Bill Warren Posted July 24, 2011 Report Share Posted July 24, 2011 I often cut slices from the thick side of the wedges, and use them for the bass bar. I like to think that those pieces are ideal for measuring a representative sound speed (with your end tapping method), as they are taken from the area of the wood that becomes central in the final table. Salve, I really like this idea. Thank you for mentioning it. Bill Quote Link to comment Share on other sites More sharing options...
Michael K. Posted July 24, 2011 Report Share Posted July 24, 2011 That fancy flexure bass bar clamp I have been using really didn't work all that well. So I took a half day to make a new bass bar clamping gizmo. The photos are really of three fixtures all together: -There's the basic plate holding frame in the bench vise -The bass bar locating fixture is attached to it (hinged and removable) -The new clamping jig is clamped to the frame The idea is lifted from aerospace bonding tooling, or sortof like a fixed-force go-bar. On the underside are adjustable arch support screws, with small segments of wooden dowels held onto the screws with plastic tubing, to prevent denting the plate. So far, this has worked quite well, and with the locating fixture removed, there is semi-infinite access to clean up the excess glue. I just have to be careful not to knock the weights off of their platforms... maybe I'll attach them next time. Is this really serious? Quote Link to comment Share on other sites More sharing options...
Don Noon Posted July 24, 2011 Author Report Share Posted July 24, 2011 Is this really serious? What part are you questioning? Quote Link to comment Share on other sites More sharing options...
Michael K. Posted July 24, 2011 Report Share Posted July 24, 2011 What part are you questioning? Nice to see how much creativity you spend for your "bass bar clamping gizmo". I understand your three fixtures all together that is fine. But I never seen the methods with the weights on it. Not important how we glue in bassbar as long the final result is perfect. I remember a maker when i was young who glue in bass-bar holding just for aprox. two minutes with his thumbs and fingers without seeing excess glue. It can work well with a perfect fitted bar without any tension and the right glue consistency. Quote Link to comment Share on other sites More sharing options...
Don Noon Posted July 25, 2011 Author Report Share Posted July 25, 2011 I understand your three fixtures all together that is fine. But I never seen the methods with the weights on it. That's the part I got from aerospace (where I worked), watching some of the bonding operations. An exact bonding pressure was specified to get the optimum bond line, and by knowing the area, they would know how much weight was required. There was an overhead gridwork of holes into which they put rods to apply pressure to the part, and the rods had platforms to put the weights on. So I didn't create anything, just copied it. Quote Link to comment Share on other sites More sharing options...
David Burgess Posted July 25, 2011 Report Share Posted July 25, 2011 That's really cool stuff, Don. Overkill for me when it comes to gluing in a bass bar, but excellent food for thought, applicable to many different things. Quote Link to comment Share on other sites More sharing options...
scordatura Posted July 25, 2011 Report Share Posted July 25, 2011 You guys need to keep in mind that Don was part of the team thay designed the Mars rovers that the original spec was for 90 sols (Martian days). The Opportunity rover has lasted 2737 sols. That is off the charts "over engineering". That gets my deep respect and could be beyond the scope of most fiddle engineering. Remember the shuttle tiles? Quote Link to comment Share on other sites More sharing options...
David Burgess Posted July 25, 2011 Report Share Posted July 25, 2011 You guys need to keep in mind that Don was part of the team thay designed the Mars rovers that the original spec was for 90 sols (Martian days). The Opportunity rover has lasted 2737 sols. That is off the charts "over engineering". That gets my deep respect and could be beyond the scope of most fiddle engineering. Scordatura, I'm familiar with all that, but that isn't what does it for me. Based solely on what Don puts up here about fiddles, I doubt that you'll ever see me trashing him. One thing that Don is super-good about is avoiding falling irrevocably in love with his own beliefs and hypothesis. Many of us pros in the fiddle business can take that as something to aspire to. Quote Link to comment Share on other sites More sharing options...
Don Noon Posted July 25, 2011 Author Report Share Posted July 25, 2011 One thing that Don is super-good about is avoiding falling irrevocably in love with his own beliefs and hypothesis. That too might be a carry-over from my previous life, where success depended strongly on how diligently you tried to prove your designs wouldn't work. Yeah, starting out with a good design helps, but you never really know until you try to make it fail. Think you made a good fiddle? Let's hand it to some professional musicians and see if they can tear it apart. Quote Link to comment Share on other sites More sharing options...
Don Noon Posted August 12, 2011 Author Report Share Posted August 12, 2011 Got it strung up today in the white That's a piece of cork between the chinrest and tailpiece, to knock out the vibration of the end of the tailpiece, so I can see the pure instrument response. Which is: In the prediction of signature modes, I win the competition. B1- actual = 435 Hz Anders: 460, or 25 Hz too high Me: 435, or exactly right B1+ actual = 553 Hz Anders: 565, or 12 Hz too high Me: 540, or 13 Hz too low The A0 is fairly high at 288 Hz, likely due to the stiff body and slightly oversized F-holes. Of course, the numbers don't matter in themselves; what's important is how it plays. And, it's pretty good, especially for first day string-up. The only real concern I have is that the G-string is relatively weak... the high A0 frequency and stiff body don't do well for the low frequency fundamentals. All the other strings seem very snappy and responsive, with significant tonal change depending on how close you bow to the bridge. The E string is interesting... not bright or harsh, but very strong and responsive. In looking at the response plot, often there is a deep valley just above the B1+ resonance, which is NOT apparent in this fiddle. That would make for a stronger, more even fundamental on the E string. Also, the "transition hill" is pushed to higher frequencies than I usually see, which would make for a strong first overtone on the E string. The "bridge hill" is relatively spotty and not very strong, which would explain the lack of brightness and harshness. I'll have to live with it for a week or two and see how things settle in, and mess around with bowed response to see how it differs from the impact spectrum. And do some comparison playing to solidify my opinions. At the moment, is seems just a bit stiff and high-strung compared to what I like and am used to, but maybe if I get used to this I'll like it better. Maybe I'll post some audio later. The top is glued on extremely lightly, so it won't be a big deal to take it off for modifications, if I so choose. Knowing me, I'll end up doing something to it. Quote Link to comment Share on other sites More sharing options...
Anders Buen Posted August 12, 2011 Report Share Posted August 12, 2011 The top is glued on extremely lightly, so it won't be a big deal to take it off for modifications, if I so choose. Knowing me, I'll end up doing something to it. I am looking forward to the data with varnish on. Is the chinrest on here? I also think that a lightly glued top also may influence the mode data. Your B1- seems weak in comparison to the B1+ which seems peaky in response to the rest. Do you have any idea why? Quote Link to comment Share on other sites More sharing options...
Don Noon Posted August 12, 2011 Author Report Share Posted August 12, 2011 Yes, the chinrest is on for this response plot. Without it, you might have won (I cheat). I doubt if the strength of the glue makes much difference; it's solid and rigid, but easy to take apart. With ultra light, high RR top wood, it is not too surprising to me that the B1+ would be strong, as the top is moving most in this mode. For B1-, the back is moving most, so wouldn't benefit from the top being light. I haven't checked your correlation charts to see if this is statistically true, but it makes sense to me. Quote Link to comment Share on other sites More sharing options...
Anders Buen Posted August 12, 2011 Report Share Posted August 12, 2011 Yes, the chinrest is on for this response plot. Without it, you might have won (I cheat). I doubt if the strength of the glue makes much difference; it's solid and rigid, but easy to take apart. With ultra light, high RR top wood, it is not too surprising to me that the B1+ would be strong, as the top is moving most in this mode. For B1-, the back is moving most, so wouldn't benefit from the top being light. I haven't checked your correlation charts to see if this is statistically true, but it makes sense to me. The balance between the B1+ and B1- is better in the earlier body spectrum plot. Does the Guarneri chinrest dampen the B1- somewhat? Quote Link to comment Share on other sites More sharing options...
Don Noon Posted August 13, 2011 Author Report Share Posted August 13, 2011 Does the Guarneri chinrest dampen the B1- somewhat? Without the chinrest, the B1- frequency is 454 Hz (+19Hz), and the amplitude is over 3 dB stronger. Nothing else changes much. That's only on the impact spectrum, though... in normal playing, clamped under the chin, I doubt there would be such a big difference in the B1-. Quote Link to comment Share on other sites More sharing options...
Don Noon Posted August 13, 2011 Author Report Share Posted August 13, 2011 I tried making recordings using this new fiddle and the Snakefiddle. I can't hear the difference between them on the recordings, yet they sound vastly different when I'm playing them. That's the way it always seems to work, every time I try it. Anyway, here's a poorly played clip of #7, for fun: #7 white.mp3 Quote Link to comment Share on other sites More sharing options...
Don Noon Posted August 22, 2011 Author Report Share Posted August 22, 2011 This fiddle has been together in the white for over a week now, and after playing it in comparison with other fiddles, in other settings, and having others play it, I'm definitely going to thin it out some more. Perhaps in another century or two it would age into another "Cannone", but I'm not willing to wait that long. It does have plenty of power and quick response as it is, but the low notes are very "dry" sounding, and the E string lacks brightness. I think I can improve these things... I spent the week tearing apart my #1 and #3 fiddles and making modifications, which generally worked out well. Back to the shop... Quote Link to comment Share on other sites More sharing options...
stephen maloney Posted August 22, 2011 Report Share Posted August 22, 2011 here's a poorly played clip of #7, for fun: #7 white.mp3 That open G at the beginning of the Bruch excerpt; I like how it slides into pitch from "F#" up to "G". How on earth did you achieve that? Try that, Strad vs. Gesu open G testers! Made my day, many thanks. Quote Link to comment Share on other sites More sharing options...
Don Noon Posted August 22, 2011 Author Report Share Posted August 22, 2011 I like how it slides into pitch from "F#" up to "G". How on earth did you achieve that? Scruggs tuner on the G string . Actually, I have no idea. Just random ineptitude, I suspect. Quote Link to comment Share on other sites More sharing options...
Don Noon Posted August 23, 2011 Author Report Share Posted August 23, 2011 I have taken 3 grams of wood out of the top, so it's as light as I'm willing to get. Still seems a bit stiff, but that's where it will stay. Top data, WITH bassbar: 58.3 grams, m1 94 Hz, m2 161 Hz, m5 378 Hz. Not much change; m5 is only down 5 Hz from before. Body modes changed only slightly too: A0 = 286 Hz (-2 Hz) CBR = 408 Hz (-13) B1- = 428 Hz (-7) B1+ = 542 Hz (-11) Today's response plot: Doesn't look hugely different from the earlier one; and the sound (from what I can recall) isn't hugely different, either. I think it's better now when I play it, but who knows for sure. The response plot is somewhat similar in overall shape to Doug's balsa fiddles... maybe 'cuz the spruce I'm using is approaching balsa in density? One curious thing I found: in the bowed response, there was a peak at 385 Hz that was stronger than the notes around A0. This doesn't show up at all in the impact spectrum, and I'm speculating it's either a room resonance or related to the CBR. Quote Link to comment Share on other sites More sharing options...
JimMurphy Posted August 23, 2011 Report Share Posted August 23, 2011 ...One curious thing I found: in the bowed response, there was a peak at 385 Hz that was stronger than the notes around A0. This doesn't show up at all in the impact spectrum, and I'm speculating it's either a room resonance or related to the CBR. Don, I notice from the response plot your fiddle takes quite an appreciable dip at ~385Hz. I'm speculating this fiddle is inducing irregular phase-shifts on played string frequencies. There're also significant dips at ~1600Hz and ~2400Hz too. Irregular phase-shifts do NOT imply 'bad sound' mind you. Jim Quote Link to comment Share on other sites More sharing options...
Anders Buen Posted August 23, 2011 Report Share Posted August 23, 2011 I have taken 3 grams of wood out of the top, so it's as light as I'm willing to get. Still seems a bit stiff, but that's where it will stay. Top data, WITH bassbar: 58.3 grams, m1 94 Hz, m2 161 Hz, m5 378 Hz. Not much change; m5 is only down 5 Hz from before. Body modes changed only slightly too: A0 = 286 Hz (-2 Hz) CBR = 408 Hz (-13) B1- = 428 Hz (-7) B1+ = 542 Hz (-11) Today's response plot: Doesn't look hugely different from the earlier one; and the sound (from what I can recall) isn't hugely different, either. I think it's better now when I play it, but who knows for sure. The response plot is somewhat similar in overall shape to Doug's balsa fiddles... maybe 'cuz the spruce I'm using is approaching balsa in density? One curious thing I found: in the bowed response, there was a peak at 385 Hz that was stronger than the notes around A0. This doesn't show up at all in the impact spectrum, and I'm speculating it's either a room resonance or related to the CBR. Is this still with the Guarneri chinrest on? From Schleskes data from a white and varnished the B1- should increase quite a bit in frequency. A 0.1mm varnish layer will increase the weight about 5g. Quote Link to comment Share on other sites More sharing options...
Anders Buen Posted August 23, 2011 Report Share Posted August 23, 2011 I have taken 3 grams of wood out of the top, so it's as light as I'm willing to get. Still seems a bit stiff, but that's where it will stay. Top data, WITH bassbar: 58.3 grams, m1 94 Hz, m2 161 Hz, m5 378 Hz. Not much change; m5 is only down 5 Hz from before. A thicker centre than the rest of the top? The m1 is pretty high, but maybe it is the low density too.. Quote Link to comment Share on other sites More sharing options...
Don Noon Posted August 23, 2011 Author Report Share Posted August 23, 2011 Is this still with the Guarneri chinrest on? From Schleskes data from a white and varnished the B1- should increase quite a bit in frequency. A 0.1mm varnish layer will increase the weight about 5g. Yes, chinrest is on. With it off, B1- goes up to 446 Hz. I too see that varnish increases B1- frequency the most, 11-12 Hz. A thicker centre than the rest of the top? The m1 is pretty high, but maybe it is the low density too.. The center is a little thicker (2.9mm) than the bouts (~2.6mm). The areas above and outside the F-holes are especially thick, around 3.5mm. Maybe the high arching also gives a high m1? Quote Link to comment Share on other sites More sharing options...
Don Noon Posted August 23, 2011 Author Report Share Posted August 23, 2011 More playing around with spectra... Here's a plot of #7 vs. the Titian. Titian data (black line) is from Strad3D, corrected for the hammer input spectrum. #7 (red line) is suspended on rubber bands, without the chinrest, strings damped. The B1+ is split by tuning the tailpiece resonance to the B1+ frequency. Since these are all relative, I fudged the level to be comparable. I'm not sure it really says much, other than it's really hard to tell from the spectrum which one is "better". Quote Link to comment Share on other sites More sharing options...
Janito Posted August 23, 2011 Report Share Posted August 23, 2011 I'm not sure it really says much, other than it's really hard to tell from the spectrum which one is "better". This offers 2 possibilities: 1. #7 and the Titian are equally good or 2. the spectral analyses miss the point. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
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.