Sensitivity of upper treble F-hole wing

[Don Noon]

Almost 15 years ago(!) I looked into what the upper treble F-hole wing might do acoustically, and generally found that the frequency range of highest wing amplitude corresponded with a zone of LOWER acoustic output from the violin.  The original MN thread is here.

This idea has been cooking in my head all this time, and I thought I'd look into it a little more carefully.  The basic test is pretty simple:  put a sequence of very small masses on the wing and see what happens to the impact spectrum.  In the chart below, I plot the DIFFERENCE from the baseline (no mass) condition, in dB vs frequency.  The masses (small dots of clay) are extremely small, and I had to buy new milligram scale to weigh them accurately.

The take-away from this is that the upper treble F-hole wing has a role in shaping the response in the low 1 kHz range.  This makes some sense, in that the wing is closely coupled to the treble bridge foot, and appears to behave somewhat as a non-radiating resonator* to absorb energy that would otherwise go into the body.

As far as practical applications, this might not be a big deal... the design of the F-hole wing has some of this effect, and is probably why the soundhole IS this F-shape.  But different designs of F-holes might behave a bit differently, and I suppose that if a given instrument has an annoying peaky resonance around this range, a little added mass might be a benefit.  But it's just in a narrrow frequency range, so I wouldn't expect a massive transformation.

* The proper term is actually a "tuned mass damper" (TMD)  or "tuned vibration absorber" (TVA), and if interested, these can be looked up online.

[LCF]

You'd think that a piece of wood flapping to and fro like that might radiate some sound so I maybe it's not so much a damping effect as an anti- phase resonance, or an antiresonance. The pitch of it does drop with added mass which supports this antiphase idea. 

But perhaps there is also some directionality? 

 

 

[Don Noon]

You have to think about the wavelength in the air that we are dealing with. It’s much much larger than the wing tip, so there is no directionality to worry about. The other thing to think about is the size of the patches on the instrument that are radiating, and at this frequency, it’s about half of the upper or lower bout that is moving, which would overpower anything being radiated by the wing tip.

 

[Davide Sora]

@Don Noon Since the added mass definitely has an effect as you demonstrated, what about the stiffness of the wing? Increasing its thickness could increase its mass, but also its stiffness, as could a longer or shorter, wider or narrower wing. What are your thoughts on this, i.e. what do you think is best to do in the construction phase to try to control (lower) the frequency at which the dip appears in the spectrum?

[LCF]

5 hours ago, Don Noon said:

You have to think about the wavelength in the air that we are dealing with. It’s much much larger than the wing tip, so there is no directionality to worry about. The other thing to think about is the size of the patches on the instrument that are radiating, and at this frequency, it’s about half of the upper or lower bout that is moving, which would overpower anything being radiated by the wing tip.

 

Reminds me of a thing we used to put in aerial feeds, a 'trap' which was a tuned circuit adjusted to the frequency of some persistent interfering signal. It had a similar effect in that it caused a dip in the response at that frequency. 

Sometime last year I was testing a 'stinger' as suggested by Oliver Rogers (via a ref. from Marty) This is a lightweight speaker driver ( 2" or so and cut away)  connected to a very weak spring which is touched on various areas of the top as the speaker is driven at different frequencies of interest. One of the areas that stood out was this specific ff hole wingtip. 

 

Davide, it is much like a free reed. Extra mass at the tip flattens, less stiffness at the 'root' also flattens and vice-versa. 

 

Perhaps one could devise a soundhole with multiple, tunable tongues, fforked tongues?

[LCF]

This area of that A. Fischer violin must be responding very differently to a regular type!

[Greg Sigworth]

I'm trying to wrap my head around this to understand it. Is it being said that the size of the upper f hole treble wing is not large enough to be the true sound radiator at this frequency, which radiator is actually the upper or lower bout region of the top plate. If this is true then the vibration energy that actually goes to these sound radiating regions can be some how controlled by the treble upper wing mass, more mass results in less energy going to the actual sound radiators in this frequency range; I guess because the energy then goes to a non-sound radiating condition of the violin. Is there possibly some oscillator coupling going on between the f hole wing and the rest of the violin? Thank you for the information.

[Marty Kasprzyk]

12 hours ago, Don Noon said:

Almost 15 years ago(!) I looked into what the upper treble F-hole wing might do acoustically, and generally found that the frequency range of highest wing amplitude corresponded with a zone of LOWER acoustic output from the violin.  The original MN thread is here.

This idea has been cooking in my head all this time, and I thought I'd look into it a little more carefully.  The basic test is pretty simple:  put a sequence of very small masses on the wing and see what happens to the impact spectrum.  In the chart below, I plot the DIFFERENCE from the baseline (no mass) condition, in dB vs frequency.  The masses (small dots of clay) are extremely small, and I had to buy new milligram scale to weigh them accurately.

The take-away from this is that the upper treble F-hole wing has a role in shaping the response in the low 1 kHz range.  This makes some sense, in that the wing is closely coupled to the treble bridge foot, and appears to behave somewhat as a non-radiating resonator* to absorb energy that would otherwise go into the body.

As far as practical applications, this might not be a big deal... the design of the F-hole wing has some of this effect, and is probably why the soundhole IS this F-shape.  But different designs of F-holes might behave a bit differently, and I suppose that if a given instrument has an annoying peaky resonance around this range, a little added mass might be a benefit.  But it's just in a narrrow frequency range, so I wouldn't expect a massive transformation.

* The proper term is actually a "tuned mass damper" (TMD)  or "tuned vibration absorber" (TVA), and if interested, these can be looked up online.

Does this graph suggest going the other way--lightening the f hole wing would increase sound output?

[LCF]

34 minutes ago, Marty Kasprzyk said:

Does this graph suggest going the other way--lightening the f hole wing would increase sound output?

Perhaps in that specific area. If you jam something against that wing of the ff hole to immobilise it, if that is possible, you might get a 1khz boost. 

[LCF]

42 minutes ago, Marty Kasprzyk said:

Does this graph suggest going the other way--lightening the f hole wing would increase sound output?

Marty, it would be a different investigation altogether on your through-post instruments since the treble foot is not connected to that soundboard area directly. 

Edited March 12 by LCF
Typo

[LCF]

51 minutes ago, Greg Sigworth said:

I'm trying to wrap my head around this to understand it. Is it being said that the size of the upper f hole treble wing is not large enough to be the true sound radiator at this frequency, which radiator is actually the upper or lower bout region of the top plate. If this is true then the vibration energy that actually goes to these sound radiating regions can be some how controlled by the treble upper wing mass, more mass results in less energy going to the actual sound radiators in this frequency range; I guess because the energy then goes to a non-sound radiating condition of the violin. Is there possibly some oscillator coupling going on between the f hole wing and the rest of the violin? Thank you for the information.

Yes, as I understand it. 

Although it is non-radiating it obviously will appear as a feature in the violin's input admittance at the bridge, so it will have (difficult to predict) effects on bowing and string function. 

[Jay Higgs]

Nice work, Don, taking the time to make objective observations.  I love to read acoustics research, but I am not at the level where I can apply it to manipulate the tone I seek reliably.   I agree with Davide, that a lump of clay on top of the wood is not the same as thicker wood.  Marty hits on my own question: what does this mean to those of us just trying to develop a decent tone? 

I suspect our ancestors accumulated a lot of empiric knowledge handed down to only their apprentices; "if it sounds like this, do that to make it better."  Empiric solutions are scattered all throughout MN's past posts.  I bet it would be popular if Maestronet compiled a tone trouble-shooting database.  "My violin (viola, cello, bass) had this tone defect, and this is how I improved it."  

[Don Noon]

I have done a number of other tests, but have not yet completed the data gymnastics to put it into presentable form.  I will get to it shortly, and I think it will help answer some questions.  I should get to it later this evening; something else is happening most of today.

3 hours ago, Davide Sora said:

@Don Noon Since the added mass definitely has an effect as you demonstrated, what about the stiffness of the wing? Increasing its thickness could increase its mass, but also its stiffness, as could a longer or shorter, wider or narrower wing. What are your thoughts on this, i.e. what do you think is best to do in the construction phase to try to control (lower) the frequency at which the dip appears in the spectrum?

Yes, lowering stiffness has a similar effect, and I have test results to show the effects of thinning a section below the upper eye.  Different shapes should do the same thing as well, although I have no direct tests.  I generally think that most modern violins have the wing resonance frequency too high, therefore the "Dunnwald dip" frequency is too high, therefore the "bridge hill" gets its start at a frequency higher than desirable.  Unfortunately, you can't really tell for sure what's better or worse until you get it together and play it.

[Marty Kasprzyk]

Maybe the wing tip fluting was done for acoustic rather than aesthetic reasons.

About 15 years ago Anders Buen also looked at the effects of weights on the f hole wing tips.  Attached is his paper.

100730Massoninnerf-holewingswithmodalanalysisandhumiditytestdata.pdf

[M Alpert]

6 hours ago, Marty Kasprzyk said:

Maybe the wing tip fluting was done for acoustic rather than aesthetic reasons.

Instinctively, I like this suggestion! I have often considered similar ideas for top "tuning", but haven't had the time or the capability/means/skill to test it.

Fascinating observation @Don Noon !! Look forward to hearing more about this sort of thing.

[LCF]

7 hours ago, Marty Kasprzyk said:

Maybe the wing tip fluting was done for acoustic rather than aesthetic reasons.

About 15 years ago Anders Buen also looked at the effects of weights on the f hole wing tips.  Attached is his paper.

100730Massoninnerf-holewingswithmodalanalysisandhumiditytestdata.pdf 1.65 MB · 10 downloads

Amazing stuff. Thanks Marty. 

[Andreas Preuss]

22 hours ago, Don Noon said:

Almost 15 years ago(!) I looked into what the upper treble F-hole wing might do acoustically, and generally found that the frequency range of highest wing amplitude corresponded with a zone of LOWER acoustic output from the violin.  The original MN thread is here.

This idea has been cooking in my head all this time, and I thought I'd look into it a little more carefully.  The basic test is pretty simple:  put a sequence of very small masses on the wing and see what happens to the impact spectrum.  In the chart below, I plot the DIFFERENCE from the baseline (no mass) condition, in dB vs frequency.  The masses (small dots of clay) are extremely small, and I had to buy new milligram scale to weigh them accurately.

The take-away from this is that the upper treble F-hole wing has a role in shaping the response in the low 1 kHz range.  This makes some sense, in that the wing is closely coupled to the treble bridge foot, and appears to behave somewhat as a non-radiating resonator* to absorb energy that would otherwise go into the body.

As far as practical applications, this might not be a big deal... the design of the F-hole wing has some of this effect, and is probably why the soundhole IS this F-shape.  But different designs of F-holes might behave a bit differently, and I suppose that if a given instrument has an annoying peaky resonance around this range, a little added mass might be a benefit.  But it's just in a narrrow frequency range, so I wouldn't expect a massive transformation.

* The proper term is actually a "tuned mass damper" (TMD)  or "tuned vibration absorber" (TVA), and if interested, these can be looked up online.

I would also try to block the vibration of the upper treble side wing to see what happens. On a VSO you could simply crazy glue a stick from the outside  over the gap. 

[Michael Darnton]

If only we had something, say on the end of our hands, a long thing with a soft damping pad on the end, that we could press on the wing with while someone else played, and listen to the difference as we put it on and off of the wing. Ah, such idealism! Never in this life. :-(

Back to the computer and measurements, then, to tell us what we hear. 

/s

[LCF]

12 minutes ago, Michael Darnton said:

If only we had something, say on the end of our hands, a long thing with a soft damping pad on the end, that we could press on the wing with while someone else played, and listen to the difference as we put it on and off of the wing. Ah, such idealism! Never in this life. :-(

Back to the computer and measurements, then, to tell us what we hear. 

/s

See previous suggestion " jam something against that wing of the ff hole to immobilise it" 

[Don Noon]

First, here’s the classic action of a Tuned Mass Damper (TMD):  vibrations of the structure are attenuated strongly at the resonant frequency, with amplification above and below the tuned frequency.  The magnitude of the induced dip and peaks are shaped by the amount of damping, among other things.

The previous plots were all based on impact response changes.  For something closer to real playing, here’s the effect of ~.1g  added to the upper treble wing, as measured by mic 1 m away from the instrument, playing a bunch of bowed glissandos on 3 different violins.  The dip is pretty apparent, with perhaps a slight rise to either side.  No, this doesn’t say ANYTHING about what sounds better or worse, or how playability has changed… but it at least hints at the existence of a knob to be turned and tested further.

Other than adding mass, the stiffness of the F wing can be reduced to lower the tuned frequency.  I cut some grooves in the top of the test VSO to test this, first ~.5mm deep and then ~1mm deep.  Then, to test the effect of the wing alone, I cut it off completely (glued back on in the photo). 

For this plot, I used the “no wing” response as the baseline, and looked at the differences due to the presence of the wing, without and with the grooves.  The wing alone gives a slight dip, and increased flex gives deeper dips at a lower frequency.

One more… my “Snakefiddle” has had a rather coarse tone to it, which I attribute partially to excessive strength in the low 1 kHz region.  The “wing” of this fiddle is very small and light, so I added .1g lead inset to the wingtip, as well as a modest groove to increase flexibility. 

I made some other major modifications elsewhere, so it’s hard to say what caused what, but here’s the before/after difference in impact spectra, again showing a slight dip in the usual spot:

It seems reasonable to me that the F-hole may have evolved early on through trial-and-error because certain shapes sounded slightly better.  I have a few examples, including a well-made violin with C-shaped holes, where the unusual soundhole shape could be a factor in the unusual and rough tone.  If this Tuned Mass Damper idea holds water, it may be possible to add one or more features like this internally to shape the tone. 

[Tim M]

Thanks for posting your experiments and results. This is fun stuff to think about. I am amazed you can measure these effects with such small masses. I have played with moving paired neodymium magnets around the top. And they can have clear effects on tone. In fact I have two sets that live on my personal fiddle. But they are an order of magnitude heavier than your lead weights. I am really interested in the weight versus flexibility versus vibration relationships. Maybe small amounts of thinning in the right spots can have effects equal to the relatively large masses I've been moving around. I'm especially interested in these relationships when it comes to weighting versus carving bridges. 

[LCF]

So, why does this specific  frequency have this tonal effect?

Is it because of sensitivity of ears in this band, or the physics of violins in that area specifically, or being in the vicinity of Bb-B roughly an octave above the  zone of peak wood response which sometimes leads to wolf notes, is that significant? 

 

Coincidentally I was looking at frequency responses of some of the popular guitar amps from the 60s and 70s last week.  Many of them have a broad mid range notch centred around 1 kHz.

 

 

 

[LCF]

6 hours ago, Tim M said:

Thanks for posting your experiments and results. This is fun stuff to think about. I am amazed you can measure these effects with such small masses. I have played with moving paired neodymium magnets around the top. And they can have clear effects on tone. In fact I have two sets that live on my personal fiddle. But they are an order of magnitude heavier than your lead weights. I am really interested in the weight versus flexibility versus vibration relationships. Maybe small amounts of thinning in the right spots can have effects equal to the relatively large masses I've been moving around. I'm especially interested in these relationships when it comes to weighting versus carving bridges. 

If you can be bothered to find and dissect them, inside the focusing mechanism of the read heads of old cd/dvd computer drives there are tiny neodymium magnets which are useful for such experiments..

[Don Noon]

8 hours ago, Tim M said:

Thanks for posting your experiments and results. This is fun stuff to think about. I am amazed you can measure these effects with such small masses. I have played with moving paired neodymium magnets around the top. And they can have clear effects on tone. In fact I have two sets that live on my personal fiddle. But they are an order of magnitude heavier than your lead weights. I am really interested in the weight versus flexibility versus vibration relationships. Maybe small amounts of thinning in the right spots can have effects equal to the relatively large masses I've been moving around. I'm especially interested in these relationships when it comes to weighting versus carving bridges. 

The treble F upper wing is a cantilevered thin piece of spruce strongly coupled to bridge movement, and therefore a very small change in mass or stiffness can have a significant effect.  The rest of the violin body isn't that way, and much larger masses are needed to cause anything significant.

The bridge is primarily an upper mass on a spring of the waist and feet, and should be pretty sensitive to small changes in mass or stiffness.  I am a bit curious about the very tiny dangly bits... the arms and whatever you call that thing in the heart.  It is possible that they could act as tiny tuned mass dampers and have localized effects in the spectrum, but maybe they are TOO small to do anything significant.

2 hours ago, LCF said:

So, why does this specific  frequency have this tonal effect?

Is it because of sensitivity of ears in this band, or the physics of violins in that area specifically, or being in the vicinity of Bb-B roughly an octave above the  zone of peak wood response which sometimes leads to wolf notes, is that significant?

Coincidentally I was looking at frequency responses of some of the popular guitar amps from the 60s and 70s last week.  Many of them have a broad mid range notch centred around 1 kHz.

The aesthetics of tone don't seem well suited to detailed technical analysis.  I just know that there are general features of spectral response that good violins usually have, and from personal experience that a response as in the experimental C-hole violin below, my reaction is somewhere in the range of "bleah", "ugh", or "gaaah"... mostly due to the red zone, but also the weakness in the highs.

[Tim M]

26 minutes ago, Don Noon said:

The bridge is primarily an upper mass on a spring of the waist and feet, and should be pretty sensitive to small changes in mass or stiffness.  I am a bit curious about the very tiny dangly bits... the arms and whatever you call that thing in the heart.  It is possible that they could act as tiny tuned mass dampers and have localized effects in the spectrum, but maybe they are TOO small to do anything significant.

I'm inspired. I need to get my USB mic and Audacity configured so I can play with these things and quantify the effects. Don, I especially like how you are studying the differences in frequency responses relative to a standard, which I imagine helps us understand relative change while avoiding a lot of the variation in measurements due to setup. 

And LCF, thanks for your tip on tiny magnets in computer drives. I've been using small ones from the hardware store, about 1 gram each. Also, thanks for your reference to tuning reeds. I have a friend who is a concertina maker. I'm going to pick his brain about these weight/stiffness/leverage/momentum/vibration issues. In the meantime, for those who want to think of violin parts like vibrating reeds, here is a nice link describing some basics of concertina reed dynamics. See the 7th post by Dana Johnson: https://www.concertina.net/forums/index.php?/topic/24307-raising-reed-pitch-by-traditional-or-other-methods/