The 'Secret' of the Strad: it's in Scientific American 1981

March 3, 2012 at 4:20 PM

This article is called the "Acoustics of Violin Plates," by the late Ms. Carleen Maley Hutchins. The findings: "Modern tests of the vibrational properties of the unassembled top and back plates of a violin reveal something of what violinmakers do by "feel" and lead to the making of consistently good violins" (page 170). The article is very specific on how the top and bottom plates of the violin are "tuned" and it goes as exact as having aluminum flakes spread over the plates and vibrated so they make specific patterns that I suppose represent the frequencies at a very high caliber of accuracy:

"We have found that the sound varies in good violins between C# 3 [the 3 indicates the octave] and D 3 for the belly, and for the back between D 3 and D # 3, so that there is always a difference between them of a half or a whole tone1."

From David Burgess
Posted on March 3, 2012 at 9:43 PM

Carleen was quite convinced and passionate about her methods. Unfortunately, others were much less impressed by the goodness of her violins than she was. ;-)

However, she helped stimulate interest in the scientific investigation of violin sound, which continues today, but with more sophisticated methods, better engineers, and better data acquired from the old instruments.

From Marc Cicchetto
Posted on March 3, 2012 at 9:36 PM

It has been my experience that Ernet Chaladni's use of vibrating particles on a plate, violin or other, will indeed show you a pattern of active and dead zones for a given frequency. However this is of little use in tuning violin plates while they are being made.
Superior tuning comes by integrating an initial frequency (of the first part made)...say the neck...into each additonal part of the instrument as they are constructed; such that the next part made is tuned in harmony with the preceding one; right through to tuning the cavity of the instrument at constructions end. The first tuned part will dictate which frequencies should follow...such as in a chord. An instrument's wood can have either three or four tunable areas during construction depending on how many assemblies the Luthier's construction methods allow.
I have great respect for the findings of Carleen, for it was working through her suggestions of tweaking the sound of an instrument by shave adjusting parts of the peg box area that convinced me to pursue a method of construction whose first tunings would be the neck assembly itself...effectively eliminating a need to tweak the wood everafter.
As to the best frequency for tuning plates? I would say this is a variable depending more upon the wood used than assuming any standard or testings of other instruments. I've found one wood sounding best tuned to C# and another tuned in the F# area or to a D or to an E, etc., etc. Woods have personalities that must be taken into account.

From John Soloninka
Posted on March 4, 2012 at 12:52 AM

Carleen Hutchins needs to be acknowledged for the contributions she made to the "science" of understanding violins. However, later research on "tap tones" revealed them to be less prescriptive of violin quality that that article suggested.

I am curious, Anna, as to why you reached back to 1981 for an unambiguous article from Carleen, rather than the mountain of more recent evidence that shows that taps tones can be interesting, but not prescriptive...and certainly not consistent in fine old instruments.

One of the best articles was by Martin Schleske, one of the most sophisticated makers (in terms of mixing acoustics with luthier craft). In the Catgut Acoustical Society Journal, he presented his taps tones for front and back of one of his instruments across 15 work steps, from 5MM uniform thickness, to properly thicknessed and graduated plates. It showed that the tap tones are not only quite naturally in the range of the numbers that Carleen suggests, but that they are relatively independent of the work step...ie that they more reflect the shape of a violin, rather than some optimal tuning of a violin.

Other research and modelling by Oliver Rogers showed that you can get a set of plates to specific tap tones...and then alter the tap tones dramatically by removing wood in sensitive edge areas. While edge thickness of plates affects free plate tap tones, in practice, the assembled corpus of the violin clamps the edges down, and their influence (which was large for the free plate), is diminished in the assembled instrument. So the "free plate modes" do in some indirect way reflect the stiffness and uniformity of the mass/stiffness distribution of the plate (which is important for good frequency response) but unfortunately the magic relationships of top and back mode frequencies are not prescriptive.

Funnily enough, many makers spend a great deal of time optimizing the shape of those vibrational modes...because it is something they can do, and is satisfying. The fact that it has little impact is of little consequence. Reminds me of the joke: A fellow is on his knees searching the ground under a streetlamp on a dark street. A passerby says "What are you doing?". He responds: "I am looking for my contact lens." The passerby says "where did you drop it?" The man responded "way over there" point to his parked car. The passerby says " so why are you searching here?!" the man answers..."because this is where the light is!"

From Robert Spear
Posted on March 4, 2012 at 2:04 AM

My God! Where do I start? I love and respect David Burgess, I really do. He's an asset to the craft to a far greater degree than he'd probably care to admit. But David, Carleen only "stimulated" interest in violin research? Man, she drove it from the start. That woman was like 100 tons of locomotive on rails-- a force to be reckoned with! No Carleen, no Catgut Acoustical Society, no Oberlin Acoustics Workshops, no *nuthin'!* :-)

Marc-- Chladni patterns of little use in tuning a violin plate? As a point of personal opinion, I happily defend your right to express it. As a scientifically proved (or disproved) fact, be careful! The violin world is awash in schemes that seek to tune the parts of the instrument to specific tones or even chords. It seems like something intuitively correct, but if plate tuning as Carleen taught it fails because of what happens when the parts are assembled, then pretty much all the other approaches fail as well.

John-- matching the modes of free plates is indeed not prescriptive, but it is indicative. Perhaps a little elbow in the ribs to makers who graduate plates because it's really all they can do and it makes them feel good is tolerable, but just consider what happens to a violin when the plates are not tuned! Reminds me of the little joke about the airplane passenger who hopes that the engines don't fail or he'll be stuck up in the sky all day!

This entry has been archived and is no longer accepting comments.

The 'Secret' of the Strad: it's in Scientific American 1981

Laurie's Books

Subscribe