Sign in to follow this  
BigFryMan

Tap tones are too low! What to do next...

Recommended Posts

I would worry about leaving a top 4.5mm+ in the center especially if the top is heavy (historical precedence is not with you here. Most old violins are ~3mm in the center). Personally, I would sacrifice some of the cross grain stiffness (thickness in the center) to lower the plate’s weight.

Share this post


Link to post
Share on other sites

For identifying the frequency and shape of the resonance modes I used the old fashioned glitter  bounce testing where the top is supported on foam rubber pads and then blasted with close  speaker using an adjustable frequency saw tooth wave sound to find the resonance frequencies.

The modern way is modal analysis such as George Stoppani uses where an accelerometer is mounted on the bridge and the plate is tapped all over.  Software then generates the mode shapes.  Laser holograms and vibration testing equipment have also been used but these are beyond ordinary maker's budgets.

Modes 2 and 5 have received a lot of attention but unless you see the actual shapes it is sometimes possible to confuse their frequencies with modes 3 and 4 while you are thinning the plates.  It is important that the placement of the supports (foam or fingers) be on the mode's node lines.  Anywhere else dampens the the ability of a particular mode to vibrate so you have to search for the best support locations to start with. As the plate is thinned the node lines of a mode often change so it important to keep adjusting your support positions.

This is also the basis for some religions.

Share this post


Link to post
Share on other sites

Just fired up my recording rig and this is what I got. It's easy to see where the modal peaks are, but perhaps I had mode 5 mixed up with another mode. I don't yet have a jig for setting my plates up for doing the tea leaf test, but I might jerry-rig something up and see what I can see.

Tap Tones.png

Share this post


Link to post
Share on other sites
51 minutes ago, curious1 said:

I would worry about leaving a top 4.5mm+ in the center especially if the top is heavy (historical precedence is not with you here. Most old violins are ~3mm in the center). Personally, I would sacrifice some of the cross grain stiffness (thickness in the center) to lower the plate’s weight.

If you are addressing me - I was following what the wood wanted, in this case. Extremely light (0.30sg) spruce. The results speak for themselves. I wouldn't do this with just any wood. 

As for historical precedent, the 1679 Stainer Hargrave wrote about is about 4.2 in the middle. My only point is that each billet is different and sometimes extreme wood requires extreme arching/grads. 

Share this post


Link to post
Share on other sites
1 hour ago, curious1 said:

I would worry about leaving a top 4.5mm+ in the center especially if the top is heavy 

(my emphasis)... I would never go that thick, but certainly with .30 density spruce you might want to go thicker than normal.  I might go to 3.5 or slightly more with super-light wood, but 4.5 seems more than necessary to me.  Still, if the top is NOT heavy, it might not hurt much.

 

25 minutes ago, BigFryMan said:

Just fired up my recording rig and this is what I got. It's easy to see where the modal peaks are, but perhaps I had mode 5 mixed up with another mode.

The 392 peak looks like it would be the M5, or what will eventually become the M5 when the F's are cut and the plate gets thinned some more.

Share this post


Link to post
Share on other sites
16 minutes ago, Don Noon said:

.. I would never go that thick, but certainly with .30 density spruce you might want to go thicker than normal.  I might go to 3.5 or slightly more with super-light wood, but 4.5 seems more than necessary to me.  Still, if the top is NOT heavy, it might not hurt much.

That belly came out at 60 g with a small, baroque bar. I worked it until it sounded well, and stopped. I've not left things that thick before or since, but for that piece of wood it seemed to be the ticket. Happy player, happy luthier. 

Share this post


Link to post
Share on other sites
54 minutes ago, JacksonMaberry said:

 Extremely light (0.30sg) spruce. 

May I ask how 30 sg. was determined and who did the determining?

Share this post


Link to post
Share on other sites
1 hour ago, curious1 said:

The only true confirmation of M5 would be by vibrating the plate and observing the modal pattern.

M5 without bass bar:

4B335E4A-BAFA-44F7-A2B1-5B07A39FCD8C.jpeg.40dcaa85c46e019025834d44c4d13097.jpeg

That way may be the best way but not the only way. 

  One other method I observed was Davide using a simple guitar tuner and tapping to hear where the pitch was at a given time and another method, which is my own personal way, is to use an online tone generator with headphones to match frequencies from my wood to what is available adjustability wise from the generator.   It's hit and miss my way while trying to cross modes, if need be.

Share this post


Link to post
Share on other sites
3 minutes ago, BigFryMan said:

So at 389 I got this I’m guessing mode 6?

 

BF4BD134-5412-4417-AF62-8C0F0E6424D0.jpeg

Nice set up! Start trying to find the mode 2  resonance first to see what that looks like and what its frequency is. Then move higher in frequency to see what  the next resonance shapes and freq. you get.

Share this post


Link to post
Share on other sites

Make sure you're pinching the plates and tapping in the right spots.  
That might help with any confusion between modes. 
Here's a good illustration courtesy of Mr. Sora.  

schema tap-tones rid.jpg

Share this post


Link to post
Share on other sites
3 hours ago, BigFryMan said:

So at 389 I got this I’m guessing mode 6?

 

BF4BD134-5412-4417-AF62-8C0F0E6424D0.jpeg

Eureka!

The cause of a wolf-note is one step closer to solution.

If you capture him on video does he "bound" across the plate?

cheers edi

Share this post


Link to post
Share on other sites

Now that I've had a bit more of a play around, I'm pretty sure my 281hz mode is mode 1. I think I've found mode 5 at 376hz by playing around with where I hold the plate. It's not super resonant and the plate is quite stiff along it's length so I guess that would make sense. I Have done some more graduating and I think I'm ready to cut the f-holes. This should hopefully get me a little more give along the length of the plate and allow me to fine tune the graduations from there.

Share this post


Link to post
Share on other sites
12 hours ago, uncle duke said:

May I ask how 30 sg. was determined and who did the determining?

I did, by establishing the billet's mass and volume, then deriving it's density. 

Share this post


Link to post
Share on other sites

Although I will continue to record taptones (as it is quick and easy), in looking at 10+ years of data I think I could do just as well without it by only going with weight and thickness targets.  My target weight is ~60g or less without the bass bar... if I hit that weight and it's a lot thicker than "normal", I'll go lighter, but not less than 55g.  If I reach my "normal" thicknesses and the weight is high, I'll go thinner.  I probably have wood that allows lighter plates, so untorrefied wood might want to be 5 - 10% heavier.

Taptones for me are more "oh, that's interesting; maybe it will mean something some day"... but it's not a target to be achieved.  And so far, taptones haven't provided any great enlightenment.  Maybe endarkenment.

Share this post


Link to post
Share on other sites
18 minutes ago, Don Noon said:

Although I will continue to record taptones (as it is quick and easy), in looking at 10+ years of data I think I could do just as well without it by only going with weight and thickness targets.  My target weight is ~60g or less without the bass bar... if I hit that weight and it's a lot thicker than "normal", I'll go lighter, but not less than 55g.  If I reach my "normal" thicknesses and the weight is high, I'll go thinner.  I probably have wood that allows lighter plates, so untorrefied wood might want to be 5 - 10% heavier.

Taptones for me are more "oh, that's interesting; maybe it will mean something some day"... but it's not a target to be achieved.  And so far, taptones haven't provided any great enlightenment.  Maybe endarkenment.

I've made fewer fiddles than yourself at this point, but this has been my experience as well, precisely. 

Share this post


Link to post
Share on other sites
34 minutes ago, Don Noon said:

Although I will continue to record taptones (as it is quick and easy), in looking at 10+ years of data I think I could do just as well without it by only going with weight and thickness targets.  My target weight is ~60g or less without the bass bar... if I hit that weight and it's a lot thicker than "normal", I'll go lighter, but not less than 55g.  If I reach my "normal" thicknesses and the weight is high, I'll go thinner.  I probably have wood that allows lighter plates, so untorrefied wood might want to be 5 - 10% heavier.

Taptones for me are more "oh, that's interesting; maybe it will mean something some day"... but it's not a target to be achieved.  And so far, taptones haven't provided any great enlightenment.  Maybe endarkenment.

Don,

I can completely understand your approach and I think I’ll end up there myself eventually. Right now though I have to look for Stradivarius’s secret for a few years :) I guess in a sense maybe I’m trying to work out how important modes are in the grand scheme of things. Tap tones can tell you something about a plate, just like weighing it, measuring it and flexing it. I don’t expect it to be a be all and end all. Being only on my third, I figure I’m going to try and measure everything I can and write it down so I have a point of differentiation and reference. Also, I like testing my numbers against someone who is successful. Unfortunately I’m a Lone Ranger (apart from Maestronet and what I can dig on the internet) so comparisons I can measure are useful to me. I’ll still keep measuring as I cut the f-holes and install the bass bar. 

Apreciate your wisdom

Share this post


Link to post
Share on other sites
7 hours ago, BigFryMan said:

@edi malinaric
What do you mean when you say "The cause of a wolf-note is one step closer to solution"? I haven't put the violin together yet so I'm not sure if there will be any wolf notes.

Hi BigFryMan - having owned a wolf-dog I immediately saw him in the tea leaves - the rest followed.

DSC02144.thumb.JPG.db8411af4437dd5167008c52cbf3684e.JPG

Wonderful companion - mother was a Siberian wolf (that's her colouring - exactly), father was a Grey wolf/Belgian Shepherd cross.

cheers edi

Share this post


Link to post
Share on other sites

Since a mode's frequency f is equal to the square root of the stiffness S to mass M ratio

f = (S/M)^0.5

it is easy to see that any mode (mode 5 for example) frequency target can be attained by an infinite number of plate stiffnesses and plate weights.  It is highly unlikely that all of these different plate stiffnesses and plate weights will all give an identical violin sound character so "plate tuning" to achieve  certain tap tone mode frequencies (typically mode 2 and 5) isn't adequate.

The above equation can be rearranged to show the plate's stiffness S is equal to the plate's mode frequency F squared times the plate's effective mass M:

S = f^2M

or the effective mass M (the portion of the plate's total mass which is moving in that mode resonance) can be found from the plate's stiffness divided by the mode frequency f squared:

M = S/f^2

Thus you can hit a target plate weight (say 60g for example) with lots of different combinations of plate stiffnesses and mode frequencies. Again it is highly unlikely that they will all sound similar.  So just hitting a target weight isn't that helpful either.

But it should be apparent that if you measure any two of the three variables (S, M, or f) that the remaining third variable is then fixed.

 

Therefore I suggest you at least record your plate's weight and mode frequencies.  If you are a control freak I suggest trying to hit the plate weights and mode frequencies of famous violins.  Unfortunately there aren't a lot of good examples published so you will probably have to generate a data history of failures and success of  your own violins to guide you.  Carefully and systematically record all this.

Or you can just make violins and skip all this stuff.

  

 

Share this post


Link to post
Share on other sites

I looked at this today, and realized I wasn't thinking. Something new? Hah!  My violins are at 3.2 - 3.5 in the middle. Maybe the one out of .3 sg. Engelmann is 3.6 or more.  The latest Gofriller may be thinner, it feels quite light. The thickness in the middle is usually set right away by the way that I work.  The guitar belly is 5 all over now, and I might go down to 4.5 mm.  I had the center thicker, and didn't like it.  It is pure catenary curve, and all over even seems to work, at least on this one.  Violin backs are always 4.5 and up; 7 mm on one viola.  Maybe I'd go thinner if it was Walnut, or Padauk.  But we aren't going crazy here; are we?  

Maybe so.  At least in spurts.  That's how I work anyway, in spurts.  Easy to stay focused intensely for 2-3 hours, but hard to remember what you did, because it just happens as soon as you think about it.  Like when you make varnish, and can't remember what you added, when, and how much.  

Share this post


Link to post
Share on other sites
2 hours ago, Don Noon said:

  And so far, taptones haven't provided any great enlightenment.  Maybe endarkenment.

What about the time period at the very beginning of your violin making journey?  Surely you thought tap tones and possibly hz readings were tops.  

Share this post


Link to post
Share on other sites
8 hours ago, BigFryMan said:

Now that I've had a bit more of a play around, I'm pretty sure my 281hz mode is mode 1. I think I've found mode 5 at 376hz by playing around with where I hold the plate. It's not super resonant and the plate is quite stiff along it's length so I guess that would make sense. I Have done some more graduating and I think I'm ready to cut the f-holes. This should hopefully get me a little more give along the length of the plate and allow me to fine tune the graduations from there.

I understand the possible confusion because when plates are very thick and stiff it is more difficult to discern the mode frequencies, one of the useful indications that can be drawn from them is precisely being able to perceive the clearest and easiest formation of the notes, which become increasingly clear and distinct, that I think can be interpreted as a good vibratory behavior of the plate.

In this video I try to explain how to play with the grip points and the listening points for your enjoyment:), provided I managed to make people understand something, or in case you missed it.

 

Share this post


Link to post
Share on other sites
1 hour ago, Marty Kasprzyk said:

If you are a control freak I suggest trying to hit the plate weights and mode frequencies of famous violins.  

Even if you have that data, unless you are using wood with the exact same properties, it will not be possible... assuming you keep the same arching.   And, as far as I know, there are no famous violins where the wood properties are all known.  

1 hour ago, Marty Kasprzyk said:

Or you can just make violins and skip all this stuff.

Probably with the same outcome, but less frustration.

 

Share this post


Link to post
Share on other sites
2 hours ago, Don Noon said:

Although I will continue to record taptones (as it is quick and easy), in looking at 10+ years of data I think I could do just as well without it by only going with weight and thickness targets.  My target weight is ~60g or less without the bass bar... if I hit that weight and it's a lot thicker than "normal", I'll go lighter, but not less than 55g.  If I reach my "normal" thicknesses and the weight is high, I'll go thinner.  I probably have wood that allows lighter plates, so untorrefied wood might want to be 5 - 10% heavier.

Taptones for me are more "oh, that's interesting; maybe it will mean something some day"... but it's not a target to be achieved.  And so far, taptones haven't provided any great enlightenment.  Maybe endarkenment.

What would be the percent weight change (loss?) from your treatment of spruce and maple from the non treated?

Sorry just reread and saw 5-10%....Does spruce or maple change more on average?

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.