Sign in to follow this  
tango

An even or uneven top thickness

Recommended Posts

Transparent composites of various thickness maps from the web.

 

Composite of thickness data on Del Gesu and Strad violin tops:

 

attachicon.gifDel Gesu tops compositr map.jpg   attachicon.gifStrad tops composite maps .jpg

 

 

These images essentially show maps of average top thicknesses across multiple instruments from each maker.    Within a tolerance that is rather gross by many modern maker's standards, the thicknesses appear to drift randomly.  To me, the 'drift' quality suggests more attention to smooth continuity of thickness than to precise measure of thickness.  Beyond this fine level of essentially random variations, the thicknessing is overwhelming even through the plates, with just the slightest tendency to thin some areas. 

 

The individual maps also show random 'drifting' variation.   The thinned patches tend to be more randomly located, which goes away when you average out the maps.

 

 

So, keeping with the fashions of current making, clearly this evidence proves that high precision 'graduation' of the tops is critical to making great playing instruments!   After all, without all that fuss, how do you know you've done a good job?   Let's keep it micro millimeter precise folks!!!!!!!!!   And don't forget to tap tune everything.  If your measurements are prefect, it's got to be good.

These are averages over data from my Excel database. The source is from Jeff Loens graduation book, with his permission, the Biddulph del Gesu book, various Strad posters and from an anonymous source.

Share this post


Link to post
Share on other sites

Yes. Only in modern times would we think of purposefully introducing such variations in a controlled way.   Like you said, just let them happen naturally.

 

 

I'm not a huge fan of tons of dampening.  Just don't think it always bad.   I don't know if those slight variations provide much dampening, but I'm guessing they provide some in comparison to perfectly clean evenness.   Also, not sure if dampening is actually the right word.   If you make a vibrating string slog through some resistance from a viscous fluid, that really is dampening.   But if you randomly attach some very light but non uniform masses along the string, is that dampening?

 

I'm calling both things dampening, probably incorrectly.  The one is plastically absorbing and wasting energy.  The other is diverting energy, but in a returnable form.

 

I think the word is visco-elastic.   If the motion lags behind the exciting force,  that means that the material needs time to respond to the force.  A prefectly elastic material responds instantaneously by Hooke's law.   A tire gets hot from visco-elastic losses,  and one can call the effect hysteresis.  If you look that up,  you will see that a cycled material makes sort of the shape of an f-hole (less the eyes !) and the area inside represents the energy loss per cycle.......... if the graph parameters are force and position.

 

I am quite interested in this.........  For example,  why does a violin not sound good at first,  and then settles down after a few days.  Perhaps more change (due to creep) in longer times.   The creep shows that Hooke's law is not obeyed.  I am thinking of ways that hysteresis can be involved in the vibrations of a well-settled violin.   I suspect it has to do with certain arching shapes ....  Michael Darnton has high faith in the importance of arching,  and I have a feeling as well.  But the math and physics is unmanageable.  I plan to do some FEA experiments.

Share this post


Link to post
Share on other sites

I think many of us have come around to believing that arching is key. I increased my arching 7% and like the result. Hopefully someone at the VSA will agree.

 

As for physics and math, I have said we are dealing with a bunch of non-linear coupled equations. That kind of system has many satisfactory solutions. Just try to find them. LOL

Share this post


Link to post
Share on other sites

Luiz Bellini told me that he adjusted the arch according to the wood properties. I mentioned this in a somewhat recent post. In typical MN fashion it precipitated an interesting debate E.g. how much do you adjust under what conditions. In a simplistic view, a higher arch should add stiffness. The shape of the arch both crosswise and longitudinally also come into effect. Edge channeling or "scoop" is another factor I would think.

Share this post


Link to post
Share on other sites

Luiz Bellini told me that he adjusted the arch according to the wood properties. I mentioned this in a somewhat recent post. In typical MN fashion it precipitated an interesting debate E.g. how much do you adjust under what conditions. In a simplistic view, a higher arch should add stiffness. The shape of the arch both crosswise and longitudinally also come into effect. Edge channeling or "scoop" is another factor I would think.

That was here.

Share this post


Link to post
Share on other sites

I do not bother too much to achieve a uniform thickness in the top.

I start with a uniform thickness of 3.5, go down by steps to 3.2 to see what happens to tap tones and flexibility trying to understand the evolution, then thin out the top and bottom bouts, than thin out the center respecting F holes and soundpost area, and so on little by little until I am satisfied.

When I reach a point where bending stiffness and tap tones frequency is almost ok for my standard and feel, I try to take away as much wood as possible reducing weight without (almost) affecting stiffness and frequency, stopping at some point where I feel that remove additional wood would be detrimental.

I seldom reach a uniform thickness, and I don't think that this is a fundamental parameter, although I do not rule out doing this if the wood requires it.

Often I reached a minimum thickness of 2.5 mm which increases up to 3.2 or even 3.5 mm in the thicker areas such as C bout side of F holes and soundpost area, usually I reach the maximum thickness near top and bottom blocks always around 4 or 4.2 mm, and this is the real max thickness of my top plates (not even my edges are so thick, excluding tips of corners).

I do not know if that makes a lot of sense, because in reality it is not possible to speak of thicknesses without considering all the other parameters (wood properties, weight, arching, model and so on) and I do not take the measures that I have mentioned as a standard, but every time I try to figure out when to stop, hoping to make decisions for the better based on feelings, and hoping that the experience does not betray me :)

Some numbers, calculations and scientific figures help me in this, but often you can also be fooled by them.....

 

Davide

 

PS Probably for  a back with even thickness or more similar to a top, you may need to put braces on it like guitars....maybe an X-brace may work.... B)

 

When I reach a point where bending stiffness and tap tones frequency is almost ok for my standard and feel, I try to take away as much wood as possible reducing weight without (almost) affecting stiffness and frequency, stopping at some point where I feel that remove additional wood would be detrimental.

Davide

Davide

Where do you remove wood to do this?

Share this post


Link to post
Share on other sites

Davide

Where do you remove wood to do this?

 

I looked for an answer to this question for several years, but unfortunately I have to admit that I never found it.

I am convinced that it is not possible to generalize, and it comes to trying to understand the behavior of each plate while gets thinner gradually.

You should try to work out a strategy observing what are the areas that are most sensitive to the lowering of frequency, respecting them as much as possible.

Share this post


Link to post
Share on other sites

Here is a guide to thinning areas of the top to control m5 and m2 posted by Marty K.   There also are some similar guides at platetuning.org but it does not show what exactly happens to m2 and m5 (whether they go up or down).  Also there are some discrepancies between two so you should do your own experimenting.  

post-24376-0-51009000-1474474678_thumb.jpg

post-24376-0-24032400-1474474701_thumb.jpg

Share this post


Link to post
Share on other sites

Here is another approach, thinning evenly so the frequencies and weight drops. Note frequencies drop when f-holes are cut out and up again when bass bar is trimmed. Thinner for denser wood and thicker for less dense +/- 0,2-0,3 mm. Same weight and frequencies.

 

post-37356-0-95552600-1474477809_thumb.jpg

 

 

Share this post


Link to post
Share on other sites

Here is a guide to thinning areas of the top to control m5 and m2 posted by Marty K.   There also are some similar guides at platetuning.org but it does not show what exactly happens to m2 and m5 (whether they go up or down).  Also there are some discrepancies between two so you should do your own experimenting.  

 

There are a lot of information on this subject around the web and in various articles, to get an idea of sensitive areas to thinning.

What is interesting is that there are always discrepancies between one and the other, though none is a priori wrong, to demonstrate the fickleness and the impossibility to assign exact zones with certainty because it depends on too many factors involved.

However, knowing as much as possible of all these methods is certainly helpful to form an  approximate idea of the general behavior.

One of the first article that have helped me to shed some light on this topic was that of Oliver E. Rodgers based on finite element analisys (Aids for tuning violin components - VSA Journal XIII N.2 1993)  that, with the detailed maps, gave quite precise directions on where to to remove wood and about the effect.

The problem is that if one gives too much credit to these things without taking into account an overall view, is likely to end up with some very weird thicknesses that are never seen on good old instruments.

Apply common sense is always imperative ;)

Share this post


Link to post
Share on other sites

Hi Melvin,

 

and once varnished and removed what is the criteria that you use to graduate your plates?

In this way you only eliminate the variable of the varnish, but you do not know yet what will happen with the string tension, the deformations and the sound quality.

I do not think that can be very different from the more usual system without varnish on the plates.

 Hi David

 

I'm with Davide... the plates are too thick initially to string the varnished fiddle up to get a sense of where it is, so this method is essentially tuning the free plates, but with varnish on them.

 

If I was going to go to the trouble of disassembling after varnishing, I would go just slightly over what I thought would be the finished thickness, assemble and play it, and then decide how much to thin things out.

 

However, I don't discount the idea that free plates can be graduated to a final dimension (with or without varnish, your choice) and still come out close enough so that regraduation is usually not necessary.  And regraduation isn't that big of a deal, anyway, especially if you're doing antiqued instruments.

I will take apart and put together as much as required...As makers we try to hit gold in one go...in reality it comes with much less naivety...you must work like a restorer

Share this post


Link to post
Share on other sites

I leave front and back about 5mm thick until all varnishing is done and cured. I then remove them and  gradate them to their desired criteria and re fit them

Hi Melvin

Is the reason you leave the plates at 5mm is to help stop distortion with the plates while carrying out the ground and varnish coats

Share this post


Link to post
Share on other sites

Averages are only averages and give only a vague indication.

They are valid only if you are able to make an average with the properties of wood you are using, and this is difficult to achieve.

But surely may suggest some directions.

 

 

These are averages over data from my Excel database. The source is from Jeff Loens graduation book, with his permission, the Biddulph del Gesu book, various Strad posters and from an anonymous source.

 

Yes.  The numbers are from the averages you posted from your database.

 

 

The color maps shown are composited from  thicknesses maps in the CT Study published in Borman's 2005 Strad magazine article.

 

 

Thickness maps for 8 Del Gesu violins:

 

post-30802-0-30386500-1475090780_thumb.jpg

 

post-30802-0-17436100-1475090808_thumb.jpg

 

 

And here is the color key for the mapped thicknesses:

 

post-30802-0-99407800-1475090121.jpg

 

A similar range of CT maps was published from Stradivarius instruments in the Strad in 2002.   I wasn't able to find the individual Stradivari image maps currently posted on the web,  Though they were a few years back when I made the composited maps.

 

 

To me, there are a number of interesting points revealed in this data:

 

1) The tops and backs are mostly diaphragms, with the center mass of the backs interrupting this.

2) The general diaphragm thickness in the backs is not greatly thicker than the tops

3) There is very significant and apparently random patchy variation in thickness

4) This patchy variation mostly goes away when averaged across multiple instruments, suggesting random variation rather than purposeful.

 

 

 

 

To me, these show a very firm plan of basically diaphragm plates, somewhat thinner for the top, with an extra area of central tapered mass in the back.

 

But that's the whole of the plan that is consistently adhered to.  Beyond that, the size, shape, and extant of the central back mass is freely varied, the overall execution for both DG and Strad is loose, and the extra thinness of the tops is very inconsistent.

Share this post


Link to post
Share on other sites

These maps of the thickness of Strad and GdG published in the two cited articles are very interesting and instructive, and I think it is more useful to see them one by one and draw their own conclusions, rather than starting from a media that makes you lose the value of the individual.

That's why I do not like very much averages, even though in some sense in the end everyone does their own averages.

I have to agree with your conclusion, as I said previously surely may suggest some directions....

Share this post


Link to post
Share on other sites

Hi Melvin

Is the reason you leave the plates at 5mm is to help stop distortion with the plates while carrying out the ground and varnish coats

I love that I'm not the only one. I definitely take the upper and lower bouts down to 3 or so, and often leave the center of the back as thick as 7 to start... I cut the ffs with the top at 5 and rescrape my final shape, put the bar in, repeat, varnish, disassemble, and final thickness a time or two. Treating the glue ng surface of the spruce with a thinned out shellac before this whole circus is a must,,,

Share this post


Link to post
Share on other sites

These maps of the thickness of Strad and GdG published in the two cited articles are very interesting and instructive, and I think it is more useful to see them one by one and draw their own conclusions, rather than starting from a media that makes you lose the value of the individual.

That's why I do not like very much averages, even though in some sense in the end everyone does their own averages.

I have to agree with your conclusion, as I said previously surely may suggest some directions....

I agree that the variations are the important thing rather than the averages.  It reminds me of the fellow who drowned in a lake that had an average depth of only 6 inches.

Share this post


Link to post
Share on other sites

One of the lessons I learned from making htese average charts with natural variation given, is that the inner and outer f-hole wing areas almost always are thicker than the central part of the top.

 

An other puzzling fact there is that there in general are more variation of the graduations in the back plates. Maybe the variation is just higher because the graduations are just thicker.

 

I think the average graduations are better to use than the exact copies of individual instruments because of the natural varaition in wood properties, runout etc. I think there will be more variation in wood properties than the arch and graduations. But that is an assumption.

 

I have mainly made these charts as a reference for comparisons to different Hardanger fiddle maker patterns. But I think that using them in mass production of violins may give better sounding instruments in general, but gives a somewhat higher risk for failures as the tops are pretty thin.

 

Personally i think the Hardanger fiddle patterns are more interesting. We have a lot more variation both in building techiques and use of the instruments. In general I think we get more experience from the variation than a regular violin maker does. At least if they stick to a low number of models.

Share this post


Link to post
Share on other sites
On 9/16/2016 at 5:57 AM, Davide Sora said:

 

I'm talking about M5, M2, M1 and so on, that I take tapping with fingers and listening.

Years ago I tried to use the system with speaker and glitters (sand) but I have abandoned it because did not seem too meaningful to me.

The only usefulness I've found is to figure out which mode is associated with the frequency you are listening, because I had some problems to recognize cello modes due to plates size and very low frequencies (sometimes I confused the M3 for M2 in cellos.....not to speak about M1 without bassbar..... :rolleyes: ).

But for violin plates it's easier not to be fooled and I think that mode shape are not so important to control in a specific way.

Iv heard of the sand method. How do you do this without the sand just vibrating right off the surface given all the curvature?

Share this post


Link to post
Share on other sites
17 hours ago, shocker828 said:

Iv heard of the sand method. How do you do this without the sand just vibrating right off the surface given all the curvature?

I don't quite understand what you mean : do you mean how not to let the sand slip away? or how do I understand the mode shape without using sand?

Share this post


Link to post
Share on other sites
18 hours ago, shocker828 said:

v heard of the sand method. How do you do this without the sand just vibrating right off the surface given all the curvature?

The plates being tested are removed from the rest of the violin and placed upside down with the sand placed on the

inside so if anything it tends to accumulate in the center.

Share this post


Link to post
Share on other sites

I was just thinking about top thickness. 

Here I have a viola, with the edge margins penciled in.

The top was around 2.8 mm, with 3-3.2 in ffs and post. 

Now, I've been kind of struggling with what to do in the edge margins area, as there seems to be 3 major schemes out there. 

One scheme is to leave that area as thick as the connecting bout, so if the bout is 2.6, then the edge is 2.6. 

Another us to thin it out so that it's ~.2mm thinner. I suppose to try and emulate a speaker cone and make a freer plate to vibrate. 

Another one it to kind of just let it naturally blend into the gluing platform, which sometimes results in a thicker edge margins than the bouts. 

I've been learning the method of thinning the edges relative to the bout. But I'm not entirely convinced that it has any consistent effect on sound. There may be some kind of phenomenon that I can't pick out every time, due to other variables making the instrument have it's own voice. 

But I've still been battling internally on what's best to do here. But I suppose that there is no clear answer, as every piece of wood is unique to some extent. 

IMG_20200428_135712.jpg

Share this post


Link to post
Share on other sites

 When dealing with the thickness of the material, that parameter is actually immaterial.

What is important, is weight and stiffness, think about how you want it to work.

Many different graduation patterns work well. I can not tell you what I do because it is always a bit different according to arches, wood, phase of moon, etc,,

There are some fantastic fiddles in the world with just those numbers you cite, You need the experience, with that wood, at some point there is no real answer,

It's you and the fiddle.

Unless you stopped by the house then things would be different.

For now,, try it there,,relax and have fun,,,still learning.

Share this post


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...
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.