Official Luthiers Forum!

Iʻm creeping along on my OM guitar project - you may remember that Iʻm a bowed string builder and this is my first guitar. Not being a very organized sort of guy, I tend to procrastinate on planning ahead. Now I have a finished body and my neck is mostly shaped and almost ready for a fingerboard and fitting. So Iʻve started to get more interested in bridge and saddle placement and intonation. Iʻve read a bunch of threads and thank goodness thereʻs a fair amount of consensus on saddle angle, setback and such. If there are a thread or three with detailed placement info that I may have missed, please do tell - nowʻs the time to absorb it all, I guess.

In the process, I came up with one question - Iʻll try to get to it without being too longwinded...
Iʻm going to use my old Larivee L10 as an example. I got all excited recently about testing its intonation carefully because Iʻve discovered a few things from playing a lot of Hawaiian slack key music on it:

1. I usually play in tunings with my bottom string down to C or even Bb because I love the rich sonority of those very wide tunings. That makes the bottom string or two pretty flabby, so Iʻve ended up stringing with light/medium string sets, or .012" - .056". The thicker bottom strings help to eliminate the "strung with spaghetti" feel on the bottom string and the splattery sounds when you accidently pick too hard (I didnʻt say, but this is all fingerpicking, no flat pick - and for me, no thumb or finger pick).

2. When I do this, my bottom string is pretty sharp when Iʻm up in the 5th - 7th fret range - and a lot of Hawaiian style playing is all over the neck and often quite far up into cutaway territory. Unless I finger very lightly on that bottom string and try to correct on other ways, Iʻm going to be sharp.

So - my question is about seat of the pants rules for intonation correction. Assume Iʻm checking intonation at the 12th fret, very light finger pressure, and Iʻm now seeing +10 with the string coming off the front edge of the saddle. How much change am I likely to see with a saddle thatʻs shaped so the string is coming off the rear edge?

Basically, since I have no experience working on guitar intonation issues, I have no sense how much change I can expect from standard saddle shaping - not talking about cantilevered saddles - yet <g>... Iʻm looking for some practical adjustment rules.

Hi Peter you are almost as long winded as I can be....

So I set-up around 600 guitars a year and many of them are for "shredders" who want a low string or even a 7th string to be tuned to B, B# and in the C's. These are electric guitars who are doing the very low, slack tunings. I work on acoustic instruments too hundreds of them but it's unusual in my experience to see someone attempting to tune as low as you are with an acoustic.

I would suggest since you enjoy the process to set-up a test rig with a dummy nut, scale length and saddles and then do the work to determine the intonation that you wish for the tunings that you plan on using most of the time. "Most of the time" is the important part here because any attempt to have great intonation with alternate tunings has to be a compromise of sorts with something perhaps not being as well served as the tuning you intonated for.

Be sure... to use the exact brand and gauge of strings that you intend to use because different brands and gauges of strings intonation differently.

So determine the scale length for each string tuned as you wish and record (write it down) same. Then you can begin to see what your saddle (or saddles you do not have to have only one....) may need to do to get you where you want to go.

Any standard scale acoustic guitar with normal intonation for standard tuning will not do what you are asking to do. The 10 cents off is less than I thought it would be and that is way too much to be off.

I play a lot too and have many of the gigging musicians that we work for over to my place often and I have a reputation for stopping in the middle of the song and getting a screw driver and correcting intonation since I absolutely cannot stand playing out of tune. So intonation snob here and if you're one too you will relate.

Bottom line you are asking an acoustic guitar to do what it was never intended to do and that means your intonation will be off. So your saddle locations will have to be different and this instrument will only play well for the tuning that you intonate it for. So be sure you want to do this because the instrument will be so specialized when it's set-up for the tunings you indicated that it will be useless to anyone else.

Back to what I do so often even the electrics with separate saddles can't properly intonate range of travel wise C or even B and when we take them in we tell our clients that we will help them but our normal, 2 cents of accuracy on intonation that we shoot for is not guaranteed for them because their chassis (guitar) does not have the range of adjustment to support it.

Be advised as well that when you check intonation how hard you fret and where are variables that can and do skew intonation. So since you are the player try to fret as you would while playing normally. I try to simulate a generic player when I set-up instruments.

There is another issue here too but it may not matter or it may matter I don't know. If it comes to pass that your saddle has to be located differently enough that your bridge is in a different location too the top bracing of your instrument may need to be adjusted location wise. But let's hope it doesn't that's getting very complicated.

Lastly my friend let's discuss one of my favorite subjects, physics. So you like lighter strings for your very low tunings but physics and that dude Isaac Newton disagrees with you. With Newton's laws of physics we know about inertia and how a body in motion tends to stay in motion.

I used to think that thicker strings would lash out more and need more relief (clearance) from the neck. Wrong..... three demerits for me.... Turns out that a heavier string (thicker too) has more inertia to overcome so they do not lash out as far as a less massive, thinner string. Go figure... Anyway I'll never make that mistake again.

Hope this helps I can be annoyingly detailed at times but I do this for a living so it's on my mind and it's a labor of love too. I think at some point I will wake up as a guitar and never revert I am thinking, playing, fixing guitars including buying them so much of my time now I'm going to turn into a guitar.

The https://www.liutaiomottola.com/formulae ... sation.htm tool is my goto intonation calculator now. I recently changed my saddle slot including width so did a pretty deep dive with this tool. You can vary your parameters (string action, thickness, tension, ect...) to see the effect on intonation / string length, in inches or mm, very directly. Every calculator I've used yields a different point. I now know this is because they use slightly different set of parameters in the computation.

I was very surprised at the difference in intonating that JUST the string action height plays in the intonation point on the saddle. Its all interrelated so it does make perfect sense. Give the tool a whirl and just vary your action height or string tension and see the difference in the intonation point. Know that even changing your tuning (string tension) will affect the intonation point.

_________________
Measure Twice,

Karl Borum

What Hesh said: when you're going for something out of the normal range, some sort of test rig can save a bunch of aggravation and time.

I've been compensating both the saddle and nut for some time, based on Trevor Gore's work. I set up a rig using a rigid beam of wood with a tuner, a movable nut, fixed first, 11th and 12th frets, and a movable saddle. The saddle has a piezo pickup in it, which I can plug into my tuner, but you can live without that.

I set it up with the nut and saddle the same distance from the 12th fret; one half of the desired string length. Set the action height right at both ends, and get the string perfectly in tune. Fret the string at the 12th fret, and see how much it's sharp of the octave. Move the saddle away from the 12th fret to adjust it. Keep re-tuning it and adjusting the saddle location until you get the octave.

Now fret the string at the first fret, and see if it makes a perfect semitone. It will probably be sharp; after all, you stretched the string a little when you fretted it. Adjust that by moving the nut [u]toward[u] the fret. It may not take much.

When you get that right check the 12th fret intonation: it will be out by a little bit. Make the appropriate adjustments. Now go back to the nut and correct that.

After a few rounds you'll be chasing your tail: making very small adjustments and not improving the result much. Note the needed offsets (I measure them from the center of the 12th fret), and do the next string. I like to do each string three times, and average the results. This is not as precise a thing as it might seem it should be. You'd think there would be single 'correct' set of numbers for any given string and setup, but I find that a little more here and less there can work as well. for one thing, you never really fret a string the same way twice. Keep in mind, too, that strings on guitars don't work exactly the same way as strings on rigid beams.

Very slack strings can need a lot of compensation, at both ends. One of my students brought in a replica of an Elizabethan 'bandora', a bass cittern with wire strings (no originals exist). They're noted for having intonation issues, and this one certainly did. The saddle was already as far back as it could go, so we tried moving the take-off point at the nut end up. It had to shift by almost 1/4" to get it right. We filled and re-cut the nut slot, and it now plays in tune far better than most of it's kind. He gets compliments from other other folks who have them, and uses it a lot more in concerts since it sounds so much better.

After we did that he looked up some old wood cuts of the instruments, and noted that they tend to show the nut set at a similar angle. The bandora was a popular instrument, but went out of use shortly after the inventor died as nobody else seemed to understand his 'secret'. Maybe it was nut compensation....

Who, me? - verbose? Nobody ever said...before... <lol>...

Thanks guys - thatʻs a lot to absorb, much as I expected. I like the idea of a test bed a whole bunch - I like empirical tests a whole bunch and that would be most interesting to me. Iʻd hoped to complete this guitar before we head back to Hawaii the 2nd week of December but maybe I should take it to the point of being ready for the bridge, take it over, and then take the time to do the testing - weʻll see...

Strings - Iʻve been trying various Elixir 80/20 nano gauges, but recently started trying DʻAddario phosphor bronzes. I was talking to one of the guys at Strings By Mail and took a suggestion to string with a 13-56 set of XS phosphor bronzes. They made the Larivee sound absolutely gorgeous but I found it was really hard to play gently and smoothly and control the sustain as I like to, and Iʻm going to go back to a set of the light/medium XS strings - thatʻll put me where I usually play.

Thereʻs no question about action height and going sharp - also heavy fingering and mashing the strings down behind the fret - very audible. I try to play as lightly as possible, both for intonation and to get smooth lines with a lot of sustain. If I have time, Iʻll even finger a bit far from the fret and push upwards toward the bridge to lower the pitch a bit as necessary.

As for trying to make a guitar do something it was never intended to do - thatʻs the whole history of Hawaiian guitar playing <g>...
In a nutshell, many years ago the Hawaiians were introduced to guitars and they loved them. The foreigners who brought guitars to the island left the instruments behind when they went home - so the locals had some guitars but had no idea how to tune and play them. So they developed their own tunings and styles - lots of different tunings! Thatʻs an oversimplification but itʻs true enough - instruments were kept in a family, and everybody learned how to play in whatever tuning the family invented. Way back when the tunings were considered to be family secrets and they werenʻt passed around - but those times are long gone.

If you feel like taking a break from whatever, here are two examples of low tunings by really good players:

Rev Dennis Kamakahi - Hilo Rag
Dennis was a great entertainer and prolific composer - he influenced all later Hawaiian players. This is a very traditional rag in the old mainland style - I wonder what he was listening to that gave him the idea - though Hawaiian players performed all over the world for decades and decades, so they werenʻt sheltered musicians at all. Tuning C-G-e-g-a-eʻ

https://www.youtube.com/watch?v=uOFWssd ... Dv&index=4

Sonny Lim - Pau Hana Rag
Sonny comes from a family of musicians and dancers - heʻs very influenced by jazz and other forms - this is no way a traditional rag! He teaches at George Kahumokuʻs summer workshops on Maui - Iʻve been to 2 of Georgeʻs workshops. Sonny plays this in BBb=F-d-f-bb-dʻ. Itʻs really a variant of C-G-e-g-cʻ-eʻ, which is almost the same as Dennisʻ tuning above except that it doesnʻt have the 6th. But he tunes it down because he likes the deep, lush sound.

https://www.youtube.com/watch?v=0vJacqP ... rt_radio=1

A bonus of nut compensation is that since you only need about half the normal saddle compensation, your saddle width gives you twice the normal range of adjustability. So you have a better chance of being able to set up for your heavy strings without having to position the bridge so far back that it will never be able to use normal tuning (although you still may need a wider-than-normal saddle to do it). Or you can make ambidextrous guitars with a non-angled saddle that can be set up for left or right handed stringing without having to make the saddle huge and heavy.

My general rule of thumb is 1 pound of tension per inch of length for steel strings. So from the D'Addario tension chart, 25.5" scale tuned down to 65.4Hz C would need a string with unit weight = (25.5 x 386.4) / (2 x 25.5 x 65.4)² = 0.000886. Then we go down to the phosphor bronze acoustic guitar strings and see PB066 is closest at 0.00087718.

And thereʻs the rub!... I canʻt conceive of playing a 66 on the bottom, or anything close to that heavy! And although some of the professional players in Hawaiʻi have rather nice guitars that are set up well, Iʻve never gotten the tiniest squeak of conversation that they spend a lot of time worrying about stringing, intonation correction and such. If I ask about things like this, the usual answer is, "we just learn to cope" or some such.

And the best of the Hawaiian performers are extremely pitch conscious - itʻs one of the reasons they like big wide tunings that are rich and resonant, sometimes with 6ths or maj7th in them - theyʻre pretty aware of pitch and how chords resonate. They also understand that tunings have mood or soul - that different tunings present different types of songs in their best light. But I digress........

Is there a good thread here on nut compensation? Iʻve never thought at all about that and Iʻm curious.

You might find something if you do a search on 'Trevor Gore'. He follows a number of the guitar fora, and posts when the subject comes up. He wrote two books with Gilet: 'Contemporary Acoustic Guitar' Vol 1 'design' and vol 2 'building', and it's all in there, particularly the first volume. With math... These are expensive books, but you might find one used or get the loan of it.

I started seeing stuff on nut compensation back in the 70s, but it took a while before it got figured out. A lot of folks, Byers in particular, focused on string stiffness as the prime factor, but I always found the reasoning hard to follow. Gore approaches it primarily from the standpoint of tension change, and it makes more sense. It also works.... This is not to say that string stiffness doesn't come in: I've done enough work with strings over the years to realize that even though they're the simplest part of the system, they're not simple.

One of the nice things about Gore's treatment of it is that he acknowledges that, and also gives several different approaches to solving the problems at various levels of precision. You will never get an acoustic guitar to play perfectly in tune on every note, and even if you did it would still have all the problems of 'temperament'. There is no 'solution' for those that is universally satisfactory. It is possible to get 'arbitrarily close' to 12-tone Equal Temperament on most guitars and it makes areal difference. I think it's worth the effort.

Ouch - I see what you mean about price! And my first searches didnʻt locate any copies that were actually available. I would like to see those books but I doubt that anybody I know would have them, and Iʻll bet the local libraries donʻt either.

Iʻd never thought much till recently about intonation issues on guitars - Iʻm really just an amateur guitar player. But in the early music world I spent a lot of time and thought. With the viola da gamba there are easy and hard aspects. With gut strings itʻs hard to get in tune and stay there for long. But viols are very low tension instruments and itʻs easy to move a string around to tune up a cadential chord. Also they have moveable gut frets, so you can tune for a particular key or piece.

Many players are very particular about temperaments - not many professional players want to play in anything close to equal temperament - the very sharp 3rds in particular are very noticeable in music played with consorts of 3, 4, ... up to 8 viols. So viols consorts usually figure out some way to set tuning and frets, and then fudge as necessary to make chords and cadences sound well in tune.

The frets are tied double, not with a single strand of gut like lutes are. So some players even split a few frets a bit so they can play sharps and flats differently.

Where it gets interesting is when you play in a group that has several viols with a violin on top. Then the tunings that favor nice 4ths make the violin player unhappy, and vice versa. Back when we were performing a lot, I spent a lot of time choosing a temperament that got both the viols and the violin good enough so that we could play together in good harmony. Luckily, the music is usually in "easy keys" that have a lot of open strings - and much of the music doesnʻt modulate far from the principal key.

So the issue of intonation and temperaments is very familiar, but applying it to guitar is completely new to me.

Many guitar players these days 'sweeten' the tunings, and there are several systems for this. There are also things like the 'True Temper' fret system, that use crooked frets (a la Perronet Thompson; there's really nothing new under the sun) that improve the usual guitar keys, to the detriment of all the others. I tend to use 12-tone ET because it's the standard, although I miss the sweet sound of a dulcimer in quarter-comma mean tone. You just can't play one along with a guitar...

Regarding Trevor Gore's books, I have implemented many of the formulas from the books, including those for nut and saddle compensation. Here's a link to my spreadsheets and some documentation on how to use them. I had one luthier report good results.

The biggest obstacle to using Gore's method is that you need to know two properties of each string: the mass per unit length, and the "stretchiness" per unit length. The first number is available only from D'Addario (although I don't think their numbers are all that accurate), but the second number you have to build rig (shown in the book) to measure the string. A few people have done it, but I'm not aware of anyone having published numbers for string products from various vendors. You can reasonably estimate it for strings with steel cores, but Nylon strings are too unpredictable (plastic deformation). So someone needs to run those tests and publish the numbers, and then the spreadsheets would be easier to use.

Greg

ptrourin:
Please check your PMs.