Moving the whole nut for better intonation ?

Please put your pickup/wiring discussions in the Electronics section; and put discussions about repair issues, including fixing errors in new instruments, in the Repairs section.
Post Reply
Chris Stadler
Posts: 4
Joined: Tue Jun 07, 2022 6:06 am

Moving the whole nut for better intonation ?

Post by Chris Stadler »

hey everyone .. i'm new here :oops: :P

been building guitars for about 6 years and just recently finished my first acoustic build. i prefer to build multiscale guitars. I'm pretty sensitive to intonation and on my electrics i can achieve pretty satisfying intonation with low action and almost no truss relief and bridge/saddle compensation only. I cut my fretboards and slots on the CNC and am pretty OCD about fret leveling and crowning so it's pretty much down to bridge and nut intonation to gain any further improbements.

On the acoustic however where super low action doesn't really work for my playing style getting it as close as i can get it just isn't close enough for my ears. i also play a lot of fingerstyle and open tunings which can amplify the intonation errors all over the fretboard. i usually get better intonation (for my ears) by sweetening the tuning in some way which works great for individual songs and chord combos but not as an overall improvement ... so i was wondering if there is another way to get a bit better compromise by moving the whole nut.

anyways, long story short my question is that i have heard through the grapevine that some guitar builders shorten or lengthen the fretboard at the nut end a tiny bit to get a better intonation compromise on the low frets with bridgesaddle compensation only. So not a fully compensated nut like the earvana or feiten thingy but just moving the whole nut a bit forward (or backward?) essentially shortening or lengthening the scale length a tad while keeping the fret positions for the original scale length. Someone in a now defunct forum once told me that PRS and Martin do this and that it was a common trick luthiers and manufacturers have been using for ages but i can't remember the specifics of that discusssion.

I can't really find anything on this anywhere and am wondering if someone here has heard of this and does this on their guitars ? the concept i think does make sense somehow since the intonation problems come from the string lengthening when fretted and moving .. so shortening the scale length a bit would probably counteract that effect .. the question is wether this does not create more problems than it solves :lol:

Note: The reason this is poppping up into my brain again now so many years after i heard this rumor is that i glued the bridge a bit too far toward the nut and am running out of space on the saddle to hit the intonation point i need. I'm about as far back as i can so if i want to claim the few cents left to match intonation at the 12th i'd have to move the bridge (the fretted notes at the 12th are just a tiny bit sharp compared to the harmonic and open string). On my electrics i never cared to investigate this issue further since i always manage to get a satisfying intonation.
Mike Conner
Posts: 235
Joined: Wed Feb 08, 2012 8:12 pm
Location: Murphy NC
Contact:

Re: Moving the whole nut for better intonation ?

Post by Mike Conner »

In my builds I locate the nut 0.013" 'forward' (towards the 1st fret). I do the standard calculations for fret positions, then reduce the nut to 1st fret distance by 0.013". The fretboard still is located so that the 14th fret is at the neck/body joint.

Some acoustic setups have achieved improved intonation by using this method, especially older guitars built in the 70's or early 80's, and this helps with intonation issues when using a capo. It's pretty easy to do - I have a razor saw with a 0.013" kerf, so I can use the nut as a guide and saw it free from the end of the fretboard. A little clean up with a fine file and the nut can be reglued or replaced with a new bone nut if the original is plastic.

Hope this explanation is helpful!
User avatar
Barry Daniels
Posts: 3186
Joined: Thu Jan 05, 2012 10:58 am
Location: The Woodlands, Texas

Re: Moving the whole nut for better intonation ?

Post by Barry Daniels »

My last few acoustics have gotten the shortened fretboard approach and I usually took off 1/32". Seemed to work pretty good. I know there are discussions of this technique on this forum if you wanted to search for them.

Another approach worth investigation is compensation at the nut of each individual string. Stephen Delft wrote the definitive article on nut compensation and it is here on the MIMF. I will post a link to that. I used this technique on a Martin that had a really bad low E intonation and it worked well.

https://www.mimf.com/nutcomp/
MIMF Staff
Alan Carruth
Posts: 1265
Joined: Sun Jan 15, 2012 1:11 pm

Re: Moving the whole nut for better intonation ?

Post by Alan Carruth »

Pressing the strings down to fret them increases the tension, and pushes the pitch up. If you plot out the amount that the pitch goes sharp it's only a little at the first fret, but it rises from there. Each string gives a somewhat different chart of sharpness, which is pretty much a straight line that rises from the first fret on up. The lines start at slightly different heights at the first fret, and have different slopes.

Normally we shift the saddle back away from the nut to compensate for string stretch, usually by comparing the 12th fret overtone with the fretted note, since they should both be one octave up from the open string pitch. This doesn't do much for the sharpening of the lower fret notes however, since the proportional change in length is so small for those. Shifting the saddle back essentially changes the slope of the 'sharpness' line. By moving the nut back far enough you can get the 12th fret note to be an exact octave, but the notes on the lowest frets will still be sharp, and the ones above the 12th will be flat.

You could get the first fret note to be in tune by shifting the fret back toward the nut, but that won't help the second fret note. For that you shift the second fret back, and so on. You could get all the notes to play in tune by shifting the frets back, but there's a problem: since each string behaves differently you'd have to shift the frets by different amounts for each string, so the frets would end up crooked. As it turns out, shifting the nut toward the first fret by enough to get the first fret note in tune drops all of the notes on that string by about the same number of cents: the whole sharpness line is shifted downward.

So; by shifting the nut forward toward the first fret you can drop all of the notes on the string down by a given amount, and by shifting the saddle back you can flatten the 'sharpness curve'. With the right adjustments you can get the first fret note in tune, and the slope of the sharpness line flat, so all the notes are in tune. In theory.

Actually, in practice this works quite well. You start by shifting the nut forward enough to get the worst string (usually the low E) in tune at the first fret, and then cut the leading edge of the nut back to get the other strings right. The saddle gets shifted back in the usual way. Note that compensating one end means you need less compensation at the other. One strategy that works pretty well, in fact, is simply to take the usual saddle-only compensation and split it between the saddle and the nut. If you needed 4mm of saddle-only compensation to get the 12th fret overtone right on the 6th string, then move the nut forward (toward the first fret) by 2mm and the saddle back by 2mm. It will be pretty close. Almost anything (within reason) that you do along these lines will probably be an improvement.

There have been several treatments of this in the past few years, with about the best having been set out by Trevor Gore in the books he and Gilet published. They are not inexpensive books, but worth it.

A final note: 'intonation' is measured by reference to the standard of whatever temperament you're using. On the guitar the default is '12-tone Equal Temperament'. It's mathematically impossible to make a a temperament that produces 'pure' tones in all intervals. We use 12-tone ET because al the semitones are the same size, so the frets are straight, and all the major keys are 'out' in the same way, so we can modulate freely and sound just as bad in every key. Once you start to notice the deficiencies of 12-tone ET there is no generally acceptable cure. :(
Chris Stadler
Posts: 4
Joined: Tue Jun 07, 2022 6:06 am

Re: Moving the whole nut for better intonation ?

Post by Chris Stadler »

hey thanks everyone, glad i' finally starting to get to the bottom of this :) so shortening the fretboard at the nut is 'a thing' ... it's weird that none of the gazillion guitar building videos and articles/posts i read mention this. the more i think about it the more it makes sense ...
Mike Conner wrote: Wed Jun 08, 2022 9:35 am In my builds I locate the nut 0.013" 'forward' (towards the 1st fret). I do the standard calculations for fret positions, then reduce the nut to 1st fret distance by 0.013". The fretboard still is located so that the 14th fret is at the neck/body joint.

[…]

I have a razor saw with a 0.013" kerf, so I can use the nut as a guide and saw it free from the end of the fretboard.
that's a great idea to use the nut as a guide for the saw, i have a collection of japanese saws, i'm sure i'l find one that's an appropriate width
Barry Daniels wrote: Wed Jun 08, 2022 10:08 am My last few acoustics have gotten the shortened fretboard approach and I usually took off 1/32". Seemed to work pretty good. I know there are discussions of this technique on this forum if you wanted to search for them.

Another approach worth investigation is compensation at the nut of each individual string. Stephen Delft wrote the definitive article on nut compensation and it is here on the MIMF. I will post a link to that. I used this technique on a Martin that had a really bad low E intonation and it worked well.
so 1/32" is 0.03125" right (as a european i'm always super confused by the whole imperial fractions :lol: )? so from Mike and your distance that's be somewhere from 0.33mm to 0.8mm ... that's much less than i thought so good to know to start small, i'd probably have taken a millimeter of to start with :lol:

as far as the fully compensated nut goes, i have a 5mm wide nut so i think i'll make a new one and try some individual string compensation once i've moved the nut and see how that works.

Alan Carruth wrote: Wed Jun 08, 2022 12:28 pm
A final note: 'intonation' is measured by reference to the standard of whatever temperament you're using. On the guitar the default is '12-tone Equal Temperament'. It's mathematically impossible to make a a temperament that produces 'pure' tones in all intervals. We use 12-tone ET because al the semitones are the same size, so the frets are straight, and all the major keys are 'out' in the same way, so we can modulate freely and sound just as bad in every key. Once you start to notice the deficiencies of 12-tone ET there is no generally acceptable cure. :(
Thanks Alan that all makes perfect sense and one of the clearest explanations of the matter i have read so far :) . on your final note: I've been fascinated by the whole 'intonation is always a compromise' thing since i was young .. i never could warm up to electronic tuners, always needded to do fine tuning by ear depending on what i was playing and never understood why certain chords or intervals sound off while others are perfectly in tune. A couple years back i attempted to tune my old CP-70 piano becuase i couldn't find anyone who would tune it (it's a very short scale piano and several piano tuners had a go at it and couldn't get it to sound good) ... that was also a revelation in tuning science .. after that i tuned my even older uprigth piano from the 50's ... and that needed a completely different stretch tuning.

i then started to do my own setups on all my electrics and when i started building electrics i was amazed how much better a guitar can sound when properly set up and intonated. so i had it all down until i started with the acoustic .. that's so much more complicated to figure out the bridge position precisely (especially for a custom multiscale) since i have only a limited adjustment range at the saddle. the need for higher action and thicker strings (i use 9.46 or 10-52's on my electrics, on the acoustic 11's and 12s sound much better) needs much more compensation and once it's glued down and finished there is no gracefull way to move it.

I have read all i could find about calculating compensation but i am severly hampered by my extremely poor understanding of mathematics. From what i've gathered and the tests i made with excell and online calculators i found the problem is that in order to calculate the compensation distance for either nut or saddles one needs to know the exact material aand deformation properties of the strings metals and windings. I have tried calculating with these by using the exact strings i found in the tables to figure out the location for my bridge (i think it was using some D'addarios who seem to be the only ones to publish this data) but the resulting compensation was completely off so i must have done something wrong or not understood something about the process.

here's my plan:

1. shorten the fretboard with shims to test the concept and figure out what a good distance would be to shorten for each string (keeping in mind that i will keep the bridge intonation as is since i don't want to move/reglue the bridge). so essentially trying to correct the slight remaining sharpness of the fretted notes by moving the nut intonation points.

2. shorten the fretboard for real by an average amount leaving room for individual nut intonation, probably the farthest i need to correct the sharpest string

3. make a new nut and then cut back the apropriate amount for each string that can benefit from additional nut compensation (this might take a couple of different blanks and trial and error)

not sure what the appropriate way to share pics here is and if it's ok to use an anonymou filesharing site like this, but i thought it might be helpful for the discussion to show the guitar/bridge in question. It's a pinless bridge and the string break angle is a bit steeper than i wanted .. thought i could get away with super low action like on my electrics, had i known that i needed way higher action i would've moved the anchor points back a bit to have a slightly shallower angle. but it sounds and plays great so i stopped obsessing over it ...

Image

Image

Image

i also found this site quite helpfull to understand the issues involved although some of it is way over my head: https://www.liutaiomottola.com/formulae ... sation.htm
Chris Stadler
Posts: 4
Joined: Tue Jun 07, 2022 6:06 am

Re: Moving the whole nut for better intonation ?

Post by Chris Stadler »

looking back at the pictures it just occured to me that 'theorethically' because it's a multiscale i can't just take an equal amount of at the nut, at least not parallel to the current nut line, it would have to be at an angle to account for the multiscale fan angle (the 'straight fret' is at the 5th fret) ... wouldn't be a big deal to cut this minute angle on the cnc during the build phase but not on a finished instrument nor by hand ... but it'll be so minute that shortening parallel or at an angle won't really matter i guess. i think removing the nut (it's not glued down anyways) and using a file or a squared piece of hardwood with sand paper to take of some of the fretboard end will be the way to go.
Alan Carruth
Posts: 1265
Joined: Sun Jan 15, 2012 1:11 pm

Re: Moving the whole nut for better intonation ?

Post by Alan Carruth »

Sometimes the simplest fix for that sort of thing is a shim under the string in front of the nut. It depends on how kludgy you want to be, but that can help you figure out how much to trim off the fingerboard.

My math chops are pretty weak as well, which is why I spend so much time doing experiments.

One of the nice things about Gore's treatment of the issue in his book is that he gives all the math, and the physical explanation, and goes into a number of ways to find the right values without having to do extensive calculations. He himself takes down the exact pitches of all the fretted notes on every string for each guitar, and uses a curve-fitting program on the computer to calculate the best way to compensate both the bridge and the nut. You can go as deep as you want with this, but any effort is worthwhile.

One simple rule that helps a lot is to remember that one semitone = 100 cents in pitch. If you divide the calculated distance from the nut to the first fret by 100, shifting the take-off point on the nut by that much should change the pitch by one cent at the first fret.

One thing I did was to make up a monochord rig to test strings. It's a beam of wood with two frets at the 'typical' locations of the 11th and 12th frets. These are height adjustable to get the right action for different strings. There is also a first fret that can be adjusted in position to be the correct distance from the 12th fret, and an adjustable nut and saddle. The saddle has a piezo pickup built in, so I can plug it into a tuner. This allows me to use the rig with nylon strings as well as steel.

In use, I set up all of the distances and action for the fret scale I'm interested in, and use calipers to find the distance of the nut and saddle from scribe marks. A string is mounted and brought to pitch, and the fretted pitch at the 12th fret is compared with the open string. The saddle is adjusted to get get that to be a perfect octave (it's possible for the overtone to be sharp in pitch from the octave, so I don't use that). Once that's set I see if the first fret pitch is exactly a semitone above the open string, and adjust the nut location to get that right. Then I go back to the saddle and re-adjust that, which throws out the nut intonation, and so on. After a few rounds of adjustment I'm chasing my tail, and I note down the offsets in my notebook. I take that string off and do the next one the same way. I've found that re-doing the string will usually give somewhat different measurements, so I do each string in the set three or four times, and average the results to get the numbers for the setup. Each end affects the other and small differences in fretting pressure and location also make a difference. Gore uses a mechanical setup with a weight to press down on the string, but I found that even that was not totally consistent.

One difference between the beam, or a solid body guitar, and an acoustic, is that movement of the top at the bridge can change the intonation of the string. When the bridge is moving the string 'sees' a stationary point that is a little in front of or behind the top of the saddle. If one of the resonances is close to a partial of a fretted note it will push that partial either sharp or flat. Since the 'main air' resonance, which causes a lot of top motion, is often close to G on the low E string it can shift the fundamental of that pitch noticeably.

In one instance like that, one of my student's guitars ended up with the 'air' pitch a little lower than A=110. This pushed the the open string pitch a bit sharp, necessitating a slightly lower tension than it 'should' have had to show the right pitch on the tuner. None of the fretted notes 'felt' that top motion, so they were all a bit flat. In a case like that you might be able to adjust the nut and saddle to be satisfactory, but the real cure is to adjust the 'air' resonant pitch if you can. Nothing on the acoustic guitar is 'simple'... ;)
Chris Stadler
Posts: 4
Joined: Tue Jun 07, 2022 6:06 am

Re: Moving the whole nut for better intonation ?

Post by Chris Stadler »

thx Alan, building a rig to investigate understand the relationships of nut and saddle compensation better is actually a great idea, i think it would do me much good to be able to play around with nut and saddle adjustment easily and understand how intonation on thew high and low frets is affected and which compromise i like best.

you've given me much to think about :P

in the meantime i spent some time measuring the intonation offsets with a strobetuner and write down all other parametrs of the guitar to get a better understanding where i'm at before i start messing around with it. the compensation amount is only rounded to half a millimeter since i don't have the tools to measure more precisely (my longest precise ruler is only 60cm long so need to get a lo ger one at some point).

As it is now when i tune each string only with the open string and a strobe tuner to standard pitch the intonation on the lower frets is pretty nasty and i have to sweeten the tuning to have a good intonation balance across the first 14 frets or so. But when sweetened a bit to be in tune on the lower frets the intonation over the first 10 frets or so is pretty spot on. No one who's played the guitar so far has complained or even noticed it but personally i need to tune around a lot until i find a nice intonation compromise for stuff that plays all over the neck with open and fretted strings together.

Image
Post Reply

Return to “Flat-Top Acoustic Guitars and Bass Guitars”