(Another) Intonation Question…

OK, I haven't got time to go through everything in detail just now but I did have a look at the data for the first concert instrument.

If this is your actual measurements,

*Ctr of 1st to 12th frets@ 166.5mm = Ctr 1st to 16th frets@ 205mm =

Click to expand...

, then the actual scale length is shorter than 380mm although I can't work it out just now.

I may have lead you slightly astray with all the fret to fret measurements (it IS a rabbit hole). The only reason that I took you down this hole is because you can't make initial accurate measurements using either the nut or the saddle.

To start with they HAVE to be excluded as they are likely wrong until we can independently confirm that they are correct.

The true zero/zero point on a fretted instrument is actually the centre of the 12th fret. I know that someone else mention the nut as the zero point and I let it go and if you KNOW that the nut is right then it does work but we don't know that yet. Also, if we do compensate the nut (which is a useful thing to do sometimes) then the nut is again completely unreliable as a zero/zero point.

So, keep on trying different (shorter than 380mm) scale lengths in the stewmac calculator until the stewmac data aligns with the MEASURED data (fret 1 to 12, fret 1 to 16). WHEN that data ALIGNS, THEN you will know the true scale length for the FRETS.

THEN, and only then can you measure to the nut and saddle from the centre of the 12th fret and get useful measurements.

@anthonyg
Thanks. By all means, don't make this a priority! I dumped a lot of info out there. .... FWIW, in the event you're unfamiliar with the Romero STC.... It's described as follows, " ST stands for Soprano and Tenor. The ST Concert combines the length of a soprano, the body of a tenor, and the scale of a concert 'ukulele. Perhaps this structure complicates the measures.

https://www.romerocreations.com/st-concert

Web-Parrot, I led you down a rabbit hole, yet you've fallen too far. I need you to come back a stage.

At this point we are measuring nothing with the stewmac calculator other than fret 1 to 12 and fret 1 to 16. Forget everything else. Anything else is a rabbit hole to the rabbit hole which we aren't going down just yet.

Back to basics. When we are talking "scale length" we are dealing with a "nominal" scale length and an "actual" scale length. Any figure you see quoted for scale length is a Nominal scale length. The actual scale length of a WELL built instrument will on average be 2mm longer than the nominal scale length yet any figure quoted will be a Nominal scale length. Actual scale lengths are something you measure.

Actual scale length = Nominal scale length + saddle compensation.
Nominal scale length is SUPPOSED to = inside nut to centre of the 12th fret x 2.

Yet, how the Hell do we know anything at all when down this rabbit hole? Great question.
The scale length you measure can be right or wrong. How do you confirm which way is up and which way is down in this crazy World?
At this level of the rabbit hole I am ASSUMING that the frets are accurately fretted to themselves. Is this correct? Who knows. Maybe not yet at this level of the rabbit hole its all we can do. The saddle position AND the nut position are BOTH in question so we need to establish the NOMINAL scale length of the frets without referencing the nut or the saddle.

Now this is where taking an educated guess as to the NOMINAL scale length (stewmac is dealing in nominal scale length) and then calculating the distances from 1 to 12 and 1 to 16 and THEN placing those distances on your fretboard to confirm OR deny you guess comes into play.

OK, on a WELL built and intonated instrument I would expect an instrument with a NOMINAL scale length of 380mm to measure an ACTUAL scale length of 382mm or close to. If your measuring an ACTUAL scale length of 380mm on a WELL intonated instrument then I would expect that the nominal scale length would be 378mm.

Yes, this is a serious rabbit hole.

So, is your MEASURED 380mm scale length, really an instrument with a NOMINAL scale length of 380mm yet lacking in saddle compensation or is it really a 378mm nominal scale length instrument that DOES have 2mm of saddle compensation?

Before we move on we NEED to answer this question. THIS is the ONLY question we are answering with doing the fret 1 to 12 and fret 1 to 16 calculations. Nothing else matters at the moment.
Based on the information you supplied above the NOMINAL scale length is shorter than 380mm. I'm taking a guess that its 378mm so enter that into stewmac, calculate the distances and lay a rule on the fretboard to confirm or deny the latest educated guess.

Is it right? great.
Is it wrong? Guess again.

ONLY when then calculated data MATCHES the measured data can we move on to the next step.

When I'm placing calculated data onto the fretboard I will try to line up the zero end of the rule with the centre of the first fret. Where is the 12th fret and 16th fret distances falling compared to to the calculated data? Try adjusting it slightly. Try placing the 12th fret distance where it is supposed to be and checking where both the first fret and 16th fret positions are.
Are they right or wrong.

A 1mm difference in scale length will be easily detected. A 0.6mm difference is scale length will be detected to without too much trouble. A 0.2 mm difference will be getting more difficult to detect.

Anyway. Once you have confirmed the real, NOMINAL scale length that the frets are fretted for then you place your rule on the fretboard with the centre of the 12th fret at exactly half the Nominal scale length distance (for a 380mm nominal scale length this would be 190m) and then see where both the nut and the saddle fall. If the nut happens to be spot on then this will make further measurements easier yet don't count on it until its confirmed.

So, is your measured 380mm scale really a NOMINAL scale of 380mm or is it shorter? The data you've given suggests that its shorter but you need to do the measurements first to confirm.
Only after we have answered this question can we move on.

"The ST Concert combines the length of a soprano, the body of a tenor, and the scale of a concert 'ukulele. Perhaps this structure complicates the measures."

No complication at all. It is a concert scale instrument. The overall length and the size of the body may be unusual features, but do not affect the relationships of the nut, the saddle and the frets.

I am following this exposition with great interest.

John Colter.

To the participants and thread watchers, thanks for hanging in with us. A second variable has entered the conversation and may allow me to bring some conclusions to this matter.

Responding to Anthony’s post at 4:05 AM (US CST):
First a re-hash of the 1-12 & 1-16 fret to fret measurements.
[To cut down on all the numbers, I’m going to limit this to the STC ‘uke.]

The numbers I provided earlier are actual:
Fret 1 to 12 measures 166.5mm
Fret 1 to 16 measures 205.0mm

I’ve re-read all the posts and presumed the next step is compare those results against a StewMac calculator. And, the calculator should be set to some subjective, “nominal” scale length. In the previous calculations I shared, I did this using a 380mm and 381mm scale length under the presumption that a premium luthier like Romero would likely use a customary 15”/380mm scale. I DO understand that as a matter of production variability the physical “actual” scale may be different hence your suggestion to do some more interpolative calculations to see if we can get close.

So, as you suggest, I tried 378mm …. the Nominal StewMac provides;
Fret 1 to 12 displays a range = 169.500mm
Fret 1 to 16 displays a range = 206.607mm

I had to return the borrowed caliper and machinist metric ruler this AM, so if I need make any more measurements it won’t happen until after our Memorial Day holiday.

Before we move to your next level I need to share some more information that came my way last night! By way of the Forums’ email notification process I received the daily summary of posts on this topic. I opened the email this AM and read through another midnight posting that suggested taking a different direction. There’s a small complication though. When I looked at the forum this afternoon, the post wasn’t there. I suspect the person who posted it either pulled it or -because it was long- “lost” it to one of the forum glitches (as I had experienced). It’s quite long, and if it was lost because the forum crashed (or whatever) I know I would hesitate re-typing it.

So, to respect the privacy of the person, I won’t mention a name, but only attempt to summarize his/her perspective and approach. I’ll refer to them as “Fluffy” ;-)

Fluffy first suggested that we make the presumption that Romero cuts the fret slots in the right place. As such, spacing won’t likely cause problems. Second, it’s hard to place the nut wrong because it butts up against the very end of the fret board. Fluffy goes on to described a number of musical terms and assumptions regarding octaves and the 12 intervals in our chromatic scale, AND how some pretty good mathematics is employed to calculate the spacing -RATIOS- between the frets, nut, and saddle. While difficult to explain mathematically, the narrowing between frets show the narrowing of the ratios as you ascend the fretboard.

Fluffy shares, “The basically its the ratio of the scale length of the open string to the length of the string between the fret and the saddle. The octave interval is exactly 2:1. This is a natural phenomena caused by how sinusoidal harmonics interact naturally. If the ratio is not exactly 2:1, nature does not work and you can hear it or measure it with a frequency meter or electronic tuner. So instead of seeing a table of numbers and mm measurements for the fretboard, you will understand it a lot better if you work out the ratios or intervals the frets create, and also this will help you understand intonation and intonation problems.”

“Two other well known intervals are called the perfect fourth and perfect fifth, they are the ratio 4:3 and 3:2 and they exist on a fretboard at fret 4 and 7.”

Continuing, “If you want to check your fretboard without all the complicated tables, just look at these intervals - octave, perfect fourth and perfect fifth. In other words check out frets 4, 7 and 12. Measure the length of the string from the front edge of the nut to the front edge of the saddle. I am going to call this the scale length for the exercise. Then measure the length of the string from the front of fret 4, 7 and 12 to the front of the saddle. Now you can work out the ratio of lengths and see how it makes the interval ratio. The ratios need to be 4:3, 3:2, 2:1 or very close. The most important is the octave it alway should be as close to 2:1 as possible. If you are really keen, you can look up the ratios in the Equal Temperament system for intervals and check the ratio for every fret.”

….which I did. Rather than cut and paste some more very very educational information I’ll just include some attachments (I hope) that displays the results following Fluffy’s process. More numbers

Finally, on one of the table you’ll notice that I’ve printed some tonal readings. I acquired another digital chromatic tuner and tuned the STC gCEA to as close as I could to the charts pitches. (lower half) I then carefully played the same pitch an octave higher (using a felt pick, pressing the string down in the center between fret wires using a rounded dowel). I repeated the octave 5 times for each string to verify the pitch. The variations are small.

So… if “Fluffy” reads this and wants me to email the missing text, please send me a PM with an email address. You may want to expand, or at least take credit for this added knowledge!

Thanks again to everyone.

Web_Parrot said:

To the participants and thread watchers, thanks for hanging in with us. A second variable has entered the conversation and may allow me to bring some conclusions to this matter.

Responding to Anthony’s post at 4:05 AM (US CST):
First a re-hash of the 1-12 & 1-16 fret to fret measurements.
[To cut down on all the numbers, I’m going to limit this to the STC ‘uke.]

The numbers I provided earlier are actual:
Fret 1 to 12 measures 166.5mm
Fret 1 to 16 measures 205.0mm

I’ve re-read all the posts and presumed the next step is compare those results against a StewMac calculator. And, the calculator should be set to some subjective, “nominal” scale length. In the previous calculations I shared, I did this using a 380mm and 381mm scale length under the presumption that a premium luthier like Romero would likely use a customary 15”/380mm scale. I DO understand that as a matter of production variability the physical “actual” scale may be different hence your suggestion to do some more interpolative calculations to see if we can get close.

So, as you suggest, I tried 378mm …. the Nominal StewMac provides;
Fret 1 to 12 displays a range = 169.500mm
Fret 1 to 16 displays a range = 206.607mm

I had to return the borrowed caliper and machinist metric ruler this AM, so if I need make any more measurements it won’t happen until after our Memorial Day holiday.

Click to expand...

OK, the short answer is that we are still not there. Try a nominal scale length of 376mm and see what we get.
Get back to it when you can.

Now I don't want to cast aspersions at your measuring yet this isn't done until you measure again with the rule over the frets and it all matches together. We must always be double checking ourselves as its not easy.

This is one of the problems. Experienced people can believe that they know what's going on yet until its measured its all a guess. One of my points is that assuming/guessing that the nut is right because it usually is is just not good enough and from my experience the nut is as often wrong as the saddle is wrong.

Measuring from inside the nut to the centre of the 12th fret and getting a nice "round" number can leave you wth a false sense of security that its correct when its really wrong.

I measured up a Soprano ukulele the other day.
Was it really a 13 1/2" (342.9mm)scale instrument with a poorly placed nut and plenty of saddle compensation? No it wasn't.
Was it a 345mm scale instrument with the nut lining up nicely and slightly less saddle compensation ? NO, it wasn't.
What it is is a 346mm scale instrument with 0.5mm nut compensation (flattening the intonation) and very little saddle compensation.

We can guess all we like yet until we measure it its still just a guess.

I’ve found this an informative and interesting thread, and one that confirms some of what I already do. Some but not all of what I do, but then as I only use Soprano and Concert Ukes (so short scale) my lack of highly accurate measurement for nut and saddle placement might not matter that much. My logic is different too in that that I’m looking for the major improvements that make an instrument ‘fit for purpose’ (pleasant to hear and easy on the fingers) rather than virtual perfection.

One point on compensation that doesn’t seem to come across that well is string end effect and the variation of it between individual strings of the same material and of different material too. Maybe I’m wrong here but with small fractions of mm’s being discussed on length wouldn’t it be helpful to discuss variations in string end effects too? How much difference does a change of string material make, etc?

Another ‘Rabbit Hole’ is the assumption that all strings are perfectly made and that there’s accurate repeatability between strings of the same manufacture. My experience (so gained the hard way) is that whether new or old no string should be trusted to give the expected results - mostly they do but don’t regard it as a certainty.

Edit. Just wanted to put these secondary issues on the ‘Radar’, there’s only so many issues that can be addressed at one time but the interactions sometimes end up being significant.

Edit. @ anthonyg “OK, for the record I AM using reading glasses for checking scales and I have a strong set just for getting up close as well. These are cheap reading glasses. Nothing prescription.”
Thanks, a useful tip that I intend to follow-up - I’m kicking myself for not thinking of it myself. That’s far better than asking someone else to hold a magnifying glass whilst you use your own two hands to mark something out, etc.

Sure. Different string stiffness through the use of different gauges and/or different materials adds to the issue. The principle is that the stiffer the string the more saddle compensation it needs and the more flexible the string the less compensation it needs. This is why wound strings are used as the string gauge gets larger and why steel strings need more compensation than classic strings do.

If strings get too inflexible though the relationships fall apart and the intonation goes bad even with lots of saddle compensation. This is another rabbit hole to go down and I was trying to address one at a time.

Intonation is indeed a rabbit hole topic. I have always worked by the rule that, get it close by adjusting, filing etc then fine tune when fretting.

I'm afraid guys your posts are way too long for me to bother read and think if any usable stuff here. Should be better if links given etc. instead musing all yourselves. Some paragraph spacings could be also used by some posters.

Jarmo_S said:

I'm afraid guys your posts are way too long for me to bother read and think if any usable stuff here. Should be better if links given etc. instead musing all yourselves. Some paragraph spacings could be also used by some posters.

Click to expand...

I’m a bit puzzled by your comment and kind of surprised - perhaps I misunderstand what you’re trying to say. Folk have gone to some considerable trouble to write out posts and share their information. Some of the posts are, for me, less readable than others but criticising them does seem to be rather like ‘looking a gift horse in the mouth’.

Jarmo_S said:

I'm afraid guys your posts are way too long for me to bother read and think if any usable stuff here. Should be better if links given etc. instead musing all yourselves. Some paragraph spacings could be also used by some posters.

Click to expand...

I tend to agree, I did wonder if some posters weren't trying to out do each other.
Is anybody seriously going to go to this amount of trouble on their uke setup.
Any luthiers here do all the steps outlined in these long and oft rambling posts? Serious question, I am not beng flippant.

I hesitate to call myself a luthier. I am an amateur ukulele builder with about thirty creations to my name. I find the approach to intonation, as described in this thread, to be obsessive, aiming for levels of accuracy that are unrealistic and seeking an accuracy of intonation that is unachievable. There is such a thing as 'near enough'. We all have to decide for ourselves where that point lies.

I prefer to work out my own fret placement, using the "Rule of Eighteen". That gives results that are perfectly satisfactory on a soprano ukulele.

John Colter.

The levels of accuracy and measuring techniques are absolutely doable and yes I do expect Luthiers and techs to be on top of it.

The measuring techniques I have described are from the point of view of a tech who is working on a random or unknown instrument who is assessing an instrument before picking up a tool to adjust the instrument.
The fact that this is unknown and considered arduous is just evidence of the widespread lack of technical knowledge in the industry.

If your a luthier then please have your fretboards slotted on a CNC machine to a known scale length and then all you have to do is create a jig to accurately place the nut and saddle.
Its simple.

It is a little more complicated for a repair tech dealing with many and varied instruments yet the measurements can be made and ready and in a data base. I currently have the common ones on paper and just refer to them each time.
Once you know the fret scale then you just place a rule down the fretboard using the centre of the 12th fret as zero/zero and see where things are.
You see how the intonation is with a tuner and then compare to the measurements.
Are you seeing predictable results?
Has an error that needs to be fixed just become as obvious as the nose on your face?

Honestly it takes just as much effort to place a nut and saddle in the wrong place as it does to place them in the right place. Why not just put them in the right place to begin with?

If your a luthier then please have your fretboards slotted on a CNC machine to a known scale length and then all you have to do is create a jig to accurately place the nut and saddle.
Its simple.

And therein lies the issue. How many Luthiers do you think can invest in a CNC machine to slot a fretboard.?
I wonder what Luthiers did before computer operated mchinery became available. No wonder all the old Martins are such crap!!
It's simple.

ukantor said:

I hesitate to call myself a luthier. I am an amateur ukulele builder with about thirty creations to my name. I find the approach to intonation, as described in this thread, to be obsessive, aiming for levels of accuracy that are unrealistic and seeking an accuracy of intonation that is unachievable. There is such a thing as 'near enough'. We all have to decide for ourselves where that point lies.

I prefer to work out my own fret placement, using the "Rule of Eighteen". That gives results that are perfectly satisfactory on a soprano ukulele.

John Colter.

Click to expand...

Well you are clearly wrong. Now invest your retirement savings in a CNC immediiately

Hi DPO, assuming that you are serious in saying I am clearly wrong, please explain how you arrive at that conclusion. Is it the "Rule of Eighteen" to which you refer? I would be most interested to be enlightened.

John Colter.

ukantor said:

Hi DPO, assuming that you are serious in saying I am clearly wrong, please explain how you arrive at that conclusion. Is it the "Rule of Eighteen" to which you refer? I would be most interested to be enlightened.

John Colter.

Click to expand...

Sorry John, too much tongue in cheek, I was hoping the smiley face gave it away. No offence intended.

I did wonder, DPO. No worries, Mate!

ukantor said:

I hesitate to call myself a luthier. I am an amateur ukulele builder with about thirty creations to my name.

John Colter.

Click to expand...

Just an opinion of the term “luthier”.....

If someone can dive into an Olympic pool and swim, dog paddle or submarine to the other end without drowning, then they are a swimmer by definition.... not an Olympic class swimmer, but a swimmer none the less. Similarly, if someone can construct a stringed instrument that is capable of being played in tune and conforms to an acceptable standard, then they are a luthier by (albeit, modern) definition..... not necessarily a Master luthier, but a luthier none the less.
None of the luthiers that I have met, have hinted that they consider themselves members of an exclusive elitist fellowship, but more as down-to-earth niche craftsmen.
I suggest, from your self description, that you should recognize yourself as a luthier.