Curtate Cycloids. Again

[violins88]

I am using the Strad plans for the Titian violin. The arching curves supplied are not symmetric. To me, it seems better to make a symmetric violin. I tried using a cycloid program, but the cycloid curves do not match the plan.

Have any of you solved this problem? How to make unsymmetrical arching curves symmetric? 

 

Thanks

[FoxMitchell]

1 minute ago, violins88 said:

I am using the Strad plans for the Titian violin. The arching curves supplied are not symmetric. To me, it seems better to make a symmetric violin. I tried using a cycloid program, but the cycloid curves do not match the plan.

Sorry not the answer to your question, it's in fact another question: Where do you get the plans for the Strad? Is it something that can be shared?

[WorksAsIntended]

The best violins I played were all unsymetrical. Thinking of a violin, I think it would be nearly strange to build it perfectly symetric, you add bass bar and sound post not in the middle and therefore it will get unsymetric at this stage anyway. Also the strings are not symetric either. 

If you want to get it symetric you can use some simple spline method with symetric boundery condition or you just look for the plane and average the heights and thickneses between left and right half. 

 

[Jim Bress]

4 hours ago, violins88 said:

I am using the Strad plans for the Titian violin. The arching curves supplied are not symmetric. To me, it seems better to make a symmetric violin. I tried using a cycloid program, but the cycloid curves do not match the plan.

Have any of you solved this problem? How to make unsymmetrical arching curves symmetric? 

 

Thanks

Hi John, I think the Titian is built off (or closely matches) the P m44 form.  I think this is Addies 1705 P form.  I used the curtate cycloid program. Half templates of the cross arch seem to fit pretty well accept for the belly c-bout.  By chance, my back c-bout half template fits well so I use that for the back and belly.  Were my choices right, no idea.

-Jim

[violins88]

Jim and Fox,

Thanks for your comments.

 

Workasintended, I got my plans from TheStrad several years ago. I don't know if they still sell them.

[ctanzio]

There are two independent parameters to the curtate cycloid equations. For a violin arch, they can be recast in terms of the height of the arch at the centerline, and the width of the arch from channel to channel (the point near the purfling where the curve bottoms/flattens out).

Are you using equations cast like this? I can post them if you wish.

[Don Noon]

5 hours ago, violins88 said:

I tried using a cycloid program, but the cycloid curves do not match the plan.

Have any of you solved this problem? How to make unsymmetrical arching curves symmetric? 

My solution is to not use the arching curves and shape primarily by eye.  The problem goes away.  I know that doesn't help much, but it is an option.

[Frank Nichols]

4 minutes ago, Don Noon said:

My solution is to not use the arching curves and shape primarily by eye.  The problem goes away.  I know that doesn't help much, but it is an option.

What features/characteristics are you looking for when arching by eye? 

Curious noobs want to know

[violins88]

Hold the phone...... I just discovered that I can alter the "shape factor" in the cycloid program by Richard Thomas.

 

Now I must do trials and error to find a kind of average fit for the left and right sides of each arch.

 

The cycloid program is available on my website jpschmidtvioliins.com. Here... http://jpschmidtviolins.com/cycloid.html

It is windows based. 

[violins88]

58 minutes ago, FrankNichols said:

What features/characteristics are you looking for when arching by eye? 

Curious noobs want to know

Consider a single arch. I measure from the plan, the middle height and then the height  halfway on each side of the center. So, for 5 arches, I have 15 data points. I gouge carefully these heights. So I have 15 marked sports. Then I carefully scrape and thumb plane the surface, using oblique sunlight to see the shadows of the contour. If I don't have sunlight, I use a shaded lamp. I was critizcized for this method a couple of years ago by a known maker. But I like this method. 

I am making templates because I am teaching hobby violin making and I think templates are easier for students to use. 

[violins88]

2 hours ago, FrankNichols said:

What features/characteristics are you looking for when arching by eye? 

Curious noobs want to know

 

Frank.  I am not a trained maker. Therefore my advice is not reliable. However, I do have a concept of how the shape near the edge looks. Also, the center bout seems to be an arc of a circle. Avoid "chicken breasted" look. Also try to make a generally gentle curves.  No puddles or bumps. I emphasize using a light source, like the sun, that shows shadows.

[Don Noon]

3 hours ago, FrankNichols said:

What features/characteristics are you looking for when arching by eye? 

The short and unhelpful answer is to remove wood that doesn't look like a Strad or Guarneri or whatever arching concept you are aiming for, and that requires having that shape firmly engraved in your head.  I'm maybe halfway there, so I use some crutches:  whenever having access to good violins, take notes on where the arching is convex and where it turns concave, and how the crossarch flows longitudinally and how things interact with the F-holes.  I also take a few arching radius measurements when I can.

But the main thing for me is to look at as many good instruments as possible to memorize the shape concepts, or what might be called "educating the intuition", and then doing a bunch of extreme experimental plates to see how they work.

[MikeC]

Don since you mentioned them, are Strad and DG arches very different?  I don't have templates to compare. 

 

[Peter K-G]

This one is really close to CC all over 

[violins88]

Peter,

 

Great pics. I wonder if other makers would post theirs? 

[Don Noon]

1 hour ago, MikeC said:

Don since you mentioned them, are Strad and DG arches very different?  I don't have templates to compare. 

 

There are some characteristic differences that I notice, mostly on the top, but I'm not a real expert on them.

[Addie]

3 hours ago, MikeC said:

Don since you mentioned them, are Strad and DG arches very different?  I don't have templates to compare. 

 

My impression is they are most different just above and below the C-bouts.  But I haven't looked at this in a while...

[uncle duke]

4 hours ago, MikeC said:

Don since you mentioned them, are Strad and DG arches very different?  I don't have templates to compare. 

 

This reply may be needed for an example for Mr. Mendenhall's Sorting Thru Conflicting Advise topic but since I'm tired of watching varnish dry and MikeC. needs info. I rounded up some old templates garnered thru the use of violin making search engines including this one so any discrepancies found we'll blame on the fore fathers of violin making who chose to put them up for viewing.  

I have a set of 1733 D.G. templates and a Strad set too though the only info. for the Strad was that it has a 15.2mm belly and 14.8 back height.  Whichever Strad that is I can't say but I do have some comparisons between the two.  D.G. - 15.8 belly, 15.7 back     

First the back plates.

Template #1 upper bout - both are similar contour wise.

#2  Similar  1 5/8" each side of glue line then Strad contour widens 2mm, then 3mm gong to recurve.

#3  Strad has tighter radius by about 1.5 mm all the way to the purfling.

#4  Strad tighter contour radius, D.G. spreads contour out a greater distance or a greater radius.

#5  Same as #4 excepting from the edges 37mm inwards contours appear the same.

Now the belly plates.

Template #1 upper bout - Strad tighter radius, D.G. more flatness, outer contours close for each other.

#2  Hard to tell any difference - DG seems to have 35 to 57mm out from the glue line a tighter radius/more contour.  The recurve areas are about the same for both depending on edge height, I'd think.

#3  DG off centered, Strad slightly tighter radius thru the bridge area, DG more contour/steepness going towards the purfling.

#4  DG off centered again, about a 1/4" or so.  Strad appears to have a wider radius 50mm in from the edge.  DG appears to have a tighter radius and recurve area is more graceful.  * * #1 Strad and #4 DG templates are almost identical from bass side edge 60mm inwards in regards to outer plate contour.

#5  Strad makes more use of the wood for this area meaning a broader, flatter area of sorts blending nicely to purfling whilst the DG contour has already started turning towards the tail block - Strad contours remains straighter before contour  makes the turn.  DG recurve area looks real nice.

 

[Johnmasters]

On 5/23/2017 at 0:42 AM, violins88 said:

I am using the Strad plans for the Titian violin. The arching curves supplied are not symmetric. To me, it seems better to make a symmetric violin. I tried using a cycloid program, but the cycloid curves do not match the plan.

Have any of you solved this problem? How to make unsymmetrical arching curves symmetric? 

 

Thanks

Just graph both on good (small divisions) graph paper and take the average.  Here is the basic math which you can also graph:

The rolling circle gives the curtate cycloid above the x axis.  Ignore that factor to simplify the expresssions.  If you wish to express the width in the formula, below is the way to do it.  At width t=W,  sine( ..) goes to zero as it should at the edge.  You will have to add on your constant edge thickness, of course.  Here, h is the maximum height of the CC.

Y = t.5*h*sin(t*pi/W)
X = .5*h*(1+cos(t*pi/W))

  
    change variables:   t*Pi/W  goes to t
    The variable is now t*W/Pi

Below is the essence:  when t goes from zero to PI,  that is half of the circle's rolling.  Notice that Y is h at t=zero.

X=    t*W/pi+.5*h*sin(t)
Y=    .5*h*(1+cos(t))

To get modified shapes,  which I believe the on-line program does,  .  It is easy to see that the CC is a cosine function mapped onto a distorted X-axis.  You can further distort this X axis with a multiplier.  You could even perhaps find other ways to make an X axis.  Finally,  as I have expressed these formulas,  you must use RADIANS where 90degrees = Pi/2 radians. 

[Johnmasters]

On 5/23/2017 at 5:44 AM, ctanzio said:

There are two independent parameters to the curtate cycloid equations. For a violin arch, they can be recast in terms of the height of the arch at the centerline, and the width of the arch from channel to channel (the point near the purfling where the curve bottoms/flattens out).

Are you using equations cast like this? I can post them if you wish.

I think I did so correctly.   They refer to half the arch,  so symmetry is there.  My "W" is the half-width of the violin.  The height is "h'"

[Bill Merkel]

I thought curtate cycloids were what football helmets were made out of, but I've taken a lot of hits to the head.

 

[Michael_Molnar]

A curtate cycloid will fit some arches on some instruments. I think this is a futile attempt at force-fitting a design where it does not belong.

 

[Frank Nichols]

I know very little about math, so this may be dumb, but I thought I had read somewhere that the "old masters" used cantilever arches - and that they checked the arching by hanging a "chain" or "thread" from one side to the other and letting it droop (hang) down into the arching. If that is a old wives tail, feel free to laugh, if not, it seems a fairly simple way to get a symmetric arch.

[violins88]

1 hour ago, FrankNichols said:

I know very little about math, so this may be dumb, but I thought I had read somewhere that the "old masters" used cantilever arches - and that they checked the arching by hanging a "chain" or "thread" from one side to the other and letting it droop (hang) down into the arching. If that is a old wives tail, feel free to laugh, if not, it seems a fairly simple way to get a symmetric arch.

Frank,

The catenary will not show the point where the curve changes from upward curving to downward curving -- the inflection point. So, no, the catenary will not work. 

 

[uncle duke]

1 hour ago, FrankNichols said:

I know very little about math, so this may be dumb, but I thought I had read somewhere that the "old masters" used cantilever arches - and that they checked the arching by hanging a "chain" or "thread" from one side to the other and letting it droop (hang) down into the arching. If that is a old wives tail, feel free to laugh, if not, it seems a fairly simple way to get a symmetric arch.

The only "old master" that I know of that checked arching with a hanging chain or thread is Ken.

 

With some luck Frank you may be able to make a set of arching templates for a violin before the holiday is through.

Here are some numbers you'll need for starters.

1.  31.5 or 32 for the c bout.

2.  2x the aforementioned 31.5 -32 for the lower bout.

3.  And 1.6, or if you follow Addie, 1.5 for the upper bout. 

Most templates sets use five templates for each plate.  Number 1 above would be template three, number 2 above would be template 5 and number three above would be template #1.

If I'm not mistaken you'll need to construct circles out of thin cardboard or plastic using the above numbers I put up above.  Those are in millimeters.  I can't find the discs I made or I would tell you more.  If I find one I'll be back with more info. or someone else can come help out too.  

The idea here is to get you started on your violin, get through with that and get you to the cello making that you want to do.