bluegrassboy4life - Posted - 07/06/2017:  10:17:20

I was wondering if everyone that has any kind of Gibson 5 string banjo could tell me the year, model, and measure the heel thickness, fret scale, nut width, width of the fretboard at the heel, scale length, peghead angle, distance from 22nd fret to end of fingerboard, wood type of the fret board, the wood the neck is made of, and everything else you can think to measure. I would like to make Gibson clones and use my name on them like Huber did with the Classic 58, but I know I can't do it till years from now, but I really like to plan ahead, and would like to get all the information on different neck measurements as I can since the neck is the cheapest and easiest part to build, and if I finish a few before I can build pots, I can lay them aside and wait patiently, but right now I'm impatient. It would be greatly appreciated!

Edited by - bluegrassboy4life on 07/06/2017 10:33:35

Old Hickory - Posted - 07/06/2017:  10:39:45

Gibson banjo nek templates available here.  Scroll down the page.

According to the listing, a 1974 Gibson has a 1-3/16 (1.87) nut width.  A 1994 RB-4 reissue has a slightly wider 1.197-inch nut.  Both have 26-3/8 inch scale -- according to this site.  Consensus on the Hangout seems to be that 26-3/8 inches was always the Gibson Mastertone scale length.

I don't get why Stew-Mac sells fretboards slotted to 26-3/16 scale.  Or why Roger Siminoff's book Constructing a 5-String Banjo shows layout for a 25-1/2-inch scale.

esullivan - Posted - 07/06/2017:  10:44:58

The necks we made for Gibson all had a 26 1/4 inch scale. Go figure....


Eric

Old Hickory - Posted - 07/06/2017:  11:40:42

OK.  26-3/8 pre-war.  After that,, who knows! 

You didn't use your vintage scale for Gibson?

Do you know how or why they spec'd 26-1/4?

sunburst - Posted - 07/06/2017:  12:41:46

AFIK, those "26-3/8" " prewar Gibsons were actually 26 1/4".

My hypothesis as to why "26 3/8" " is reported so often is: someone measured from nut to bridge, and with the bridge set for good intonation, it was pretty close to 26 3/8" from the nut. The correct way to determine scale length, of course, is to measure from the nut to the 12th fret and double the number.

Edited by - sunburst on 07/06/2017 12:44:55

rb4player - Posted - 07/06/2017:  12:55:10

My experience is that Gibson neck dimensions have been all over the place, whether modern or vintage. Hence not much value in trying to sort out fine differences. YMMV.

Best of luck.

Jim

sunburst - Posted - 07/06/2017:  12:57:29

Should have mentioned in my earlier post:
Best thing to do is get your hands on the necks you want to copy, take your own measurements, make neck profiles, photocopy inlay patterns, etc..
Measurements from someone else are useful, but there is no substitute for actually holding the thing you want to copy.

bluegrassboy4life - Posted - 07/06/2017:  13:02:41

So perhaps if I can buy a Gibson Banjo sometime get all of my measurement from that?

esullivan - Posted - 07/06/2017:  13:10:35

Old Hickory,

I really think they went with that scale because that is what we offered and we could have them inlaid by Custom Inlay.  I do not believe the the thought process went any deeper than that.  The 26 1/4 scale works good enough and is basically accepted as a Gibson scale but the 26 3/8 scale we use now came directly from an original pre-war Gibson 5-string neck and actually seems more like a, for lack of a better term, tempered scale as the math does not work out as a geometric progression but it notes out great.  I think the old guys knew what they were doing.

Eric

pickin_fool - Posted - 07/06/2017:  13:12:57

"So perhaps if I can buy a Gibson Banjo sometime get all of my measurement from that?"

if i am not mistaken i think stewart-macdonald has a poster..exploded view..of a mastertone that gives you all the pertinent dimensions you need

Muskrat - Posted - 07/06/2017:  13:28:51

quote:

---

Originally posted by esullivan

… the 26 3/8 scale we use now came directly from an original pre-war Gibson 5-string neck and actually seems more like a, for lack of a better term, tempered scale as the math does not work out as a geometric progression but it notes out great.  I think the old guys knew what they were doing.

 

Eric

---

That's fascinating Eric. A few more details about the geometric progression (or lack of same) would be appreciated.

I've always been interested in scale length and it's effects on tone, string tension and intonation. In the world of guitars, a lot of people use the terms Gibson scale and Fender scale. But not many people know that Gibson used the 25.5-inch scale years before the Fender company existed. For example on the famous L5. Gibson has always used the longer scale and still does. It's only because Les Pauls and 335s are 24.75 (approx.) that folks seem to think all Gibson guitars use the shorter scale.

rcc56 - Posted - 07/06/2017:  16:18:11

My experience with Gibson's "tempered" scales on prewar guitars and mandolins has not been positive.  Most of those instruments have intonation problems.  Some are worse than others, because Gibson's tooling or assembly practices did not consistently result in the frets landing in the same place from one fingerboard to the next.  I have found it necessary to fill and recut fret slots on a number of Gibson boards over the years to improve the intonation of an instrument.  In a few cases, the frets were so badly placed on the original boards that it was necessary to replace the board.  I have no reason to believe that the banjos are any better than the guitars.

Fret location is calculated by using a chain equation.  The first fret location is calculated by dividing the scale length by 17.817.  The second fret location is calculated by subtracting the distance between the nut and the first fret from the scale length, and dividing that result by 17.817.  For the third fret, subtract the distance from nut to second fret from the scale length, and divide by 17.817.  And so on.  In the old days, we had to figure this out by hand.  Now there are online fret calculators that do the work for us with a few key strokes.  One of them is on the Stew-mac website. 

I have written at length on this subject before, but all I will add here is that Gibson's "tempered" scales are not very well tempered at all, and are sometimes positively ill-tempered.​  I have never moved frets on an old Gibson board without achieving an improvement in the intonation of the instrument.  It's tricky work, though, and I only do it when an instrument is pretty far off.

  

rudy - Posted - 07/06/2017:  16:22:12

quote:

---

Originally posted by esullivan

Old Hickory,

I really think they went with that scale because that is what we offered and we could have them inlaid by Custom Inlay.  I do not believe the the thought process went any deeper than that.  The 26 1/4 scale works good enough and is basically accepted as a Gibson scale but the 26 3/8 scale we use now came directly from an original pre-war Gibson 5-string neck and actually seems more like a, for lack of a better term, tempered scale as the math does not work out as a geometric progression but it notes out great.  I think the old guys knew what they were doing.

 Eric

---

One of the time-tested methods of developing fret placement positions was geometric projection.  I've never heard the term "geometric progression" used, but I'm certainly not familiar with all of the terminology used in the past.  

My own strong suspicions are that Gibson used geometric projection (well known to established luthiers of the time) to develop fret placement "storyboards".  The simple act of copying these storyboards from someone else's bench most likely resulted in the small discrepancies in fret placement that can be seen in many early factory instruments.  I'm positive that they simply wrote off the need for accuracy, as their general market wouldn't know the difference.  The consumer of fretted instruments is much more educated today.  I've got a chapter written titled "Locating Frets, The Story Of Equal Temperament, How Frets Have Been Located In The Past, and An Alternative For Proper Placement " that explains a bit about these old techniques that I'll eventually get posted here on BHO.

Old Hickory - Posted - 07/06/2017:  17:58:05

quote:

---

Originally posted by esullivan

I really think they went with that scale because that is what we offered . . . .  The 26 1/4 scale works good enough and is basically accepted as a Gibson scale but the 26 3/8 scale we use now came directly from an original pre-war Gibson 5-string neck and actually seems more like a, for lack of a better term, tempered scale as the math does not work out as a geometric progression but it notes out great.

---

That's consistent with what I was trying to say about the consensus on the Hangout that Gibson scale was 26-3/8.  I should have specified "pre-war."  There have been numerous discussions of scale length on the Hangout in which people who have seen and measured many banjos said that every original old Gibson they had ever measured had a 26-3/8 scale.  That's close enough to 26-1/4 that I imagine it sounds and feels the same.  And the bridge can end up in the same spot on the head with just a slight change in the amount of wood beyond the 22nd fret.

Two question if you're still following this:   Was the profile of necks you made for Gibson the same as the profile on your own Sullivan banjos, such as the Radcliff?  And were you making necks for Gibson as far back as 1994?  Would you have made the necks for the RB-4 reissue?



As to use of the term "geometric progression" or "geometric projection" our friend Google says:

In mathematics, a geometric progression, also known as a geometric sequence, is a sequence of numbers where each term after the first is found by multiplying the previous one by a fixed, non-zero number called the common ratio.

esullivan - Posted - 07/06/2017:  17:59:38

Rudy,

I bet you are right about geometric projection and not geometric progression.  Not a super duper math whiz  here but the correct terminology may help someone do some research on their own, thanks.  As far as Gibson scale is concerned, I do not believe anyone outside of engineering even cared what the scale was and I have personally experienced the faults of a copy of a copy of a copy.  You may want to check on the old banjo scale though.  The math does not work out at all, period, but the banjos that we have used it on note out much better than the "geometric" scale we used to use.  For me proof is in the final product and I have finally come to the point where I do not have to question everything and just accept the things that work.  BUT, any idea when the practice of geometric projection and fret placement on the banjo began?   Again, thank for the proper terminology.

Eric

rudy - Posted - 07/06/2017:  18:19:32

quote:

---

Originally posted by Old Hickory

 

quote:

---

Originally posted by esullivan

I really think they went with that scale because that is what we offered . . . .  The 26 1/4 scale works good enough and is basically accepted as a Gibson scale but the 26 3/8 scale we use now came directly from an original pre-war Gibson 5-string neck and actually seems more like a, for lack of a better term, tempered scale as the math does not work out as a geometric progression but it notes out great.

---

That's consistent with what I was trying to say about the consensus on the Hangout that Gibson scale was 26-3/8.  I should have specified "pre-war."  There have been numerous discussions of scale length on the Hangout in which people who have seen and measured many banjos said that every original old Gibson they had ever measured had a 26-3/8 scale.  That's close enough to 26-1/4 that I imagine it sounds and feels the same.  And the bridge can end up in the same spot on the head with just a slight change in the amount of wood beyond the 22nd fret.

Two question if you're still following this:   Was the profile of necks you made for Gibson the same as the profile on your own Sullivan banjos, such as the Radcliff?  And were you making necks for Gibson as far back as 1994?  Would you have made the necks for the RB-4 reissue?



As to use of the term "geometric progression" or "geometric projection" our friend Google says:

In mathematics, a geometric progression, also known as a geometric sequence, is a sequence of numbers where each term after the first is found by multiplying the previous one by a fixed, non-zero number called the common ratio.

---

Google is correct about the mathematical relationship of geometric procession, but geometric projection is actually a mechanical method of developing fret placement, and as in any mechanical procedure, the results are only as good as the person wielding the compass used to generate the fret positions.  I have graphics to demonstrate how it's done, but the explaination is also widely available on the web.

rudy - Posted - 07/06/2017:  18:45:20

quote:

---

Originally posted by esullivan

Rudy,

 I bet you are right about geometric projection and not geometric progression.  Not a super duper math whiz  here but the correct terminology may help someone do some research on their own, thanks.  As far as Gibson scale is concerned, I do not believe anyone outside of engineering even cared what the scale was and I have personally experienced the faults of a copy of a copy of a copy.  You may want to check on the old banjo scale though.  The math does not work out at all, period, but the banjos that we have used it on note out much better than the "geometric" scale we used to use.  For me proof is in the final product and I have finally come to the point where I do not have to question everything and just accept the things that work.  BUT, any idea when the practice of geometric projection and fret placement on the banjo began?   Again, thank for the proper terminology.

 Eric 

---

Hi Eric,

I don't have any solid dates, but the practice was developed and widely known and used thoughout the ages of early manufacture of musical instruments.  The basic premise was the result of mathematicians who understood the basic ratios that resulted in the "pure intervals" as related by Pythagoras' early experiments with harmonic structure of sound.

One of the earliest applications by luthiers was the "Rule of 18".  Derived as possibly the first (and crudest) mathematical ways of scale generation, it may have been what some of the earliest designers used for initially finding the "starting point" for geometrically projected fret positions, and may have been where some of those "false intervals" were derived from.  This number was indeed crude by today's standards, but it was a much simpler method than figuring out the exact numerical representation of the 12th root of 2.

Some of this is only some wild speculation on my part, but it dovetails nicely with the oddities of instruments made by the largest early "factories".  You have to be able to come up with some sort of logical answer why frets were obviously out of position according to what we can now calculate by punching in a number on a spreadsheet.

True historians would most likely have access to a closer version of the actual workings, but I know I'm pretty close.  I don't believe anyone actually knows the minutia associated with exactly HOW Gibson (or any other early manufacturing entity) determined fret placement.  I think anyone who believes that they had anything more than a crude understanding of fret placement as related to equal temperament is deluding themselves, plus I just don't think that it was a strong selling point at the time.

esullivan - Posted - 07/06/2017:  19:07:27

One issue I could never get a handle on was how you account for neck relief in your geometric projection. It seemed to me that the frets in the middle of the neck would be further away than frets at the beginning or end. Any ideas?

Eric

rudy - Posted - 07/06/2017:  19:20:46

quote:

---

Originally posted by esullivan

One issue I could never get a handle on was how you account for neck relief in your geometric projection. It seemed to me that the frets in the middle of the neck would be further away than frets at the beginning or end. Any ideas?

Eric

---

The deviance in distance between frets resulting from neck relief are so miniscule that they can be ignored.  The ACTUAL amount of frets being out of position is much greater than what you would suspect before being a factor.  The average pitch deviation for most players can perceive is something on the order of three cents.  What that translates to be is approximately 1/16" at the first fret and about 3/64" at the octave.  That type of accuracy is achievable with a common tape measure used carefully.

Obviously, we as makers choose to go for a little higher accuracy than that, but folks obsess over the numbers far too much.

Relating to the discussion about what Gibson did or did not do, there's an archived topic that has some comments from Bill Porter and he's about as close as we'll ever come to hearing it directly from the horse's mouth:

banjohangout.org/archive/225965

sunburst - Posted - 07/06/2017:  19:46:59

A few things:
Old Gibson mandolin fret boards were notoriously inaccurate, including the Loar-signed mandolins. People didn't have much problem with them in those days. People are much more "picky" now.
I once had pre-slotted mandolin fingerboards from 3 different sources. One day I casually laid them side by side and found that they were all different. That was before CNC and before people were so "picky". It is also the reason I made my own fret slotting jig.
When I built my first guitar, I re-measured the frets and found that some of them were over 1/32" off. I figured the guitar would be horribly out of tune, but when it was finished it played (and plays) in tune just fine.
I once adjusted the intonation on a guitar for a player who plays in DADGAD tuning all the time. The guitar was, of course, set up for standard tuning, so the saddle had to be re-cut for intonation in DADGAD. I also adjusted the nut and had to add a small piece of bone to the saddle. When I was done, every string at every fret was within 3 cents and that was as close as I could get it because further adjustments would have made it worse somewhere while making it better somewhere. In other words a string might be 3 cents sharp on one fret and 3 cents flat on another... no way to improve that. Another string might be different, so even moving the frets, if that were possible, couldn't have helped.
Bottom line: frets cannot be perfect, tuning inaccuracies are par for the course, and slightly misplaced frets are not a problem because of all that.
Even with perfect strings noting on perfectly spaced frets, we would still be playing in equal temperament and thus, in one way of thinking, we're out of tune from the get go.

esullivan - Posted - 07/06/2017:  20:31:35

John and Rudy,   Thanks for the further explanations.  You guys make the hard to understand somewhat easier.  Thanks,  Eric

rcc56 - Posted - 07/06/2017:  21:09:19

I have found that +/- 1/64" of ideal fret placement [the 17.817 thing] is suitable for most of my more sharp eared clients.  Above the 12th fret the tolerance probably needs to be a little tighter for best results.  Some of the Gibson boards are nowhere near that close.  Mandolins are more of a problem because of the shorter scales.  [Closer fret spacing means tighter tolerances].  I have found that Gibson fret placement was usually tolerable in the 'teens, and started getting worse around 1920.  The worst I ever saw was an early '30's mandolin.  I had to replace the fingerboard on that one.  I did include the original in the case when I sold the mandolin.  The new owner can hang the board on the wall for a trophy if he so wishes-- it was useless on the instrument.  If it wasn't for the obsession with originality, it would have been of more use as shim stock in my shop.

The names of a few prominent players who have had boards corrected on Gibson instruments are Chris Thile, John Reischman, and Norman Blake.  I believe Lynn Dudenbostel does Chris Thile's work.  I don't know who did John Reischman's mandolin.  John Arnold has done some of Norman's work, and I have also. 

Another cause of intonation problems is deeply grooved frets.  Sometimes these can be recrowned, but it is often necessary to replace the offending frets.

Sometimes a cantankerous Gibson guitar or banjo can be brought into reasonable temper by adding a sliver to the end of the fingerboard and moving the nut back about .015".  Conversely, I moved the nut forward about .045" on a '23 Vega style M that was playing sharp on the lower frets.  It had been cut off a bit long at the factory.  I rarely see fret placement or fingerboard cut-off length problems on Martins.  However, Martin has been really bad about their bridge saddle placement on and off since about 1946 or so.  Some of the '70's guitars are especially far off, and the problem continues even on some recent issue guitars.  Whenever somebody brings in a Martin with a loose bridge, the first thing I do is measure the saddle placement-- it's easy to plug and recut the saddle slot when the bridge is off the guitar.  It's amazing how good the intonation can be on a modern Martin with a CNC cut board can be when you get the saddle in the right place! 

And then, there's the issue of learning to temper an instrument when you tune it . . .  anybody want to start another thread??

Edited by - rcc56 on 07/06/2017 21:19:30

rcc56 - Posted - 07/06/2017:  21:17:05

If anybody wants to see how laying out a scale geometrically is done, there is a great diagram with explanations in Hideo Kamimoto's "Complete Guitar Repair," page 91 in my old dog-eared copy.  This book is still in print and is highly recommended by better men than me for anyone interested in fretted instrument repair.

monstertone - Posted - 07/07/2017:  15:02:38

The formula for working out a fret scale doesn't care whether the scale is 26.25, 26.375, or 26.50. It's a formula,,,based on the 12th root of 2 (1.059463094) or the reciprocal (0.943874313), depending on whether you divide or multiply. Numbers are numbers. With the exception of the 12th fret, they are not gong to "work out" to even fractions, decimals, millimeters, or anything else. The proof that you got it right is when the 12th fret is exactly half of the scale length.

In days of yore, figuring this out was extremely difficult, not to mention time consuming. Once you crunched the numbers, they became a highly guarded trade secret. And every banjo manufacturer proclaimed their banjo's to be superior to everyone else's for various reasons, not excluding their scale length! Now you can put an app on your phone & plug the numbers into a CNC machine in less time than it takes to explain how to do it.

Edited by - monstertone on 07/07/2017 15:14:19