#5 and #6 under way

[Don Noon]

I have managed to get some work done on the next set:

On the left is something different for me: naturally aged, unprocessed wood. The top is 20 yr old Engelmann, which I am using unprocessed because the measured properties were very good. The back and sides are 40 yr old wood from a retiring maker, European maple I think (written "Goetz" on it... from the German maker?). It's gorgeous, which made me extra-perturbed when I trimmed 40% into the button area before I realized I forgot to lay the button out. Oh, well... this will be my opportunity to impress you all with a fabulous undetectable button splice. Thank you, Matthias, for leading the way. The top and back are both on the thin side, so I will make the ribs 1 or 2 mm taller to compensate for the lower arching (14 mm looks like the most I can get).

On the right is the beginning of the new Snakefiddle, using the neck and "scroll" from the original. I expect to keep this one as my personal playing fiddle, as it will be too odd to sell. I also hope it will be killer sound-wise, using my ultra-low-density processed top with a radiation ratio approaching 18. The back and sides are also from the stock of the retired maker... bigleaf, 20 yrs old, originally purchased from Strobel.

Some other things I will be doing differently this time:

- arching templates, cycloids for cross arching. Darnton influence, to some extent. It DOES match amazingly well to Cremonese stuff, and looks good. For consistency and record-keeping, I think I should be using templates.

- top plate stiffness/thickness: Previously, I had been going quite light and thin; there might be some benefits to a less extreme top.

... shoulda done this before. Might have saved the button splicing project.

Specialized lining channel chisel, made to exactly the width of the lining. HSS blade made from part of a jointer blade. Handle is a walnut towel peg. I really like this tool, and maybe over my remaining lifetime I might recover the hours invested in making it. If not, so what? The pleasure of using nice tools is worth it.

[COB3]

"... shoulda done this before. Might have saved the button splicing project."

Don, I can't tell what your template is made of-- can you splice a "half-button" onto the template, and avoid a repeat?

It might be worthwhile.

template_repair.bmp

[Don Noon]

The template didn't have the engraving or red highligher on it before... I think that (and the memory of the splicing) will be sufficient to wake me up in the future.

The template is gold anodized aluminum (door kickplate).

[COB3]

Yeah, splicing anodized aluminum would not be worth the effort. From the photo, I thought it might be door-skin plywood, but I wasn't sure. Door kick-plate material sounds great. I'll have to remember that.

BTW, even if you are tempted to do so, don't try welding that anodized stuff-- it will turn the anodizing in the heat affected zone to a dead black, I am told.

But you are right-- I doubt you will forget again. I have done this, and have come close to a repeat, but caught myself (same template). Since then, I make all my templates with a button. The notches, and drilled hole are so you can lay out the front plate.(minus the button). Making a detection-proof button graft is fun...but not the kind of fun you want to do more than once or twice.

[Bill Yacey]

I know exactly what you are saying about your own hand made tools. It's almost like you form a bond with your tools and reminisce every time you use them.

I use the same template for top and back plate with a button. It's easier to remove the button from the top than trying to remember to allow for it when cutting the back outline. I use plexiglass for my templates.

Do you have any close up photos of your spindle sander guide for trimming the plate edges? I made one and am curious to see if we both had the same approach.

[Don Noon]

<!--quoteo(post=477045:date=Aug 26 2010, 10:45 AM:name=Bill Yacey)--><div class='quotetop'>QUOTE (Bill Yacey @ Aug 26 2010, 10:45 AM) <a href="index.php?act=findpost&pid=477045"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Do you have any close up photos of your spindle sander guide for trimming the plate edges? I made one and am curious to see if we both had the same approach.<!--QuoteEnd--></div><!--QuoteEEnd-->

I trimmed away the lower part of the PVC guide tube... to get the contact point nearer to the plate being trimmed. I also needed to wrap some tape around the tube to get the diameter bigger.

[Don Noon]

Just a status update...

#5 scrollwork is started and the body is closed temporarily. The top is glued on very lightly, as I expect I might want to do adjustments (possibly change the bass bar) when I find out what is going on acoustically. There are so many new/unusual things in this one:

- Top arching is low: 14 mm - due to a thin billet

- Back wood is very high density (haven't measured it yet, but it was very obvious)

- Back arching is slightly low: 14.5 mm - also a thin billet

- Ribs height was increased ~1 mm to compensate for low arching

- Back thickness and weight left heavy (121 g)

- Cross arching has more recurve/scoop at the edges to allow tighter arch in the middle

- Bass bar is from high density, processed Sitka, and made thinner and lower

- Graduations were given some experimental asymmetries to try to get more acoustic output from the CBR resonance and reduce the output of a common resonance around 950 Hz.

So I'm delaying the arching and graduation of #6 until I see what happens with #5.

The deep oxidation of the old back wood doesn't look too bad in the final arching, just a little "burnt looking" in the center. I'm pretty sure I can shade that out in varnishing.

The grafted-on button came out nicely, especially on the left side where it lines up well with a grain line. Couldn't quite get both sides to come out like that, but it should be OK.

[Torbjörn Zethelius]

Just a status update...

#5 scrollwork is started and the body is closed temporarily. The top is glued on very lightly, as I expect I might want to do adjustments (possibly change the bass bar) when I find out what is going on acoustically. There are so many new/unusual things in this one:

- Top arching is low: 14 mm - due to a thin billet

- Back wood is very high density (haven't measured it yet, but it was very obvious)

- Back arching is slightly low: 14.5 mm - also a thin billet

- Ribs height was increased ~1 mm to compensate for low arching

- Back thickness and weight left heavy (121 g)

- Cross arching has more recurve/scoop at the edges to allow tighter arch in the middle

- Bass bar is from high density, processed Sitka, and made thinner and lower

- Graduations were given some experimental asymmetries to try to get more acoustic output from the CBR resonance and reduce the output of a common resonance around 950 Hz.

So I'm delaying the arching and graduation of #6 until I see what happens with #5.

The deep oxidation of the old back wood doesn't look too bad in the final arching, just a little "burnt looking" in the center. I'm pretty sure I can shade that out in varnishing.

The grafted-on button came out nicely, especially on the left side where it lines up well with a grain line. Couldn't quite get both sides to come out like that, but it should be OK.

Hi Don, It's looking good! How does your "arching gage" work? The sawed off button is a once in a lifetime type of mistake. Let's hope...

[Don Noon]

Hi Don, It's looking good! How does your "arching gage" work?

I carve with the plate screwed to a turntable, so it's inconvenient to unscrew it to use a caliper. This gage straddles the center bout, you use a tapered stick feeler gage to measure the gap between the plate and the middle of the gage. Subtract that from 20mm, and that's the arch height.

[Don Noon]

Finally got back to the shop, after VSA trip and subsequent few weeks of back and sciatic problems which prevented sitting or standing for more than a couple of minutes at a time. And progress ensued.

#5 playable in the white. As usual, I am creating a little more work for myself by getting it together before doing the detail work on the neck and body, but I was impatient to see how it works acoustically. And the answer is:

It's a bit early to claim success, but so far, it looks and sounds very good. I'm sure Anders will notice that the L value for this fiddle appears to be ~22.5, which I think might be due to two things: Higher A0 from flatter plates and more edge recurve in the arching (maybe taller ribs help?), and reduced peaks in the "transition hill"... by adding some thickness to the antinode that seemed to be the worst offender (no, I didn't add wood... just used info from other sources to make my best guess where to leave it thicker).

I also tried something I thought would increase the CBR acoustic output; can't say if it worked or not, but at least the CBR response is visible. On the previous plot, the B1- peak looks a bit lopsided; at higher resolution a split is visible between the CBR at 417 Hz and the B1- at 435 Hz. The CBR frequency is quit high, likely due in part to the taller ribs.

The high-frequency area looks a little weak and rough at this time; 20 minutes after first tune-up I wouldn't expect anything fabulous yet.

So much for analysis. PLAYING it, I think it's going to be very good when it opens up. The brightness and sparkle isn't there yet, but it is very strong and even everywhere. The D string sounds particularly nice to me.

Of course, this could all change with varnishing and time. But, so far, so good.

[Wm. Johnston]

I also tried something I thought would increase the CBR acoustic output; can't say if it worked or not, but at least the CBR response is visible.

So far my testing is showing that stiff violin backs produce a violin with a fairly strong CBR. By stiff I mean Cannone level thicknesses, thinning the back seems to reduce the violin's CBR output very quickly. As is usual in life there is a catch, the stronger CBR is at a higher frequency than the CBR of a more normal violin. For a stiff back the CBR amplitude and frequency ends up close to where the B1- is in a violin with a thin back so you don't actually gain low frequency response.

[Don Noon]

I will have to take the top off later and re-glue it (I used extremely thin glue, as I was fairly sure I'd have to fix something), and I might take some wood off here and there to try to reduce the CBR frequency without affecting the other resonances too much. Probaby not going to work, though, and I don't want to get too radical... it sounds pretty good the way it is.

[jezzupe]

http://www.strad3d.org/demo/modalvid.html

If we look at the graphs, we see the "red" violin. To me the most could be learned from this violin. I wonder, do they give dimensions and specs for this "bad" violin?

I find it funny, I had never seen this before, yet many of the animations are just as I would have imagined, and many of the things they say are things I have said before about the motion of air and such.

To me your probably going to end up with a better sounding one than the previous if you have lowered your arch.IMO a slightly lower arch will be weaker and if supported right with the bassbar will add to the "warble" effect that helps billow the air in and out. I am not convinced that a thick back is going to always be a good thing. I will say again that to me this motion we seek is much like two people of a trampoline trying to always hit the sweet spot as they go up ad down in order to get the best bounce. I think as we think of the violin in motion we should think of it much like these animations, very exaggerated and slowed way down, at least I do.

To me there are several motions that we are trying to harness

1. up and down

2. the side to side "slosh" {as if in a bathtub moving the water back and forth east to west

3. the back and forth "slosh" {same thing north to south, think childhood, moving about in the tub}

4. the rug snap recoil, this of two people hold the end of a blanket, one snaps his end the the other in the other direction in rapid succession, or like a "rolling" earthquake

5.the "twist" as if trying to twist the violin in half one way, the the other

6. the teeter totter snap, each end, the top by the neck block and the bottom by the end pin get a "snap" right at the end at the edge of the nodal area, kinda like a rocker valve in a car.Really this happens all the way around the edge, thus the importance of the re curve and how it works with what is going on right underneath it on the inside.

To me, beyond the actually consistency of the wood {mass, density. radiation, stiffness, grain/cell structure} what we are trying to exploit is the woods flexibility.

There is that short video Of Sam Z. working in his shop, now, while he goes into nothing of detail about his work, they show one shot of him twisting his plates, THAT is the key, or a darn big part of it. Or understanding how symmetrical/asymmetrical flexibility related to thickness/thinning of wood in select areas influences the above listed motions.

If I then start thinking of building bad sounding and or wrong violins I try to think of what motions we don't want and what types of dimensions and where they are that would hinder proper motion.

Rib thickness is more important to me than slight variations in height, taller ribs may add to the swaying a bit more but could also become to flabby and end up as a trampoline "hole" or "arms out front with the mitt, easing the fast ball in when catching by bring back the arm as the ball contacts the mitt"

I like top 1.5/6, middle C's 1.2/3, bottom 1.7/8 mm, this makes the middle like a hinge point for the slightly stiffer top and bottom. The thinner top than the bottom is good because the tops bouts are less span, therefore slightly thinner ribs

allow for a little more flex while the thicker bottom with the wide span makes it "shake its booty" with more "junk in the trunk". Both however are thicker than the middle and therefore exert more stress to the middle C bouts area.This is very helpful for the "twist" and the "rolling earthquake". It helps send it back one end to the other.

I hope you post sound samples of some of them so we can do a side by side listen. Maybe 3.4.5 and 6.

[JimMurphy]

I like top 1.5/6, middle C's 1.2/3, bottom 1.7/8 mm, this makes the middle like a hinge point for the slightly stiffer top and bottom.

Jezzupe,

Are those numbers typos? Those values are awfully thin compared to the best Old Master fiddles which have had 'trouble' with fallen arches, bulging LowerBouts, etc.

Jim

[James M. Jones]

Jezzupe,

Are those numbers typos?

I think juzzupe is talking about rib thickness/ michael

[Don Noon]

Those are either really thick ribs or really thin plates. And who ever uses fractional metric units? Haven't we evolved to decimals yet?

[jezzupe]

Rib thickness,

upper bout area 1.5 to 1.6 millimeters

middle C bout area 1.2 to 1.3 mil,

lower bout area 1.7 to 1.8 mil'

I hope that clarifies, no where did I say "plate thickness or graduation", not sure where the confusion came from

I feel "even" 1.2 to 1.5 ribs and or consistant rib thickness is a bad idea.

[James M. Jones]

Those are either really thick ribs or really thin plates. And who ever uses fractional metric units? Haven't we evolved to decimals yet?

I think he means to say 1.7 to 1.8 mm

[jezzupe]

Those are either really thick ribs or really thin plates. And who ever uses fractional metric units? Haven't we evolved to decimals yet?

I never really evolved, I'm much more of a spawn'er. the 1.7/8 area is reserved for the end block area and about 1.5 inches from it in both directions

[JimMurphy]

I hope that clarifies, no where did I say "plate thickness or graduation", not sure where the confusion came from

'My bad', Jezzupe. Variable rib thickness can be interesting.

Thanks,

Jim

[Don Noon]

Just a note to myself: almost all varnish disasters are derived from trying to speed up the process.

Therefore #5 will experience a slight delay while I strip off and re-do the varnish. In this case, there were 2 instances of trying to do too much with one coat; the first oopsie was annoying, but correctable, the second was complete disaster. The color was coming out very nice, though.

On the positive (?) side, the stripping left a nice looking afterglow, with a suggestion of age. This has opened up the opportunity (or necessity) of making this a lightly antiqued finish, which it wasn't going to be originally. I think it will actually end up better than if the disaster never happened ... if I take my time.

[Don Noon]

#5 IS COMPLETE!

I rather like the look of this one, much more interesting and refined than the RED of my first four. I didn't use any red nano iron oxide this time, but I did use some of the yellow nano iron oxide in the first color coat. The rest of the color comes from zinc and aluminum rosinate/alizarin (Michelman resin M based) and Gilsonite (a hard asphaltum). It was all cooked into the varnish, not a cold mix.

I didn't try to get too real or complicated with the antiquing, just a bit here and there, with some shading. I'm happy with the result, although all the stripping and messing around may have contributed to more crap than usual in the varnish. It kinda fits the antiquing, though.

The button splice came out fine; not totally invisible, if you know it's there and look for it, but not glaring.

I got a bit of a scare when I put on the junk strings for fit-up, and (natrally) had to play it... the A string was horribly weak. I had accidentally grabbed a wound E string and put it on instead of the A.

Sound-wise, it is quite enjoyable to play, with obese (that's really fat) G and D strings. It's even and easy to play, but nowhere near as bright and penetrating as the ones I made from processed wood. So it may take a while to open up, and then it should be even better.

For the technogeeks, the CBR resonance is just under 400 Hz, and shows up as a slight lump on the side of the B1- resonance in the spectral response plot. This seems to give the open G string some much needed fullness. In the "transition hill", everything appears to be pretty even and spread out, with no big peaks. I think that's a good thing.

[match]

Congrats, looks very nice! Nice curves and sharp lines. Enjoy the playing!

[Roger Hill]

Very nice work, Don. I really like the varnish. what pattern did you use for the outline/ff holes?

[catnip]

Very nice indeed! I like your corners and edge work. I have finished #7 and #8 in the white but up here in Canada it is too cold to varnish in the basement without ventilation in the winter. I also like your golden brown amber finish. Congrats!