Sign in to follow this  
jezzupe

scale and acoustics

Recommended Posts

Anders Buen wrote:

The reflectivity of the ribs determine the A2 and A3 modes across the bouts, so they do play a role

Do you happen to know the frequency range for these two air modes? I expect them to be pretty high. I also expect their ability to radiate to be quite low and probably indistinguishable from local (non air) radiating modes.

I do sometimes attempt to couple air modes with corpus modes but have never attempted anything beyond A1

Oded

Share this post


Link to post
Share on other sites

Anders Buen wrote:

Do you happen to know the frequency range for these two air modes? I expect them to be pretty high. I also expect their ability to radiate to be quite low and probably indistinguishable from local (non air) radiating modes.

I do sometimes attempt to couple air modes with corpus modes but have never attempted anything beyond A1

Oded

Just around 1kHz I think. They may be able to smooth excessive peaks in that region.. I think that is important.

I will look the data closer up when I come home.

Share this post


Link to post
Share on other sites

I have some qesutions that some of you may be able to help me understand.

If one were to stand in an empty auditorium and shout, one would here the echo slapback and reverb as the sound reflects of the room surface and bounces back to our ear.

If a hypothetical "shrink ray" were able to shrink the building and the person inside to scale, say to the size of a violin, and then the person shouted [with scaled lungs to meet the small person, as well as I assume scaled volume} would the same laws of acoustics apply? would the little tiny man inside of the little tiny auditorium hear reverb and echo?

this is relating to some "theory's" I'm experimenting with related to the rib structure and lining them with balsa. The Fox violin is really developing into something quite special sounding, and I'm starting to think what I did with the ribs {lined entirely with balsa} has something to do with it. It seems to do two things at the same time. It acts as both as dampening surface and a "booster" for frequency's that may need accentuation. It is very much like thinking of the interior of a high end recording studio, the internal surfaces are well thought out, some dampen, some reflect. It is almost seeming like a clip/booster/compressor for electronic recording. Where the signal is reduced when volume is to high and boosted when too low.

But seemingly for this "theory" to seem logical in my mind, the laws of sound would have to apply in scale.

So, when a spider screams inside of a milk carton, does he hear his own echo? :huh:

The time lag depends on the size of the auditorium. The outgoing and incoming waves make a "standing wave". A violin in a small box would show resonances of the box driven by the violin. You would hear a violin, but with enormous peaks associated with the box.

In a large auditorium, the returning wave is smaller than the outgoing one, and the net wave is a wave traveling slower, but outward. (Physics talks about phase and group velocity of a wave)

I read a number of the replies, and could make sense of only a few. But idea of standing waves comes from the superposition of all waves. When two waves in opposite directions are added, they give a standing wave (if there is no absorption or damping.) Physics applies simply through summing all waves at any given point.

Summing waves or "superposition" is applicable for low intensity waves that do not really change the properties of the medium, air. This is true for acoustics of music, but not for explosions and large shock waves.

Edited by Johnmasters

Share this post


Link to post
Share on other sites

Here's a list of the first sevenair modes

Their frequencies are:

284Hz, 499Hz, 1077Hz, 1190Hz, 1340Hz

Oded

Yes, I think they miss that there is two modes (1.0) one across the lower and one across the upper bout. There might even be one across the central bout, but the f-holes will dampen that quite much, if not totally.

I think these modes will end up at varying resonace frequencies in different violin models and in violin bodies where the plates does vibrate. More typical for the A1 is 460Hz ish not 500Hz as Rossings work indicate. Maybe they measured in a stiff cavity?

Share this post


Link to post
Share on other sites

I tested one instrument and spotted what appears to be an air mode at ~3KHz, have you ever seen this?

OK

I looked, and didn't see anything there. What leads you to believe it's an air mode? Unless that one instrumnet was very odd in dimensions, I would expect all violins to have very much the same air modes, within a few percent.

Share this post


Link to post
Share on other sites

I tested one instrument and spotted what appears to be an air mode at ~3KHz, have you ever seen this?

OK

They will come denser and denser for higher frequencies. Lothar Cremer is discussing the air modes and their modal density in his book "The Physics of the Violin".

Share this post


Link to post
Share on other sites

"If a hypothetical "shrink ray" were able to shrink the building and the person inside to scale, say to the size of a violin, and then the person shouted [with scaled lungs to meet the small person, as well as I assume scaled volume} would the same laws of acoustics apply? would the little tiny man inside of the little tiny auditorium hear reverb and echo"

No, he wouldn't. Although there would still be nodal resonances, above perhaps 300 hz, their Rt would be short enough to be completely disguised by the Haas effect. The brain would not PERCEIVE them as anything but part of the original sound.

Share this post


Link to post
Share on other sites

"If a hypothetical "shrink ray" were able to shrink the building and the person inside to scale, say to the size of a violin, and then the person shouted [with scaled lungs to meet the small person, as well as I assume scaled volume} would the same laws of acoustics apply? would the little tiny man inside of the little tiny auditorium hear reverb and echo"

No, he wouldn't. Although there would still be nodal resonances, above perhaps 300 hz, their Rt would be short enough to be completely disguised by the Haas effect. The brain would not PERCEIVE them as anything but part of the original sound.

The Haas effect times and distances are based on the brain and nervous system of a full-scale person. Presumably these would scale too.

Share this post


Link to post
Share on other sites

The Haas effect times and distances are based on the brain and nervous system of a full-scale person. Presumably these would scale too.

I assume?, crap now not only do we have to figure out this violin thing, but now we need to invent a shrink ray, Don, I'm sure your on it.

Again, thanks for all the replys

Share this post


Link to post
Share on other sites

The only way I know of testing radiation of air modes is to hum a note into the Bass side F hole then cove and open the treble side F hole. The radiation is very obvious for the Helmholtz. I usually test for air modes (if I'm not using reciprocity ;-) is to drop a lapel mike into the cavity and blow into the F hole or sometiems gently slap the back of the instrument. I'm not too clear about how to separate plate and corpus modes from possible air modes.

Any advice?

Oded

Share this post


Link to post
Share on other sites

The only way I know of testing radiation of air modes is to hum a note into the Bass side F hole then cove and open the treble side F hole. The radiation is very obvious for the Helmholtz. I usually test for air modes (if I'm not using reciprocity ;-) is to drop a lapel mike into the cavity and blow into the F hole or sometiems gently slap the back of the instrument. I'm not too clear about how to separate plate and corpus modes from possible air modes.

Any advice?

Oded

They will couple to give a set of composite modes. How they do this would depend on their relative strengths and frequencies. A corpus mode in vacuo will not be the same as in a real violin, in other words.

There is another way to excite these air modes. Put a computer microphone near the f hole and turn up the speakers. As you approach and recede from the speakers, you will get a feedback loop. (try about 5 feet) The resonances are very loud if the speakers are up. As you move, the resonance stays until it suddenly jumps to another one. This is VERY interesting. You can do it with a beer bottle or any cavity oscillator. It is a must try for anyone.

I have not done anything with it. My FEA has no capability to handle these acoustic contributions.

Share this post


Link to post
Share on other sites

My FEA has no capability to handle these acoustic contributions.

There are acoustic elements in Abaqus. Maybe a less than 1000 node model could be built with just stiff boundaries? Could start with a rectangular box as Cremer uses in his theoretical calculations. Or maybe a trapezodial box? There is a simple example of a loudspeaker box with acoustical elements inside and I believe shell elements in the loudspeaker box. Therre are elements for the loudspeaker cone too. Maybe something could be learnt from that?

Your feedback idea is smart. I watched the modes while moving the oudspeaker around while Iused TV holography. The resulting vibration should be the one that contribute to the sound level in the loudspeaker position for that given frequency, by the reciprocity principle.

Share this post


Link to post
Share on other sites

just wondering in theory would air modes exhibit transients? would those transients perhaps look a bit different?

OK

I think how the transients look will depend on the damping of the air and body modes around that perticular frequency. I do not know if the air modes are more or less damped than the body modes. But from Janssons experiment measureing the modes inside a violin body with platered outside and closed versus open f-holes indicate that the damping of the interiour modes is increased with open f-holes.

With the body free to vibrate they should be damped even more, but also the body mode vibrations should appear.

I think it should in theory be possible to measure the interiour modes with a tube extended from a micprophone into the violin with the tube not being in contact with the violin body. Then one can tap the bridge and see. Havent tried it yet.

I think that some of the air modes should be visible in modal analyses as the phase regions in the top and back plate should match for the air modes. I have done two such modal analyses and could pick out iar mode candidates. The plates does also have modes close nearby coupling to the air modes, so the picture can be complicated.

One way to asess the air modes effect on violin tone is to fill the interior with a heavier gas. The A0 will be affecetd too, but for the played notes some distance above the A0 the effect of the air modes should be audible. Helium would escape the f-holes, while a heavier gas than air will easier stay inside. Anyone with CO2 available in a gas bottle out there?

Share this post


Link to post
Share on other sites

I scanned Cremers chapter on cavity modes. He is very theoretical, but you see Janssons experiment results and a couple of formulas that can help anybody to assess where to find the modes, just by the dimensions of the body and some simple calculations.

You will see that the cavity resonances look more rounded when the f-holes are opened than when they are closed. More rounded and weaker resonances indicate more loss, and thus increased damping. You also see the A0 appear when the f-holes are opened.

Formula 11.63 give you estimates for the resonance frequencies. The "n numbers" are integers between 0 and anything and give the number of nodal lines in that direction. Numbers larger than say 4-5 will give very high resonance frequencies.

His theory also indicate that the difference in dimension across the upper bout, central and lower bout does not introduce three modes as I anticipated. I am not quite sure yet if I accept that. Would like to look abit on measured or calcualted modes first.

Cremer On natural cavity modes in the violin.pdf

Share this post


Link to post
Share on other sites

Anders,

I hope to be able to measure transients very soon with the new program I have been developing with a friend. (in beta form now and still under development)

Oded

Interesting. I know that Sam Z is very interested in the transients, I believe especially how fast the harmonics build up.

What are you looking at with your program? The buildup time and decay?

Share this post


Link to post
Share on other sites

... .

There is another way to excite these air modes. Put a computer microphone near the f hole and turn up the speakers. ... .

I measure A0 this way. Nevertheless, when my violin has the right dimensions A0 falls around 275-7 Hz, the canonical value for classical violins.

I recall Bissinger noted at an Oberlin Acoustics Workshop how the strength of A0 is indicative of a good sounding instrument. I presume that would be due to a stiff, rigid body - one with little parasitic damping.

Stay Tuned.

Mike

Share this post


Link to post
Share on other sites

I measure A0 this way. Nevertheless, when my violin has the right dimensions A0 falls around 275-7 Hz, the canonical value for classical violins.

I recall Bissinger noted at an Oberlin Acoustics Workshop how the strength of A0 is indicative of a good sounding instrument. I presume that would be due to a stiff, rigid body - one with little parasitic damping.

Stay Tuned.

Mike

I think already Meinel in the 1930ties noted the importance of the A0 level, later Dünnwald put an emphasis on the A0. As you mention, also Bissinger has noted that the strength of the A0 was measured to be stronger in the two Strads and a del Gesu he has measured in his program. He has data from some 17 insruments in his base.

I do not measure the stiffness directly, but I think that the opposite of what you suggest is true. The less stiff the plates are, the stronger the A0. I think the driving mechanism is the important one for the relation, not the losses.

In my data mining work the preliminary conclusion is that the lower the top plate weight and mode 1 top plate frequency you have, the stronger the A0 will tend to be. I have about doble as many data points as Bissinger, but measure using a simpler setup and fewer measurement points. I also measure on mass produced instruments and regraduate them. They are not bad, I think they are bi tri octave tuned, but they tend to be thick and somewhat stiff in comparison to what we may expect from cremonese instruments in general.

Share this post


Link to post
Share on other sites

[qu

++++++++++++++++

A0 discussion is too difficult to understand for most consumers or players. My method is so simple, that is, play it if I like it (its price and its tone) buy it .

If A0 is good but I don't like the tone, what I am going to do? My ears must be bad. I never know.

Share this post


Link to post
Share on other sites

[qu

++++++++++++++++

A0 discussion is too difficult to understand for most consumers or players. My method is so simple, that is, play it if I like it (its price and its tone) buy it .

If A0 is good but I don't like the tone, what I am going to do? My ears must be bad. I never know.

The concept is very easy to understand if you have a good model for it. If you take two bottles of about the same volume and necklength but one being made of a thin material while the other is stiff. If you tap the side of those two, from which of them do you expect to hear the strongest air mode?

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.