Official Luthiers Forum!

When compensating the nut , is the string compensation based off of compensated nut position ?

_________________
MK5acoustics.com

Wes, Can you please re-frame that question, then I'll have a go at answering it!

There's a variety of ways of going about compensating a nut and a variety of reasons you might want to be doing it, so it would be great to have a little more information. Thanks in anticipation.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

There have been discussions in the past which left me pondering. Suppose I compensate the nut .020 , should I use this Shortened distance from nut to 12th fret then double and compensate the bridge normally or add the .020 back into my bridge compensation?

_________________
MK5acoustics.com

OK. If you're using that approach, move the nut .020 towards the first fret then proceed as normal (i.e. as if you hadn't moved the nut), adjusting the saddle until you get zero error on the 12th fret between the harmonic and the stopped note. I'm assuming you're talking classical guitars here, as that approach will give a reasonable result. If you're talking steel string, it will give you a barely perceptible improvement.

If the result you get is still not sufficiently in tune for you, you may want to consider more complex options, e.g. http://www.goreguitars.com.au/main/page_10929_5.html or http://www.goreguitars.com.au/main/page_innovation_summary_nut__saddle_compensation.html

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

The nut compensation only has a real effect on the first fret or two. Beyond that, it's all in the compensation at the saddle. Read the article on Greg Byers' website. It's a bit simplified, but straight forward and easy to understand. Averaging works fine though, and it works for me. I have plenty of weaknesses in my building, but the folks who have played my guitars, seem to like the way the compensation works. I use a nut compensation of about .5 mm, which is about the same as what you said - .02". Then, I double compensate my saddle which is 2.5 mm thick, a la David LaPlante.

You have to look closely, but you can see it here. What I like about it, is that it works for both nylon and carbon strings without having to fiddle with additional nuts and saddles. It's just enough to keep it close all the way up the board. Not perfect, but close.

_________________
Waddy

Photobucket Build Album Library

Sound Clips of most of my guitars

Maybe I'm misunderstanding but it's a common (but wrong) assumption that compensating the nut only effects the open note. If compensating the nut causes one to reposition the saddle (and it should), then it effects all notes, not just the ones close to the nut.

On steel strings, instead of compensating each string at the nut, I come in about .020" on the bass side and about .012" (half the fret saw kerf) on the treble. Works good enough for me. Unfortunately, I can't even tell you what effect it has on the saddle because I got there by trial and error and it's all built in to my jigs now and I don't have to measure.

_________________
http://www.chassonguitars.com

Regardless of what is done at the saddle, nut compensation has an exactly equal effect at every fret. If you flatten the first fret note by three cents, you are also flattening every other note on that string by three cents. Compensation at the saddle has a progressively greater effect as you move up the neck. You can combine these two to come closer to correct overall intonation. What doesn't work is dialing in the twelfth fret with saddle compensation, and then adding compensation at the nut to bring the first fret into tune. The need for nut compensation is really sensitive to action height at the first fret (nut slot depth). Don't start compensating the nut until you're sure you've got the slot depths where you want them.

Ok, here's one...
I set up a guitar last night that had 1.5 mm taken off the FB at the nut (per Gore's book). I compensated the saddle as normal to about 1-2 cent each string at the 12th fret. The first fret plays dead in tune, but then each fret progressively get flatter til about the 7th, then starts climbing back to in tuneness by the 12. That's about average for each string, though there is some variance. One thing, may saddle height is about 1/16 higher than I meant, so the break angle is kind of steep. I wondered if it might be a body resonance thing, but it seems like that should only affect one note. Top is 167hz, air 94hz, back 110hz. Any help would be appreciated, I had been hoping to gig with it in Germany and I'm leaving tomorrow AM...
Slotting jig was LMI's 640 mm one. I've never measured it for accuracy, I've just taken it for granted they made it right...

Yep, that's the "quickest and dirtiest" nut intonation fix for steel string guitars of the 5 methods I discuss. I'm assuming you're talking SS here, even though 640mm is normally a classical scale length.

If the T(1,1)3 ("back") really is at 110Hz, that's an issue and even if you're just tapping the back and measuring the frequency, it's still very low. Are you sure it wasn't 210 or 220Hz, which would be more usual? Whilst an issue, I suspect it's not causing your problem.

The first question, of course, is how does it sound? If the errors are moving "in formation" it may well sound OK (or at least OK enough). If you're totally allergic to intonation errors (some people are) you'll need to do some more work, but be pragmatic about it.

The modelling in the book for SS was based around a .012" top E string set (lights, Martin MSP 4100 set) with a medium action (3mm for the 6th and 2mm for the 1st, 25.4"/645mm scale; see footnote 43) . Other sets will react differently and if your strings are lighter/lower, that would explain some of what you report. To alleviate the problem (and you don't actually say how far flat it goes) you could try adding a bit more neck relief, which may result in having to lower the saddle a bit to give your preferred action at the 12th. The down side of this is that you might end up with a "ski jump" effect over the higher frets. However, if you're 1/16" higher than intended at the saddle and your neck set angle is correct, you shouldn't get any "ski jump".

If you're looking for really accurate intonation, then there's no real alternative but to use the more sophisticated methods.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

I attended the seminar at this years symposium on the matter of compensated nuts conducted by Bob Meltz. He used Earvana nuts, which are already compensated. You reintonate the guitar as usual. The difference was really striking.He had guitars with different bridges/saddles and nuts that could be changed in about 3-4 minutes. Same guy (some one from the room), same song, same guitar, what a difference in sound. He had a strobe there to back it up. Made a difference in intonation all the way to about the 12th fret. The problem as he explained it was in the break angle of the string from the nut when fretted , especially at the first positions. Also of note is that intonation adjustments made at the nut move in the opposite direction of those made at the bridge.

_________________
Brian

You never know what you are capable of until you actually try.

https://www.howardguitarsdelaware.com/

Agreed. Once you've heard a properly in-tune guitar, which usually requires a compensated nut, there's no going back!

Nut compensation makes a difference to every fret. Eric, above, has it pretty much right. A 1mm shift of the nut towards the saddle will flatten all notes by ~3 cents relative to the in-tune open string. A 1mm shift of the saddle towards the tail block will flatten the first fret by ~3 cents and progressively more up the neck (around 6 cents at the 12th fret). One way of intonating a guitar is to put the equations for that into an optimiser and let it shift the nut and saddle compensation around until the intonation error is minimised.

Sort of. The issue is that the stopped string has a longer path length, so has to stretch and so is pulled tighter, so it plays sharp. The effect is more pronounced on frets where the elongation is greatest.

BTW, my measurements, backed with the maths, show that string bending stiffness on normally strung guitars has little to do with intonation problems. This might not be true on bass guitars, but I've not tested that.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

A lot of different things could be going on, but what you describe is exactly what you would expect to see on a guitar with too much nut compensation (and probably high action at the first fret). If you fret at the third fret, what's the gap at the first? How far off would things be if you split the difference, so that the first fret was slightly sharp, and the seventh the same amount flat? I'm sure Trevor covers this in his book as well, but you can move the first fret towards the nut part of the distance of your minimum nut compensation, and reduce the problem you're dealing with.

Don't forget that if your guitar has perfect intonation, all of your major thirds will be out of tune by fourteen cents. Your minor thirds and your major sixths will be sixteen cents off.

If perfectly intonated to the equally tempered scale, the thirds and sixths will be "off" compared to Just intonation. But that's a compromise that we have to live with unless we want Lacote or Panormo style movable frets, or revert to tied frets, and don't modulate.

It's the 14 or so cents on the thirds and tenths, adding with other intonation errors, that can have you uncomfortably 20 or so cents from Just that really rankles. Most people seem to be reasonably happy with accurate equal temperament. Not too many people complain about piano's being "out of tune".

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Ok then, so damned either way. Minor thirds and tenths (which are the same note I think?) are my bread and butter for my style. Lots of high fretted notes over open strings in open/altered tunings ala Page and the Tea Party guy. No, not the political group, the Canuck Moroccan roll band...
On this guitar, I leveled and crowned the frets again, and added a bit more relief. It tamed it somewhat, to about 2-3 cents flat at the 7th from 7 cents. But it's not consistent, and there are occasional non linear jumps that are frustrating.
This guitar would be a perfect science project as a diagnostic challenge.
I think there is something deeper going on. It has so many chimey overtones going on that it's very difficult to actually hear it. Like it's out of phase or something. It also has a bewildering number of mysterious rattles and buzzes, and a g string that sounds like a sitar on all but four frets, even after good fret leveling and multiple nuts and saddles. It's pretty close to wearing a bus. I make no claims to being an expert, but it's my 56th guitar and it's doing things I've never encountered before that I just don't get.
Oh well, iOS 5 is out today, maybe there's a new app that can fix it.

Why not just make a fretless guitar? (Sorry I just had to...)

_________________
"Preoccupation with an effect gives it power and enhances the error"
from "Your Owner's Manual" by Burt Hotchkiss.

Part of the problem is the way we taper most fingerboards at the nut end. We measure all the fret placements perpendicular to the centerline of the fretboard, right?
Then by narrowing the string spacing at the nut, we've now made the string length longer than the measurement between frets.

That's why Fender electrics are never as "in tune" as Gibson electrics seem to be.
The Fenders have a wider string spacing at the bridge, and for the most part have narrower string spacing at the nut.
Gibsons have less string spacing taper as a rule.

I built a chambered electric back in the early 90's for a local jazzer who was always complaining sbout out how out of tune his Stratocaster was. I explained the problem, and he requested that I build him an instrument with no string spacing taper at all (like a classical). All the keyboardists who played with him afterwards started commenting about how much more "in tune" he was sounding, and he attributed the improvement to the wide neck.

_________________
"Act your age, not your shoe size" - Prince

I think you may have just discovered why it is so difficult to make really responsive guitars that will still play in tune. My guess with the "jumps" is that the string note (or one of its harmonics) is coupling with a body resonance, which shifts the string's frequency. The out of tune, due to this cause alone, can be as much as 30 cents. There's been a discussion over on Delcamp here:
http://www.delcamp.us/viewtopic.php?f=11&t=62935 talking about coupling and frequency shifting (repulsion) between the air and top resonances. The frequency shifting happens whenever two (or more) resonators couple. That's why modal tuning is so important. You have to manage the frequency of the resonances and the peak density to minimise the frequency shifting.

If you don't make responsive guitars, you don't have these problems! And that's the cop-out most manufacturers employ.

A tenth is just an octave 3rd. e.g. Lennon and McCartney's "Blackbird". 10ths all over. But in that tune they mostly appear across the same string pair (at least they do how I play it!) so you can tweak the interval into Just (or closer, anyway) using the tuners.

If you have a top and back pitched too close you can get an "out of phase" sort of a sound. I'm still baffled by your number for the back resonance.


That accounts for less than 1 cent of it.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Couldn't he add a brace to the back to 'lessen' the coupling effect?

_________________
There is no difference between the man that thinks he can....and the man that thinks he cannot.

But, Todd, we want to make sure we intonate so closely that we can never put a new set of strings on the guitar. String consistency is no where near close enough for that kind of intonation. Especially, if you are talking nylon or carbon strings. Steel is closer, but they have their own variations.

_________________
Waddy

Photobucket Build Album Library

Sound Clips of most of my guitars

Oops,
Sorry I meant to mention, back is 208 hz, though it seems that can change a hz or two depending on where I hold the gadgetphone. Top is 167ish, air is 94/5 ish. Being within a hz or three is close enough, I hope. I found when tuning the back that it jumped from being above target to below target without a commensurate amount of wood removal. It went from above to below without passing 'on target'.
By modal tuning, do you mean tuning the top modes? Is there a way to change, say, the cross dipole without affecting the monopole or tripole etc.?
Dang, I though making guitars better would make better guitars!
Well, the guitar in question is going to have a few weeks in a cold dark studio to think about it's errant behavior while it's 'subpar' but far more effective predecessor goes on a trip to Europe. I get to go be on the fun side of a guitar for a while. It's wierd, this will be the first time in a decade that I've been away from a workshop for more than 4-5 days.
See y'all in November!

The 'sitary' G string could be a problem of the longitudial wave coupling with the transverse. Note here that I'm not talking about the 'tension' signal, that occurs twice-per-cycle as the string vibrates, but the high frequecy longitudinal compression/rarefaction wave within the string itself that Novack called the 'zip tone'. It's a function of the frequency match between the two: it seems to happen when the longitudinal wave frequency is an exact odd multiple of the transverse pitch. Since both are related to the length of the string it can show up at all of the fretted notes as well as the open string. The match has to be pretty exact for this to happen, and sometimes just changing the string, or altering the way you install them, can help by changing the frequency of the longitudinal wave.

That, to me, sounds like you found the "repulsion" effect, even though the numbers you give don't support that. But it depends what numbers you're actually measuring and how you're doing the measurement.
When oscillators couple, the resonant frequencies repel each other (as discussed in the Delcamp thread). If you bring the back down in frequency and then it drops below the top frequency the repulsion of the top on the back becomes downwards rather than upwards and you see what you saw, a jump from above to below. For the sceptical (always be sceptical!) you can see this on a live back guitar by adding mass to the back (Blutac and coins/fishing weights) and tapping to take the frequency response curves). It's when you have things tuned in that cross-over region that you can get the "out of phase" sort of sound that you mentioned. But from this distance, I could be barking up the wrong tree altogether!


Yes. The ones you have to get right at the T(1,1)1, the T(1,1)2 and the T(1,1)3 ("air", "top" and "back"). The others will change the sound of a guitar but generally won't compromise it's functionality. There's a book on how to do the whole modal thing.


Yes, its nice when you stop! Have a great time!

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

hi,
I spent some time with this men and play this guitar

http://www.google.rs/url?sa=t&source=web&cd=1&ved=0CBsQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DMYK_PF9WTRE&rct=j&q=tolgahan%20cogulu%20microtonal%20guitar&ei=59WXTt2zCYrssgbc5NSGBA&usg=AFQjCNG0YlsHmfZByTid-8sQAufZi7A7wg&cad=rja

it has common sense
but after recalculation after all of this years only few people complain to me for non-perfect intonation so I can live with Greg B. method for classical guitar.
On the first next I'll follow Trevor's instructions , so will see...
by the way I'll make separate post about books

best
fric