Jump to content
Maestronet Forums


  • Posts

  • Joined

  • Last visited

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

xyz1000's Achievements

Junior Member

Junior Member (2/5)

  1. By your results on the aluminum of accuracy better than 0.5% precision better than 0.1% the measuring device is very good. If measuring a wedge-shaped wood gives 10-20% variations (measuring one wedge-shaped wood several times), then something is not right with the measurement, not the wood. Perhaps the act of measuring is having an effect -- maybe the hammer impulse ruins the spot where you hit the wood, maybe something else. If the peak is broad or asymmetric, that's fine, but it should be repeatably broad and asymmetric and be at the same place all the time. If you're looking for 50% variation between different wood, then the 20% variation on one wood may be acceptable. (I can't believe no one has responded to Mr. Burgess's antics -- good one. If I ever get rich enough I'll buy one of his violins just so I can say ..yeah the violin maker on that one, wood, hairs, the rustle test, localized response, all had different meanings for him.)
  2. Great discussion. I don't read this site regularly, so pardon the late response. I like the device, Don Noon. Very much. To have a measuring device like this, you'd probably want to measure its accuracy and precision. Most of the time it's precision that's more important and fortunately easier to test. Measure the same thing 30 times consecutively. What's the standard deviation? Is it 5700 +/- 100 m/s or 5730 +/- 10 m/s? For longer term precision (other terms used are repeatability for short term, reproducibility for longer term), repeat the measurement three consecutive times and then do that every hour or two. After about 10 hours again calculate the standard deviation. Usually this will be greater than the repeatability standard deviation. As for accuracy, measure an aluminum plate that's the same size. Measure 30 times consecutively. Compare with accepted/published/literature values. Quick google search says aluminum is 4877 m/s, which probably doesn't differ much from machine shop stock aluminum. Curious what the results are. Should tell how "good" the measuring device is.
  3. quote: Originally posted by: Michael Darnton Since this line is being followed, does anyone have a handle on someone who could test a small piece of wood for me? You can see what's on it, in the form of a white eflorescence, and I want to know what it is. I don't have enough for more than one shot at it, probably. While my expertise is not on wood, I have access to the analytical instruments used for microanalysis. I've always thought about countering some of Nagyvary's claims, but certainly I have no access to wood chips from a Strad and I'm no luthier. If any of you are willing to send samples, think about that and I will do a little survey on whether the instruments here can be used on organics (wood) and I will check whether the staff here is confident in analysing such samples. Got IR, Raman, x-ray (EDX), Auger, TEM. The catch is that the scientists here must be assured that there's a possibility of publishing in a major journal. You can post (preferred) or send PM (I don't normally check this).
  4. Clarinets are not violins. I am, however, intrigued by the paper mache guitar. How about cement, coconut tree, or heavily starched blue jeans? The point of course, is that you cannot go the extreme one way or the other.
  5. Quote: A friend of mine once referred to the "pleasant fantasy that the parts of a musical instrument should themselves be musical". On the other hand, there's a reason why luthiers don't make violins out of particle board, bamboo, or the rubber tree. Non-musical components are unlikely to produce a musical system. Having said that, ctviolin's finding that both ringing and non-ringing components can produce a violin that sings well is proof that the assembled system has more influence than the sum of individual components. In such a case, the test for individual components is then how it would perform in a system (not how it would perform as an individual component). That is, if you want to test how good a top plate is, remove the top plate of a violin that you know and then replace it with the test top plate. Test other top plates that way so that in the assembled system (violin), only the top plate is the variable.
  6. EdnaNavo – based on what I read here, I am almost certain you will get good advice. I’m rather curious to how your situation unfolds and to read the advice of some on the boards, so let me try to help in a simple way. If you send me the photos by email, I will host them and post them for you so that others can see the extent of the damage and what needs to be done. Then I’m sure the master luthiers here will debate and discuss the violin instead of the politics. PM or send me email.
  7. Quote: CHICAGO (Reuters) - Luxury jeweler Tiffany & Co. has sued eBay Inc., claiming the online auctioneer has contributed to violations of the Tiffany trademark by letting counterfeit items be sold on its Web site, a Tiffany spokesman said Monday. http://money.cnn.com/2004/06/21/technology...s.reut/?cnn=yes http://www.baltimoresun.com/business/bal-bz.tiffany23jun23,0,3332042.story?coll=bal-business-headlines Not directly violin making related, but I figured readers would be interested to know about this story. Now if there was one entity analogous to Tiffany that could sue ebay over the sale of fake violins......
  8. Okay, how about some specifics -- what do you guys think of the cornerless violin? I find it aesthetically pleasing, and engineering-wise, in producing sound it's better to have a sound-generator with less sharp corners (that's why speakers are circular instead of rectangular). I've never actually heard one, but when I told my daughter's violin teacher about it, she frowned and said it was a waste. I'd like to hear other opinions.
  9. Quote: " I bring this up because I have been in a position to observe the results of people who claim to occupy both camps first hand. My observations lead me to believe that those people who utilize science and instrumentation in order to guide their construction principals, do not generally construct violins that are in any way superior to those who use tradition and/or intuition. In fact in the majority of cases I have noticed a trend in the opposite direction. " I completely agree with this statement. From what I've seen, I would say that the question that doesn't get asked is "SHOULD everything be controlled" not "CAN everything be controlled". Generally I find the instruments of people who claim to control their instruments via acoustic technology to be not actively bad, but simply boring--the kind of art you'd get if you could teach a machine to paint. ….. This is unfortunate, but something I can easily believe. A lot of scientists jump into conclusions without considering the secondary or more subtle effects because the science becomes more difficult. There’s also the fact that the scale from bad to good is subjective. There’s no quantitative, measurable parameter that makes one violin good and another bad. At least nothing definitive that I’m aware of. The sound is described in terms of “rich” or “dark” or “full-bodied” or “sexy” or other adjectives that don’t mean a thing to the scientist. There are a lot of things to be gained from using science and technology, but it’s a matter of using the tools properly. I don’t know enough about Chladni modes and tuning, but if there’s a database of Chladni patterns for 1000 violins, from the Strads to the factory junk, I contend the information would be very useful for luthiers. There is control and there is measurement. If you’re controlling anything, you’d better know exactly what it is you’re controlling and that you’re measuring the right thing. The question I’d like answered by those using science for violin making is, “are the important measurements being made?”
  10. Here's the link: http://applied.math.utsa.edu/~gokhman/ecz/duffing.html then go to page 12. You can also do a search on "Duffing's equation" 1. nonlinear = change in response with the same force/excitation The fact that the violin changed in character is a nonlinear phenomenon. 2. In nonlinear systems, there are other things besides resonant frequency location (spring constant) and damping coefficient. Interpreting the response solely on these parameters will no longer be sufficient. 3. Yes, the graph is triple valued. If you start at one end (increase the frequency) you won't get the same map as if you start from the other end (decrease the frequency). There will be jumps. This is also where hysteresis comes in, as alluded to in previous posts. I'm not saying this is exactly what the response will be. This is an EXAMPLE of a nonlinear system and is one of MANY possibilities. 4. This is where I'm going to stop. Continuing this analysis is no longer useful to understanding why the violin is changing in character and will drive all the luthiers to red wine or beer. We wouldn't want that now, would we?
  11. Quote: ... If you can attribute the behavior to nonlinear behavior, you need to elaborate. Here are a few graphs I’ve hand drawn. The first shows the classic linear harmonic oscillator. Initial condition is in red. When the damping coefficient decreases, then the amplitude increases. In the second, it’s a case where non-linear effects come in. The system can no longer be described as a harmonic oscillator. The response to a force is no longer linear (not kx anymore) but will rather have other terms (kx + cx^3, for example). Why other terms? Because that’s the way things are. Force proportional to displacement is only valid for very small displacements. The concept of the resonant frequency having the max amplitude no longer applies. (for those interested in more, see http://applied.math.utsa.edu/~gokhman/ecz/duffing012.html In the third figure, it’s a case where the “resonant” frequency in the non-linear system changed (to w1). This has the effect of moving the whole response (from red to black), but if the system is still being excited at w0, then the amplitude increases. The point in all this is that all of these parameters cannot be separated in the real world. Personally, I don’t think that any of this is helpful at all in understanding the concept of why the violin response is changing. (… but you asked for me to elaborate) This is where what engineers do is much more useful – measuring the response of the violin when you excite (sweep) it through the range of audio frequencies. I saw in the movie “The Red Violin,” a scene where the restorer was measuring the response of a violin and comparing it to a fake. I can only guess how such a device is implemented, but that’s an excellent way of measuring what’s going on.
  12. Quote: That wood experiment appears to confirm my hypothesis that playing in is caused by a decrease in damping. Assuming that the results are as you said, the only possible explanation is that the Q of the wood increased. [Q is explained below.] Since the wood was excited at the resonant frequency, a change in stiffness could not have caused the increase in amplitude. That would have shifted the resonant frequency and reduced the amplitude instead of increasing it. Since the mass didn't change in a couple of hours, the damping coefficient must have decreased and the Q increased. In a simple (linear) harmonic oscillator, all that you’ve said is true. A change in the spring constant (stiffness) will not change the amplitude*, but rather the resonant frequency. The damping constant will change the amplitude and the Q. Unfortunately, the violin (or even the wood pieces) is not a simple simple harmonic oscillator. And this is where the model is limited. The real world is non-linear where stiffness and damping are not separate. In the described experiment, I would find it unlikely that after the observed amplitude increase, the damping constant is the only thing that changed. I’m willing to bet a bottle of red that the resonant frequency changed and that the same happened to other resonances. Going back to the original question of why a change in sound character, both are surely involved. To me it’s enough to see in the wood experiment that the response changed after exciting it for a long time. That’s a non-linear phenomenon. I’ve been focused on the stiffness because intuitively for me that’s the most physical effect. *amplitude of the response at the resonant frequency. The amplitude of the response at the driving frequency will decrease.
  13. Still here, Alan_Coggins. Got a glass right here, too. Cheers!
  14. Quote: It will be sufficient for me if you talk about creep, hysteresis, and resonances in general. Undamped ones or ones whose damping may eiher increase or decrease. But please say something about the physical processes, and which ones are most closely related to the effects discussed by the makers. Let me state a second time that I am no expert when it comes to polymer science and I’ve never studied wood or violin (or any instrument) sound scientifically. So talking in general is really all I can do. I’m intrigued about the interplay between violin making and science. If you’re looking for expert explanation, I ain’t it. Creep is the change in strain (length) over time for a given stress (force). If you put a 10 pound weight on top of a piece of wood, the wood length will change. If the weight remains there for one year, the length will change more compared to when the weight was there for only one second. It’s not a far extension that this same phenomenon (not necessarily called creep, but the same phenomenon) applies to how the violin might sound after playing it continuously from when new. When playing the violin, the back vibrates to some amplitude. The vibration is the equivalent of the stress. Over time, the amplitude increases. If a player plays a note the same was as when the violin was new, but this time the violin vibrates with increased amplitude, then certainly the sound will be different. And that is where I think there is a connection between creep and the violin opening up. Of all the effects discussed by makers, I believe that this and the stick-slip ones are the most dominant. On a microscopic level (cells and bonding), I will have to defer that to experts. Quote: So, as to the materials you are familiar with, do you find that hysteretic losses become less when they are vibrated for a while? That is, when they are cycled through many cyles of a normal mode by being excited or vibrated ? (Please answer at least this one for the record. That seems to be your area of expertise.) I am familiar with metals, but I’ve never vibrated them nor have I looked into the “cold flow” described by a previous poster. Okay, where's the wine? Don't leave me hanging here.
  15. Quote: The 'science types' answered this question years ago. Strips of spruce taken from violin tops were mechanically excited to resonance at their fundamental vibrations and it was found that the amplitude of vibration increased steadily over an hour or two. Leaving them for a day or two allowed them to revert to their original stiffness. In other words, the fibers of the belly really do loosen up with playing but the playing needs to be regular otherwise you are back to square one. From theory to experimental verification all within a matter of days – that’s pretty good . The experiment you describe is a good one since it separates the two regimes I was thinking about. This one is involved only with the wood stiffness and not the stick-slips associated with joints. References would be good, although it’s unlikely I’d ever take the time to find a library that carries Catgut Association Society Journal or an equivalent one. What I’m surprised about is the time involved – that the amplitude increased after an hour or two. Perhaps the testing was done with accelerating the effect in mind. The posters here describe a longer-time effect. Thanks for the info, GlennYork.
  • Create New...