Novak Multiscale Patent "Superceded"

[pilotjones]

I just read something very interesting. I thought that a new thread in the Luthiers' forum might be in order.

Paul Everard Booker of the UK was, on October 18, 2005, granted US Patent # 6,956,158. This patent covers methods of determining the fret positions on a f*nned-fret fretboard. As well as a few related areas.

It references the Novak patent, and it is demonstrated within this patent that the Novak patent results in an improper fret positions, except in the case where the strings are parallel. (This is a point that myself and many others, including some luthiers here, have previously discovered independently.)

It describes two basic methods, with variations, to lay out f*nned frets. Each uses two reference scales to lay out the frets. The first of these methods results in fret positions with an acknowledged maximum of a one-cent intonation error; the second, while not stated as such, results in perfect positioning.

The first method is somewhat strange, and is impressive if you're one to be impressed by geometry. The guy definitely has some intuition, and some analytical sense.

The second method is, in my interpretation, an extension of the parenthetical mention in the Novak patent of a method of laying out two scales for the outer strings, and drawing lines between them. This part of the Novak patent however is not part of the "what is claimed" part of that patent.



My interpretation of all this (IMO!) is that this patent reinforces the following points:
- the Novak patent only covered the "converging to a point" method of creating a fan;
- the Novak "converging to a point" fan produces bad intonation, and should not be used (except if the strings are parallel);
- both of these patents cover specific methods to lay out the frets, cover a fingerboard produced by using the method, and the instrument containing the fingerboard. But
- the basic idea of a multiscale instrument, with nonparallel frets, is not specifically patented by anyone, which makes sense since the concept has been in use for centuries.


The Booker patent also covers jigs, and a nut-end string anchor.

The Booker patent is here:
http://patft1.uspto.gov/netacgi/nph-...S=PN/6,956,158
I found Paul Booker's website. He seems to have produced 1 1/2 basses so far, and gives lessons.
http://www.btinternet.com/~paul.ever.../Splaybass.htm
He also has some trussrod ideas, and has made a spreadsheet to calculate neck deflections based on profile and reinforcements.



So what do you guys think?

[.matthew e wengerd.]

nothing to add...just wanted to say that this is quite interesting.

[Rodent]

for those wanting to have 'vaneaxial' frets, or build instruments utilizing them, this is potentially a huge step forward. it'll be interesting to see if anyone currently paying mob money to Novak challenges the previously paid extortion monies and legally demands a refund.

all the best,

R

[THIRSTYGUMS]

So call me thick, but i cant get my head around this. can anyone tell me in plain terms (with diagrams if possible) what the solution is? i was just about to build my first Fanned neck too

[pilotjones]

Originally Posted by THIRSTYGUMS

So call me thick, but i cant get my head around this. can anyone tell me in plain terms (with diagrams if possible) what the solution is? i was just about to build my first Fanned neck too

The Novak patent claims a fretboard with the frets converging to a point. You can utilize the method of the Novak patent by:
- marking points down the centerline of the fretboard, using a properly designed fret scale,
- make your fret lines by drawing lines that project from these points to a common point.

The problem is that
- if your strings are not parallel, as is the case with most instruments, then
- only a string along the neck centerline will be intonated properly. The frets for the other strings will be in the wrong place, at the wrong angle.

One solution, that produces an instrument with properly angled frets, is to use two properly designed fret scales aligned along the inner and outer strings, and connect the pairs of points to make the fretlines. This is mentioned in the Novak patent as an alternate way of producing frets converging to a point, but it is not one of the claims. And if the strings are not parallel it does not produce frets converging to a point, rather, it produces a proper fretboard.

Another solution is to use one of the methods in the Booker patent.

[Scott French]

I think I'm missing something, now two people have a patent on process that been around for hundreds of years? I thought the whole argument against Novak was that he was allowed to patent old news.

[Volk]

But the Novak patent is in the US right? This could lead to trouble.

If so many people here already knew that, someone should've patented it....

[erikbojerik]

So am I correct in thinking that the "dual-scale then connect the dots" method is not (strictly) covered by ANY patents??


Intonation off by one cent? Meh...I am going to patent the method of intonation correction via finger pressure, including any and all string-bending methods. Prepare to open your wallets every time you strap it on!

[.matthew e wengerd.]

Originally Posted by erikbojerik

Prepare to open your wallets every time you strap it on!

There's a phrase worth repeating.

[pilotjones]

Originally Posted by Scott French

I think I'm missing something, now two people have a patent on process that been around for hundreds of years? I thought the whole argument against Novak was that he was allowed to patent old news.

IMO this new patent demonstrates that the Novak patent covers the method of producing the fan by having the fret lines converge to a point.

On a fretboard where the nut spacing and the bridge spacing are not equal, that is to say where the strings are not parallel, if it is set up properly, the fret lines do not converge to a point. Therefore, IMO, Novak has no patent protection.

Booker describes two ways to lay out a multiscale fretboard properly, even with non-parallel strings. But IMO his patent does not cover a properly-done multiscaled fretboard; only one produced by his patent-protected method.

[pilotjones]

Originally Posted by Volk

But the Novak patent is in the US right?

The Booker patent is also a US patent. He has also filed for one in Great Britain.

This could lead to trouble.

If so many people here already knew that, someone should've patented it....

This is my interpretation: the patents only cover methods of producing a multiscale board, and the boards thus produced-- but the idea of a multiscale fretboard is not patented by anyone, and is not patentable.

[pilotjones]

Originally Posted by erikbojerik

So am I correct in thinking that the "dual-scale then connect the dots" method is not (strictly) covered by ANY patents??

It is mentioned, but not claimed, in Novak's patent. It is mentioned, as a trivial case in the Booker patent, but not specifically claimed. (Booker uses two scales, neither of which is the scale of, or in the position of, an actual string, and then connects the dots. He mentions that if x = 1 in his equations, these two "figuring" scales will coincide with the outer two strings.)

I think that the reason that no one has specified "dual-scale then connect the dots" as a patent claim is that it is not claimable. Surely IMO if Booker could have, he would have.

[THIRSTYGUMS]

Is being one cent out on intonation a big problem? i mean i was about to start work on a fanned six string using the FRETFIND 2D program ... www.fretfind.ekips.org/2d/index.php
, are you saying this is a bad idea? Do i just work out two sets of fret layouts and find the average distance?

[Phil Mastro]

Originally Posted by THIRSTYGUMS

Is being one cent out on intonation a big problem? i mean i was about to start work on a fanned six string using the FRETFIND 2D program ... www.fretfind.ekips.org/2d/index.php
, are you saying this is a bad idea? Do i just work out two sets of fret layouts and find the average distance?

From what I've understood, fretfind uses Booker's method, cuz it asks for 2 different scales, and your string spacing. Hence, it's the connect the dots style system. Which I think is Booker's method.

Can someone confirm this? (either fretfind using the connect the dots method, and/or the connect the dots method being Booker's method)

Regarding the Novax system, the one cent variation can probably be fixed by moving the bridge saddles a bit, so I figure it's not that big of an issue.

[pilotjones]

Originally Posted by THIRSTYGUMS

Is being one cent out on intonation a big problem? ...

Originally Posted by Phil Mastro

Regarding the Novax system, the one cent variation can probably be fixed by moving the bridge saddles a bit, so I figure it's not that big of an issue.

Let me correct a misperception, quickly here:

Booker shows two methods (with variations). One method is perfect (no error), the other results in a maximum one cent error (which I personally could easily live with if there were other benefits).

Novak's method results in HUGE errors (like half a semitone!!) if used on a normal, non-parallel-string instrument.

The plain, two scales coincident with the outer strings method, which is used by Fretfind, and which is mentioned but not claimed in both patents, is perfect (no error).

[pilotjones]

Now that I think of it, if the two-scales-on-outer-strings-then-connect-the-dots method had been covered by the Novak patent (that is, had been a patent-protected claimed invention rather than just something mentioned in the patent), Novak might have gone after the Fretfind author years ago. I hadn't thought of that before.

[Trevorus]

I originally though that the Novak method was the 2 outside string scales. I suppose he went for geometry instead of accuracy. This is a fantastic breakthrough. I really only thought the only method was calculating the proper fret positions by the scale length. But, little do I know, eh? This is pretty cool. I would love to see mutli scale instruments in more mainstream manufacturers lineups. Maybe get them in the reach of my wallet.

[pilotjones]

Originally Posted by Trevorus

I originally though that the Novak method was the 2 outside string scales. I suppose he went for geometry instead of accuracy.

The possibility exists--this is speculation--that he laid out the two different scales on the outer strings of a parallel-string instrument, discovered that the frets all converged to a point, and then just plain missed the fact that that they don't converge to a point if the strings aren't parallel.

But as far as I can tell his patent claim is for the frets, bridge, and nut converging to a point. (The 1904 Edgren patent #652,353 has the nut and frets converging to a point, but forgot the bridge!)

This is a fantastic breakthrough. I really only thought the only method was calculating the proper fret positions by the scale length. But, little do I know, eh? This is pretty cool. I would love to see mutli scale instruments in more mainstream manufacturers lineups. Maybe get them in the reach of my wallet.

I think the reason you don't see lots of multiscale instruments is not just their unusual appearance and the lack of knowledge of their benefits, but also, perhaps primarily, because you can't cut all the slots simultaneously on a gangsaw in 20 seconds, as you can with "normal" fretboards.

[seanm]

Does anybody know how big the error is with "normal" straight frets? I realize it depends on where you fret but maybe an average or a max/min.

[jay_t]

Not in a mathematical sense but think about this:

The strings get higher from the frets as you go further up the neck. The higher is goes, the sharper the note will be in relation to the scale. Unless you take this into consideration when you mark your scale... it will always be off depending on the action.