I Think My Frets Are In The Wrong Place....

[VesQ]

Calculating with this number 1.05888229 for all the other frets gets very close to the Gibson 24 3/4 scale

Measuring to the 12th fret, doubling and using it in the standard formula is no-where near as accurate

If the same number is calculated for all the frets it suggest me that either nut or bridge was moved.

[Crusader]

If the same number is calculated for all the frets it suggest me that either nut or bridge was moved

I don't follow what you mean, its just a calculation as follows

(628.65-(628.65/1.05888229^1) = First fret

(628.65-(628.65/1.05888229^2) = Second fret

(628.65-(628.65/1.05888229^3) = Third fret

- and so on

I did these calculations very quickly and the results show that you could use kingfishers idea with some degree of success. But you would have to actually make an instrument to see how well it works. At a closer look I might find I was completely wrong LOL!

crudaser feels gibson have some secret technique but i have not found it...

Its on post 9 on this thread but like I say it is not necessarily how they worked it out. Its just a method I found that matches the measurements I took

well i stayed out the gibson thing so far - but my own research has not found the differences crusader talks about... on new or vintage instruments

I thought it was common knowledge that the Gibson fret spacing is based on a different rule-of-thumb, but what methods and guitars have you used to take your measurements?

I took measurements off two different guitars in 1997 and 2009 with the same steel rule, with a magnifying glass and/or two or three pairs of glasses under all different kinds of lighting. I must have measured every fret a hundred times and when I measured the second one I did not look at the figures from the first one. I measured in inches and millimetres

The result is they have a different amount of compensation at the nut but apart from that they match each other within 0.1mm

And some other "fun-facts" -

The first one measured exactly 12 9/32" to the 12th fret

The second one was slightly more at the 12th fret but the 1st fret measures 1 3/8" - That is exactly 1/18th of 24 3/4"!!!

[WezV]

last one i checked was this:

http://projectguitar.ibforums.com/index.ph...ic=42380&hl

i will have to wait till it gets back from the sprayer to give you accurate measurements. obviously these old juniors have quite an angle on the bridge and with the uncompensated wraparound design i will be interested to see how well it actually intonates once together with new strings. it seems to be near to 24 9/16" but not quite, about 1 mm longer

it is common knowledge gibson do things differently and that their scale length has never been the same as 24 3/4" anywhere else, its also common knowledge they have used different versions over the years. I dont check every one that comes through the doors but the few newer gibsons i have checked all seemed to be fretted based on a 24 9/16" measurement, or near as damn it. two that stick in my head are a late 90's studio and a 2003 standard LP

as stew-mac say as clear as mud

Gibson's standard 24-3/4" scale is a compensated length based on a true scale of about 24-9/16". Our fret spacing matches Gibson's most common spacing. Many Gibson's have other scale lengths

so there is always some room for debate on this

[Prostheta]

Nooooo...not a scale length discussion!!

I think zero frets or Earvana nuts are as far as I would go to compensate an instrument. Granted, there are advanced systems out there to take it that one step further - each with their own inherent shortcomings - but none of them easy translate to the hobbyist or lay player who buys the "standard plank" off the wall, or builds/modifies with less than commercial equipment. Buzz Feiten is achievable also. I think we have to accept the idiosyncracies of our chosen instruments and the realistic improvements. Just my thoughts.

Or stop being such girls and play fretless instruments instead ;-)

[Crusader]

last one i checked was this:

http://projectguitar.ibforums.com/index.ph...ic=42380&hli will have to wait till it gets back from the sprayer...

as stew-mac say as clear as mud

Gibson's standard 24-3/4" scale is a compensated length based on a true scale of about 24-9/16". Our fret spacing matches Gibson's most common spacing. Many Gibson's have other scale lengths

so there is always some room for debate on this

You might find a Junior has a standard scale length, I'm quite sure Epiphones do

Just to give you a bit more info, I first encountered this different scale length after I bought my first LP in 1977

I tried to do the intonation on the 1st string and found when I made it correct at the 12th fret it was progressively sharp up to the 22nd fret

Next time I was in the shop I brought this up and thats when it was explained to me that the Gibson has not just a different scale but a different method of fret spacing. One thing that didn't help, I was using 38 to 8 guage strings. I'm sure a set of 52 to 12s would intonate a lot better

So heres a suggestion

Instead of measuring each fret you could just set the intonation to be correct at the 12th fret, on the 1st string using a 9 or 10 guage string (with very low action, with a nice straight neck, with that teeny bit of relief)

Then if its correct at the 22nd fret, it is surely a standard scale

If it is sharp it is likely to be the special Gibson 24 3/4 scale

Regarding that quote from the stewmac website

I see it as just a way of side-stepping the issue because they don't want to get into complicated explanations

On the first guitar I took measurements from, the 12th fret was exactly half of 24 9/16" - not 'about' 24 9/16" and yet the spacings do not match what you get from the standard formula

Here are two clues

The Gibson 1st fret is closer to the nut than a standard 24 9/16" scale

The distance from 1st fret to 22nd fret is more than the standard 24 9/16" scale

Hey I forgot I had this. I was going to make a better diagram but who's got time...

Here is the difference between the standard 24 9/16 and SG scales (I also chucked in the standard 24 3/4)

It is very minimal but it is there

If you only had one or two frets out a tiny bit you would consider them to be outliers

But this diagram shows a consistent slight difference between scales

Anyway as you say there is always going to be debates, and thats because we are never going to hear official word from Gibson (unless they think it will improve sales!!!)

Prostheta

I can understand your frustration (lotsalaughs)

mmm fretless...it would take a lot of practice to put your fingers in the right place when playing chords. But you could aim at "just intonation" which sounds a lot nicer than the tempered system

[WezV]

You might find a Junior has a standard scale length, I'm quite sure Epiphones do

Its an american made guitar from the gibson factory in 1960, made from incredible mahogany and a brazilian rosewood fretboard. They used the same materials as the high end instruments of the time, surely it will follow whatever other method for fret slotting they were using at the time ??? although it was a time when gibson was onyl just starting to make the correct design decisions (which is why a 59 les paul is worth more than a 53). A 1960 LP junior is not comparable to a cheaply made far eastern equivilent

anyway, i will make sure to get you accurate measurements when i get it back, i did some quick ones to check but cant find them now... and i am willing to be proven wrong

Shorter nut to 1st fret spacing is not uncommon. Its a bit of a trick to give instant nut end compensation, something PRS do at an angle now and i think alot of luthiers actually remove a small amount from this end. It still doesnt change the sums used to work out fret position though - although there are a few numbers used for this over the years, for example some instruments use 17.835 instead of 17.817. There is another way in one of the acoustic building books but i cant remember it off the top of my head

bare in mind kingfisher was talking about fret compensation so that you need no bridge compensation - that would be very hard to achieve in a consistent way just by using a different number to divide the scale length up. Bridge (and nut) compensation is a much more versatile method for playing somewhat in tune

[Crusader]

I was just thinking they might reserve that special scale length for the higher-end guitars, but what you say also makes sense

Shorter nut to 1st fret spacing is not uncommon...

Ah yes but then the 12th fret won't be half the scale length from the nut

...bare in mind kingfisher was talking about fret compensation so that you need no bridge compensation...

I don't think Kingfisher was talking about not moving the bridge at all, in the third post he says "...so less movement of the bridge will be necessary"

...there are a few numbers used over the years, for example some instruments use 17.835 instead of 17.817

As far as I can see, this is exactly what Kingfisher was suggesting, and its like I commented "no matter what you think of it has already been done"

Anyhow I think the ultimate solution is fanned frets

Look forward to your results from the Junior

Keep in mind I found measuring frets to be very difficult but I am a fanatic, I'm crazy. A normal person would not go to such lengths. Its not something I would want to do again. Getting out of bed at 3am to get the guitar out to check something over again...no no I'm getting too old for these crazy things!!!

cheers!

[WezV]

Shorter nut to 1st fret spacing is not uncommon...

Ah yes but then the 12th fret won't be half the scale length from the nut

yeah, thats right. That method doesnt work as well if you have a compensated nut, but generally the compensation is quite small .

Will let you know on the Lp junior. as you can see from my thread i have already been quite thorough with recording its details. One of the best sounding guitars i have played!!

in post 13 kingfisher says

What we need is a magic number to divide by that doesn't need bridge adjustment - that is the holy grail.

to me this is the wrong way to look at it, because bridge compensation is not correcting maths errors, but correcting for the physical act of fretting (and therefore bending sharp slighty) a string.

[WezV]

just done some slight modifications to my excel spreadsheet for working out fret positions. essentially a low budget fret calculator - but the formula's used are visable and changable.

Since Kingfisher is interested in numbers other than 17.817 i altered it so this number can be changed and it will work out the new fret positions - but i also added an extra page so you can compare either different scale lengths or different calculations for the same scale length, like the sloane 17.835 or even the primitive 18. wfret does these methods but does not compare automatically and tell you the difference.

Also a brief description of Fret factors as discussed in the Cumpiano & Natelson book and some common scale lengths in fractional inches, decimal inches and metric so hopefully its usefull to someone

http://www.freedrive.com/file/1156655,fret-positions.xls

edit: should point out that no nut or bridge compensation is included... so dont use the 'fret to bridge' measurement to actually locate the bridge!!!

[low end fuzz]

last one i checked was this:

http://projectguitar.ibforums.com/index.ph...ic=42380&hli will have to wait till it gets back from the sprayer...

as stew-mac say as clear as mud

Gibson's standard 24-3/4" scale is a compensated length based on a true scale of about 24-9/16". Our fret spacing matches Gibson's most common spacing. Many Gibson's have other scale lengths

so there is always some room for debate on this

You might find a Junior has a standard scale length, I'm quite sure Epiphones do

Just to give you a bit more info, I first encountered this different scale length after I bought my first LP in 1977

I tried to do the intonation on the 1st string and found when I made it correct at the 12th fret it was progressively sharp up to the 22nd fret

Next time I was in the shop I brought this up and thats when it was explained to me that the Gibson has not just a different scale but a different method of fret spacing. One thing that didn't help, I was using 38 to 8 guage strings. I'm sure a set of 52 to 12s would intonate a lot better

So heres a suggestion

Instead of measuring each fret you could just set the intonation to be correct at the 12th fret, on the 1st string using a 9 or 10 guage string (with very low action, with a nice straight neck, with that teeny bit of relief)

Then if its correct at the 22nd fret, it is surely a standard scale

If it is sharp it is likely to be the special Gibson 24 3/4 scale

Regarding that quote from the stewmac website

I see it as just a way of side-stepping the issue because they don't want to get into complicated explanations

On the first guitar I took measurements from, the 12th fret was exactly half of 24 9/16" - not 'about' 24 9/16" and yet the spacings do not match what you get from the standard formula

Here are two clues

The Gibson 1st fret is closer to the nut than a standard 24 9/16" scale

The distance from 1st fret to 22nd fret is more than the standard 24 9/16" scale

Hey I forgot I had this. I was going to make a better diagram but who's got time...

Here is the difference between the standard 24 9/16 and SG scales (I also chucked in the standard 24 3/4)

It is very minimal but it is there

If you only had one or two frets out a tiny bit you would consider them to be outliers

But this diagram shows a consistent slight difference between scales

Anyway as you say there is always going to be debates, and thats because we are never going to hear official word from Gibson (unless they think it will improve sales!!!)

Prostheta

I can understand your frustration (lotsalaughs)

mmm fretless...it would take a lot of practice to put your fingers in the right place when playing chords. But you could aim at "just intonation" which sounds a lot nicer than the tempered system

this picture to me looks like a fanned fret layout;

[SwedishLuthier]

this picture to me looks like a fanned fret layout;

with a misplaced nut on the 24 9/6" scale!

Actually a misplaced nut would really result in exactly that specific fretting pattern. However this:

it was explained to me that the Gibson has not just a different scale but a different method of fret spacing

is something that I would like to send to the mythbusters department. I have never heard about this and I have a really hard time believing that I should have been in this business without picking it up. I suspect that it is the commonly known low quality of the Gibsons, like that misplaced nut above that started this myth.

[Crusader]

low end fuzz

do you know you can delete unecessary paragraphs from the quotes? Leaving the whole lot in uses up a lot of space

this picture to me looks like a fanned fret layout;

with a misplaced nut on the 24 9/6" scale!

Actually a misplaced nut would really result in exactly that specific fretting pattern

I thought the diagram was self-explanatory but perhaps I should have said more about it

It is just 3 different scale lengths compared to each other

The 24 9/16 and the SG both start at the nut

The 24 3/4 starts at the 22nd fret (ie and works backwards) to show that the SG scale matches 24 3/4 from about the 12th fret upward

The main purpose of the diagram was to show how the SG scale is not 24 9/16"

(I should have left the 24 3/4 out but I would have had to take the time to modify and re-load to photobucket)

The only place where they really match is the 1st and 12th frets

Starting at that point you can see that the SG frets creep closer to the nut

and in the other direction they creep closer to the bridge

The fact that the SG 1st fret is closer to the nut indicates there is compensation at the nut (it is not misplaced)

If you put the compensation back on, the 12th fret would no longer be half of 24 9/16

So like stewmac says, "...it is...based on a true scale of about 24-9/16"

However I would put it like this "...based on a STANDARD scale of about 24 9/16"

Now looking at the two scales again

The SG frets above the 12th fret creep closer to the bridge

I find this results in much better intonation right up to the 22nd fret on the 6th string

A standard scale becomes very flat above the 12th fret on the 6th string, if intonated perfectly at the 12th fret

(and this is one of the main points that kingfisher was talking about, although I dont' think he actually stated which string he was referring to)

I was just trying to point out that I think his idea is a good one, yet he is not the first to think of it and I used the Gibson 24 3/4 as an example. Also, I have actually tried making a couple of guitars with the bridge straight across (no compensation) by using a different scale method on the 6th string side and a standard one on the 1st string side. I made them about ten years ago but checking the intonation is something which is still in progress - when I have the time...it takes so much time!!!

it was explained to me that the Gibson has not just a different scale but a different method of fret spacing

is something that I would like to send to the mythbusters department. I have never heard about this and I have a really hard time believing that I should have been in this business without picking it up. I suspect that it is the commonly known low quality of the Gibsons, like that misplaced nut above that started this myth

Yes there are some things which seem to be very hard to prove/disprove. Have a look at this discussion and scroll down to the post where a guy starts "Love the site (especially the rants)

http://www.edroman.com/guitars/gibson.htm

He says:

"On most Gibson electrics the scale length from the nut to 12th fret is 24.562"

No problem with that...but the scale length above the 12th fret is 24.75.

That means OK intonation from the nut to 12th fret (especially with Buzz Feiten Tuning System or similar), but really BAD intonation above the 12th fret"

(He doesn't say it but I'm sure he is referring to the 1st and 2nd strings)

I was trying to find a single formula to figure-out the Gibson fret spacing for years and it was this comment that lead me to my final conclusions by using a combination of three (Any new readers please read post 9 on this thread) Putting aside his anti-Gibson approach, he is saying the same thing as I am, that the Gibson 24 3/4 is not a standard-formula scale. I have searched and searched and to this day have never found another person saying this on the internet

Talking about Gibson low quality, I am well aware that it comes and goes. My first LP was a dog but the later two were/are very good. Talking about fret spacings, according to the measurements I took they match each other within 0.1mm except for the 5th fret on the SG. It was 0.7mm out...but the funny thing, that guitar intonated very well over the whole fretboard, except that little bit sharp on the 1st and 2nd string above the 12th fret

Anyway it seems that kingfisher has left the discussion and in any case talking is not going to solve anything. If you take the information I have put forward I am confident you will find it to be correct

cheers

[kingfisher]

Anyway it seems that kingfisher has left the discussion and in any case talking is not going to solve anything.

No and yes in that order.

post removed till I think a bit more

Edited March 30, 2010 by kingfisher

[SwedishLuthier]

This is an extremely interesting discussion. Crusader: Do you have the actuall numbers for the distances of the 24 9/6 scale (and possibly the SG measurements too, I have the the standard 24 3/4 of cause) . I'd love to put them into the cad probram and play around a bit. I'm still not 100% convinced that the 24 9/6" scale isn't in reality a standard, even dividor sclae with a missplaced nut. But I need to tinker a bit with the data first.

OT: Just my 2 cents regarding Ed Roman: Take everythnh that man says or writes with more than two grains of salt. Just my personal opinion.

[WezV]

i was talking with my wood supplier last night as i had ordered a preslotted 24 9/16 fretboard to save myself some time on a firebird I project i am starting. he offered me a better deal on some 24 5/8 ones which he had been asked to cut specially for some vintage LP clones. Its also closer to the 24 1/2 PRS is adopting on a lot of models aimed at the gibson market

the point for me:

* we know gibson have used different scale lengths over the years all referred to as 24 3/4"

* We know they have had varying quality over the years

* we know their scale is referred to as a 'compensated 24 3/4"' scale length, rather than referred to in the industry standard way (instant confusion!!!)

My understanding of some of the anomalies in gibsons fret placement (i dont deny they can exist) is that they were factory errors - iirc this was reported on some 70's models, but thats memory rather than proven fact. but its just like them making guitars with incorrect neck angles (early 50's), super low frets (60's-70's), **** fretwork (until plek), bone dry fretboards (most of the 90's) and various other mistakes they have made over the years (i only mentioned some of the ones that define era's)

I also very much doubt they would want to keep a purposefull attempt to improve intonation quiet... go to those lengths to design a compensated fretting system and you can be sure it would be included in the marketing

[Samba Pa Ti]

My understanding of some of the anomalies in gibsons fret placement (i dont deny they can exist) is that they were factory errors - iirc this was reported on some 70's models, but thats memory rather than proven fact. but its just like them making guitars with incorrect neck angles (early 50's), super low frets (60's-70's), **** fretwork (until plek), bone dry fretboards (most of the 90's) and various other mistakes they have made over the years (i only mentioned some of the ones that define era's)

ive got a guitar with all these features, hybrid lol

[kingfisher]

In the spirit of modern experimentation, I thought I would build a guitar to try see what happens -

As you can see, it is built from exotic timbers and is minimalist in form, having just one string (9 thou). The action is normal to low, and requires pressing the string down. (9 is our best selling gauge by the way)

First I set the bridge and nut 650mm apart (- they are both fixed), and tuned it to E, using the tuner in the picture.

Next I slid the movable frets to about halfway, and moved back and forth, until I hit E again. Nut to fret distance 323.8mm

Up another octave, nut to fret 486mm

Next I wrote a php program <a href="http://www.rockfactory.co.uk/fretcalculator.php">fret calculate</a>

results here -

Scale is 650mm

divisor is 17.91

1 36.2926 36.293

2 34.2662 70.559

3 32.3529 102.912

4 30.5465 133.458

5 28.841 162.299

6 27.2306 189.53

7 25.7102 215.24

8 24.2747 239.515

9 22.9193 262.434

10 21.6396 284.074

11 20.4314 304.505

12 19.2906 323.796

13 18.2135 342.009

14 17.1966 359.206

15 16.2364 375.442

16 15.3299 390.772

17 14.4739 405.246

18 13.6658 418.912

19 12.9027 431.815

20 12.1823 443.997

21 11.5021 455.499

22 10.8599 466.359

23 10.2535 476.612

24 9.681 486.294

As you can see a divisor of 17.91 seems to provide a fairly close match to my experimental results

Here is the php if anyone needs to check

“

<p>

<?php

$scale=650;

$divisor=17.91;

echo "Scale is " . $scale ."mm" . "<br /> <br />";

echo "divisor is " . $divisor . "<br /><br />";

for ($counter=1; $counter<25; $counter++)

{

$fret=$scale/$divisor;

$total=$total+$fret;

if ($counter==12)

print "<br />";

echo $counter . "&nbsp; &nbsp; &nbsp; &nbsp;";

print round($fret,4). "&nbsp; &nbsp; &nbsp; &nbsp;";

print round($total,3) . "<br />";

$scale=$scale-$fret;

}

?>

</p>

“

Loads of potential errors here, - new string, maybe more elastic than a used one - measuring equipment (cheap tuner and steel rule) a bit primitive - but nevertheless I will be going with this until convinced otherwise.

Still working on this when I have the time - will post if any more insights.

Edited March 30, 2010 by kingfisher

[WezV]

interesting - do you have the string a 'normal' playing distance above the fret, i.e allowing for action

now you need to try the rest of the string set, different gauges, and maybe repeat with a few different brands to see if you can get some consistency in your results

I imagine you will get a different number with the different string gauges which i would expect to take away the mainstream use, although it would have potential for very specific set-ups

quickly shoved it in my spreadsheet next to the normal calculations

[kingfisher]

interesting - do you have the string a 'normal' playing distance above the fret, i.e allowing for action

now you need to try the rest of the string set, different gauges, and maybe repeat with a few different brands to see if you can get some consistency in your results

I imagine you will get a different number with the different string gauges which i would expect to take away the mainstream use, although it would have potential for very specific set-ups

quickly shoved it in my spreadsheet next to the normal calculations

Yes normal action. This is why there are 2 frets on the slidy bit.

I too expect to get different results with different gauges, as we know the usual bridge compensation differs with gauge, probably a fan fret system could compensate for some of this. (ie a different divisor for top E and bottom E), but then we are getting into the realm of specialist instruments (harder to sell?). Consistency will be difficult without more accurate measuring equipment.

For now, as I say, I regard this as an improvement, invisible to the user, but tunes a bit better. If anyone would care to duplicate my experiment, please do and tell us your findings.

Do you think I should enter this instrument in guitar of the month? (this is a joke by the way for the humorously challenged.)

Edited March 30, 2010 by kingfisher

[Prostheta]

Do you think I should enter this instrument in guitar of the month? (this is a joke by the way for the humorously challenged.)

Yes. No, really. It is a Gibson copy though. Just don't use the logo.

Edited March 30, 2010 by Prostheta

[WezV]

the problem with doing it fanned fret is that i would not expect the number to change gradually from high to low - since string gauge doesnt change gradually/evenly, especially not going from plain to wound strings.

this leads you down the route of individual fret placement for each string, which then leads you down finding the optimum place for each fret the way you have done for 2 so far. test it thoroughly and you will arrive at the true-temperament system. all well and good

so you develop something like that and realise the flaws inherent in it and have to produce 4 more versions for different situations, because 1 right answer is not enough

http://www.truetemperament.com/site/index.php?go=1&sgo=0

and then you can consider how much brilliant music has been written on scale lengths based on the 17.817 divisor with often imperfect intonation and realise how good it can sound. I remember my grandpa teaching me all about discord when tuning pianos... and i remember how nice a little of it could sound

and yes GOTM, all the way

[kingfisher]

Wes, I agree with everything you say - but - (you knew there would be a "but" didn't you) - I am not looking for perfection because I don't think it exists, I am just trying to involve in the equation the fact that we have to press a string down to get any other note than open.

17.818 (or 17.817) takes no account of that at all, and I think it should. In fact my experiment (within experimental error) proves it should.

Do you think the neck profile needs some work?

[SwedishLuthier]

OK, let me try to see if I understand you correctly: You have made two different measurements with that jig, the first and the second octave. And you used a tuner to hit the right spot. And then you made a calculation based on those two measurements. Am I right?

It would be extremely interesting to do things the other way around. Place the sliding fret in the position that you original idea would place it, fret the string and measure the note. I would love to do the same experiment as you have but I have too much going on to be able to prioritize experimentation right now, unfortunately…

Do you think the neck profile needs some work?

Not necessary, I like necks a bit bulky.

[WezV]

this is where the idea falls down for me - if you could create a 'perfect' system that worked well for you then it might be worth it, if that isnt even the goal then i dont know what is.

and dont forget, you can already get close to perfect with a normal scale, using the bridge to compensate for string action and gauge and a compensated nut to even out some of the issues inherent in the equal temperament system - and most players are perfectly happy with a lot less than that

Needs more research to see what happens with each string, try some different gauges, scale lengths and action heights to see if you can get a number that will consistently work, shouldnt take too long with that set-up you have there. a lapsteel would also be usefull for testing it... and then actually implement it in a actual guitar fretboard and start testing that and seeing if players are happy with the changes, or even notice.

so no, dont carve the neck but make a lapsteel (everyone should anyway), would be perfect for testing this. Make it a long scale one and you can always shorten the scale and test different ones by using a dummy nut/lap steel capo. you can then use you slidy fret under the strings like shown above

[supplebanana]

another variable that no-one has mentioned is how the magnetic pull of the pickups can affect the frequency at which the string vibrates.... let's see somebody factor that in too.... lol