Loar Fret spacings.

[Ivan Kelsall]

Myself & a few Cafe contacts have been viewing with great interest,the awsome worlk done by Steve Gilchrist on the first known 'signed' Lloyd Loar mandolin. On his website SG shows the original Loar fingerboard side by side with the new one that he made, & the incorrect fret positions on the original are clearly shown.
I think that most of us on here know that other 'Loars' have required a new fingerboard because of the same reason - but how did Gibson get such a fundamental thing as the fret positions wrong ?.

Gibson had been producing 'fretted instruments' for many years & i can't understand how such an error occurred. Did the original owners not notice anything odd in the intonation ?,
Ivan
Pic.courtesy of Steve Gilchrist's website.

[Beanzy]

Our 12 tone equal temperament compromise system is equally out of tune in all keys.
Although used and known about for a long time it was seen as too much of a compromise artistically by many and was just one among many possibilities, This currently popular system was not adopted very widely until after Loars time at Gibson. Interestingly our ears are now so universally educated to the 12TET system that many people are thrown when they hear playing in a given key with a more perfect system.

I've not ever heard anyone describe any study of which of the systems were preferred by Gibson at the time. I presume it would have been chosen with the repertoire &popular keys for orchestral playing at the time they chose. It would be good to see if any records of the choices were available.

From your photo it looks like there was an alignment at the 5th of each course. Players in the past would often also tune with 'narrow 5ths' between courses, centred on an A at whichever tuning fork pitch was preferred. So you would match the A then slightly flatten the e, raise the D a couple of 'cents'and the G would come up a couple more as well. So visually interpreting the fretboard would also have to find out if they were going with the narrow 5ths tuning too.

[Hendrik Ahrend]

Our 12 tone equal temperament compromise system is equally out of tune in all keys.
Although used and known about for a long time it was seen as too much of a compromise artistically by many and was just one among many possibilities, This currently popular system was not adopted very widely until after Loars time at Gibson. Interestingly our ears are now so universally educated to the 12TET system that many people are thrown when they hear playing in a given key with a more perfect system.

I've not ever heard anyone describe any study of which of the systems were preferred by Gibson at the time. I presume it would have been chosen with the repertoire &popular keys for orchestral playing at the time they chose. It would be good to see if any records of the choices were available.

From your photo it looks like there was an alignment at the 5th of each course. Players in the past would often also tune with 'narrow 5ths' between courses, centred on an A at whichever tuning fork pitch was preferred. So you would match the A then slightly flatten the e, raise the D a couple of 'cents'and the G would come up a couple more as well. So visually interpreting the fretboard would also have to find out if they were going with the narrow 5ths tuning too.

Sorry Beanzy, I don't agree with that. There is no "perfect system" to begin with. There were typical systems for certain types of music or eras. In Gibson's time, everything was equal temperament, which basically means that all major thirds are 13.7 cents off (too wide). Around 1700 meantone temperament was in use (on keyboard instruments): 8 pure major thirds, 4 hardly (or un-) usable. It was a necessary compromise, because back then the instruments generally sounded quite bright (lots of overtones; think of the colorful sounds of pipe organs and harpsichords) and not many composers, with Bach and a few others being the exception, composed in "off" keys. (I won't touch upon the many in-between, "well-tempered" systems invented by Werckmeister, Young, Neidhardt and other theorists.)
Equal tempered major thirds sound weird on those older brighter instruments - even to modern ears - whereas a modern piano (with felt hammers!) doesn't generate as bright a tone, and a major third 13.7 cents too wide (from being pure) sounds okay. That "our ears are now so universally educated to the 12TET system" is a tendentious myth repeatedly used by those, who are not fond of older musical styles, also by organists, who can't transpose and desperately have to play any Bach piece in the original (proper) key - of course, without knowing the original pitch or temperament. I better stop here, since there is way more to this 500 year old ongoing discussion.

The Gibson company no doubt made their instruments for the modern system of the day, which was equal temperament. Their instruments were all about the latest, newest and progressive inventions.

Gibson's fret spacing problem was a technical one. As MC member and ex Gibson employee Bill Halsey pointed out somewhere here on the Café, the fingerboards were cut on a gang saw with an arbor and multiple circular saws. Every now and then those saws had to be taken off the arbor and sharpened. And it may have happened that the spacers between the saw blades weren't set back in the proper position, hence the weird fret spacing.

[Hany Hayek]

This discussion is great, although you are already starting to disagree.
Beanzy, I had never heard of the narrow fifth tuning. I looked it up and the first thing that came up is that a violin is tuned in a slightly narrower tuning, seems to compensate between G and E. Could you explain that. Thanks.

[Ivan Kelsall]

Whilst understanding that there is no 'perfect system' as Henry Eagle explains,my question was more about the ''mechanics of getting the fret positions wrong''. I agree that 'maybe' as Bill Halsey apparently explained,that the gang saw might have been stripped down for re-sharpening,but surely the spacers between the saws would (should) have been identified with their position
on the arbour ie.1 / 2 / 3 etc. in order to be correctly re-assembled ?. Seeing the photos.from Steve Gilchrist's website,& noting that the 12th fret is at the same position on both fretboards,makes me think that it was more the fact that the fret position calculation was incorrect ?.

The specification quoted for the F5 body size top & back was as per the F4 model. The spec.for the neck outlined a 3 piece neck - ''The design according to drawings. Dimensions standard''. Which 'standard' dimensions were they, & did they include the fret positions ?. I'm wondering if some dims. were carried over to the F5 from the F4 that should have been adjusted & weren't.
It is pretty weird that a company such as Gibson could make what seems to be such a fundamental error. Could the fact that the F5 neck joined the body at the 15th fret rather than the 12th fret as per the F4, have muddled things up - Ivan

[HoGo]

I think thre is even more than just what is mentioned above...
If you align the boards at 1st fret instead od 12th you'll get most of the lower frets OK slowly departing up the board (and nut position cut a bit short - which many makers do even these days intentionally), that is sign that the scale length they used was slightly longer than what is is compared to on the pic (the boards are edge to edge and each has different angle of edge to centerline so this throws it off and on old Gibsons frets were often cut not perpendicular to centerline but all askew - like when you don't align the board with the gang saw). Gibson advertisements mention 13 15/16" scale while modern is 13 7/8" - that's what you get when you take 0-12th measurement and double it on modern board but back then thay didn't have calculators and there were various simplified methods of calculating fret position which were not precise and error accumulated towards higher frets (when compared to 12th root of 2 method simply calculated on your iPhone...). The error will not go in just one direction and the result will float because if you round the numbers, sometimes you round up sometimes down ...
Someone posted line graph of fret positions measurements of various Gibson tenors and few other instruments and they all showed the same schematic error when compared to perfect board so my first guess would be this one follows the same formula.
I will try to find time and blow it up in photoshop to find out what exactly is happening here...

[HoGo]

Here is old discussion about this...
http://www.mandolincafe.com/forum/sh...mpromise/page2

[Tobin]

I think it's also important to keep in mind that in Loar's day, they used different strings than we have today and tended to use a much higher action than modern players. Intonation up the fretboard has a LOT to do with the tensile nature of strings used, as well as nut/bridge height, neck relief etc. What we perceive as "incorrect" on his part may not necessarily be incorrect at all. What's "incorrect" is to judge it by our modern standards. It was likely just a matter of Loar making intentional adjustments to fret spacings to achieve pleasing and correct sounding notes, using the string types and action heights that were common during that time. Keep in mind that players in the mandolin orchestra heyday had a completely different set of expectations from their instruments than we do today, and they "heard" the musical notes differently. They did not have electronic tuners; they used their ears. And fret spacings may have been adjusted to "sweeten" certain notes or make them fit in better with the rest of the orchestral instruments that were in play.

Loar was no dummy. He didn't spend all that time perfecting his F5 design just to muck it up with the fret spacings. If it had sounded wrong at the time, he would have corrected it. I believe that his fret spacings were indeed correct for the standards and materials being used at the time. And I don't know if it's even possible to reproduce it today to check.

[Ivan Kelsall]

I just had a quick read through the link that Adrian posted, & having read that the ''theoretical reasoning'' for incorrect fret spacing maybe being the actual ''tuning frequency used'' ie. 435 Hz etc. not 440Hz as today,Paul Hostetter had already posted my thoughts = it wouldn't have any effect on fret spacings. Tune your "A" strings correctly & they should vibrate at 440Hz. Tune them down by a few 'cents' & the notes will still be the same intervals apart as you play the strings up the fingerboard. David Collins also stated that this would be the case.

Years ago (1963) i fretted a fretless banjo. Somebody gave me the formula to work out the fret spacings & i did the division by longhand - like we used to do & it worked out ok. Maybe this article could go some way to explaining the differences between 'then' & 'now' :- http://www.liutaiomottola.com/formulae/fret.htm ,
Ivan

[Hendrik Ahrend]

Loar was no dummy. He didn't spend all that time perfecting his F5 design just to muck it up with the fret spacings. If it had sounded wrong at the time, he would have corrected it. I believe that his fret spacings were indeed correct for the standards and materials being used at the time. And I don't know if it's even possible to reproduce it today to check.

I don't think Loar had anything to do with the fret layout, since there was only one 13 15/16 mandolin scale. Why should he have changed it (for the worse) on the F5? So what Adrian says only makes sense to a degree. BTW a similar weird fret layout appears on Gibson tenor banjos.
And I remember a teens A model with the second fret being off, the most audible effect being a good B note in a G-chord, and a flat E note (on the D-string) in an A-chord.

[gtani7]

This discussion is great, although you are already starting to disagree.
Beanzy, I had never heard of the narrow fifth tuning. I looked it up and the first thing that came up is that a violin is tuned in a slightly narrower tuning, seems to compensate between G and E. Could you explain that. Thanks.

I think the issue is Pythagorean intervals vs just intonation vs equal temperament (12TET or 12 equal tone divisions). There's some videos on youtube and blogs about this: http://joostdevblog.blogspot.com/201...tuning-in.html

[ProfChris]

Has anyone calculated whether the original fretboard was rule of 18, rather using than the current (mathematically correct) divisor of 17.817?

This rule of thumb method of fret calculation makes all the fret spacings a fraction smaller, though not by uniform amounts, and produces built in compensation. It was in common use for centuries.

[David L]

I think it's also important to keep in mind that in Loar's day, they used different strings than we have today and tended to use a much higher action than modern players. Intonation up the fretboard has a LOT to do with the tensile nature of strings used, as well as nut/bridge height, neck relief etc. What we perceive as "incorrect" on his part may not necessarily be incorrect at all. What's "incorrect" is to judge it by our modern standards. It was likely just a matter of Loar making intentional adjustments to fret spacings to achieve pleasing and correct sounding notes, using the string types and action heights that were common during that time. Keep in mind that players in the mandolin orchestra heyday had a completely different set of expectations from their instruments than we do today, and they "heard" the musical notes differently. They did not have electronic tuners; they used their ears. And fret spacings may have been adjusted to "sweeten" certain notes or make them fit in better with the rest of the orchestral instruments that were in play.

Loar was no dummy. He didn't spend all that time perfecting his F5 design just to muck it up with the fret spacings. If it had sounded wrong at the time, he would have corrected it. I believe that his fret spacings were indeed correct for the standards and materials being used at the time. And I don't know if it's even possible to reproduce it today to check.

Different strings and/or string height would change the compensation incrementally up the neck and would not affect different frets differently. "Sweetening up" certain notes to compensate for equal temperament doesn't work because the same note needs to be lowered for some chords and raised for others.

[pops1]

My '22 A2 wouldn't play in tune and drove me nuts. Finally I decided to make it work and started measuring frets to replace the board. The nut was 1mm to close to the first fret. After moving it back the mandolin played wonderfully in tune. I though it was a worn template being used during the transition to Loar specs as mine has the wider board yet, but truss rod, adjustable bridge. After reading this I think it was due to the strings at the time. I have changed other '22''s that had the same problem and each played in tune after moving the nut back about 1mm.

[dhergert]

Very interesting discussion... So, Pops1, could this imply that the fingerboard was cut 1mm too short at the nut?

[Mandoplumb]

I don't see why string height wouldn't change intonation, although very slightly. If nut height remains constant and strings arer raised wouldn't you be stretching the string more at the 12th fret.yes you could move the bridge back but wouldn't that make the first fret note flat, you haven't changed the height there. I know that's the way we set intonation but all intonation is really incorrect and a compromise.

[pops1]

Very interesting discussion... So, Pops1, could this imply that the fingerboard was cut 1mm too short at the nut?

I think so, the measurement from the first fret to the nut was less than the first fret to the second fret, and that can't be. Shimming the nut back make it play wonderfully in tune, and saved me from putting on a new fingerboard. I hadn't measured before and was measuring to figure out scale length to replace the board. The shim was kind of an experiment, but it worked so well I haven't taken the maple out and put in ebony, just colored the maple shim black. Shhhhh.

[Tom Wright]

I had some sympathy for the "intonation is really incorrect and a compromise" feeling until I fixed the spacing to the first fret on my Buchanan (reduced by about .010"). It was too far, and even with the nut bottoming out the first few frets played sharp. This made tuning it very frustrating, as I had to squeeze those fifths mercilessly in order to make the octaves acceptable, and then finger oh so carefully to avoid playing sharp.

Now it plays easily in tune for all fingered intervals. I find it seems to stay in tune better, which really means it is more forgiving of slight drift. I find fingering so much easier because I am not leaning this way and that to make a chord acceptably in tune.

The only test that matters is playing. I have never been unhappy with tempered tuning---who really hates pianos? Or organs? They sound fine to me, and now so does my mandolin.

There is a correct fret system, and when it works the tempering complaint goes away. I think many high-nut intonation issues might be this problem instead.

[dhergert]

So the shim did two things -- it lengthened the scale length, and it essentially moved all the frets further from the nut... The scale length change could be accomplished by moving the bridge. But if I'm understanding correctly, the scale length change isn't what fixed the general intonation; it was the distance of all of the frets from the nut that fixed that.

Very good!

[pops1]

I think the board was just cut too short when it was made. It made the scale length proper to what it should have been and was fretted for. If you intonated at the 12 fret it would not play in tune. In fact it just would not play in tune, especially bad on the A strings. A friend of mine and fellow mandolin player said "why don't you sell that and get something that will play in tune" I told him it sounded so good if I have to put on a new fingerboard to make it play in tune so be it. Having the nut too close to the first fret made if off on every fret. There was no place I could set the bridge and sound good. This was too many years ago to even remember how many years it's been. Anyway several people said their A2 would not play in tune, all 22's. That little fix worked on all of them. That's why I thought worn out fingerboard jig and transition years, who knows.

[Willie Poole]

One thing that hasn`t been mentioned here is that on a lot of older mandolins had a dark place where their bridges were placed and when we string them up with our modern gauge strings we have to move the bridge about 1/4 in. toward the fingerboard to get correct intonation at the 5th and 12th frets...SO there must have been some huge differences in what Loar found with strings that were available in his day...As mentioned above I don`t think the older players played very much up the neck and were mostly interested in what a mandolin sounded like when playing on the first 5 fret positions and they only had their ears to try and tune one properly...

I know that doesn`t answer Ivan`s question and I doubt if anyone will, we can all speculate as to went on though...

Willie

[Jim Garber]

I brought my former '24 F4 to Brooklyn luthier Bob Jones many years ago and he showed me the very same thing. It was the norm for some Loar era instruments. I never actually looked close to my '23 A2 to see if that is also the case.

[FLATROCK HILL]

I brought my former '24 F4 to Brooklyn luthier Bob Jones many years ago and he showed me the very same thing. It was the norm for some Loar era instruments. I never actually looked close to my '23 A2 to see if that is also the case.

Hey Jim...I've followed this thread (and this subject for some time now) with great interest.
Could you clarify, just for my own understanding...by "the very same thing" are you talking about a short fingerboard, mis-cut fret slots, or the bridge being placed incorrectly?

[MikeEdgerton]

Was this also exacerbated by the worn spacers on the gang saw cutting the frets? That's been discussed here several times over the years.

[Ivan Kelsall]

Mike - I understand your point,but to have the spacers 'wear',they'd have to be loose. It was the fact that the sawblades on the arbor were clamped very tightly that enabled them to actually cut. If they were loose,they'd freewheel & most likely jam. I don't know if it was the case at the time,but many modern circular saw blades have a 'key slot' that locates on a 'key ridge' on the arbor, the same goes for the spacers in order to prevent any wear through friction between the saws & the spacers. However,if that wasn't the case - then you'd be perfectly correct,
Ivan