Del Gesus unsymmetric top archings - What do we know about the functionality of unsymmetric archings?

[Andreas Preuss]

So here we go for my next question to members of MN.

Having copied some of the late period Del Gesus I noticed that all those I have copied have an unsymmetric top arching. All have the zone below the bass side f hole sunken in. (The worst being the Leduc) My first thought was of course that it came from the bass bar but later I questioned this idea because Del Gesus late period tops are pretty solid and second if the bass bar pulls in so much a bass bar crack would be inevitable. 

I am trying not to find an answer to why Del Gesu made it that way. I rather take the fact that an unsymmetric arching can function. Therefore: Is there any reason the arching must be symmetric?

To avoid any misunderstandings right from the beginning: I am not thinking of anything exaggerated. What I mean with this is, if someone looks at it from 2 meters distance he/she wouldn't notice anything. 

 

[Dave Slight]

I would say there is not a reason the arching must be symmetrical at all, it just seems natural to try and make it so.

Depending on ones diligence, methods of work, tools for checking the resulting shape, it's clear to see that the end result is very variable. Some makers worked very accurately and made great instruments, for some accuracy was less important than the overall result. Add in the passage of time, and almost three centuries of string tension, then things can be all over the place.

None of the 17th & 18th century instruments that I get to work on have symmetrical arching's, but it can be hard to tell what was simply the product of working method, and how much is a result of other factors.
As you say, a healthy thickness is less likely to distort significantly compared to a thinner one, but also we know that Del Gesú was using quite fresh wood too, which could have moved in unpredictable ways as it dried.

With this being the case, I think you could make quite unsymmetrical arches which weren't really obvious. Soundholes, scrolls, outlines etc. are hardly symmetrical anyway. How many times did you give a client a very distorted antique instrument to try, and they even mentioned the arching?
As to whether there could be any benefit from this is a point for discussion, I have observed several times on Vuillaumes and others that a very flat arching around the bass F hole and a bad wolf note coincide. In these cases, maybe some sinkage might even help.

Plenty of makers have copied archings warts and all, so clearly it will work, but I think carved in, is quite different from natural distortion, and I wonder how it might fare over time, with some of this inevitable distortion creeping in too. I think the hard part might be trying to judge what is the right amount to begin with.

[Don Noon]

Below the F on the bass side is a soft spot...  outboard of that, the long grain goes from rib to rib, and is very stiff out there.  Inboard of the F, again there is more length to prevent crossgrain bending.  In addition, there's arching, which tends to be convex and stiff inboard, and not so much (if at all) below the F.

In one set of detailed photos I have of a Strad, there appears to be a significant convex arch just below the F.  In a similar set for a delGesu I have, it appears to be more flat.  I continue to be impressed at the structural efficiency of Strad's arching.

I can think of no reason why archings need to be symmetrical, other than aesthetics.

[joshuabeyer]

A very talented maker in Chicago once told "never sacrifice shape over symmetry".
That's all I have to add for now...

[Michael_Molnar]

1 hour ago, Don Noon said:

Below the F on the bass side is a soft spot...  outboard of that, the long grain goes from rib to rib, and is very stiff out there.  Inboard of the F, again there is more length to prevent crossgrain bending.  In addition, there's arching, which tends to be convex and stiff inboard, and not so much (if at all) below the F.

In one set of detailed photos I have of a Strad, there appears to be a significant convex arch just below the F.  In a similar set for a delGesu I have, it appears to be more flat.  I continue to be impressed at the structural efficiency of Strad's arching.

I can think of no reason why archings need to be symmetrical, other than aesthetics.

This convex arch below the ff's is due to making the ff's flat. Minutes ago I finished a plate and happened to remember noting this feature. 

[Don Noon]

1 hour ago, Michael_Molnar said:

This convex arch below the ff's is due to making the ff's flat. Minutes ago I finished a plate and happened to remember noting this feature. 

... or maybe making the ff's flat is due to wanting to keep convex arching below the ff's.  A lot of makers didn't make them flat, and we only think flat is cool because Strad did it.

[Torbjörn Zethelius]

7 hours ago, Andreas Preuss said:

So here we go for my next question to members of MN.

Having copied some of the late period Del Gesus I noticed that all those I have copied have an unsymmetric top arching. All have the zone below the bass side f hole sunken in. (The worst being the Leduc) My first thought was of course that it came from the bass bar but later I questioned this idea because Del Gesus late period tops are pretty solid and second if the bass bar pulls in so much a bass bar crack would be inevitable. 

I am trying not to find an answer to why Del Gesu made it that way. I rather take the fact that an unsymmetric arching can function. Therefore: Is there any reason the arching must be symmetric?

To avoid any misunderstandings right from the beginning: I am not thinking of anything exaggerated. What I mean with this is, if someone looks at it from 2 meters distance he/she wouldn't notice anything. 

 

I haven't noticed this because -I'm guessing- I focus mainly on the internal arch. If he did use a chain then I'd say the internal arch should be fairly symmetrical and the asymmetry would be due to difference in thicknesses. The asymmetry could be because the way he worked (i.e. held the plate while planing) when doing graduations. I think he had an idea of what thickness he wanted but he wasn't too fussy about making it perfect. Because it doesn't need to be.

[Michael_Molnar]

25 minutes ago, Don Noon said:

... or maybe making the ff's flat is due to wanting to keep convex arching below the ff's.  A lot of makers didn't make them flat, and we only think flat is cool because Strad did it.

I think your interpretation is the correct historical one. Modern makers spot the inclination of the ff's before thinking about arching. 

BTW, I suspect that the added thickness for some graduation maps at the bottom of ff's reflects this missing arch. Maybe. 

[David Beard]

7 hours ago, Andreas Preuss said:

So here we go for my next question to members of MN.

Having copied some of the late period Del Gesus I noticed that all those I have copied have an unsymmetric top arching. All have the zone below the bass side f hole sunken in. (The worst being the Leduc) My first thought was of course that it came from the bass bar but later I questioned this idea because Del Gesus late period tops are pretty solid and second if the bass bar pulls in so much a bass bar crack would be inevitable. 

I am trying not to find an answer to why Del Gesu made it that way. I rather take the fact that an unsymmetric arching can function. Therefore: Is there any reason the arching must be symmetric?

To avoid any misunderstandings right from the beginning: I am not thinking of anything exaggerated. What I mean with this is, if someone looks at it from 2 meters distance he/she wouldn't notice anything. 

 

My first thought is that it's random, simply a looser precision than we expect today.

As for meaning, I believe reflects a maker putting an extreme priority on certain details and aspects of character of work and shapes, but almost no priority on the exact measure or exact shape.

[Don Noon]

44 minutes ago, David Beard said:

My first thought is that it's random, simply a looser precision than we expect today.

My first and second thoughts were that it's creep and a weak spot in the structure, and third thought was that arching may contribute.

[duane88]

I think that the goal was mostly to make things as symmetric as possible. Then they distort.

Amati/Strad, and a few others, perhaps were able to achieve something much closer to symmetry. del Gesu, Storioni, and others in that vein, either saw something different, or just didn't think that it mattered.

There isn't anything to look at that has escaped the ravages of time and tension, not to mention "repair" and "restoration". Who knows, they may have all been crooked when they were made!

[Andreas Preuss]

16 hours ago, Dave Slight said:

I would say there is not a reason the arching must be symmetrical at all, it just seems natural to try and make it so.

Depending on ones diligence, methods of work, tools for checking the resulting shape, it's clear to see that the end result is very variable. Some makers worked very accurately and made great instruments, for some accuracy was less important than the overall result. Add in the passage of time, and almost three centuries of string tension, then things can be all over the place.

None of the 17th & 18th century instruments that I get to work on have symmetrical arching's, but it can be hard to tell what was simply the product of working method, and how much is a result of other factors.
As you say, a healthy thickness is less likely to distort significantly compared to a thinner one, but also we know that Del Gesú was using quite fresh wood too, which could have moved in unpredictable ways as it dried.

With this being the case, I think you could make quite unsymmetrical arches which weren't really obvious. Soundholes, scrolls, outlines etc. are hardly symmetrical anyway. How many times did you give a client a very distorted antique instrument to try, and they even mentioned the arching?
As to whether there could be any benefit from this is a point for discussion, I have observed several times on Vuillaumes and others that a very flat arching around the bass F hole and a bad wolf note coincide. In these cases, maybe some sinkage might even help.

Plenty of makers have copied archings warts and all, so clearly it will work, but I think carved in, is quite different from natural distortion, and I wonder how it might fare over time, with some of this inevitable distortion creeping in too. I think the hard part might be trying to judge what is the right amount to begin with.

You made good points there and I don't deny that fresh wood and shrinkage can distort the arching . But if I look at the degree of distortions I would almost expect cracks which are more or less unavoidable so I am tempted to think that some of the sunken in area must have been there from the beginning.

However, besides the causes of deformations I was more thinking about what we can do with slightly unsymmetrical archings when making new instruments. I was quoting the example of Del Gesu as an example for apparent unsymmetrical archings which seem to have no negative effect on the sound. (Or ?)

I was more or less thinking of two types and maybe one is working better than the other. The first would have the central mountain shifted in direction of the treble side, the other arching would be the reverse. From considerations of pure mechanics one would function different than the other. If this is audible or not is another question but I think it should. 

with sound post on one side and bass bar at the other side the structure of the violin is not symmetric from the beginning.i am looking on all those irregularities like 'contaminations in crystal' which  are often cause for a characteristic color, so maybe an unsymmetrical arching can create a characteristic sound color. 

[David Beard]

Regardless of idea or method while making, an arch subjected to a compression load from the apex must contain a good catenary shape within its thickness, or else the force will distort the shape.  

Don't know or if the relates to the force the back experiences from the post?  Seems like that would always tend to distort the shape?

But my understand (potentially insufficient book learning on this point), a well shaped top isn't necessarily going to distort, though it will of course flex.

 

[HoGo]

2 hours ago, David Beard said:

Regardless of idea or method while making, an arch subjected to a compression load from the apex must contain a good catenary shape within its thickness, or else the force will distort the shape.  

Don't know or if the relates to the force the back experiences from the post?  Seems like that would always tend to distort the shape?

But my understand (potentially insufficient book learning on this point), a well shaped top isn't necessarily going to distort, though it will of course flex.

 

Catenary is good for load evenly distributed all along the curve (think of chain where all links weigh the same will form a catenary), I think straight curve would do best for point load. From my experience with arched-top mandolins, the more you go away from straight line between bridge and tailpiece the more distortion the instrument will undertake over time in that wide open (more or less unsupported by braces) area. The further away from straight line you are the more thickness you need there to avoid "folding" of the top at recurve or heavy bulging out behind the bridge sometimes leading to collapse of top.

[Artiom]

I have seen many 100-150 y.o. violins made in the "correct" symmetry era, otherwise symmetrical. The amount of distortion is very small, if ever can be noticed by eye. 100+ years is enough for a plate to be noticeably distorted. This brings to a clear conclusion that the old guys did it with a certain intention right from the start.

[edi malinaric]

6 hours ago, HoGo said:

Catenary is good for load evenly distributed all along the curve (think of chain where all links weigh the same will form a catenary),

Hi HoGo - I think I'll politely disagree with you.

That's me putting a 70 kg point load on a 6 day old, 20mm thick cement catenary arch.

OK - we did add another 20mm layer later that day  - and I'll admit that I held my breath.

That was 54 years ago and it's still standing.

I wonder if a tailpiece, bridge and ...?

cheers edi

 

[deans]

9 minutes ago, edi malinaric said:

Hi HoGo - I think I'll politely disagree with you.

 

That's me putting a 70 kg point load on a 6 day old, 20mm thick cement catenary arch.

OK - we did add another 20mm layer later that day  - and I'll admit that I held my breath.

That was 54 years ago and it's still standing.

I wonder if a tailpiece, bridge and ...?

cheers edi

 

Edited

[Marty Kasprzyk]

1 hour ago, deans said:

Edited

The question isn't whether or not a cantenary curve can support a point load.  The question could be: What is the best shape for supporting a given point load to minimize deflection.

But a shape that doesn't  move much statically won't move much dynamically either (static is the same as dynamic at zero frequency) so it won't produce much sound.

So there is a trade-off between life and sound output: weak body & strong sound or strong body & weak sound and various combinations in between chosen for your own priorities.

Since different people have different priorities it then follows that there is no single best shape.

 

[David Beard]

I think perhaps the function of the classical top arch shape and roughly diaphragm thicknessing is that to a short enough wave it looks like an even flexible diaphragm, but to a long enough wave the central arch section looks like a cupped and unified shape (thanks to the arching), sitting on the channel as its buttressing.   But the channeling is a sort of flexible designed cantilevered from the edge, so for a long enough wave we have a relatively unified/stabilized/undistorting large shape sitting on a bouncy unstable platform.  So that allows the dynamic element.

 

[Andreas Preuss]

59 minutes ago, David Beard said:

I think perhaps the function but to a long enough wave the central arch section looks like a cupped and unified shape (thanks to the arching), sitting on the channel as its buttressing.   But the channeling is a sort of flexible designed cantilevered from the edge, so for a long enough wave we have a relatively unified/stabilized/undistorting large shape sitting on a bouncy unstable platform.  So that allows the dynamic element.

 

Sorry, David, this time that's  a bit beyond my comprehension in English. Could you make your explanation somehow easier? 

Where are the waves coming from? Isn't it rather that the frequency of the strings is moving the neck in the first place and this is transformed into flexing waves on the body of the entire instrument. Low frequencies generate bigger patterns higher frequencies smaller patterns. 

.hmmmm.......  

Maybe if we try to understand from this side it's just getting too complicated. 

What would be better? Have the bass bar side of the arching rounder (stiffer) or the treble side?

I'd say to give the bass bar an easier movement the arch there can be slightly flatter than on the other side. And this is the reason for me to think that the deformed DG arching acoustically still works.  (Regardless the cause)

 

 

[Andreas Preuss]

16 hours ago, edi malinaric said:

Hi HoGo - I think I'll politely disagree with you.

That's me putting a 70 kg point load on a 6 day old, 20mm thick cement catenary arch.

OK - we did add another 20mm layer later that day  - and I'll admit that I held my breath.

That was 54 years ago and it's still standing.

I wonder if a tailpiece, bridge and ...?

cheers edi

 

The difference to a violin is that there are no f holes on your concrete arch. (Cool picture though!)

Ironically in the spot where the top would needs most support we cut the catenary arch thrpugh with the f holes. So the stability comes rather from the uncut grain/ fiber in the center connected in both ends to the blocks.

Therefore I don't know if is catenary arch so important. 

 

[Andreas Preuss]

22 hours ago, Artiom said:

I have seen many 100-150 y.o. violins made in the "correct" symmetry era, otherwise symmetrical. The amount of distortion is very small, if ever can be noticed by eye. 100+ years is enough for a plate to be noticeably distorted. This brings to a clear conclusion that the old guys did it with a certain intention right from the start.

In principle I think so too. The question is how does it influence the sound?

[edi malinaric]

4 hours ago, Andreas Preuss said:

The difference to a violin is that there are no f holes on your concrete arch. (Cool picture though!)

- snip -

Dammit Andreas! Why didn't you say something earlier - f-holes would have looked so much nicer. :-)

There is a matching window on the other side - in the same arch.

Note the distributed load on the front arch.

cheers edi

 

[HoGo]

17 hours ago, edi malinaric said:

Hi HoGo - I think I'll politely disagree with you.

That's me putting a 70 kg point load on a 6 day old, 20mm thick cement catenary arch.

OK - we did add another 20mm layer later that day  - and I'll admit that I held my breath.

That was 54 years ago and it's still standing.

I wonder if a tailpiece, bridge and ...?

cheers edi

 

All in due respect, but from what little I learned about arches I'm pretty certain catenary isn't the best one to support point load.

Here is quote from Wikipedia (I know internet is not best source but I believe this one is correct):

"What makes the catenary arch important is its ability to withstand weight.[4][5] For an arch of uniform density and thickness, supporting only its own weight, the catenary is the ideal curve.[6]"

So for your concrete "tent" tha catenary would be ideal because the weight is distributed uniformly across the arch.

If you think of point load you can think of single weight hanging on a string, it will create two straight lines (almost straigth as gravity will make the lines a bit bent as well) pointing right to the point where the load is hanging. The gravity is negligible if the point load is large enough so straigth lines are optimal. It's just the same principle as used in catenary arch vs. hanging chain.

In stone architecture the ideal arch will be closer to catenary because the heavy weight of the arch material itself and the parts above that arch are more or less uniformly spread across the arch.

[David Beard]

2 hours ago, Andreas Preuss said:

Sorry, David, this time that's  a bit beyond my comprehension in English. Could you make your explanation somehow easier? 

Where are the waves coming from? Isn't it rather that the frequency of the strings is moving the neck in the first place and this is transformed into flexing waves on the body of the entire instrument. Low frequencies generate bigger patterns higher frequencies smaller patterns. 

.hmmmm.......  

Maybe if we try to understand from this side it's just getting too complicated. 

What would be better? Have the bass bar side of the arching rounder (stiffer) or the treble side?

I'd say to give the bass bar an easier movement the arch there can be slightly flatter than on the other side. And this is the reason for me to think that the deformed DG arching acoustically still works.  (Regardless the cause)

 

 

Hi.  Did you mean bridge?  I don't think much happens from the neck.

I don't know how to say these things well.

The string is vibrating at a particular frequency, and causing the wood flex. The wood is forced into moving by the vibrations of the string.  

The string vibrations set up standing waves in the body of the instrument.  These are the patterns you mention. Smaller for higher frequencies.  Larger for lower frequencies.

I don't know why, but I tend to think about the indivudual presure pulses of the vibrations, and how these pulses travel into the instrument and set up the standing.

Even though the end result is standing wave patterns, each pulse of pressure from the string and bridge creates a deflection of the top plate's wood.  The deflection at the bridge foot from a single pulse of the driving vibration quickly restores. But like a disruption of the surface of a pond it then travels through the wood. 

This deflection traveling through the wood has a speed.  And the cycles of pulses from the driving string vibration will create cycles wood deflection at the same frequency.  These cycles of deflection travel through the wood, and will therefore have a physical wave length in the wood.  And this is what leads to the various patterns of standing waves that can be seen in the motion studies people have done.

To me, it seems valuable to consider these details.    I feel it can contribute insight into how parts function in a violin, and how this depends on frquency.

In the motion studies you can see the physical size of the standing waves that develop at different frquencies.

You can see that with higher frquencies many small patches of motion develop. This is what meant by 'for short wave lengths the plate looks like a diaphragm'.

In contrast, you can see in motion studies that with lower frequencies (which also means long wave lengths) the standing waves form as only a few large portions of the plate. Thus the motion is in larger more unified parts.

So what does this have to do with arching?

For one, it suggests that for higher frequencies the arching like doesn't much matter. The wave lengths are short and the patches of the standing wave patterns are small. For these freqencies the plate might as well be a flat sound board.

But we can also see that for low frequencies we want relatively large portions of the plate to move in a more unified wsy.

I think the arching shape contributes to this. The arching in effect creates a cupped shape. And cupped are a way to stiffen an object by its geometry. This cupped shape, and the bass bar, both stiffen the plate.  This means it will take higher frequencies for the standing waves to break up the plate small patches than was shaped as just a flar soundboard.

Flipping that around, the arching allows the plate to move in larger more unified standing waves for higher frequencies than would be possible without the special shaping.

 

 

 

Sorry. I fear this actually just longer and more confusing.