Official Luthiers Forum!

Finally got my last video in the Scale Length / Temperament / Intonation series uploaded.

A bit longer than I usually aim for, but it's a complicated topic, and I think I compressed it about as much as I could.

_________________
Eschew obfuscation, espouse elucidation.

I can't wait to get home to watch this. Thanks, Dave.

Another fantastic and informative video!

Thanks very much for sharing your knowledge and tremendous experience David.

I actually found the whole thing quite riveting to watch, as I've spent many years thinking about these issues and tried out many of the things you mentioned.

Great stuff!

Great video. Great clear explanations with very practical information.

Thank you for producing and sharing the video.

_________________
http://www.Harvestmoonguitars.com

Thanks a lot guys! I'm never really sure if others will find the minutia as fascinating as I do, or where to shoot the line between too complicated and too simplified. It's good to hear others find in interesting, and hopefully that at least some of what I say makes some clear sense.

_________________
Eschew obfuscation, espouse elucidation.

Thanks David, If I'm being really picky I would suggest using the terms "Bending Stiffness" and "Longitudinal Stiffness" rather than "stiffness" and "elasticity"

David, that was probably the most informative and well laid out "expose" on intonation I have ever seen.
Thank you.

_________________
The name catgut is confusing. There are two explanations for the mix up.

Catgut is an abbreviation of the word cattle gut. Gut strings are made from sheep or goat intestines, in the past even from horse, mule or donkey intestines.

Otherwise it could be from the word kitgut or kitstring. Kit meant fiddle, not kitten.

Dave,

Thanks so much for the video. Very helpful in understanding more about how the guitar works.

Max

_________________
Max Bishop
Brighton, Michigan

Thank you Dave - your depth of knowledge, and ability to break it down for us in a very accessible way is hugely valuable. Thank you for sharing.

Lots of good info. Thanks!

Pat

_________________
formerly known around here as burbank
_________________

http://www.patfosterguitars.com

I think this is going to be a classic. Well balanced for all interested.


Sent from my iPhone using Tapatalk

_________________
Ken

Dave, the videos are fantastic. Thank you.
So many information video's are brutally slow, with ramblings that make your lids heavy almost immediately. Yours verge on too much density to handle. Grab a fresh cup of coffee, sit up straight, take a breath and hit play, because here we go!

I'm curious about the card over the sound hole isolating the string pitch. Is this effective for just the particular issue on this instrument? Air resonance? Do you have other methods like it when you suspect influences outside of normal set-up and fret placement are going on?
Anything that helps with a diagnostic flow chart, like blocking the sound hole , would be useful for those strange hair pulling instances.

Thanks again guys!

David - yes, this is a particular issue I find particularly poignant with this individual guitar, which is of course why I picked it for the video. Most of the time it's not nearly so bad, although on a lot of dreadnoughts I find similar issues (often less extreme) around the a-flat to g range.

As far as troubleshooting the source, I start with a good hug to the gut to rule out back resonance. Then my little cardboard over the sound hole is a quick check on chamber frequency, and then a couple of magnets on the top near the bridge do well to see if changing the top plate frequency has any impact.

Then there are basses, where bar vibration frequencies in the neck can often come in to range of influence, andasmall c clamp on the headstock can help with this test.

As to solving these problems, much as I hate to say it, I don't often go there. If a change so slight as a few grams difference in bridge pins may work, great. More than that, and I'm not keen to get involved with revoicing. Too much liability, too much room for subjective (and unpredictable) preference in the final results.

Here I just like to be aware of odd interferences in making adjustments, and will occasionally recommend slight tuning offsets to dampen the final effect.

Admittedly, my motive in including this in the video was not so much to instruct on how to solve the problem, but more to increase awareness of where limitations outside the control (or perceived fault) of the technician may be inherent to the instrument itself.

_________________
Eschew obfuscation, espouse elucidation.

Thanks David. I see these tricks as possibly useful demonstrations for customers who have fallen into the intonation rat hole. Offering a glimpse of the abyss might get them to climb back out. Or, more importantly to me, convince them that their luthier doesn't suck.

Fantastic! Thanks David!

Another great video, David. Thank you! I will certainly be recommending that my customers and colleagues take a look at this. As others have said, you've packed a huge amount of useful information into a small space. I agree completely with your description of the relationship between nut slot adjustment and the need for nut compensation.

One critique: I wish you'd made it explicit that compensation at the nut has an equal effect (in cents) up and down the fingerboard. When I first listened, I thought you were saying that moving the nut forward flattened notes at the first fret, and sharpened notes beyond the twelfth. Thinking it over, I presume you mean that moving the nut forward doesn't do enough to flatten the higher notes, and so they remain significantly sharp. I hope I have this right.

I'm glad that you touched on the effect of body resonances and air cavity resonances. That's a rabbit hole that would swallow any attempt I might make at explanation. Likewise the player's potential control. You got the bending stiffness issue in there too. Excellent! Thanks again.

Thanks Eric - as it would be, I did actually talk more about that effect when I was shooting. Moving the nut forward alone will flatten all frets equally, but then if you move the saddle forward to bring intonation at the 12th back to neutral (or as you would set without a compensated nut), you end up with this net effect of sharpening beyond the 12th.

In editing though, while trying to shorten it down from the hour+ I had recorded, it was one of those spots I didn't feel was explained clearly enough. Given that the main goal was to demonstrate glimpses in to the complexities of the task rather than teach in great depth, that was one of the bits that got cut.

Good catch though - maybe some day I will do more focused works where I'm able to go in to more detail on specific topics. The biggest challenge in the end on this one was editing down to reasonable length, and figuring where cuts could be made.

That, and uploading the video. The Mac I use is a few years old, and between the rendering on Final Cut Pro and uploading, this one took 25 hours from the time I hit "share"!

_________________
Eschew obfuscation, espouse elucidation.

All I can say is wow and thank you! What a great lecture David. It feels like a master class to me. I think I'll watch this a couple of more time to let it all sink in. Thanks Again.

This was amazing. My brain is fried.

_________________
Support the OLF! Bookmark our STEWMAC link Today!
Lance@LuthiersForum.com

Thanks David.

Sent from my HTC Desire 626s using Tapatalk

_________________
Another day, another dollar.

Nice job Dave, thanks for putting this together!!
L.

_________________
Cut to size.....Beat to fit.....Paint to match.

A few points:

You say that string bending stiffness has a major influence on intonation. Not my experience, UNLESS you have a clamped end condition for the string. You illustrate this effectively with your vise demo. However, in all my testing (and I've done plenty) the end condition of a real string on a real guitar is better represented by a pinned end condition. The bit of string between the finger and the active fret or the saddle and the pins seems to be sufficient to prevent the clamped condition (which you clearly have in you rig at time 3.28). This bit of string allows "rocking" over the support (nut, fret or saddle) and thus makes the pinned condition a much better approximation. So bending stiffness has virtually zero effect on the intonation of the fundamental but can have a significant effect on the higher partials, particularly on monofilament nylon strings. All the details of testing, experiments, modeling etc. in the book.

But think about it: why on earth should applying saddle compensation so that the open string and 12th fret play in tune make all the other frets play in tune? It doesn't, of course. Saddle compensation is there all the time, whether you need it or not. The open string doesn't need it, so is tuned sharper to obviate the effect, which leaves all the other frets up to the 11th sharp, then the 12th is made to be right. You get sharpening below the 12th fret and flattening above, with only the open string and 12th accurate.




You definitely left the impression that nut compensation results in sharpening of the higher frets, which is incorrect. Yes...

If you apply proper nut compensation, the saddle is compensated in the same sense as normal (i.e. moved towards the tail). Done right, the correct amount of nut compensation together with the correct amount of saddle compensation on a per string basis leaves you with pretty much zero error compared to the equally tempered scale (ignoring body coupling effects, which shifts notes as you showed, but that issue is dealt with in a different way)

If proper nut compensation produced no benefits, I wouldn't be doing it. The benefit is more accurate reproduction of equal temperament, the best compromise, not any other temperament, such as the Feinten system etc.. Modeled results below:



Typical nut compensation:


...and we haven't even mentioned the effects of neck relief yet....

Edited: Edited out the following incorrect comment pointed out by David Collins: Re: nut compensation, around 15.45 you say a fret-plane nut doesn't have sharpening at the low frets because the sharp string angle to the nut is eliminated. Not so. There is still a progressively decreasing break angle at the nut as you fret higher up the neck, even with a leveled zero fret. This still results in a contribution to sharpening on the low frets.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

I suppose I should have made clear in the video that my focus was entirely related to steel strings, which constitute 99%+ of the instruments I deal with. I'm sure there are significant differences in weighting of factors with nylon, and this was entirely focused on steel strings.

I should also add the disclaimer that the video was intended more as a brief view of larger concepts. Many details were grossly glossed over, some perhaps generalized too far, but the intent was to raise awareness of the complexities more than instruct in the finer details.

That said, it does seem that you and I Trevor, have experienced what seem to be some irreconcilable observations. I am certainly open to reviewing my portrayal of the theory, but stand quite solidly confident in the accuracy of my observations of results.

For over 20 years, my full time job has been working on guitars, adjusting intonation for thousands of players and instruments over the years. In that time, I don't believe I have ever seen an instrument with standard nut placement and a traditional proper setup show intonation anywhere close to resembling the chart above. The study of theory is somewhere between a professional interest and personal hobby of mine, but observation and adjustments is my day job, and I've just never encountered errors which follow that trend on any steel string instruments.

On string angle to the nut, I stand with what I said. The example was simplified with the straight neck, and things can change a smidgeon with relief, but if the string leaves equal to the fret plane from the nut, my assertions are quite plain and true. A simple model of a triangle sans frets demonstrates this quite easily. With a striing for your hypotenuse and long side as your fret plane, push down on the string at various positions along your board...

As to sharpened upper frets with compensated nuts relative to uncompensated, this is true as well, assuming you maintain the same intonation at the 12th region. It seems you agree with this in your models though, so I may be misunderstanding. We both show that the upper frets would be sharper with nut compensation, with the exception that your charts show this change as relative to upper frets being excessively flat with standard nut placement, which I have not found to be the case in my years at the bench.

Regarding the effect of stiffness, it's possible that I may have overemphasized its effect on fretted notes in the video. On a well cut nut with any fair break over angle however, an open string undoubtedly behaves closer to a clamped string than one which rocks over a boundary point. A notable demonstration of this is that locking nuts and conventional ones with decent break over angle behave quite similar, whereas Fenders without string trees (or the G on single string tree models) and LSR roller nuts tend to act as exceptions and behave differently, where the string is not locked as solid against the boundary point and tend to rock more.

It's a head scratcher though, as we seem to not only have reached very different conclusions, but observed what seem irreconcilably different results. My models of changes resulting from nut compensation seem to match yours quite perfectly. The observations of the initial errors however, seem to differ quite significantly. When I check a standard spaced guitar after setup and intonation, I tend to find much more of a flat line start to finish, and nothing resembling the lower sharpening and upper flattening shown in your charts.

_________________
Eschew obfuscation, espouse elucidation.

Regarding Trevor's points: it would be great if there happened to be a Gore or Gillet-made guitar (incorporating the nut and bridge compensation system detailed in the books) in Michigan that David could measure.


Sent from my iPhone using Tapatalk

On a different note, David have you ever had the opportunity to measure an Evertune equipped electric?

Such a bridge design has a floating saddle for each string which eliminates any change in tension from fretting notes, regardless of action.

I've played one and the stay in tune performance was phenomenal. I didn't have a chance to measure all the notes on each string though.

I would expect better intonation performance over any conventional bridge design.


Sent from my iPhone using Tapatalk