Graduations for Response to Hard Bowing

[Michael Darnton]

1 hour ago, Don Noon said:

2) There is no supporting theory for why that particular mathematical curve should be the ultimate crossarch.

No supporting theory, but I believe I have found some strong circumstancial evidence that They were using this method at least as the initial design. That doesn't mean everyone followed it but I see much more use of it than you think you have seen. Still trying to figure out what to do with this. . . An article for someone, maybe, and I've been feeling people out about how they read these facts.

Mainly this is for my own entertainment.

[Michael Darnton]

By the way, John Masters had a pretty good supporting theory for cycloids which you may not remember.

[David Stiles]

2 hours ago, Michael Darnton said:

The only reason to make the edges of the f-hole 3 mm is to have them not look thin. That only needs to happen for the 5mm or less immediately around the opening. I was taught to treat that area completely independently. For the post an area about 25mm diameter with a gentle approach is sufficient.

Thanks Michael, makes sense, I can see what you mean here.

2 hours ago, Michael Darnton said:

I think you got yourself into this mess by trying to tune.

 

Thanks for empathizing this, I have a lot to learn.  I am not sure if I am in a 'mess', I still think this violin could well be better than my last.  I have not been focused on mode tuning, just trying to get to a target weight.  If we had had this conversation earlier, I wouId have been comfortable taking more weight out from the centre.  I probably still could take a little. 

I am not keen on mode tuning for it's own sake, just as a crutch while I learn what a plate should feel like.   Mode frequencies may be a useful as a measure to communicate what a late is like to others.

[Michael Darnton]

I would suggest to you that you consider the idea that flexibility is controlled by arching and that graduation is about weight distribution. In that context tuning is irrelevant.

The originators we're much more likely to have used a mechanical model rather than an acoustical model which did not exist at that time, and they seemed to do just fine.

[Don Noon]

3 hours ago, Michael Darnton said:

By the way, John Masters had a pretty good supporting theory for cycloids which you may not remember.

The only thing I could find was this thread by John Masters which mentioned cycloids... but in no way related CC's to anything connected to sound or playability... or anything I could determine, other than that it's a mathematical curve that can have an inflection point.  Feel free to find a "supporting theory" anywhere in there.

Again, I'm not saying CC's are bad... it's a convenient way to get a smooth curve that inflects, and you can diddle with the parameters to get a range of shapes.  I'm only saying that there's no rational reason to think that this precise curve is the holy grail of perfection.

[Michael Darnton]

I am not John and cannot speak for him, but his conclusion regarding his model (did he call it an FET model, I don't remember) was that put under the forces that a violin has the arching naturally converges into cycloids as ? the most stable shape? It wasn't about the strongest shape. . . it was about the most natural shape for the circumstances. That's pretty close to a Holy Grail of perfection, but he would have to explain it. He may have been strange, but he was not an idiot.

That's my bad memory of something I didn't understand but he did, so don't argue with ME about this.

[David Burgess]

What John Masters told me, when I discussed various things with him at a VSA Convention in Cleveland, was that it was pretty easy (for him) to program his CNC to carve the various permutations of curtate cycloids into plates, and then test the various outcomes by playing and listening to the instruments. So my impression at the time was that this was done more for experimental expediency, than out of a fundamental belief in cycloids.

And once he found a shape that worked, it would be easy to repeat. Just run the program again.

[Michael Darnton]

When we talked about it, he was talking about his scientific model, before his CNC excursion, not CNC expediency. That probably influenced his choice of what to program.

[Don Noon]

50 minutes ago, Michael Darnton said:

That's my bad memory of something I didn't understand but he did, so don't argue with ME about this.

I mostly understood what he was doing, and argued with him about it.  Without rehashing obscure structure and vibration stuff that is meaningless to most folks, my engineering opinion is that he did not prove that cycloids were ideal for static stress (they aren't), and didn't attempt to show any theory about vibration and tone.

If precise cycloids were The Holy Grail, it makes me wonder how so many of the top makers today can do so well without using them.

[Christian Pedersen]

.

[Michael Darnton]

I have never said that you can't build a good violin without cycloids. I'm just maintaining that just as you can run a marathon with 10 lb of weight on your legs and actually finish it you can also build a violin with an inefficient system and muscle it into sounding good but this is not the most effective way to do it and may not yield the best results because of the compromises it enforces.

The players who use the fine old instruments would maintain that the modern makers who think they're equal are delusional and that their judgment may be overwhelmed  by self interest.

Dismissing the needs and opinions of the users seems like the ultimate in disrespect and arrogance.

[Christopher Jacoby]

This is good Maestronet. I've missed good Maestronet!

[Don Noon]

 

[Marty Kasprzyk]

18 hours ago, Anders Buen said:

Otto Möckel seen to have a similar idea, to use stiff borders. I have been working on flat plates and their mode shapes related to their effective masses. The way the edges are fixed play a role. Eg clamped edges give a smaller first mode shape, but a lower effective mass than simply supported plates where the edges are free to «tilt» over the edge. The result is probaly a louder resonance.

Bernahrd Richardson describes this in one of his later articles on the guitar, which do have almost flat plates. The above experience made me understand his idea on this. And I think your comment here support the same idea, although it is not described in acoustic terms.

I think George Stoppani may have informatiuon on this as he do measure mode properties on violin plates on «pinned» edges, almost immobile. One shoould see effects from a stiffer edge region on these mode shapes.   

An arched plate is stiffer than a flat plate of the same thickness but the violin's history has shown that an arched plate still has inadequate stiffness and that an added stiffening bass bar is needed.

Classical guitar history shows that adequate stiffness of a flat plate could be achieved by using a  multitude of added bars going in various directions so arching wasn't needed.

 

[Anders Buen]

1 hour ago, Marty Kasprzyk said:

An arched plate is stiffer than a flat plate of the same thickness but the violin's history has shown that an arched plate still has inadequate stiffness and that an added stiffening bass bar is needed.

Classical guitar history shows that adequate stiffness of a flat plate could be achieved by using a  multitude of added bars going in various directions so arching wasn't needed.

 

I think arching is also a matter of stability over time for climate variations. The flatter arched instruments after Stradivari behaves better for humidity variation than than the older higher arched instruments. Personally I think this is more important for the development of the instruments than their sound, although old instrument arch and thin plates does sound a bit different from flatter thicker arched instruments. 

[Michael_Molnar]

Curtate cycloids are a huge family of curves. It’s like saying  you can find circles on a violin.

[JacksonMaberry]

2 minutes ago, Michael_Molnar said:

Curtate cycloids are a huge family of curves. It’s like saying  you can find circles on a violin.

Right. The thing that finally convinced me that cycloids weren't the end all be all was comparing the cycloids with the actual arching images on various old Italians on Luthiers Library. Some fit perfectly, some sorta fit, and most don't track. 

[Michael Darnton]

To say that curtate cycloids are a huge family of curves is misleading. They are a family of curves generated by a single formula which yields the exact same curve every time the same specific heights and widths are entered. A violin has varying heights and widths, but the curves are mathematically predetermined and tied together via the formula. They are not grabbed out of a bag of different curves any of which would fit. In a sense the curves *appear* different while being mathematically identical.

Violin arches are strange in that the wider and lower areas have a lot of concavity relative to convexity, and in the high, narrow middle convexity nearly takes over. You assume this because you've grown up seeing violins, but this assumption comes from the fact that this is what cycloids do. This progression is not inevitable, as can be seen in other schools and in amateur made violins. The change is dictated by and follows the cycloid progression and is definitely not just intuitive.

Regarding fit, I have said before that Roger Hargrave was the first to note that you could make virtually all del Gesus with the same set of half templates. I don't believe that whole templates where used in a workshop situation because that would require each violin be absolutely identical, which was probably not their intent and is much more difficult to achieve--it would require a new set of templates for every single change in arching height, for instance. Half templates can be jimmied around to fit the varying realities of shop procedure. So the Luthier's Library whole templates are pretty much useless in trying to reveal what was probably done on those violins. If you would experiment with half templates you would find them fitting well quite often. My working theory is that a shop would have one set and would make them work on the situation at hand, and actual half template tests on classical violins hasn't contradicted the idea that this was probably done.

Point of fact, you can find circles on a violin. Lots of them. Many people still don't get this. The whole of 19th century French makers didn't realize the extent of it, for instance. What might surprise you is that in Cremonese violins there isn't anything except circles, except for a couple of specific inches of outline which were sketched in to fill gaps. With modern technology this is easy to see. This is an important fact, not trivial, and again, people who make violins based on their own memory of what a violin looks like don't achieve this detail.

The problem here is people arguing against a procedure they very obviously don't understand and haven't thoroughly experimented with, having already rejected the idea without testing because it doesn't appeal to their prejudices. I'd put money on it that neither Mike Molnar nor Jackson has made sets of half templates and tested them on a representative range of actual Cremonese violins, not pictures, to see what happens.

[Michael Darnton]

6 hours ago, Marty Kasprzyk said:

Classical guitar history shows that adequate stiffness of a flat plate could be achieved by using a  multitude of added bars going in various directions so arching wasn't needed.

 

Quite a few classical guitar makers both modern and historical have used slightly arched tops, and backs are always arched in the best instruments. Hauser even used a strongly curved bridge to clamp the section of the top directly under the bridge into a cylindrical shape in an ungraceful way to give added stiffness there. They have not been unaware of the effects of arching. In a very general way, guitars are often arched in about a 16 foot radius on the back and around 24 foot radius on the top. Hard to perceive, but there.

In the 80s and early 90s I wrote a couple of things for the Guild of American Luthiers about easy shop methods for establishing these arches that didn't involve the traditional method of using spherically shaped workboards used to press the tops and backs to spherical when gluing in braces.

[David Burgess]

13 minutes ago, Michael Darnton said:

The problem here is people arguing against a procedure they very obviously don't understand and haven't thoroughly experimented with, having already rejected the idea without testing because it doesn't appeal to their prejudices.

Are you claiming to be devoid of prejudices?

[Michael Darnton]

I'm claiming that before I shoot off my mouth about something I do my homework to confirm or disprove my initial ideas, then I proceed to further work on a basis of that rather than my intuition. That's how you hold prejudice in check.

You're certainly correct: people who don't do the homework and experiments depend on prejudices and their thought experiments, and I agree with what I take as your possible implication that this is a bad way to work. So I guess we're on the same side here.

[Davide Sora]

It seems to me that chasing curtate cycloid is not very different in principle from chasing the tuning of tap tones, both can be found in ancient violins, or not, and both can be useful, or not. It is up to the luthier to know how to use them and what their limits and potential are, and when to deliberately ignore them, or not...

[Marty Kasprzyk]

56 minutes ago, Michael Darnton said:

Quite a few classical guitar makers both modern and historical have used slightly arched tops, and backs are always arched in the best instruments. Hauser even used a strongly curved bridge to clamp the section of the top directly under the bridge into a cylindrical shape in an ungraceful way to give added stiffness there. They have not been unaware of the effects of arching. In a very general way, guitars are often arched in about a 16 foot radius on the back and around 24 foot radius on the top. Hard to perceive, but there.

In the 80s and early 90s I wrote a couple of things for the Guild of American Luthiers about easy shop methods for establishing these arches that didn't involve the traditional method of using spherically shaped workboards used to press the tops and backs to spherical when gluing in braces.

Yes sometimes flat top guitars are built with back and top plates with these large radii. This produces arch heights of about 2 and 3 mm respectively.  A typical flat top guitar is roughly twice as wide as a violin so a violin would have around 1mm arch height on the top plate which would give negligible increase in a flat plate's stiffness and its various mode frequencies.

 

 

[JacksonMaberry]

1 hour ago, Michael Darnton said:

put money on it that neither Mike Molnar nor Jackson has made sets of half templates and tested them on a representative range of actual Cremonese violins, not pictures, to see what happens.

You're right, I haven't, but I'm more than willing to try it

[Michael_Molnar]

Let me share briefly some of my findings for Cremonese cross arches. Most back plate cross arches can be fitted to a CC which is a tangential cosine and sine function. However, fewer top plate cross arches are true CC’s. They are often cosine and sine functions tangential to a connecting straight line of varying length. These can be found on some back plates too. On some Strad’s I found a compound sine (two sines) connecting tangentially to a straight line that runs to the cosine trough which are 4 functions.  My take away is that it is a complicated story. Calling these curves CC’s is not an accurate picture.

Pardon me if I drop off this thread. I have bigger fish to fry.