William Fry Internal Scraping Method

[Quadibloc]

3 minutes ago, Marty Kasprzyk said:

What does the red line show?  The only German I can remember is ein grosse bier

In the diagram, the line showing how much actual signal strength is required for a subjective loudness of 60 dB is made red so that one can easily see that the orange arrow and the blue arrow both reach up to the same line - thus illustrating that a 20 Hz sound with a physical energy of 100 dB and a 4 kHz sound with a physical energy of 50 dB both have a subjective loudness of 60 dB.

[Marty Kasprzyk]

1 minute ago, Marty Kasprzyk said:

What does the red line show?  The only German I can remember is ein grosse bier

I forgot to say that if you flipped over the red line vertically it would resemble a good violin's frequency response curve with strong output in the signature modes and in the "bridge hill" region.

[Violadamore]

2 hours ago, Marty Kasprzyk said:

What does the red line show?  The only German I can remember is ein grosse bier

Michael is obviously inputting from something that spellchecks on the fly.  For "isoprene" read "isophon".  The graph is explained here: https://en.wikipedia.org/wiki/Equal-loudness_contour

[Michael Szyper]

2 hours ago, Marty Kasprzyk said:

What does the red line show?  The only German I can remember is ein grosse bier

These lines are isophones. Sorry,my auto correction thought i would mean isoprene...!?

so the graphs show the human ear sensitivity to different frequencies at given sound pressure levels.  For example, at 20 dB,  you need a sound pressure of 100 dB to achieve the same loudness impression as 50 dB at 4 kHz. 

[Michael Szyper]

51 minutes ago, Violadamore said:

For "isoprene" read "isophon".  The graph is explained here: https://en.wikipedia.org/wiki/Equal-loudness_contour

Thank you for the reference, it was my auto correction on the iPad. Seems to  do mostly things I don't mean.

[Michael_Molnar]

I was a grad student at U. Wisconsin - Madison where "Jack" Fry was a physics professor. Our paths crossed only in the hallways. Years later my colleagues at UW told me about his violin work and I got the book about his findings. I tried contacting him to no avail but did communicate with the author, Kameshwar Wali. He had little to add to what was in the book. I then went to the UW Library and retrieved a copy of his course notes about violins. There was little in them useful to a maker - just lots of theory from a physicist's point of view.

Anyhow, I did try to test Fry's methods and found little to help me. His graduation map is similar to what other people have suggested. As for his scraping to change a violin's response, I could not repeat it. I did notice some ephemeral changes probably due to wood relaxing from the scraping. I really do not know what to say to those audiences that claim he substantially changed a violin's output. I suspect that some of these violins under test may have had overly thick plates and that any wood reduction would have improved them.

Sadly, he is not here to defend his findings. 

[David Burgess]

3 hours ago, Michael_Molnar said:

I really do not know what to say to those audiences that claim he substantially changed a violin's output.

Nor do I, aside from the observation that beliefs can have major impacts on perceived outcomes.

[Violadamore]

4 hours ago, Michael_Molnar said:

I suspect that some of these violins under test may have had overly thick plates and that any wood reduction would have improved them.

IMHO, good suspicion.  Fiddles of that description aren't difficult to find. 

[Quadibloc]

7 hours ago, Michael_Molnar said:

Sadly, he is not here to defend his findings.

Yes, that is true, but we can't let that prevent us from honestly attempting to assess their merit.

As a naive outsider, naturally I started out with the assumption that violins by Stradivari, Guarneri, et al, were as unsurpassed as it is often claimed. Having read statements by acclaimed concert musicians that new, promising violin talents were not being developed properly because the supply of instruments of this high quality are so limited, I accepted their statement of the problem at face value.

So when I hear about people like Fry - or Nagyvary, or Kreit - my interest is piqued. So is my skepticism, because they seem to be making big claims.

On the other hand, many of the luthiers who post here are convinced, from their own personal experience, that making a fine violin requires good wood, craftsmanship, and hard work - and that these will produce results, without the need of any magic formula to bridge a stubborn gap that would still remain, fully comparable to what the old masters had achieved.

If you aren't selling a magic formula, then you aren't selling snake oil - but if the claim "I can make a violin as good as Stradivari" makes us suspicious of William Fry, Joseph Nagyvary, or Anatoly Leman - then it ought to be understandable that the same will be true of those who aren't selling magic formulas... only violins. Even when it's clear that one is dealing with an able craftsman who is honest - perhaps he could be mistaken about what's really causing all the excitement about Stradivari.

It's clear that modern makers can produce violins that project just as well as anything by Guarneri, and that sound as sweet as anything by Amati, Stainer, or Stradivari. So, if there's anything to the Stradivarius myth, it would have to be related to the side of the violin not experienced by the audience in a blind listening test.

The claim has been made that the violins of Stradivari, Guarneri, and Guadagnini exhibit tonal changes in response to the way they are played to a greater degree than those by modern makers.

If that were true, it would make it worthwhile to find out how to imitate what it was they did right. And if it's not true, it would also be helpful to stop this groundless worry among concert musicians, and connect violin students to the appropriate quality modern violins that will help them in their learning.

And while it may not be as easy for scientists to devise experiments to objectively test and measure this, I don't think it's impossible. Frequency spectra of notes produced on diferent violins, in a test bed that draws a bow across them at different spots, at different speeds, with different pressures, and so on... that can be done.

Of course, in the case of a negative result, it could always be argued the tests weren't sensitive enough to find the magic... but there's always the chance of a positive result that would tell us what remains to be done.

[dpappas]

It’s also possible to demonstrate that method X or product Y changes a violin tone even if nothing is new.  One can intentionally or subconsciously play near the fingerboard, scrape a little, and then play near the bridge.  Success!  A change in tone.  

[David Beard]

51 minutes ago, dpappas said:

It’s also possible to demonstrate that method X or product Y changes a violin tone even if nothing is new.  One can intentionally or subconsciously play near the fingerboard, scrape a little, and then play near the bridge.  Success!  A change in tone.  

Yes. A masterclass produces tone change without altering the instrument.

You simply don't get to evalute the performance of a violin without a player driving it.  Passive tests aren't able to measure all or the same things that matter in actual playing.  

So to evaluate modern violins, acceptance by good players is the thing that is truly meaningful. Even though it remains complicated and imperfect.

By this measure, some makers have gained ground.  But it's not so much the same ones who self proclaim.

 

 

 

[Pate Bliss]

I'm curious if the asymmetries in thickness relative to the sp position, and the thinned area he calls the tongue really exist. 

It's not hard for me to believe that a small amount of wood (couple 10ths of mm) removed from exactly the right places could make a big improvement while much more wood removed from random places might make no difference at all.

[edi malinaric]

On 8/14/2018 at 3:18 PM, Don Noon said:

- snip -

In practice, thinning does move some mode frequencies downscale, and the lower frequencies become stronger.  However, I still stand by my experience that the overall shape of the middle and higher frequencies, and thus the major tonal character, remains recognizably the same, except for really massive thinning (say over 10 grams for a top). 

- snip -

Hi Don - could you comment on any differences you have noted between the violin in the white and after varnishing?

many thanks edi

[Michael_Molnar]

3 hours ago, edi malinaric said:

Hi Don - could you comment on any differences you have noted between the violin in the white and after varnishing?

many thanks edi

I'm not Don, but my experience is that the ground can either enhance or diminish the overall tone. As for varnish, less is best. 

[Don Noon]

9 hours ago, edi malinaric said:

Hi Don - could you comment on any differences you have noted between the violin in the white and after varnishing?

many thanks edi

I think of varnish as tightening up the instrument, which is a good thing if it's wild and loose in the white.  With my current making sequence, I don't have instruments in the white to string up, but earlier instruments showed that varnish decreased the very high frequencies (above 4 kHz), low frequencies moved around just a little, and the middle frequencies varied in no consistent pattern... but overall there were no big changes.

And I'd say less is best too.  Some of the best violins have almost no varnish.

[Quadibloc]

On 8/17/2018 at 12:05 AM, Bill Merkel said:

It's not hard for me to believe that a small amount of wood (couple 10ths of mm) removed from exactly the right places could make a big improvement while much more wood removed from random places might make no difference at all.

It's not hard for me to believe that either. But if people have looked at videos of him demonstrating the improvement in sound by removing material... and heard no difference, then they can certainly be skeptical of him without having to maintain that what he claimed to have done is impossible in principle.

I suspect, too, that removing material from a certain spot might end up making a difference to some notes, and not others.

As far as I know, all we have is tantalizing hints of his method. He did explain here and there some of the things he did, and some of the reasons why, but he did not leave behind a comprehensive guide to his technique.

Given that, even if he knew how to regraduate a Skylark into a Stradivarius - or, at least, nearly so, as the wood wouldn't be as good - we don't know how to do it the way he did, and, therefore, except by assessing some of the individual instruments he modified, we can't tell if he could do it.

Worse yet, he claimed that he had examined multiple instruments by Stradivari, and found complex graduation profiles which, while they differed from one instrument to another, followed a common scheme. Several Stradivarius violins have now undergone computerized tomography scans, and they appear to have very smooth and regular interior surfaces.

One of his claims, though, a thin region on the bass bar side, could have some substance from what I think I've seen, although not consistently in every Strad.

Not having a CT scanner available for his work, he measured the wood thickness of Stradivarius violins non-invasively using an apparatus involving a steel ball bearing, and a magnet which was gradually pulled away from the violin until the bearing was no longer held in place.

This seems like a rather crude device, and so I would not find it implausible that his thickness maps were in error. However, the CT scans that look so smooth also look blurry - because their resolution is limited. So it's not absolutely certain, as far as I know, that the details found by him don't exist.

Edited August 18, 2018 by Quadibloc
corrected typo

[Stephen Perry]

I hesitate to step in.  However, tools for removing tiny bits aren't at all difficult.  Bent sharpened coat hanger.  I have yet to find anything that greatly alters what wood and arch do.  I have managed to get a certain amount of noise out and otherwise gently move response a little bit this way and that.  About as much as adjusting a bridge through a few minor cuts.  

Bad violins still sound like bad violins, however.  And the really responsive and highly pleasing ones I wouldn't touch.  It's the in-between with a few things that could be acting better which appear most worth mulling over.

[David Beard]

The thing that gets me about such tiny microadjustments of the violin itself is that if there's any effect at all, it will be swamped by the next changes in setup.

Setup is the right place to do the finest grain adjusting.  

[Stephen Perry]

David, the pattern I use for mandolins circles around and around.  The basic work stays put, with minor shifts.  Years later, still mostly there.  Same for violins.  I likely have a much different conceptualization and lower expectation of impacts!   I started out trying to debunk "Ears of the Angels" and found much that persistently "improved" things.  What I hear is a reduction in noise/more clean sound.  Maybe a bit more consistent and snappier response.  The main use for me is in improving balance and getting the high partials to come out and sparkle.  Really does a great job on mandolins.  Violins prove devilishly complex, but there are some consistently effective things I use.  I also see the same things showing up in some very nice instruments I've examined, so I'm not alone.  No miracles in turning a 2x4 into a magic wand.

I tend to think of setup and tweaking like working on cars.  Go through and blueprint, make everything where it should be, then ponder a bit and start working on all the interconnected bits.  Can't cut the head and bump the compression without thinking about timing and fuel metering, and then the manifolds, and back pressure, etc.  

[David Beard]

On 8/16/2018 at 11:05 PM, Bill Merkel said:

I'm curious if the asymmetries in thickness relative to the sp position, and the thinned area he calls the tongue really exist. 

It's not hard for me to believe that a small amount of wood (couple 10ths of mm) removed from exactly the right places could make a big improvement while much more wood removed from random places might make no difference at all.

You can find CT thickness scans on the web a look yourself.

My take away from studying those is that most of the assymetry is random drift.  To me this means there was more priority on smoothing then on precise thickness beyond about .2mm.

But further, if you look statistically across many plates, you can see some directed choice dispite the random drift.

This directed choice includes:

* leaving wood/thickness outside the soundholes 

* a little thinning/relief along or near the major bout width

* a tendency per maker to make a choice of thickness between the upper soundhole eyes.  Some makers choosing thicker. Some choosing thinner (Strad).

Around the post, a very large number of the old instruments have a patch. And tend to be a bit thicker here.

 

These are the 'patterns' I've noted.  And they are on a scale of about .15 to .3mm.  So on any one instrument these patterned choices aren't very conspicuous amid the pervasive random drift on a scale of about .2mm.

[Stephen Perry]

For another fun discussion, Bud Purvine's ENABL and similar treatments for audio drivers and enclosures are just as much fun.  Tragically, the system works nicely, but requires (as always), quite educated ears and clean input.  My former brother in law and I worked up a number of speakers, a couple of which were spooky good, such that digital music sounded digital, and the background voices in live recordings could sometimes be made out.  Typically, the amount of work per unit gain proved insane.

Back to the original question, re Fry graduation.  I did a couple of white violins according to the pattern provided in the Wali book.  They worked well.  So did every other systematic pattern.  The Fry system gave what sounded to me more of a fiddle tone, but it wasn't a big deal magic thing, and I can't for a moment pretend it has anything to do with 18th C working methods.

I suspect that tying numbers to specific positions isn't what counts.  It's manipulating the plates to do what one wants on a large and small scale.  That's bound to vary from instrument to instrument, and perhaps there's some consistency when it's hung on an appropriate map, which would take into account everything.  Not going to hold my breath on that research.

[Oded Kishony]

I would like to add my two pennies to this discussion.

Scraping or removing wood from the surface of the instrument in some random way, even if it is an 'educated' guess, and hoping for a positive outcome is like putting your elbows on a piano keyboard and expecting to hear a Bach fugue-not likely.

The timbre of an instrument or a string is dependent on the harmonic structure of the sound, the brain re-arranges the sound to produce the timbre. A well-documented example of this is that the violin produces a very weak signal at 196 Hz (open G string) the listener assembles the harmonics which produce the effect of 'open G' ie 196Hz. If some component of the harmonic structure is missing or is weak the character of the sound (timbre) will change. If, for instance the higher harmonics are missing the sound may seem to be dull. If on the other hand,  lower harmonics are weak but higher ones have a greater amplitude the sound may seem to be harsh or too bright.

Even though we talk about discreet modes, the modes (areas of vibrations) don't function separately under playing conditions. When an instrument is played, any frequency generated by the strings or bow that 'finds'  a friendly spot on the surface of the instrument will be amplified. However, most of the vibrations on the surface do not get projected into the far field (apx 3ft) because of 'phase cancelation', which works the same way as sound canceling headphones (see google for details).

When any given area of the violin is thinned it will affect a great many frequencies at the same time, so selecting where to thin is tricky and in my experience often requires some compromises. For instance, if I have a too bright G string I may tap the surface until I hear a clear G fundamental vibrating, then while scraping that area I listen for changes because as you scrape, the tonal landscape is changing under your scraper. I often find myself 'chasing' a note or harmonic around the surface. In addition, you should keep in mind that on a violin, all frequencies below ~ 1kHz are produced by the whole corpus, which is why scraping the ribs can produce dramatic changes in the sound (see caveat above re psychoacoustics).

If you are computer fluent you can explore this by sampling the surface of an instrument with a simple computer mic or a contact mic. just tap the surface of the instrument with the microphone while running an FFT spectrum analyzer on 'continuous' with 'peak hold'. You will see the frequency spectra are completely unique for any spot on the instrument. (dampen the strings)

What I find exciting is this can be done without using a computer, because the information embedded on the surface is transferred to the strings, when the surface is tapped or being scraped. You can hear when you're thinning an area that affects the harmonic of the string because the string starts to vibrate at that harmonic frequency.

 

Oded Kishony

PS entire sections of material that I've written seem to erase automatically when I post here. Am I doing something wrong or is there a glitch in this program? Never had this issue before on Maestronet. 

 

 

[Oded Kishony]

The deleted section was a discussion of the effect of aging on the instrument, the possibility that playing an instrument over a long time period enhances and increases the amplitude of the harmonics.

OK

[Ken_N]

I've read that before.  It seems possible that bending and flexing over the years in certain areas would do that.  

I've done tuning for overall evenness.  Don't know if that does anything.  But I did do some massive cutting away over the end blocks of a belly to get a violin to play easier.  It needed to be cut up quite a bit more on the inside. It helped it tremendously, and I learned a lesson.  Too stiff is too stiff.  The violin is very clear and easy to play.

Ken

PS I have problems with writing programs at times too.  The thing bothering me lately is my little finger has some nerve damage for a cut, and it wants to push the caPS LOCK and not the A.

[David Burgess]

8 hours ago, Ken_N said:

PS I have problems with writing programs at times too.  The thing bothering me lately is my little finger has some nerve damage for a cut, and it wants to push the caPS LOCK and not the A.

LOL, I too have a "rogue" finger, due to nerve damage. Mostly, it makes minor typos, but occasionally, it will delete entire pages or posts, in some way that doesn't allow me to recover them. I haven't figured out where where Mr. Badfinger is going to do that yet, since I don't look at the keyboard when I type.

Wah.