For Peter KG, Don Noon, and others interested in engineering.

[curious1]

Torbjorn's method seems to do just that. Though I wish he would re-write the article with more photos and "how to's".

Yes, and maybe some metrics to prove that they actually match "the various types of arching you see in Cremona".

[captainhook]

Since what Torbjörn is offering is free, I'd say it is up to you to disprove it. If you don't like it, don't try it. But it works for me, with great flexibility.

[curious1]

Since what Torbjörn is offering is free, I'd say it is up to you to disprove it. If you don't like it, don't try it. But it works for me, with great flexibility.

No, since Torbjorn is the one making claims of matching "the various types of arching you see in Cremona" it is up to him to prove it. Otherwise just call it an interesting and useful technique. That's called intellectual rigour.

[captainhook]

Oh, sorry. I misunderstood. I thought we were talking about making fiddles. "Intellectual rigour" is way out of my league.

[GlennYorkPA]

I'm not sure the historical examples support the notion of letting the materials govern the shaping of the arching.  Of course, if you're aiming to engineer a superior result without first understanding what worked for the old masters, then I guess the historical practices don't matter?

 

It seems that virtually all the variation in the old Italian examples lies in the approach to edges and channel, rather than the central section of the arch.   The central arch sections seem to be extremely consistent in shape, across different regions and centuries of Italian making, and even looking at other arched bowed instruments.   Through the central section of classical Italian arches, you find pretty much the same shape on any instrument for a given distance from the center line and a given fall/rise from the level at the centerline. 

 

This suggests that some fixed method of making did indeed set the shape for any given amount of fall fo a given distance from the center line.  The apparent differences in various old Italian arches lay not in the central curve of the arches, but in the shape and height of the edge, the level at the centerline, the channel, and in the transition from the channel into the central section of arching.

 

This shape in the central section of the arches is very very similar to a catenery curve, but I don't the evidence supports an actual match.

David,

 

Stradivari was often required to produce sets of violins so I can understand that they should be similar in appearance. His use of forms allowed him to maintain strict control over the outlines but what about the critical archings? 

 

Maybe I'm wrong but I don't think arching templates have survived.

Might this suggest that some principle apart from templates was used to guide the arching?

As luthiers all know, if you want acoustic consistency, you must adapt arching and thickness to account for variations in wood.

 

Glenn

[Don Noon]

Maybe I'm wrong but I don't think arching templates have survived.

Might this suggest that some principle apart from templates was used to guide the arching?

As luthiers all know, if you want acoustic consistency, you must adapt arching and thickness to account for variations in wood.

I'm perfectly comfortable with the idea that talent and experience would allow someone to create archings that look consistent and work well.

And I'm fairly sure there are no existing Cremonese arching templates, which can not prove anything, but reduces the possiblities to either they were ALL lost, or that they didn't use them.

[David Beard]

Yes.  I'm inclined to believe there was a well established workshop method for controling the central portion of the arch, given any particulart fall/rise across any particular distance from the centerline.  I don't believe templates were the way.  And while a catenary is very close, I don't believe it is dead on to what they did.

 

From what one can read in relatively early texts about woodwork and cabinet making, you see lots of workshop design and layout methods based on simple geometry using straight edges, edges fixed at set angles, compasses, dividers, etc.    The had many methods for controling curves using such simple means.   I've been looking at such simple straightedge and proportion means to find a workshop control for the Italian arching.

 

As Curious1 points out, I've not examined enough arches to defend such a broad claim as I made.   However, I have access to a complete set of the Strad posters, images of all the Ashmolean posters, and a good number of laser, xRay, and Scans of arching that have appeared on the web.   This is certainly not an entirely reliable set of reference data, but it's far from worthless.  What I've been to examine so far gives good support to the idea I suggested.

[curious1]

I'm perfectly comfortable with the idea that talent and experience would allow someone to create archings that look consistent and work well.And I'm fairly sure there are no existing Cremonese arching templates, which can not prove anything, but reduces the possiblities to either they were ALL lost, or that they didn't use them.

This is what intellectual rigour looks like.

[curious1]

Yes.  I'm inclined to believe there was a well established workshop method for controling the central portion of the arch, given any particulart fall/rise across any particular distance from the centerline.  I don't believe templates were the way.  And while a catenary is very close, I don't believe it is dead on to what they did.

 

From what one can read in relatively early texts about woodwork and cabinet making, you see lots of workshop design and layout methods based on simple geometry using straight edges, edges fixed at set angles, compasses, dividers, etc.    The had many methods for controling curves using such simple means.   I've been looking at such simple straightedge and proportion means to find a workshop control for the Italian arching.

 

As Curious1 points out, I've not examined enough arches to defend such a broad claim as I made.   However, I have access to a complete set of the Strad posters, images of all the Ashmolean posters, and a good number of laser, xRay, and Scans of arching that have appeared on the web.   This is certainly not an entirely reliable set of reference data, but it's far from worthless.  What I've been to examine so far gives good support to the idea I suggested.

Hi David,

Yes, those are all valuable resources. People tend to speak in sweeping statements, I'm as guilty as the next, to prove our point when more specific observations would be more compelling.

No doubt there is an underlying methodology to Cremonese arching. That is how workshop/factories operate. My feeling is that as more and more archings are digitally scanned and compared an understanding of that system will emerge. What we do with that knowledge is a different question.

[James M. Jones]

I have been particulay impressed by Francois Denis' artical in the Feb 2010 article "Golden Arches" where he explains his idea of an enginerring aproche related to building and boat design/ to arch design of the Bresion and cremonesse, essentialy a series of ratios of the mximum thickness of the plate,placed at key points to establish eight points within the plate that govern the overall shape of the arch. A system such as this would allow for flexability in design,outline, arch height, inflection, centeral hill shape, could all be modified and variable according to the whim of the maker ....yet share famial traits.

  I have compaired blind resultes made with arching templates of Strad and DGU ...finished arces ..to the ratio,s proposed by Denis,and the results  are remarkably close to the ratio's he gives.

  The system boiles down to

   marking an X from widest points of bouts center A

   centering points on the legs of the X points B = 1/2 A

  horizontal lines @ The widest points of the upper and lower bouts and thru the center point A

   the upper bout center point C = 7/10 A

   lower bout center Point D = 4/5 A.........slightly higher than the upper bouts

  c bouts 1/4 width  point's E E' = 4/5 A

  curiously, the narrow of the c bouts seems to coinside with the point A

  I do the layout on the inside with only compass and straight edge so that I can measure and adjust the outside without having to re-layout after every change.

   I'm very intrested in what others think of this approche.

[curious1]

I have been particulay impressed by Francois Denis' artical in the Feb 2010 article "Golden Arches" where he explains his idea of an enginerring aproche related to building and boat design/ to arch design of the Bresion and cremonesse, essentialy a series of ratios of the mximum thickness of the plate,placed at key points to establish eight points within the plate that govern the overall shape of the arch. A system such as this would allow for flexability in design,outline, arch height, inflection, centeral hill shape, could all be modified and variable according to the whim of the maker ....yet share famial traits.

  I have compaired blind resultes made with arching templates of Strad and DGU ...finished arces ..to the ratio,s proposed by Denis,and the results  are remarkably close to the ratio's he gives.

  The system boiles down to

   marking an X from widest points of bouts center A

   centering points on the legs of the X points B = 1/2 A

  horizontal lines @ The widest points of the upper and lower bouts and thru the center point A

   the upper bout center point C = 7/10 A

   lower bout center Point D = 4/5 A.........slightly higher than the upper bouts

  c bouts 1/4 width  point's E E' = 4/5 A

  curiously, the narrow of the c bouts seems to coinside with the point A

  I do the layout on the inside with only compass and straight edge so that I can measure and adjust the outside without having to re-layout after every change.

   I'm very intrested in what others think of this approche.

Can you make a quick sketch?

[Christopher Jacoby]

I love the exposure I've had to Denis' methods. I walked through a viola extrapolation that Jason Viseltear made this summer. Incredible.

[James M. Jones]

Can you make a quick sketch?

maybe this helps.

[curious1]

DSCN8142.JPGDSCN8145.JPG

maybe this helps.

Perfect, thanks.

And the height of the point half way to the edge in the c bout is 4/5 A?

[James M. Jones]

Perfect, thanks.

And the height of the point half way to the edge in the c bout is 4/5 A?

Yes, according to Denis, Max thickness = A

                                      upper bout center 7/10 A

                                       lower bout center 4/5 A

                                        C bout quarters 4/5 A

                                       X arm centers       1/2 A

   Pretty simple to figure and adjustable to new models and ideas.

   Also acording to Denis is that the 0 point on a bresian is the outside surface,i.e.The edge crest ,wheras the 0 point on cremonesse is on the rib side of the plate .... these diffrent 0 points give a variety to the arch, the Bressian having a fuller arch and the Cremonesse with more scoop..

[Marty Kasprzyk]

I have been particulay impressed by Francois Denis' artical in the Feb 2010 article "Golden Arches" where he explains his idea of an enginerring aproche related to building and boat design/ to arch design of the Bresion and cremonesse, essentialy a series of ratios of the mximum thickness of the plate,placed at key points to establish eight points within the plate that govern the overall shape of the arch. A system such as this would allow for flexability in design,outline, arch height, inflection, centeral hill shape, could all be modified and variable according to the whim of the maker ....yet share famial traits.

If there is a lot of scatter how does one know if there is an underlying method in which the maker deliberately changes or if there is actually no method and just a lot of scatter?

[Don Noon]

I'm very intrested in what others think of this approch.

It looks like a scalable system that would lead to a very roughly consistent overall shape.

One might think that a mathematically constructed shape might be appealing to me as an ex-engineer.

It isn't, though.

Why? That took me a while to figure out: basically I am biased against recipes unless there is some underlying physics that looks promising to me, or empirical evidence that it works better than what I do now, which is occasionally use a template that I sortof vaguely follow, but end up blending and modifying everything as I go. Not very well-controlled, but I'm not convinced that more precision = better. Yet.

[David Beard]

Perhaps the physics is that the old arching is very close to catenary, which has special charateristics for arches and load bearing in a thin shape.  But the old arches are slightly flatter than a true catenary.  The emperical evidence in favor of finding and using the old method is the relative success of these instruments with the player/investor/collector/dealer communities.

[FredN]

Thank you Fred for describing your methods. They are elegant in their simplicity. I would suggest an even simpler method. Using a similar grid and a calculator the heights of the various points can be generated from known formulas. The advantage here is that instead of defining a surface that may be in compression or tension a median point can be defined and perhaps a truer arching created. It also eliminates the intermediate step of creating templates.

 

I think the reason I liked the method because it spread out any stress, but it certainly has nothing to do with making a class A inst.  fred

[Marty Kasprzyk]

I think the reason I liked the method because it spread out any stress, but it certainly has nothing to do with making a class A inst.  fred

Fred,

 

Could you also show us your drawing which shows the longitudinal and cross arches.

[Don Noon]

The emperical evidence in favor of finding and using the old method is the relative success of these instruments with the player/investor/collector/dealer communities.

This presumes that "the old method", if there was one, is somehow responsible for the success.

I sense an undertone of magic bullet in these discussions, where following a particular arching method, if you can only find the right one, will yield the holy grail of the Cremonese sound, or do some other optimal thing with stress distribution, or something.

Let's just say, for the sake of argument, that they did all of the arching by eye and experience. Why isn't anyone searching for THAT method? (come to think of it... that's what I'm doing) Why does it have to be some obscure mathematical construction? And what does it really matter how you make arching, as long as it ends up close to the arching of stuff that is known to work?

What matters, in the end, is what you are most comfortable using, and gets the results you want. I prefer the easiest way.

[James M. Jones]

I did my first eight or so strictly by eye,There must be others who are exploring this idea.  After meeting and showing my work to one multi gold winner the comment came that everything on a violin is done with patterns and or measurment,scrolls, forms,corners, bass bar ff holes, ff hole,placement,mensure, heal and button, ...why not arches? So I built a few sets and started using them. What I have found is a distinct incresse in both the speed of the work and flow of the final job, less thinking, more doing. NOT that arches are a magic bullet, At the very least they do play a big role in how the work is initaly percieved..

  It doesn't Have to be a "obscure mathmatical construction"however I think it's a fair bet to say it could have been.,  I am empathetic to the idea of geometric/ mathmatical derived formulas for violin design purposes because of the methods and practices I see in my own trade(tradtional blacksmith),simple precice and good looking. The historical trend of the day was to build to propotions as a method of creating harmony in structures. A method such as Denis proposes would allow for simple targets to act as guides.Much as the grad punch speeds the work on ther inside, this set of nine refrence points would speed the work as well, whether by eye or arch template. The real magic is probably in the graduations.But the arch is what people see.

[francoisdenis]

Yes, according to Denis, Max thickness = A

                                      upper bout center 7/10 A

                                       lower bout center 4/5 A

                                        C bout quarters 4/5 A

                                       X arm centers       1/2 A

   Pretty simple to figure and adjustable to new models and ideas.

   Also acording to Denis is that the 0 point on a bresian is the outside surface,i.e.The edge crest ,wheras the 0 point on cremonesse is on the rib side of the plate .... these diffrent 0 points give a variety to the arch, the Bressian having a fuller arch and the Cremonesse with more scoop..

Hello every body

 

To be correct, the maximum thickness of the arching doesn't need to match the point A  (even if all the measurements depends of this maximum thickness) We often notice that the maximum height of the back arching is set slightly lower.

 

Francois Denis

[James M. Jones]

 

Hello every body

 

To be correct, the maximum thickness of the arching doesn't need to match the point A  (even if all the measurements depends of this maximum thickness) We often notice that the maximum height of the back arching is set slightly lower.

 

Francois Denis

 

Francois,Thank you for your input,Sorry but I have a question,

Do you mean that the backs max height is often noted to be slightly lower than the top.?..or slightly lower than the ratio's would indicate? i.e. flatter. Also I assume the max height of most plates to be in the area of the stop or sound post,and that point A will be above ,(north of) that point .... following this method a fellow would figure from the Max height and NOT the height @ A point? The diffrence would be relitavely small but little details do make up the ensamble .

  I have noticed that when following my arching templates of the Plowden and Messie for shapping, upon compairision to your ratios,they apply remakably well, usualy within a few tenths of a mm. This leads me to think that you could be onto something.

 One more borring thought on Ratio's and work methods, When splitting wood it is VERY important to split down the center, or the split will vere to the weak side. 1st grade stuff, I realize..... a lifetime of splitting wood for violin tops ect. would produce a "ratio" tuned way of seeing or processing, be it thoughts or wood.  Another observation....the glass is always either half emty or half full ...it almost never has 3.145  cm cu of wine in it.

[FredN]

Fred,

 

Could you also show us your drawing which shows the longitudinal and cross arches.

 

Hi Marty- about 15 or more years ago I sent a paper to VSA on  bent sticks for arching inst's and they rejected it. Unfortunately, the editor of VSA, not being a maker sent it out for editing and sadly sent it to a person who absolutely believed plates were bent, not carved and the paper had no interest and should be rejected. If I find it, would you care to try and submit it again. I'm just too old  to try again. There is detail that would save someone lots of time. I can't recall if there was such a drawing, but i'll check.

 

I think this is important no matter how you do your arches, but in my case where I used templates I found out in about 2-3 months you cannot match the shape of the template to the  arch of the inst  where it was used. I recall trying to match the template station behind the bridge and it would not fit the shape of the top. I did this to maybe 2-3 inst's.  I can't recall where the change in shape occurred. This was done on violas, and it might not occur with violins. Honest, I never drank more then 2 beers when working.   fred