Nut job

[Nelson Guitars]

P.S. It looks like you build some beautiful guitars~!

Thank you. I try my damnedest to put out a quality instrument, but as all who build these things will attest I know where all the warts are. The best part about it though is handing the finished ax over to the new owner. The second is seeing it played on stage.

Greg N

 

[Arx]

\
Note: if you are doing a f*nned multiscale fretboard, it is not simply a matter of choice. Deviating from equal centers will make the instrument unable to be properly intonated, to a degree determined by the fret slant angles, and by the amount of deviation from equal centers on the "worst" string, which is in turn governed by the differences in guage across the string set, and the number of strings.

On a standard parallel fret fretboard, this is not an issue.

Assuming you're using single string bridges, I don't see how it would be a problem. If you move both ends of the string the same distance off of their centres all you're doing is changing the scale length very slightly. Of course, if the bass is already built, and set up one way, you can't change to the other unless you have a way to adjust the saddles horizonally as well. (Which is probably what you meant)

 

[pilotjones]

Think about it some more. Most basses are set up with equal centers at the bridge, regardless of how they are at the nut. Even if they were equal gaps at the bridge, this is half the amount of shift, percentage-wise, as compared to the nut. So, the middle strings are no longer carving out equally-proportionally-spaced segments at each fret. The now slanted string, having been moved more at the nut than at the bridge, has a reduced its scale length considerably, and yet most of the change in distance between frets has occurred up near the nut, not at the 24th fret area. (Realize that if you rotate a string on a f*nned fret instrument with respect to its "natural" position, you are moving to an area of wider frets at the nut and closer frets at the body (or vice versa)).

I've figured this out in cad for a luthier, to check his suspicions that this might be an issue. On a 12-string bass, the angle produced on the sixth or seventh string was sufficient to produce a note that was (if memory serves) 21 cents off at the 24th fret. I don't have access to my email or design spreadsheets right now to give you the proper number, but there's another luthier on the forum with whom I've also shared the information. If he reads this, maybe he can dig up the numbers.

 

[Nelson Guitars]

21 cents off at the 24th fret.

That much? Wow I would never have suspected that given what seems a very minimal change in overall scale length/fret position. Do you recall if it was 21 cents off flat or sharp? Or was it flat on the bass side and sharp on the treble?

This brings up another thing I have always wondered about f*n frets. (Why won't anyone here spell the word without blanking out letters?) To get the proper note don't you have to be very careful to fret the string exactly perpendicular to the FB? even a slight push to the bass side will shorten the fretted length as well as stretch the string giving a double whammy on the sharpening. Pulling towards the treble side would still stretch the string, but lengthen the fretted length of the string lessening the sharpening effect on the string.

Greg N

 

[pilotjones]

That much? Wow I would never have suspected that given what seems a very minimal change in overall scale length/fret position. Do you recall if it was 21 cents off flat or sharp? Or was it flat on the bass side and sharp on the treble?

Should be flat, IIRC.

This brings up another thing I have always wondered about f*n frets. (Why won't anyone here spell the word without blanking out letters?)

While Novax no longer has an in force patent, they still have an in force trademark on Fanned-Fret(tm). I've read (not seen first-hand, though) that at some point they made MIMforum remove all instances of the phrase, or modify them to include the (tm) symbol.

To get the proper note don't you have to be very careful to fret the string exactly perpendicular to the FB? even a slight push to the bass side will shorten the fretted length as well as stretch the string giving a double whammy on the sharpening. Pulling towards the treble side would still stretch the string, but lengthen the fretted length of the string lessening the sharpening effect on the string.

As you say, on the bridge side of the "parallel fret". Opposite in the nut side. I don't think the amount of real-world side pulling is enough to have a significant effect for most players, but that's a guess/opinion.

 

[Arx]

Think about it some more. Most basses are set up with equal centers at the bridge, regardless of how they are at the nut. Even if they were equal gaps at the bridge, this is half the amount of shift, percentage-wise, as compared to the nut.

I'm not sure you want to be keeping it proportional "percentage-wise " though. I'm a total amateur, so feel free to correct me if I'm wrong, but is there actually any correction in place for the angle between the strings? I don't see how it could really be done without either a very pronounced radius on the fretboard, or curved frets. (maybe next project ) So, if that's the case, and you do have the ability to adjust the bridge horizontally, I would think that you would want to move them parallel to keep them to the same degree of error they originally were. (though not moving them at all at the bridge would probably bring them towards correct, wouldn't it?)

Am I missing some form of correction here? Is there a correction built in somewhere to compensate for the divergence of the strings?

This brings up another thing I have always wondered about f*n frets. (Why won't anyone here spell the word without blanking out letters?)

They're afraid of (or in some cases poking fun at) the patent laws. "fanned fret" is apparently a Novax trade mark. Personally, I don't care, because:

A. I'm not currently planning on selling any of the instruments I build.

B. If I do, It'll probably be in Canada where, contrary to what a lot of Americans believe, US patents don't apply. Of course, If I were to sell one to the US, I would pay the $75 or whatever licence fee.

C. I really think patent laws are counterproductive. As much as I would like to think I could make a profit on an invention of mine, I know that in general patents work better for large corporations who have the cash to back them up legally (and have the $$$ to file the patents to begin with). The original intent has been flipped on its head.

 

[Nelson Guitars]

Hmmm. My understanding is that you can copyright a set of words in a style or design sort of way but not the words themselves or the order in which they are used, especially if they are descriptive terms for what the thing is. I could see an argument if they called it a "John Doe's exclusive fret system" and everyone wanted to use the term, but "Fanned Fret" seems to be a description and not trade mark-able .

I really hate hair splitting legal bull shot.

Greg N

I just noticed that I posted that in a thread titled "nut job". How appropriate.

 

[Arx]

Hmmm. My understanding is that you can copyright a set of words in a style or design sort of way but not the words themselves or the order in which they are used, especially if they are descriptive terms for what the thing is. I could see an argument if they called it a "John Doe's exclusive fret system" and everyone wanted to use the term, but "Fanned Fret" seems to be a description and not trade mark-able .

I really hate hair splitting legal bull shot.

Greg N

I just noticed that I posted that in a thread titled "nut job". How appropriate.

Lets drop the legal stuff right now.. If someone disagrees, I don't care. I really don't want to see this thread hijacked like so many others.

So, back to the previous subject. Does anyone apply any kind of correction for the divergence of the strings on the fretboard? If so, how is it done?

-Nick

 

[Nelson Guitars]

Is it a hijack or a natural flow of a conversation? Hmmm. I guess it's up to the originator of the thread and I say as long as it stays somewhat luthier in nature and un-troll like let's have at it.

Personally I always set up the spaces at the nut and the saddle as equal. It is less noticeable at the saddle because the spaces are bigger. I'll have to think on the whole thing about fan frets for a while and maybe do an experiment or two. I think I understand the point though. Not much of a difference at the parallel fret but because of the extreme angle the frets take on at the upper registers being "off center" of the strings scale line will really make a difference.

Greg N

 

[Arx]

Is it a hijack or a natural flow of a conversation? Hmmm. I guess it's up to the originator of the thread and I say as long as it stays somewhat luthier in nature and un-troll like let's have at it.

Personally I always set up the spaces at the nut and the saddle as equal. It is less noticeable at the saddle because the spaces are bigger. I'll have to think on the whole thing about fan frets for a while and maybe do an experiment or two. I think I understand the point though. Not much of a difference at the parallel fret but because of the extreme angle the frets take on at the upper registers being "off center" of the strings scale line will really make a difference.

Greg N

I just think the legal stuff has been pretty much covered elsewhere. I've seen two other really great threads on fanned fret basses turn into debates on the moral/legal issues regarding the patents, and I don't think any more need to get derailed over it. IMHO of course.

I get your point about the upper registers + fanned frets. I think it's pretty much an unavoidable effect though unless you keep the strings quite parallel. I think Ill start a new thread about it. Maybe we can get some of the more experienced fanned fret builders talking.

 

[Arx]

Ok, I measured/did the math on a standard parallel fretboard. The error due to the string divergence on my ibanez btb555 works out to approximately 0.02%
AKA insignificant. 0.2 mm of difference in the length of the string.

I was thinking about the Fanned fret situation some more, and it's not any worse, assuming you design it with the divergence in mind (I wonder how many people don't?)

If you change the nut spacing, you would need to adjust the bridge spacing a proportional amount (as you suggested) not an equal amount (as I suggested).

But it will work, given that you have a way to adjust the saddle horizontally. You'll merely change the scale slightly on that string, a small adjustment to the saddle length adjustment, and you're back in tune.

-Nick

 

[Dirk Diggler]

That is the great thing about adjustable bridge pieces, and what makes a good setup on any bass. Now if you were talking about a non adjustable bridge like an acoustic guitar I could see the issue then.
Dirk
p.s. I'll say Fan Fret anytime, I used to call it "Sex Scale" on my six strings.

 

[Arx]

That is the great thing about adjustable bridge pieces, and what makes a good setup on any bass. Now if you were talking about a non adjustable bridge like an acoustic guitar I could see the issue then.
Dirk
p.s. I'll say Fan Fret anytime, I used to call it "Sex Scale" on my six strings.

Most of them don't adjust horizontally, though. If you move your strings on the nut, with out a proportional move on the saddle, it's going to mess things up pretty badly.

So, it's pretty much a build decision on a fanned fret. You gotta put your saddles in the right spot for either equal spaces at the nut, or equal centres at the nut. You can't just adjust the nut afterwards without having a way to horizontally adjust the saddles.

note that equal spaces at the nut, and a proportional change at the bridge does NOT give you equal spaces at the bridge, since string thickness is a constant. It'll be close, but not quite. (someone else might want to confirm this. I'm just figuring it in my head.)

 

[pilotjones]

Ok, I measured/did the math on a standard parallel fretboard. The error due to the string divergence on my ibanez btb555 works out to approximately 0.02% AKA insignificant. 0.2 mm of difference in the length of the string.

Quite true in this case, but insignificant for a much more important reason than the small amount. Insignificant because the frets are parallel, so as you lengthen the scale, you also proportionally lengthen the fret distances, by the law of similar triangles. If you had a super-wide parallel fret neck, and ran a string at a 45 degree angle across it, the fret spacings would still be correct, due to this proportionality.

This does not happen on a f*nned fretboard. Imagine: a string dead on center of a f*nned fretboard. The progression of fret distances is correct. Now, angle the string by rotating it clockwise a little (nut end towards the treble side, bridge end towards the bass side). What has happened? The fret spacing on the lower frets towards the nut has decreased, while the fret spacing on the upper frets towards the bridge (and in most cases, the scale length) has increased! It cannot play in tune.

Now consider that most basses have equal-centers spacing at the bridge. If you start with equal centers at the nut (which will produce a proper scale with f*nned frets, it can be demonstrated) and then you change to equal gaps, the outer two strings will stay in place, while the inner strings shift towards the treble side to varying degrees. This induces angles on the inner strings, which will knock off the scale.

I didn't know whether the amount of out of tune was significant or not, until I did a CAD mockup (using actual string gages) and numerical analysis on a 12-stringer. I just looked up the old email. The results were not as bad as I posted before (glad I put an "IIRC" in! ) but they are still significant.
For the 6th string on this 12-stringer, the error at the 5th fret would -3.7 cents; 12th fret, -6.5 cents; 24th fret, -8.9 cents.

This is a real, but somewhat extrem example, and a bass with fewer strings will have less of a max angle. Likewise, an 8-string guitar will have a smaller angle than an 8-string bass, because the variation in string gages is smaller, so the shift to produce equal gaps is less. So, it's up to the luthier. One builder making a mildly fanned, many- (but not 12-) stringed guitar, knowing how this all lays out, did a nut spacing that I think was between equal gaps and equal centers, and was happy with the results. But the luthier building the 12-string bass decided to go with equal centers, because it made no sense to him to make an instrument that from the start had bad tuning.

 

[randbguy]

Things that make you go, Hum..........? I'm not sure I'm with you pilotjones on the deviation with the f'd fret system. Th first fret would be out with the nut of course but the fretted notes would be OK if the bridge was placed properly. Did I miss something?

 

[pilotjones]

While it's probably because I didn't explain something properly, you are definitely missing something.

Since I'm not sure exactly what you mean, I'll just say
- where I spoke about rotating the string, I was talking about changing its anchor point along the nut or zero fret, not simply moving it off into space beyond the nut.
- since all pitches are based on the pitch of the open string, if you move the nut position (and then of course re-tuning!), you are in effect moving all the frets.

 

[Lucas G]

that headstock is just STUNNING! wow!

 

[Nelson Guitars]

that headstock is just STUNNING! wow!

Ya like that? Here's some more.

Greg N

 

[Arx]

This does not happen on a f*nned fretboard. Imagine: a string dead on center of a f*nned fretboard. The progression of fret distances is correct. Now, angle the string by rotating it clockwise a little (nut end towards the treble side, bridge end towards the bass side). What has happened? The fret spacing on the lower frets towards the nut has decreased, while the fret spacing on the upper frets towards the bridge (and in most cases, the scale length) has increased! It cannot play in tune.

Now consider that most basses have equal-centers spacing at the bridge. If you start with equal centers at the nut (which will produce a proper scale with f*nned frets, it can be demonstrated) and then you change to equal gaps, the outer two strings will stay in place, while the inner strings shift towards the treble side to varying degrees. This induces angles on the inner strings, which will knock off the scale.

One builder making a mildly fanned, many- (but not 12-) stringed guitar, knowing how this all lays out, did a nut spacing that I think was between equal gaps and equal centers, and was happy with the results. But the luthier building the 12-string bass decided to go with equal centers, because it made no sense to him to make an instrument that from the start had bad tuning.

Yes.. Good point on the equal triangles thing. I hadn't actually considered that, but was just calculating the difference in string length.

I'm nit-picking on this one, but you wouldn't typically move the string in the opposite direction on the nut from the bridge. Usually you would leave the bridge fixed(if it's not adjustable), so what would really happen is that the scale length would get shorter over the entire length of the string, but it wouldn't be linear. At the nut (where you're presumably moving the string) you're reducing the scale slightly. At the half way point, where the 12th fret should be, you're only reducing the scale by 1/2 as much as at the nut, and the fret won't be under it. You might be able to fudge it to some degree if it's not too extreme out by moving the saddle towards the head, but that'll just balance out the error so you're a little out at the top and bottom of the neck, instead of being a lot out at the top (bridge) end. (kind of like the opposite of a normal intonation adjustment you might need with a high action) Anyhow.. I'm getting off on a tangent here....

Regarding the reasoning for the equal centres though, I think that only applies if you can't adjust the bridge horizontally, right?

If I'm building a new bass, and using single string bridges, I could offset the bridges by a proportional amount and the only effect would be a very slightly decreased scale length of everything but the highest and lowest strings, right?

-Nick

 

[pilotjones]

Problem is in adjusting a proportional amount. If the string is shifted .030 at the nut, now you have to move it .060 at the bridge (assuming something like 3/8 baseline spcing at the nut and 3/4 @ bridge). This is overcompensating at the bridge, actually getting wider gaps at the low strings than the high strings.