Recommended Posts

There seems to be an assumption here that the structure of the violin can't resist the longitudinal tension exerted by the strings. That sounds suspect to me.

I know it's often postulated, but has it ever been measured to occur in a well built instrument?

 

Link to post
Share on other sites
  • Replies 113
  • Created
  • Last Reply

Top Posters In This Topic

On 2/18/2021 at 1:44 PM, Owen Morse-Brown said:

But nobody talks about inside profiles. What if they're more important than the outside? I think we're so obsessed with profiles when we don't even know if the old masters carved inside or outside first. The inside shape looks so much simpler than the outside without the recurve and scoop and if you carved that first, then the scoop would be a result instead. 

You can't have one without the other.

From my perspective, there isn't any good reason for carving the inside first. Kind of like building a house from the inside outwards.

Link to post
Share on other sites
12 hours ago, Dennis J said:

There seems to be an assumption here that the structure of the violin can't resist the longitudinal tension exerted by the strings. That sounds suspect to me.

I know it's often postulated, but has it ever been measured to occur in a well built instrument?

What is well built? If you put strings on anything it's gotta bend. Laws of physics. Even thick steel profile would bend with violin strings stretched over it, though just fraction of a mm. Keep the force for hundred years and it will even stay partly in that bent shape after releasing the tension. Wood is no different and it will necesaarily give a bit under tension and if you measure new instrument without tension and after stringing (and settling for some time) you'll find where the stresses go.

I mostly work on mandolins and they tend to form bulge behind bridge as that area is unsupported by tonebars and often makers carve new tops with highest spot behind bridge (because that's what they see on old precious mandolins) and that causes even more bulging over time. Some of the better preserved old "Loar" mandolins that didn't have too thn tops seem to retain he highest spot under bridge and thus much less deformation.

Link to post
Share on other sites

If the flatter area of the front was caused purely by deformation over time, you would see quite a big curve in the plane of the ribs but all the pictures David Beard posted show flat ribs. To get that much rise in the upper and lower bouts, the top and bottom blocks would have to rise in relation to the corners. Surely the rib garland once glued to the plates creates a rigid enough structure to resist that. 

Link to post
Share on other sites
13 hours ago, Dennis J said:

There seems to be an assumption here that the structure of the violin can't resist the longitudinal tension exerted by the strings. That sounds suspect to me.

I know it's often postulated, but has it ever been measured to occur in a well built instrument?

 

 

1 hour ago, HoGo said:

What is well built? If you put strings on anything it's gotta bend. Laws of physics. Even thick steel profile would bend with violin strings stretched over it, though just fraction of a mm. Keep the force for hundred years and it will even stay partly in that bent shape after releasing the tension. Wood is no different and it will necesaarily give a bit under tension and if you measure new instrument without tension and after stringing (and settling for some time) you'll find where the stresses go.

I agree with HoGo, with string tension there will always be an immediate deformation, more or less significant depending on the stiffness of the body (arching shape, thicknesses, ecc.). A well-built violin will be the one that will be able to contain these deformations within acceptable limits (i.e. that they do not influence the setup and playability too much) for a long period of time (optimistically centuries) before a restorer is forced to intervene. But if these deformations were absent (well, too small) it would mean that the violin is not well built at all because it would be too stiff and probably a not good sounding one. This is violin making, finding the right balance by staying on the razor's edge without cutting yourself;)

This image refers to a violin a few moments after being tuned to 440Hz, I don't know if it's a good or bad one and obviously the numbers will change from violin to violin, but the areas will be more or less the same.

408273665_Deformazionicontensionecorde.thumb.jpg.3af147ae9d75b50b5c46436972ec16be.jpg

Link to post
Share on other sites
15 hours ago, Dennis J said:

There seems to be an assumption here that the structure of the violin can't resist the longitudinal tension exerted by the strings. That sounds suspect to me.

 

Marty Kasprzyk has a design which puts no longitudinal compression on the top. It's far from being a "conventional" fiddle though.

There have also been some fiddles with a straight bracing rod running between the upper and lower blocks, which I presume could carry most of the longitudinal compression load if properly designed and positioned.

Link to post
Share on other sites

Remember,  this distortion somehow magically but accidentally transformed every Cremona arching into the well formed range of archcungs that the world now covets.  

And, of course, since this instruments were adooted as the best quite soon after their making, they were coveted before, during, and after this magical transformation.

Link to post
Share on other sites
1 hour ago, David Beard said:

Remember,  this distortion somehow magically but accidentally transformed every Cremona arching into the well formed range of archcungs that the world now covets. 

This shape also "somehow magically but accidentally" imitates the distortion which normally happens with age, under prolonged string tension, particularly under moist climactic conditions. ;)

What's holding you back from performing the experiment I have suggested? If you wish, I am willing to perform it on a fiddle you have made, as long as I am granted permission to post the outcome here.

You live in Santa Barbara, which is one of the friendliest climates in the world, for fiddles. Unless you have fiddles all over the world, from Alaska to the tropics (as many other makers do), I wouldn't expect your fiddles to have been challenged much.

Link to post
Share on other sites
55 minutes ago, David Beard said:

Remember,  this distortion somehow magically but accidentally transformed every Cremona arching into the well formed range of archcungs that the world now covets.  

And, of course, since this instruments were adooted as the best quite soon after their making, they were coveted before, during, and after this magical transformation.

The distortion is real, nothing magic about it, plain laws of physics. Just like any old car will magically and accidentally transform into pile of rust if left in the weather.

The Cremonese makers (and especially Amati's) were prized mostly because they were the biggest (and for some time sole) producers of these beautiful instruments for nearly 3 centuries. Many of the guys further from Cremona made just VSO's of the time (now prized, but then they just filled demand for cheap instruments for general public).

One should remember that for part of their life they were strung with gut strings that (I guess) have lower tension than modern strings so the range of archings may be result of their stringing history (along with the other factors mentioned in previous posts).

Re: Don Noon. I think there is much less width deformation that can be attributed to string tension than we may believe at first glance. The soundpost tension and pressure of bridge feet playing the major part. The plastic deformation of the wood works in all directions and the wood fibers may "slip" along each other as well keeping the long time effect deformation caused by string tension just in the central parts of bouts. The movement of tail block can be really tiny relative to the visible bulging.

One more side note. On mandolins with traditional "two foot" bridge after some time a small bulge appears right between the feet. Many builders say this is the edges of bridge feet digging into top but I watched this on my first mandolin for over 20 years now, and came to conclusion that majority of this is caused by the string tension. I made a new bridge base with one continuous foot replicating the original arch shape and after few months the arch was back with just tiny impression of original two feet bridge.

Link to post
Share on other sites
5 hours ago, Owen Morse-Brown said:

If the flatter area of the front was caused purely by deformation over time, you would see quite a big curve in the plane of the ribs but all the pictures David Beard posted show flat ribs. To get that much rise in the upper and lower bouts, the top and bottom blocks would have to rise in relation to the corners. Surely the rib garland once glued to the plates creates a rigid enough structure to resist that. 

No big curve, as I noted before the deformation in length is tiny and so is any long bend of violin body. The back is rigid and being stretched it (mostly) keeps the flat plane while the top gets the compression load and tries to "wrinkle" under the load. Also along the long life of the violins the ribs were commonly adjusted or re-fitted by restorers, sometimes planed...

Link to post
Share on other sites

Hogo, there is no question that some distortion occurs.

David, I believe your experiment. And I've been aware of it for many years.

The question is 'are these distortions the cause of the classical ling arc shapes?'

I still say no.  The distortions you point to are consistent enough or extensive enough to be the cause.

I don't have time today, but sometime in the next 36 hours I post some about some of the specific the lead me to believe simple distortion is insufficient to be the cause of those classical shapes.

Link to post
Share on other sites

When you look at the violins witch "camel back" it is clear they didn't start that way so similar amount of distortion can be part of any other violin of similar age and similar top stiffness. Of course it is impossible to tell exactly where they started and how much they deformed over time. The arches of many(?) were corrected to form that restorers believed to be correct and many were thinned or breast patched and modified so we can only speculate.

We can probably look at some of the healhier "thick" del Gesu violins as they certainly spot less distortion versus those that were thinned early in their life and used extensively and that would show some trend. Today's best restorers/makers with access to those instruments could guess from flexibility of body during tuning up/down how much distortion can be in the instrument. All that we can be certain now is the tops were built less flat then they are now.

Link to post
Share on other sites

Davide Sora's distortion images show just that, but no evidence of any shortening of the top plate caused by upward bending. The tension of the strings twists the body of the violin more than anything else.

If linear string tension buckles the top why are there so many examples of older violins with no buckling whatever.

 

 

Link to post
Share on other sites
29 minutes ago, Dennis J said:

Davide Sora's distortion images show just that, but no evidence of any shortening of the top plate caused by upward bending. The tension of the strings twists the body of the violin more than anything else.

If linear string tension buckles the top why are there so many examples of older violins with no buckling whatever.

Part of it is because restorers keep undoing the buckling.

What sort of twisting are you referring to?

Link to post
Share on other sites
8 hours ago, David Burgess said:

Marty Kasprzyk has a design which puts no longitudinal compression on the top. It's far from being a "conventional" fiddle though.

There have also been some fiddles with a straight bracing rod running between the upper and lower blocks, which I presume could carry most of the longitudinal compression load if properly designed and positioned.

I got the idea from piano sound boards which aren't under any string tension.  Violin makers should get out more often.

Link to post
Share on other sites
1 hour ago, David Burgess said:

Part of it is because restorers keep undoing the buckling.

What sort of twisting are you referring to?

I think someone previously said that there is a lot of inbuilt asymmetry in a violin which I think is pretty obvious.

And I've always assumed that sideways bending or twisting is far more likely than upward bending under string tension. And that could involve the neck to some degree.

So there must be plenty of ways a violin can bend before the violin top's length is involved.

 

Link to post
Share on other sites
12 hours ago, Marty Kasprzyk said:

I got the idea from piano sound boards which aren't under any string tension.  Violin makers should get out more often.

It’s interesting how piano builders and violin makers think of soundboards. When a piano soundboard gets old and warped and develops cracks, a piano restorer will rip the thing out with a claw hammer and throw it in the trash, and replace it with some fresh, new pieces of Spruce! If violins had lids we would probably do the same.

Link to post
Share on other sites

It took longer than I anticipated to get back to this.

 

OK.

We're considering Burgess's hypothesis that characteristic flattish stretches in Old Cremona and Old Italian and generally much or most pre-modernization violin making didn't start that way, but was carved originally with more continuous curvatures rather like the backs, and that the flattishness and associated extended height of these long top arches was primarily acquire by distortion from string pressure over time.

Let's start by noting that this hypothesis solves a problem that doesn't need solving.   There is no reason why the could not have just directly carved more or less the shapes we see now.   And, Yes.  Of course string pressure can lead to various distortions, but that does not automatically mean they are the primary cause of the main shapes we see today. 

Further, such deformations express themselves in many ways.  We can plainly see bulging and distentions in some plates around the areas where the post and bridge make contact.   However, there is no obvious way to justify believing that even larger, yet somehow much more uniform and beautifully shaped distortions will occur in every single classical example at the far ends of the plates?

The principle evidence for this hypothesis is an experiment the demonstrates that a non classical top long arc, if made with a good through curved but full height extending from bridge toward the end blocks can be deformed into a more classical looking shape (with flat and then sharper change to descending curve) than it started with.

archingdistort.jpg.e340b7ec21f035a7fc14574fad17b5b9.jpg.7cd69a7d30b88a3f947ed09b1aaf1a98.jpg

 

But notice, the distortion demonstrated here changed the shape only a small amount, and the basic dimensions have changed only very little or not at all.

The distortion did not make the arching fuller, it all ready was full.  It substantially lift the overall long arch, it all ready was rather high, and kept fairly high until after the lower bout line.

What did happen is a non classical top arch shape got deformed and pushed toward a more typically classical shape.

 

This does not demonstrate that such deformation could happen in every case.  It does not demonstrate would unavailing happen every case that wasn't carved with a classical shape to start with.

And, it doesn't demonstrate that any classical examples ever weren't made with a flattish stretched carved into the top arch from day one.

 

Again, it is a solution to a non-existent problem.  Other than the maker's choice and carving, there is no need to explain the current shape.  There is no reason they could not have been carved basically as we see them now. 

Unless, you are someone who for the last 25 years plus has chosen to carve your top arches without this feature.   Then, if you don't want to acknowledge you've just been ignoring or defying the classical examples, then you have a motive to claim that old arches weren't originally the way we see them now.  No, no, don't believe your eyes and evidence. No.  'Those arches were really made the way I do today.'    No.  Those old instrument used to be as good as mine, but now their just deformed and distorted.

Really?

There is no evidence that the classical top arches were not substantially as we see them today, minus the obvious deformations of bulges around bridge and post, and actual collapsed portions around some bridges.   But, as for the beautiful and close to symmetric shapes we typically see in the upper and lower bout long arcs and cross arcs, there is no cause to believe they all greatly altered from the starting points.   Small amounts of course.  Large amounts and basic character of shapes, no.

Let's look at the scale of the issue.

 

653365878_1693violin362StradHarrisonbassside.thumb.jpg.76f6c3b64fcb78c1d4b75a61da8a4d56.jpg

In some cases, as in the Harrison Strad shown here, the fullness of the top and back long arches is very comparable.  The differences amount to subtleties of the shape.  The back arch sustains most of its height nearly as long as the top arch. But, unlike the top, its curve begins almost immediately around the bridge area. In contrast, the top looses almost no height until clear of the corners, and then comparatively abruptly curves rapidly down.

But, in such well balances cases, the Burgess hypothesis seems feasible.  We CAN think maybe the top start off original a little more curved, a little higher and deformed into this shape.

But why get all convoluted like that?   Why couldn't they just carve this shape. No reason.  They could just carve this shape. And, there is no reason to believe they didn't.

If look at more than just a few examples, we will immediately realize that the long arches in classical examples aren't all nicely balanced like in the Harrison.

To succeed, a hypothesis has to account for the full range of examples, not just a few.   And, if we survey classical examples, we'll a great range of variety.   We will find that they made back arcs the retain their height well as the head to the outer bout areas, and we will find backs were most of the height drops away before then.  And, with tops the same.  Sometimes the flat area continues far into the bouts, sometimes it begins coming down considerably sooner.  We'll also see that both tops and backs can begin with low, medium, or extra tall height.  And we'll the classical makers mixed and match these choices as pleased them.    We'll find examples where the classical makers experimented with introducing an overall tilt to either a top or back long arch, or both.  They occasional put the high point more toward either upper or lower bout, instead of the typical location near the bridge, and most typical just in front of the bridge line.

Some of these combination really stretch the imagination if you try to cling to the distortion hypothesis.

Here are a few examples directly comparing how different classical back and front long arches can be.

1666c Andrea Amati:

286968179_1566cAAmatioverlay.thumb.jpg.e6946c03cc3b813af1918b72d52326cc.jpg

 

2127903891_1566cAndreaAmati351362_top_back_ribs.thumb.jpg.da0411eea9ff3aac0013ab559a9b876e.jpg

 

1721 Lady Blunt Strad:

1657121471_1721LadyBLuntoverlay.thumb.jpg.087176a5cf994ca080258718541ebd76.jpg

160802467_1721violin357StradLadyBlunttside.thumb.jpg.42e2bb3377caac9ca3b82225a3f476a6.jpg

 

1750 Peter of Venice:

1525960264_1750PeterofVeniceoverlay.thumb.jpg.85e8f4ac97d3d653d7fdc17961dce8a6.jpg

985899403_1750PeterofVeniceviolin093123mge4ipu0i8505ptm.thumb.jpg.da74bda6b21d3349353f084aaf2df5a6.jpg

These example begin to show the extent of top and back differences and the range of cases the 'distortion hypothesis' must carry to be a valid explanation. 

There's a big differences between saying 'Look at this dynamic. It might be a factor in some of the shapes we see today.' and saying instead 'Look at this one experiment.  Obviously none of the shapes we see today originally looked remotely like they do now.'

Again, there is no need for the 'distortion hypothesis'.  Certainly some of the flattish long top arcs are clearly collapsed around the bridge.  That doesn't mean they didn't start out basically flattish on day one. Just that they collapse some since then.     And, it's certainly possible that some flattish examples might have started out slightly more crowned, but that doesn't mean they didn't start basically flattish on day one.    

 

Let's also consider the cross arches associated with the long arc.    When the height of the long arc sustains, that also means the cross arches stay fuller.

It's hard enough for me not to laugh at the notion that the long arcs lifted and changed as much as the 'distortion hypothesis' would require, but when also consider fullness of the cross arching, the idea is absolutely ridiculous.

The areas most concerning are here:

113832115_viottiregions.thumb.jpg.8b906426675ee4b1ea66af62b29e3a4d.jpg

The amount of long arc lift and cross arc expansion to convert some of the back long arc profiles into some of the top long arch shapes is absurd.   And, to do so would require actually stretching the surface area of the wood, which is not a sane proposition.

Unfortunately, my collection of cross arch xRay cross sections is limited.   Most public images of these focus on the main arch locations, and this problem is more likely to be maximum a bit closer toward the ends. 

From side views of long arcs, it's clear that many more extreme cases exist, but I don't have the corresponding cross arc xRays.

This 1728 Del Gesu example shows the greatest difference in cross arc fullness I have in my image collection.

1671790954_crossdistortion.thumb.jpg.9d54ee4e699ead28a72c079cc2ab33ed.jpg

This example obviously isn't as dramatic as some.   Nevertheless, the paths of these arcs are about 2% different in length.  For more dramatic examples, the idea of one shape distorting into the other would be impossible.

*****

Okay, enough for today.    We live more in a culture of believe than reason, so I'm not likely to convince anyone who wants to believe the 'distortion hypothesis' is true.

To me, it's obviously insufficient to meat the range of examples that exist.

To salvage the hypothesis, you'd have to concede that the carved the basic fullness of the arching we see, with all the variations we see.   But still , there is simply no need to resort to this idea.  Distortions certainly did and do occur, but mostly taking shapes in clearly unplanned and deformed direction.

To believe that none of the classical instrument we have today had the characteristic long top arcs originally, but that instead uncontrolled deformation consistently changed all of them from one well form shape into a different well formed shape is fantasy.   But a fantasy that some want to believe.

Over the coming years, I will make a point to gather the specific data that will resolve this definitively.    Studies of grain direction particularly will ultimately settle this question in an absolute way.

 

 

Link to post
Share on other sites
8 hours ago, David Beard said:

There's a big differences between saying 'Look at this dynamic. It might be a factor in some of the shapes we see today.' and saying instead 'Look at this one experiment. 

One experiment? Heavens no! There are many more, including the image Davide posted above, showing the distortion of a violin immediately after being strung up. This well illustrates the forces and the general trend, and this distortion doesn't magically stop after minutes or days.

Bruce Carlson was also involved in distortion measurements of the Cannone Guarneri, which show the same trend, but I believe that publication is behind a pay wall.

Wood is a plastic, deformable material. Do you remember the threads about cold-bending ribs? All that's required to do some major bending is force, moisture, and time.

The "one experiment" I suggested was intended as something you or others could easily duplicate, without needing decades of experience in the old instrument restoration trade to begin to understand how instruments distort and how extensive this distortion can be, even if they haven't had enough contact with the hard-core restoration trade to know how frequently this distortion is "corrected" to one degree or another.

Link to post
Share on other sites

David (Beard), you probably missed it but no one stated here that the top and back arches on Cremonese violins started with the same shape or height (you know this from flat back viols...). The only thing stated was that they do deform in a somewhat predictable way (though hardly predictable amount) for their whole long life. When there is force there will be reaction, the wood is not perfectly stiff or fully elastic material. The hypothesis (and supported by laws of physics that are not to laugh) is thatthe top arches started with less flat top arch than what is generally considered "classic".

David Burgess as an "insider" knows much more first hand info on them than most folks here and since many (most?) Cremonese violins have SP patches, breast patches, bridge feet patches etc. and that all comes hand in hand with pressing against cast that can be easily adjusted to correct any changes in arching... Only few folks know how much of this really happens (or happened)... just like the "unantiquing" of Messie mentioned in other thread.

Link to post
Share on other sites
19 minutes ago, HoGo said:

 Only few folks know how much of this really happens (or happened)... just like the "unantiquing" of Messie mentioned in other thread.

One famous restorer told me of a cello top that he re-arches every couple of years. It goes "saddle-backed" and the sound deteriorates. After re-arching, it's good for another year or two.

Link to post
Share on other sites

Some of you may have seen this pic before, but anyway. Since the discussion is about what the old makers did or did not, consider this excerpt from Dizionario delle Arti e de Mestieri, by Francesco Griselini, Marco Fassadoni, Venice 1770

LIUTIERE, o FABBRICATOR. DI VIOLINI, ED ALTRI STROMENTI.

2CCC1E75-3C50-4306-ABBF-42674F1CBD50.thumb.jpeg.643a890fb6af3ad3e26a4831f7ce11ed.jpeg

”The principal point for the goodness of the instrument is to find good, old and sonorous spruce for the belly: the best is from Tyrol. The cavity shapes given to this belly in a vault shape more or less high, the diverse thicknesses to be observed, the way that the bassbar is placed inside, to the side of the cordone, which is the thickest string of the violin, the height of the ribs, and finally the excavating of the back which has to correspond perfectly to that of the belly; all this together with the true way of positioning the two holes in the shape of S which are carved in the violin belly, the placement of the soundpost and the bridge, contribute in an essential way to the goodness of the instrument." 
Dizionario delle arte e de’ Mestieri, volume 8, page 196.

Note that the long arch isn't mentioned at all. The internal curves should reflect each other, there's no reason to make it more complicated than that. The difference in the outer shapes of the back vs belly is due to different distribution of thicknesses, that combined with distorsion from 300 years of constant string tension. Thin bellies tend to distort more than thicker ones, to the extent that they sometimes cave in under the bridge.

Link to post
Share on other sites
On 2/21/2021 at 2:43 AM, Dennis J said:

So there must be plenty of ways a violin can bend before the violin top's length is involved.

Every time a solid object bends there is a neutral axis where there are no stresses but the further in crossection the material is from this axis the more it gets stressed, the outside is under tension while inside under compression. So the outside stretches and inside of the bend compresses. On every material, even with tiny force. There can be elastic bend with smaller forces where materal is still able to get back to original shape after force is released but with long time stresses the deformation becomes permanent. Wooden archery bows will weaken after some time and even snap eventually due to material fatigue.

The violin is not solid but the bending force of strings is distributed through whole structure and since back is much stiffer than top the neutral axis (or plane/surface) of the whole violin is close to back and while the bending visible is minimal and with the shape of ribs hardly identifiable at all, the top takes major share because it is further from the axis. This is of course somewhat simplified but it corresponds with visible deformations.

Link to post
Share on other sites

Apart from the sinking just forward of the bridge common in a lot of instruments I would say that most older instruments' top profiles are probably about the same shape as they were originally made.

What might look like upward buckling of the upper bout might have more to do with the neck angle changing over time reducing the clearance between the top and the underside of the fingerboard. And that could be explained by cross grain shrinkage of the upper block along with the forces of compression from string tension.

 

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    No registered users viewing this page.