Official Luthiers Forum!

Rossy's thread on different scale length and Joe B's post there got me thinking on an area that has (and still does) confuse me greatly). Is there a "standard" definition of scale length?

Some use the distance fron the nut break point to saddle break point measured along the centre line of the fretboard/guitar - but this includes both nut and saddle "compensations".

Others measure from the nut break point to the middle of the 12th fret along the centre line of the fretboard and double this. This would include nut compensation but not saddle compensation.

I would prefer a definition based on the position of the frets excluding both nut and saddle compensation (which are variables used in subtly different ways by each builder) - say the distance between the middle of the 1st and 12th frets measured along the centre line of the fretboard multiplied by 2.252892 (or the formula for this factor in terms of the square root of 2).

What do y'all think?

_________________
Dave White
De Faoite Stringed Instruments
". . . the one thing a machine just can't do is give you character and personalities and sometimes that comes with flaws, but it always comes with humanity" Monty Don talking about hand weaving, "Mastercrafts", Weaving, BBC March 2010

I think the proper, standard definition of scale length is the purely mathematical one: the number you use to determine fret spacing, so twice the distance from (uncompensated) nut to 12th fret. Compensation at either end is added later, to overcome the limiations inherent in the system/the strings. That some use it incorrectly (ie, Gibson, who's '24.75"', if I understand it correctly, includes compensation, at least sometimes) doesn't mean we should.

The fret spacings are all related by the 12th root of 2, so the distance
between any two of the frets can get you the scale by straightforward math,
and you can ignore the nut and saddle.

For example, if the scale length is X, then the distance between frets 1 and 2
is the 6th root of 2 times X, minus the 12th root of 2 times X. If Y is the
distance between frets 1 and 2, then the scale is Y divided by (the 6th root
of 2, minus the 12th root of 2).

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

I agree with Howard & Mattia. Scale lenght is simply the distance from the nut to the saddle. Compensation should be considered later.

The confusion comes in because not all the big manufactors do it this way, and to make it worse they are not consistent in the way they figure it form decade to decade. For instance, Gibson has added compensation to it's announced scale lenght in the past but then sometimes they do not.

_________________
Joe Beaver
Maker of Sawdust

[QUOTE=Joe Beaver] I agree with Howard & Mattia. Scale lenght is simply the distance from the nut to the saddle. Compensation should be considered later.

[/QUOTE]

Joe,

This is why I am confused - the distance from the nut to saddle WILL include saddle compensation (and nut if any)if the saddle is compensated. If it isn't the guitar won't intonate properly.

Howard,

Thanks, that is what I was trying to say in my preferred definition.

_________________
Dave White
De Faoite Stringed Instruments
". . . the one thing a machine just can't do is give you character and personalities and sometimes that comes with flaws, but it always comes with humanity" Monty Don talking about hand weaving, "Mastercrafts", Weaving, BBC March 2010

Dave,

You are right. What I should have said is the scale of an instrument is the distance from the nut to the saddle before compensation. The confusion comes into play once the scale is divided into half tones (frets), and compensated. The scale should not be defined by these subdivisions or compensation. It is just the overall lenght you have chosen before you do anything to it. At least that is how I look at it.

Another source of confusion is added when the term scale is used to define a series of half notes, such as a chromatic scale.

_________________
Joe Beaver
Maker of Sawdust

Forget the nut and saddle. The scale length is an arbitrary figure used to determine the placement of the frets from a fixed starting point. As Mattia wrote, it's just the math. Don't confuse it with STRING length, which in the physical world, must be longer than the scale length for a guitar to play in tune (as nearly as we can get it!).

Most guitars are not compensated at the nut. Measure from the nut to the 12th fret and double it. That should be the scale lenght without the saddle compensation. What if the nut is compensated? Then it is difficult to find a single measurement that will give you the scale length without doing a bunch of match later.

I tend to be a little lose with scale lengths when I talk to muscians and even other luthiers. A 25.4 scale is actually 25.340 according to Stew-Mac. This only matters when you are actually laying out the frets or rplacing an existing fretboard. In all other cases, .06" is irrelevant. I can't imagine a musician saying that the scale felt "long" if you gave him a guitar with a true 25.4" scale instead of 25.34. That is less than 1/4 percent difference.

is this supposed to be about the definition of the scale length or about attempting to determine an unknown scale length on an existing instrument. it would appear that many posts have got off on a tangent about the latter.

and a great many instruments do have compensated nuts, e.g., all taylor recent production have compensated nuts.

[QUOTE=crazymanmichael] is this supposed to be about the definition of the scale length or about attempting to determine an unknown scale length on an existing instrument. it would appear that many posts have got off on a tangent about the latter.

and a great many instruments do have compensated nuts, e.g., all taylor recent production have compensated nuts.[/QUOTE]

Michael,

My original post was about a "definition" that could be easily interpreted when people use the term as this was always a confusion to me when I first started up. My suggested one was the mathematical one that Mattia, Howard and others have confirmed.

But I also had an eye to what you suggest - about determining the "scale length" of an existing instrument - particularly old ones. The only complication here is that the frets could have been laid out innaccurately, so I suppose you would have to measure each and every fret interval (from fret one upwards) and compare this against different mathematical scale lengths to find the best fit.

I have an old parlour guitar that I will go an do this on.Dave White38831.5380208333

_________________
Dave White
De Faoite Stringed Instruments
". . . the one thing a machine just can't do is give you character and personalities and sometimes that comes with flaws, but it always comes with humanity" Monty Don talking about hand weaving, "Mastercrafts", Weaving, BBC March 2010

[QUOTE=Joe Beaver] I agree with Howard & Mattia. Scale lenght is simply the distance from the nut to the saddle. Compensation should be considered later.

[/QUOTE]

I agree...when I lay out a pattern for a classical scale length is distance from nut break to theoretical saddle break without compensation...I mark that point and then add on appropriate comensation for the actual saddle break position.

In their books Campiano, Courtnal and Williams all talk about a 650mm scale length but the actual distance between nut and saddle break is slightly longer.



Cheers Martin

All this talk of compensation intrigues me. How far does one go to achieve exact intonation? Comments:

1. You must reach a point where the player/listener can no longer detect that the intonation is not perfect.

2. The way I see it you can never achive perfect intonation while there is variation in each player's technique. IMO part of learning to play guitar well is learning how to adjust your technique to achive best intonation.

Cheers Martinkiwigeo38831.6264814815

i would say scale length, and its determination depends on whether you are building from scratch, or doing somehting to an existing guitar, like replacing a fretboard. If you are building from scratch, you get ot pick whatever scale length you want, and usingthe 12th root of 2, determine your fret spacing. How you compensate at the nut and saddle is again up to you. If you are trying to replace an existing fretboard, I would atke a couple measurements to try to determine what you are dealing with. For starters, take the nut to 12th fret. As well take the 1st fret to 12th fret, and compare both of these against known fret scales. You will most likely find one that fits, be it 24.9, 25.4 25.5, 24.75 whatever. If none of these seems to work, then try playing around with a fret program until you get one that lines up as close as possible with your measurements.

_________________
Tony Karol
www.karol-guitars.com
"let my passion .. fulfill yours"

[QUOTE=TonyKarol] For starters, take the nut to 12th fret.[/QUOTE]
Just to clarify--make sure you're measuring to the MIDDLE of the 12th fret or fret slot.

Right, any measurement to a fret needs to be to its centre. Or if measuring fret to fret, you can go near side to near side, or outside to outside minus the fret width.

_________________
Tony Karol
www.karol-guitars.com
"let my passion .. fulfill yours"

    Here is a fast understanding to what is happening
SCALE LENGTH
    this is the actuall fret spacing. The scale length while determined by a mathmatical calculation can be determined by measuring from the nut to the 12th fret and mulitplying by 2.

COMPENSATION
    Though scale is a fixed value , the string isn't. A string at rest has a longer length when played. To help keep the string in tune as it is played the saddle is moved to accomodate this . Though there ins't a perfect intonated guitar an accepted angle of about 3 degress is used on most steel string guitars. This helps to take into account the (working length) of the string.

I use about .125 when using medium strings and .1 for lights.

INTONATION
   This is the actuall tone created by the string that is freted up the scale length. The saddle compensation helps to allow intonation so the notes will be in key.

I hope this helps
john hall

As far as I'm concerned the only measurement that doesn't include a 'fudge factor' is the mathematical one, and that should be the defined 'scale length'.

After that things get much more problematic. There are several different ways that people compensate strings, and all of them seem to work pretty well for somebody and terribly for other folks. I'm of the opinion that there probably is no way to get every note on an acoustic guitar to play perfectly in tune unless you're willing to use six-piece frets, and even then.... We usually settle for 'close enough', and there seem to be several ways of getting there.

Alan
   I can't agree more. I doupt there is a perfectly intonated acoutic. Each string will ad its own problem. Maybe it is best to say scale length is static and comp length is dynamic.
John Hall

Yep, pretty much nothing plays perfectly, and I find that 3rds and 6ths are the worst offenders - for those that arent sure what those are - tune a guitar to play a G chord , as true as you cn. Now play a D, and doesnt the f# (high E, 2nd fret) sound sharp a touch ??? Bruce Hornsby says - Thats just the way it is.

_________________
Tony Karol
www.karol-guitars.com
"let my passion .. fulfill yours"

[QUOTE=TonyKarol] Yep, pretty much nothing plays perfectly, and I find that 3rds and 6ths are the worst offenders - for those that arent sure what those are - tune a guitar to play a G chord , as true as you cn. Now play a D, and doesnt the f# (high E, 2nd fret) sound sharp a touch ??? Bruce Hornsby says - Thats just the way it is.[/QUOTE]

That's probably why I play almost entirely in modal tunings where the first thing you usually do is take out the 3rd and make it a suspended 4th (eg. DADF#AD becomes DADGAD - ind you the A 2nd string is a real pig to get to sound right in this tuning)

_________________
Dave White
De Faoite Stringed Instruments
". . . the one thing a machine just can't do is give you character and personalities and sometimes that comes with flaws, but it always comes with humanity" Monty Don talking about hand weaving, "Mastercrafts", Weaving, BBC March 2010

I agree guys. I don't know if it's common practice but I have been cutting my "0" fret line a tad short which places the nut up a bit from center 0. To my ear, this seems to compensate for some of the annoyances with open chord problems.
Maybe it just shifts the problem up the neck, I'm not sure.
It seems to me you could always carve away some of the nut if it is over compensated on some strings.

Wade

_________________
Wade
Nashua, NH
http://www.wadefx.com

I am helping a friend re-neck a Washburn electric, with the Feiten system. Seeing as it has a straight locking nut, there is no individual compensioation going on at all. After measuring the fret spacing to see the scale length, I determined that the nut was in fact 1mm short, ie closer to the first fret. I have been cutting mine about .75mm all along (based upon the old GAl article on classical nut comp) !!!!

_________________
Tony Karol
www.karol-guitars.com
"let my passion .. fulfill yours"

hello, i'm new to this forum, found it searching for scale length and compensation.

i dont want to thread jack this, but you guys are talking about what i'm wanting to know.

anyway. i picked up a 12 string guitar a while back.. epiphone FT165 bard. decent guitar but it was outta tune with itself. i bought a bone saddle and shaped it to fit into the bridge, now i'm wanting to compensate the strings for the best intonation i can get.

i've looked at other guitars like talors and other high end guitars, i noticed that the B string is compensated longer than the high E string, then the G string is shorter and then tapers back out longer for the bass strings.

i could do it the hard way and compensate a little, then measure with my strobotuner, then do some more.. but i hate removing/replacing strings a bunch of times to get it right.

i'm wondering if there is a formula or given compensated lengths for each string that i could just start out with and not have to screw around a ton going back and forth.

thanks,
Richard

ps: i love my electric guitars, they're alot easier to set intonation!

Hi Richard,
First of all, Welcome to the OLF. I'm sure you will find what you need here.
Before you start carving away at your new saddle for intonation, you may want to check your nut first.
I'm trying to guess what you mean by "out of tune with itself". Is it when you play open chords? Or up the neck?
For example, if you tune your strings so it sounds in tune when you play an "E" major chord; then play a "D" chord and it sound out, then you may have issues with the nut. But if its out of tune as you play up the neck, the saddle is likely the issue.
A little complicated I know but first things first.

Also, I don't know of any formula for each string but you should know it changes with differing gage strings as well.

Wade

_________________
Wade
Nashua, NH
http://www.wadefx.com

Some moons ago, I wrote to W. Cumpiano asking about the compensation distance that should be applied for a 24.9" guitar and he nicely advised that, from his experience, 0.15" figure is applicable for any of the modern guitar scale lengths, in which 24.9" is included.

More interestingly at the end of his remark, he mentioned about a luthier's solution to this eteranl compensation dilemma and gave this link for reference which I think it's interesting enough to be shared here; http://www.chouard.de