Scale length
 

I believe when measuring the scale length of an acoustic guitar you measure from the fingerboard side of the nut to the edge of the saddle IE- the sound hole side.

This divided by 2 should give you 2 equally divided measurements from the nut to 12 fret then 12 fret to saddle.

My question: Because the string breaks at the back of the saddle bridge pin side then considering the thickness of the saddle, does this not then influence the true scale length if we are measuring our scale as stated from the front of the saddle.

Thanks

Brian

None of the strings are true scale length, they are all a little bit longer by what is called compensation. When a note is fretted, the string is stretched a bit, making it a bit sharp. Each string is a little different, so the saddle is slanted on the bridge to compensate for this tendency to play sharp. The breakpoint on the saddle may also be adjusted to fine tune compensation and intonation, which is the ability to play in tune all the way up the neck. It's not quite that simple, but that's the basic idea.
To get the scale length, measure from the nut to the 12th fret and double it.

Yep, this.

Ed

p.s. Rodger, I’m a fellow Baltimoron, er, Baltimorean.

When measuring scale length, all that matters is the point the string leaves the saddle, not where the back edge of the saddle is. If you try to put the break point all the way to the back of the saddle, the string will be contacting a knife edge, and will wear into the saddle very quickly.
On a steel string guitar, compensation on the first string should be about 0.075" (~ 2mm). In other words, the distance from the 12th fret to the saddle break point should be 0.075" longer than the distance from the nut to the 12th fret. All the other strings should be more than that, because the thicker the string, the more the tension is increased when it is fretted.
That is generally true, but not with Martins. Martin chose to include first string compensation in their scale lengths. For example, Martin's 25.4 scale actually figures to be 25.34 when you double the nut to 12th fret distance.

Thanks for the advice.

One more question: When I measure my guitar from 12 fret to saddle I get 12 and 3qtr inches.

Measured from the nut to the 12th fret 12 ich and 5 eights a difference of 1 eighth.(shorter)

Should I try to equal the two lengths by pushing the nut forward 1 eighth or is this where adjustments/compensation (intonation) at the saddle comes in to play to adjust for such variances.

This is a chinese made guitar are variations like this common.

Thanks again

Brian

https://www.youtube.com/watch?v=U6KtnjoD1N0


Just found this, Here Stew Mac talks about the break angle when calculating the scale as my previous enquiring question.

Brian

The video seems to make it unnecessarily mysterious.

Simply, but for the exception John noted, the scale length is the theoretical string length used to calculate the placement of the frets. In practice, but for the exception John noted, the actual vibrating string length is always longer than the scale length.

What is typically done to correct for the sharpness of the pitch of the fretted notes is to move the saddle away from the nut, making the the actual vibrating string length longer than the scale length. The amount by which the saddle is moved away from the nut is referred to as compensation at the saddle, or saddle compensation. Practically, this makes the distance from the nut to 12th fret shorter than the length from the 12th fret to the saddle.

The amount of compensation required varies with a number of factors, including string type, string gauge, string tension, string length, string mass and string height (i.e. amount by which the string is deformed when depressed against a fret). With variations from one string to the next, to play well in tune, generally, each string requires its own amount of compensation, with heavier bass strings requiring more than lighter treble strings. On many guitars, this is approximated by slanting the saddle slot in the bridge. These days, you will often see the B string individually compensated, since it is not usually adequately compensated by a straight-line approximation.

No. Equal lengths will produce sharp intonation, for the reason already explained.
1/8 inch is too much for the first string, and unless the action is really high, intonation will be flat. You need about 0.075" compensation on the first string. That means moving the saddle 0.050". When I say move the saddle, it is the contact point that is important. You may be able to improve intonation by reshaping the top of the saddle, instead of moving the whole thing.

Roger Knox wrote:
"To get the scale length, measure from the nut to the 12th fret and double it."

Unless, of course, the nut has also been compensated. In that case the nut will be closer to the first fret than it 'should' be, and you'll come up with a scale length that's too short.

Ultimately, the scale length is the length used to calculate the fret positions. After that you apply whatever compensation you feel will correct the intonation so the guitar plays in tune, more or less. It's never going to be 'perfect', and even if it is, some folks will still complain about it. ;)

Yeah, but who does THAT? ;);)

There is probably a way to figure out the scale length by measuring from the 1st fret to the 13th or something, but I don't fell like figuring it out right now :D

It's not that unusual to knock off a few thousandths in the nut to first fret distance. Some go to greater lengths in calculating nut compensation, there's a procedure in Trevor Gore's book.

Fascinating. I never knew about this exception. Thanks for sharing your knowledge, John!

I was kidding.

When slotting a fingerboard, if one cuts centred on the zero mark, one shortens the distance from nut to first fret by half the saw kerf, typically about .011". The Buzz Feiten system, for example, advocates 1/32".

I've read Mr. Gore's fine books and his process for optimizing intonation.

John, so you are saying that the scale length Martin states is based upon the first string total length...??