.
User avatar
octavedoctor
Regular
Posts: 115
Joined: Sat Sep 06, 2008 11:00 pm
Location: Bridgend, Cardiff and Bristol

Re: Intonation: Nut compensation

Postby lyounkins » Wed Sep 10, 2008 4:16 pm

While I'm no fan of the Buzz Feiten system, I think nut compensation does have merit.
Greg Byers has presented some research on the subject (http://www.byersguitars.com/research/Intonation.pdf) and his work is worth reading for more technical details. He created a model for fretting a string that includes travel stretch and fretting stretch. Travel stretch is the distance the string travels to meet the fret. Fretting stretch includes an additional distance required to actually fret a note with a player's finger. As might be expected, fretting stretch is player dependent and effects the amount of compensation required.
String inharmonicity effects (due to string stiffness) are also included in the model.
Although his results are presented for a classical guitar, I think similar measurements could be made for an electric guitar.
I hope this helps.
Larry
lyounkins
Poster
Posts: 12
Joined: Tue Apr 11, 2006 11:00 pm

Re: Intonation: Nut compensation

Postby zenguitar » Wed Sep 10, 2008 11:36 pm

octavedoctor wrote:
I'm not sure what you mean when you say that the averaging of equal temperament works better for some strings than others. Modern string manufacturers produce strings of remarkable consistency and reliability.

Sorry, I should have made myself clearer here. I wasn't referring to different makes, types, or gauges of string. I was referring to the individual courses on a guitar. The best example being the G-string. The equal temperament pitch being very different from the harmonic equivalent. Every guitarist has their own compromise tuning for the G-string. And it usually depends on the keys, chords, and voicings they prefer.

And as you say, ultimately you need to trust your ears when setting up. The theories and science informs what we do, but no more so than our experience and the flappy things on the side of our heads.

Andy :beamup:
User avatar
zenguitar
Moderator
Posts: 9251
Joined: Thu Dec 05, 2002 12:00 am
Location: Devon
When you see a fork in the road, take it.
Yogi Berra

Re: Intonation: Nut compensation

Postby octavedoctor » Wed Sep 10, 2008 11:37 pm

Yes, i read this while back; it's very good, although some of the conclusions about inharmonicity are based on some research by a couple of Swedes I think and I'm not sure that I have confidence in the conclusions because outside of the lab it appears that most people tend to assess enharmonic tones as slightly flat, which is perhaps why piano tuners tend to pull up the pitch of higher notes to compensate.

Fretting stretch isn't something I regard as a sufficiently constant variable to be worth compensating for. It will vary from player to player, fret to fret and even the same player will exert different levels of pressure on the string at different times. Someone once attempted to use the effect of a capo to demonstrate the effect of fretting stretch so I challenged trhem to reproduce the effect with a single finger bar; of course, they couldn't do it, because over six strings it is virtually impossible to compress the strings that much with a straight finger. A single digit might be able to exert enough pressure on one string to distort the pitch significantly but if you build compensation in for this (even if you could) then you have caused problems for the same string if fretted as part of a barre chord.

The best solution to fretting stretch is not to find a way to compensate for it, but to promote good technique which involves only the minimum pressure being required to bring the string in contact with the fret. The trouble is that guitarists are often lazy musicians and want instant technical solutions in place of a developed skill set.

I suppose what I'm trying to say is that people worry too much about it. Guitars sound OK if they are intonated properly and tuned to equal temperament. Getting obsessional about accuracies of a two or three cent resolution when a guitar string has a stability window of up to 12 cents is a bit silly, in my opinion
User avatar
octavedoctor
Regular
Posts: 115
Joined: Sat Sep 06, 2008 11:00 pm
Location: Bridgend, Cardiff and Bristol

Re: Intonation: Nut compensation

Postby Saberthorn » Thu Oct 17, 2019 2:12 pm

If you use a capo, you don't need compensation, because the frets are theoretically all the same height. If the nut has a zero fret, you wouldn't need compensation. The compensation is for the fact that the nut slots are higher than the frets.
Saberthorn
Posts: 1
Joined: Thu Oct 17, 2019 2:10 pm

Re: Intonation: Nut compensation

Postby Watchmaker » Thu Oct 17, 2019 3:11 pm

This conversation is nuts!

I have noticed most manufacturers cut nuts to a reasonable compromise that allows/expects/accommodates the modicum of customization required to achieve a personally satisfying solution.

There's no meaningful way to mass produce items that suit everyone's perceptual biases. If general relativity has any meaning at all, the cures are many and none, depending on the observers location in time and space which is neither here nor when...point being the first fret usually intonates worse than other frets - at least we perceive the dis-harmonic content arising from variations in scale (both mechanical and musical) with greater acuity.

As pointed out, the benefits of compensating the nut are context dependent, up to the point that one begins fretting after which it is moot.
User avatar
Watchmaker
Frequent Poster
Posts: 625
Joined: Tue Apr 12, 2016 11:00 pm
Location: Upstate NY, USA
Take my advice, I'm not using it.

Re: Intonation: Nut compensation

Postby Music Wolf » Thu Oct 17, 2019 4:08 pm

Wow! A second revival / third life.

I've just realised that this thread started before my son was born and the little bugger is now, in a most disrespectful manner, towering over me.
User avatar
Music Wolf
Frequent Poster
Posts: 1770
Joined: Fri Feb 17, 2006 12:00 am
Location: Exiled to St Helens

Re: Intonation: Nut compensation

Postby Sam Spoons » Thu Oct 17, 2019 4:21 pm

Welcome to the forum sabrethorn :thumbup:

This is an old thread though and that point was first made way back in 2004.

To add my 2p, zero frets, IME, are always higher than the rest primarily to stop the non speaking part of a fretted string rattling against the frets behind the fretting finger (same reason a nut is cut slightly higher than the first fret).
User avatar
Sam Spoons
Jedi Poster
Posts: 10330
Joined: Thu Jan 23, 2003 12:00 am
Location: Manchester UK
Finally taking this recording lark seriously (and recording my Gypsy Jazz CD)........

Re: Intonation: Nut compensation

Postby Wonks » Thu Oct 17, 2019 5:24 pm

Too may loose terms here like 'compensation'. Compendation where? At the nut or the saddle (intonation adjustment). You'll always need intonation adjustment because the thicker teh string, the longer the non-vibrating part of the string is. and so its sorter overall working length. And different strings have different mass/unit length and people fret with different pressures which puts more/less tension on the string. The lower the nut slots are, the less extra tension you need to push the string down to the fret (especially on the first few frets) and the higher the bridge height is, and the higher the subsequent action, the more pressure you need to push down on the string to fret it and so raise the tension compared to the open string position.

Even an open string when plucked will rise up and down in pitch slightly, so it's impossible to ever get a guitar fully in tune. Whatever system you use, it's a compromise.
User avatar
Wonks
Jedi Poster
Posts: 9734
Joined: Wed May 28, 2003 11:00 pm
Location: Reading, UK
Correcting mistakes on the internet since 1853


Who is online

Users browsing this forum: No registered users