3D violin scroll.

[Mauricio Sartori]

Has anyone here ever ventured to model a violin in a 3D software?

In the image below, I have a small attempt whose choice was the scroll because it is obviously the most difficult part of the endeavor. The software used is Rhinoceros 7. I also started this to try to better understand how a scroll is designed.

Of course, it's not perfect and it's a bit over the top. I'm also using this project to dig deeper into the software in question.

[lvlagneto]

I've done this a few times for different animation props, but never for something that was going to be carved/machined (not necessarily your goal). 

I'm not proficient with AutoCAD, only Maya and ZBrush.  I was usually given stringed instruments, so that we ended up with reasonable representations of things, instead of violins without bridges and inverted f holes.

I'd be curious to know how much bloat/offset the form of the scroll would need for roughing out with CNC (assuming people would hand finish).  Maybe this would interest Don.   

[Ron1]

My avatar includes about half of a violin, and was done in 3D.  I'm not sure if it will load/view here, but if so I can upload the file, which can be turned & viewed from all sides.

[Ron1]

the file size is 165 MB

[HoGo]

I've done some work in Rhino (ver.5 only) but it contained mandolin scrolls mostly :-) Like these:

[Renegade Violin]

https://youtu.be/5iP2bkFUj94

[Mauricio Sartori]

On 5/12/2023 at 4:46 PM, iburkard said:

I've done this a few times for different animation props, but never for something that was going to be carved/machined (not necessarily your goal). 

I'm not proficient with AutoCAD, only Maya and ZBrush.  I was usually given stringed instruments, so that we ended up with reasonable representations of things, instead of violins without bridges and inverted f holes.

I'd be curious to know how much bloat/offset the form of the scroll would need for roughing out with CNC (assuming people would hand finish).  Maybe this would interest Don.   

Thanks for answering. I don't consider myself proficient in any modeling software. What I used the most so far was Blender. I started using Rhino and I liked the flow of work. So I decided to learn a little more. I have no goals with machining. Just a challenge to try to better understand the proportions and something like that: how to go from 2D curves to a 3D model. For the violin spiral I used what I could understand from Kevin Kelly's method.

[Mauricio Sartori]

On 5/12/2023 at 9:34 PM, Ron1 said:

My avatar includes about half of a violin, and was done in 3D.  I'm not sure if it will load/view here, but if so I can upload the file, which can be turned & viewed from all sides.

Hi Ron1. Nice avatar. Thanks for answering. Did you do it yourself?

[Mauricio Sartori]

14 hours ago, HoGo said:

I've done some work in Rhino (ver.5 only) but it contained mandolin scrolls mostly :-) Like these:

Hello, HoGo. You really did an excellent job! Still creating things with Rhino?

[Mauricio Sartori]

13 hours ago, Spongebob said:

https://youtu.be/5iP2bkFUj94

Thanks for sharing.

[Renegade Violin]

1 hour ago, Mauricio Sartori said:

Thanks for sharing.

https://sketchfab.com/3d-models/original-stradivari-from-1737-f6a9ac1b7e7c452c99bee567c77e3906?fbclid=IwAR0yze2Wb5369ODIz-Vgx-XCQ27anGJjFmXfUFpIp9q9xx7125hc4br8yYk

[HoGo]

22 hours ago, Mauricio Sartori said:

Hello, HoGo. You really did an excellent job! Still creating things with Rhino?

Thanks. This was actually my first and only big RHino project- I pretty much learned to work with Rhino onthat. I've been working on mandolin drawings for a decade or so in 2D so I jumped into 3D just for fun. I've done some smaller projects, like designing (and actually building) a gazebo for our summer house or some instrument related forms or fixtures.

Right now I'm working on generic Plowden del Gesu model. Still in Photoshop/ Illustrator phase preparing full size photos and CT scans and correcting any distortion caused by photography. I will eventualy go to 3D with basic shapes of scroll and arching just to check for any errors.

BTW I've studied geometry of few scrolls and I see some of the drawing theories are a bit far-stretched. There are simpler methods, IMO. But for my model I will start with the original and try to detect worn areas and also where the maker carved curves a bit smoother where original possible circle curves meet. Simple half/ quarter arcs (aligned with the axis of neck) will give you great approximation and in some well executed and preserved scrolls they are pretty obvious choice. Of course we don't know whether these scrolls are produced right from the drafted "master" pattern or just tracing of a tracing of a... which blurries the original drafting methods.

[Mauricio Sartori]

57 minutes ago, HoGo said:

Thanks. This was actually my first and only big RHino project- I pretty much learned to work with Rhino onthat. I've been working on mandolin drawings for a decade or so in 2D so I jumped into 3D just for fun. I've done some smaller projects, like designing (and actually building) a gazebo for our summer house or some instrument related forms or fixtures.

Right now I'm working on generic Plowden del Gesu model. Still in Photoshop/ Illustrator phase preparing full size photos and CT scans and correcting any distortion caused by photography. I will eventualy go to 3D with basic shapes of scroll and arching just to check for any errors.

BTW I've studied geometry of few scrolls and I see some of the drawing theories are a bit far-stretched. There are simpler methods, IMO. But for my model I will start with the original and try to detect worn areas and also where the maker carved curves a bit smoother where original possible circle curves meet. Simple half/ quarter arcs (aligned with the axis of neck) will give you great approximation and in some well executed and preserved scrolls they are pretty obvious choice. Of course we don't know whether these scrolls are produced right from the drafted "master" pattern or just tracing of a tracing of a... which blurries the original drafting methods.

Good to know. I just started with Rhino and the violin is my biggest challenge so far.

How do you get the full size photographs?

I don't have easy access to books and importing them would cost me a lot. Are there any public documents that could help me with geometries and drawings? Do you have any tips in this regard?

[HoGo]

Search the internet. There is vast number of bad pictures but sometimes gems occur. You can search for measurements of violins (that will also vary a bit and you have to judge the source for precision of measurements)

The LOC has nice pictures and some of them also contain measurements, Tarisio has some pics that look like the ones in Biddulph book etc. Some makers posted nice good pictures on their websites and the Strad 3D project is online. You can find several CT scans or parts of them online and If you can work with these they can be great resource. Even with their low resolution in Photoshop you can resize them pretty correctly to full size and they often have the least distortion so you can use them as crossreference for photographs. Sometimes even the tiny poster previews resized to full size can be used as the "master reference" to be able to find good photographs to assemble the violin from. Always check all the dimensions (and their possible relative errors) before resizing. I use excel sheets to claculate best possible factor/s for resizing from at least four basic measurements and check calculations to detect distortion. If I find that widths do match but length doesn't there may be either some scanning imprecision or perspective distortion, both of them can be corrected if not severe and you can identify what you are looking at.

The most important is to select the few pics that are good enough to work with, they are not always the nicest looking or sharpest. Some of thenicest pics have way too much distortion from improper photography that they are useless for taking outlines (often you'll find out only  after you try unsuccesfully to match published measurements and other photos of the same instrument).

Think about what low resolution does to the original picture and resizing back to big blurries (especially with CT scans). Think about distance from which pics were taken - it's unknown to us but some details give you hint that it was taken from close or far distance. Scrolls especially are often taken at close distance that shows nic edetails but the turns of the volute get exagerrated measurably (and if you have measurements from poster or such) the enlarged picture will never match the numbers anywhere close enough.

I've been doing this on and off for two decades mostly with mandolins or other instruments. Now I got to violins again but my time doesn't permit sitting for several hours daily at the PC.

I see that Rhino7 has some new features in surface editing/creating. I had pretty hard times especially working on final surfaces of plates on my Rhino5.

[Mauricio Sartori]

42 minutes ago, HoGo said:

Search the internet. There is vast number of bad pictures but sometimes gems occur. You can search for measurements of violins (that will also vary a bit and you have to judge the source for precision of measurements)

The LOC has nice pictures and some of them also contain measurements, Tarisio has some pics that look like the ones in Biddulph book etc. Some makers posted nice good pictures on their websites and the Strad 3D project is online. You can find several CT scans or parts of them online and If you can work with these they can be great resource. Even with their low resolution in Photoshop you can resize them pretty correctly to full size and they often have the least distortion so you can use them as crossreference for photographs. Sometimes even the tiny poster previews resized to full size can be used as the "master reference" to be able to find good photographs to assemble the violin from. Always check all the dimensions (and their possible relative errors) before resizing. I use excel sheets to claculate best possible factor/s for resizing from at least four basic measurements and check calculations to detect distortion. If I find that widths do match but length doesn't there may be either some scanning imprecision or perspective distortion, both of them can be corrected if not severe and you can identify what you are looking at.

The most important is to select the few pics that are good enough to work with, they are not always the nicest looking or sharpest. Some of thenicest pics have way too much distortion from improper photography that they are useless for taking outlines (often you'll find out only  after you try unsuccesfully to match published measurements and other photos of the same instrument).

Think about what low resolution does to the original picture and resizing back to big blurries (especially with CT scans). Think about distance from which pics were taken - it's unknown to us but some details give you hint that it was taken from close or far distance. Scrolls especially are often taken at close distance that shows nic edetails but the turns of the volute get exagerrated measurably (and if you have measurements from poster or such) the enlarged picture will never match the numbers anywhere close enough.

I've been doing this on and off for two decades mostly with mandolins or other instruments. Now I got to violins again but my time doesn't permit sitting for several hours daily at the PC.

I see that Rhino7 has some new features in surface editing/creating. I had pretty hard times especially working on final surfaces of plates on my Rhino5.

Thank you for the tips. I'll look into it more carefully.

I'm not familiar with previous versions of Rhino, but I imagine a lot of good stuff has been added. I was more used to meshes and the change of approach to NURBS and booleans takes a while, at least for me. I feel that Rhino still lacks some features, but nothing that can't be worked around with some strategy. For filleting, for example, I find it very limited.

[HoGo]

I was thinking about Fusion but I already spent hundreds of hours at Rhino and don't have the extra time to switch.

[Mauricio Sartori]

34 minutes ago, HoGo said:

I was thinking about Fusion but I already spent hundreds of hours at Rhino and don't have the extra time to switch.

I've already tried a little of Fusion 360. I liked it a lot, but it has advantages and disadvantages. The way of working changes significantly. I like the layers and snaps in Rhino. On the other hand, the possibility of something parametric, I don't know to what extent this is valid in the F360, is very useful. Measure definitions are not better in Fusion because they are dynamic.

I think it would be worth your while to try it out without compromise - you'll like it.

[Ron1]

On 5/15/2023 at 6:02 AM, Mauricio Sartori said:

Hi Ron1. Nice avatar. Thanks for answering. Did you do it yourself?

No, I worked with a very capable digital sculptor in Bucharest, Romania.  Super easy to work with, very reasonable pricing, and she's fluent in english.  She worked from photos I sent her of my violin.