Machine and CNC Carving Opinions

[mrbadger]

How common a practice is machine carving (duplicator or CNC) of violin plates among individual makers? Are there any ethical  issues involved? What do folks think of the practice? Does anyone on here use them?

[Johnmasters]

How common a practice is machine carving (duplicator or CNC) of violin plates among individual makers? Are there any ethical  issues involved? What do folks think of the practice? Does anyone on here use them?

Why does it matter if you get what you want ?  Comparing a CNC with a duplicating router,  I have come to a conclusion.  I built a CNC and canabalized both a violin and cello duplicator.  And it works great.  I programed to have it cut curtate cycloids.

 

The CNC is slow as it carves only about 2-3 mm per pass.  If I intend to make a lot of instruments with it,  I would invest in some more linear bearings and rods to make a duplicator.  I could make a pattern on the CNC and use that on a duplicator to save time.  It is much easier to hog on a duplicator.  Then put the work back in the CNC for a final trim of a mm or so.

 

The people who complain that it is not "hand-made" are the sorts who use a handsaw instead of a bandsaw.  Besides I made my own design for a  CNC and wrote all the programing.  I don't think I can be criticised for using it.

[James M. Jones]

As a sort of tool junkie , with a bigger hammer complex , the things offer a certain ..fascination to me , I don't have one though , but can see a certain viability in there use as production tools ...the large company's are using 8 head routers ...

I am pretty sure quite a few individual makers use them to some extent , and a few here on MN . I know Steve Sirr and John Waddle in Minneapolis did quite a bit of development with the tech ,using CT scans to generate inputs . ...The Betts Strad.as I recall.  I saw it and it was fantastic. That said I have also seen more than some CNC products that was not so fantastic ...as in pretty bad . the problem lies in the Gcode programing where model sections are created , If, lets say, one were to use geometric constructions and CC curves , the results can come out rather sterile, and exhibit bends where various curves come together. mechanical and  precise and with out a human touch .  

 The CT scans and touch probes offer a translation of the human input by capturing the very small nuance of construction ,the odd corner , the bit of hollow in the wings , the slightly lopsided shoulder. the dropped treble f , the irregular purfling groove ect that all work together to tell a story .  however the programs still need to have some modifications made so that things like raised grain lines ,scratches and wear of the soft lines are not translated into a new plate resulting in miss matched grains to wear patterns and the like , also distortion over time should be factored in as well , an arch that starts out @16 ...moves to 17 and gets reproduced the  turns into 18 .... and other dimensions become similarly affected as well . So in short it seems to be a craft in and unto it's self , and one that does not circumvent or replace any of the other skill sets associated with Violin making .

 Personally I do not see any ethical concerns on the face of it , however making statements like "hand carved" ," unique" and "one off", become targets for criticism.

  We have a saying in iron work that makes the point ,....who swings the hammer .makes the work... it is possible to do very good work with high tech approaches, equally possible is to do bad.  

  These days I hear a lot of talk about a makers movement .... a renaissance of sorts, where traditional skill are becoming highly sought after for the ...for lack of better terms ...'soul' embodied .  ,from spoon carving and treadle lathe , to birch bark canoes and  hand made prosciutto, people with the $$$$ seem to want to have at least something that was made by a skilled artisan/craftsman.  

[DGV]

There is only an ethical issue if you call something 100% hand-carved when it is not.  Otherwise, I have no problem with roughing out plates with a machine though I still haven't found one that I can afford and does a good job.

 

I am using a purfling router to save my hands since I need to play violin as well.  I don't see why it should be a problem as it does take some skill to use a purfling router properly.

[kevin Prestwich]

And don't forget about the hand cut artisanal ice cubes. $1 a cube, adds great quality to any cocktail. I see no shame in using a duplicator to do roughing out work. Saves lots of pain on maple cello backs. That said, I made a violin a few years back with no electricity whatsoever and while entirely naked. I even heated my homemade bending iron in the vent ports of a wood stove. I think it depends on what your goals are. As James mentioned above the successful use of these machines is a craft unto itself and getting it right can take considerable time and effort. There are a few people out there that can help you with the programming of a CNC machine. If you are not so technically inclined seeking out such a person may be very helpful.

[Conor Russell]

Years ago, I bought some cello plates, pre routed to about 15mm thick, from a retired maker. They'd been cut in the late seventies I think, and they fitted my Montagnana form. The thing I liked most was the colour of the wood, darkened through, and I've thought about getting a duplicator, and seasoning cello wood this way.

 

There's so little work in a violin plate that I wonder would a router save any time at all.

 

I've been making a very tight and accurate duplicator for patches. I think that the time saved by making a cast of the patch bed and copying it will be enormous, and I have a lot of patches to do, some of them very big. 

[Michael Richwine]

I don't think you can do anything close to finished work on a CNC or a duplicator.  I have quite a bit of experience around them, and if nothing else there is a phenomenon called cutter compression, like the ripples left by the planer or molder in commercial lumber.  These can't be taken out with water, but need to be scraped or cut away. They look awful under varnish.  Also, there are nooks and crannies on the scroll that are really difficult to reach with normal tooling.

 

Best bet for me would be to use the CNC to establish the basic outline and arching to within a millimeter or two, and clean the surfaces, sweeten the curves, and do the final graduation by hand.  You still end up with sort of a monotonous sameness from one instrument toanother that way, due to greater uniformity, but that's a plus in commercial instruments. In essence, you take out the trash with the machine (mechanized apprentice), and do the work that counts by hand.

[DGV]

A mechanized assistant/apprentice is exactly what I need.  

[Ron1]

These responses are primarily from makers, and represent that viewpoint.  I  think responses from players might be different, and responses from dealers and collectors might show an entirely opposite viewpoint.  It's kinda the makers vs the market.

[DGV]

I am sure there are people who hire accountants who do all calculations by hand.

[nathan slobodkin]

My own experience is that machine roughing leaves you to close to finished to use your big tools and too far to scrape.  I Also find that I don't really know what to do with the wood until I start roughing the plate. The feel of the wood under the gouge helps me decide how high and what kind of arch andI  have an idea of how thick to make it as well.

[Davide Sora]

There is only an ethical issue if you call something 100% hand-carved when it is not. 

 

 

My own experience is that machine roughing leaves you to close to finished to use your big tools and too far to scrape.  I Also find that I don't really know what to do with the wood until I start roughing the plate. The feel of the wood under the gouge helps me decide how high and what kind of arch andI  have an idea of how thick to make it as well.

 

I agree.

[Davide Sora]

These responses are primarily from makers, and represent that viewpoint.  I  think responses from players might be different, and responses from dealers and collectors might show an entirely opposite viewpoint.  It's kinda the makers vs the market.

 

 Interesting point of wiew.

In your opinion, what they think about this topic?

I think it would be more interesting to know the views of the "market" rather than that of violin makers.

[jim mcavoy]

the cnc  is basically a dumb device  ~  (in the end) it will only do what you tell it to do  so think about what is being asked

 

Jim

[mrbadger]

Are there any aesthetic questions involved when using a duplicating machine or CNC router for the plates?

[steveperry]

The aesthetic question is an interesting one.  I can see a production argument for just cloning something you like.  Not much to do there except figure the constraints.  How close can you cut, what cutter leaves minimum print through to the final surface after cleaning, production type thinking.  Gets into size of cutter, amount of force allowed on cutter and so on.  The design is left out of it.

 

I'll make up something different.  The machine is set to allow mathematical manipulation of everything.  The art then becomes the design process, but it's still a bit cold.

 

Another scenario.  The machine, whatever it is, blocks out the highest arch and lowest interior of plates.  So the channel area is pretty close, with the plate getting thicker and thicker towards the middle.  Then the worker starts cutting from outside and maybe inside to match that piece of wood.

 

Another scenario.  The designer works with planes and lines and a specific arch this way and that. In response, develops a series of patterns that duplicate what he would do typically.  Piece of wood and design concept show up in mind.  Looks at low, medium & high arches with short medium & long central runs.  Picks pattern that cuts that long arch.  This could be done, for example, with twin patterns on each side of the plate and a router used to profile it.  Or with similar but upside down on a router table.  Or by marking distances and setting up a Wagner power planer.  

 

Now that I think of it, patterns could be used to develop a symmetrically faceted arch.

 

Where in that complex array of possibilities is not allowed to be artistic?  In my world, the purchase of a "good" instrument and the duplication of it's components by machine without consideration of why and where one is heading has little or no aesthetic value.  On the other end, using equipment to set up a standard array of symmetrical points and then working from those seems perfectly acceptable.  I have made arches this way, only using a plane and chisels to make the facets and lines that guide the arching.

 

This might be more of a philosophical question.  Regardless, it's worth considering the aesthetic.

[Christopher Jacoby]

You can certainly do finished surfaces with a CNC built to do so, and the right bit. Many makers do arch, channel, ff hole, edgework and outline in one go. I think the knowledge you get running a thumbplane over the long arch of a piece of wood informs the choices you make to have a great-sounding instrument. I'd happily use plates routed out a couple of mms too fat everywhere, on a large arch beneath that, and work down. 

[Peter Lynch]

I do almost everything by hand / without power tools except rough band saw work. A hand cut purfling channel looks much more interesting  and adds to the effect I am after.   I cook my own varnish and make purfling by dying pearwood in a traditional way.  I advertise my work as such and at least some clients appreciate this  way of approaching it.   To say this is as silly as making expensive hand crafted ice cubes, well to be honest I don't even know how to respond to that.  On other threads you hear the idea that there are many makers with machines "hidden in the back of their shop".  You have to ask yourself why this is such a universal phrase and why everyone reading it nods and understands what is being said. 

 

I think that this discussion should also include David Pye's ideas of "Workmanship of risk" vs "workmanship of certainty" as well as other ideas he writes about.......

[James M. Jones]

I really have got to get a copy of that book.

  One thing in my mind as I try and grow up , is the power of story , for the most part , it seems like people connect to a good one.

[Bill Yacey]

 Personally I do not see any ethical concerns on the face of it , however making statements like "hand carved" ," unique" and "one off", become targets for criticism.

 

I've never seen a violin carved by hand. Invariably the hand always held a saw, gouge, plane or scraper, a tool of some sort. The quality of the work is determined by how well the hand and eye manipulates the tool. So it goes for using powered machinery.

[Johnmasters]

There is only an ethical issue if you call something 100% hand-carved when it is not.  Otherwise, I have no problem with roughing out plates with a machine though I still haven't found one that I can afford and does a good job.

 

I am using a purfling router to save my hands since I need to play violin as well.  I don't see why it should be a problem as it does take some skill to use a purfling router properly.

 

There is only an ethical issue if you call something 100% hand-carved when it is not.  Otherwise, I have no problem with roughing out plates with a machine though I still haven't found one that I can afford and does a good job.

 

I am using a purfling router to save my hands since I need to play violin as well.  I don't see why it should be a problem as it does take some skill to use a purfling router properly.

Yes, I rough out the CC only,  The finish job still needs removal of tool marks.   The edges, f-hole wing scooping and purfling are all done separately.  Mike makes no secret of having a small CNC and does all, including the purfling groove.  I don't know where he gets his patterns.

 

Mine programs  the CC equations and calculates cross-sections every 1/20" (for the final cut) with 100 transverse points.  The points are connected by a sloped line,   not stair-step fashion,  and that is quite sufficient for a smooth cut.  I wind up with G-code that has over 20,000 points to pass through.

 

I was operating under Michael Darnton's idea that the arching was the most important aspect of the plates.  It is true that the guitar-shaped line which the inflections pass through is difficult to place by eye.  It is very difficult to see where the curve goes flat from a picture of the transverse arch.  I don't know how important it is,  but where the arching cross-section goes flat would have a unique property.  Bending caused by compression should go to zero as curvature goes to zero.   (Or at least minimize.)  It would be like observing the buckiling of a curved stick compressed from the ends.

 

With FEA experiments on domed disks, of varous radii shapes,  it seems that the inflection moves the least when compressed in the radial direction.  I believe the compression deformation tends to have somewhat of a boundary at this curve on the surface. It might tend to minimize sheer in the centers of the plate.  There is SOMETHING sensitive in arching....  I believe it is connected to the locus of zero-points of curvature generated by the CC.

 

Nobody wants hand-carved.  Whatever that means.  It is mind-hand carved.

[Johnmasters]

I don't think you can do anything close to finished work on a CNC or a duplicator.  I have quite a bit of experience around them, and if nothing else there is a phenomenon called cutter compression, like the ripples left by the planer or molder in commercial lumber.  These can't be taken out with water, but need to be scraped or cut away. They look awful under varnish.  Also, there are nooks and crannies on the scroll that are really difficult to reach with normal tooling.

 

Best bet for me would be to use the CNC to establish the basic outline and arching to within a millimeter or two, and clean the surfaces, sweeten the curves, and do the final graduation by hand.  You still end up with sort of a monotonous sameness from one instrument toanother that way, due to greater uniformity, but that's a plus in commercial instruments. In essence, you take out the trash with the machine (mechanized apprentice), and do the work that counts by hand.

Yes, that is right.  But still,  marking the shape of a CC transverse curve (if that is what you want) is done well if you plane off a uniform layer of tool marks.   A CNC is time-intensive as opposed to a duplicating router.  That is why I said I would use the CNC to make patterns that had the shape I wanted.  It is more of a precision curve generator than a roughing tool.  You can make your own CC curves for transverse templates,  but maybe a good idea to make a dozen,  and not just five. 

 

And use the equations and a math function plotter instead of a wheel with a pencil hole along the radius.  You will need some kind of program to plot these.

[Don Noon]

I don't think you can do anything close to finished work on a CNC or a duplicator.  I have quite a bit of experience around them, and if nothing else there is a phenomenon called cutter compression, like the ripples left by the planer or molder in commercial lumber.  

 

You can certainly do finished surfaces with a CNC built to do so, and the right bit. Many makers do arch, channel, ff hole, edgework and outline in one go.

 

I haven't seen any cutters that can cut cleanly at shallow angles against the grain, and that HAS to happen somewhere in the toolpath of CNC cutting.  There's always some amount of raggidy edge, which is not a big deal for purfling grooves, but a big deal for trying to finish the edgework..  I have used fresh, sharp spiral carbide bits running at high speed; perhaps that is not the "right bit", but I'd be interested to know what's better.

[Johnmasters]

I haven't seen any cutters that can cut cleanly at shallow angles against the grain, and that HAS to happen somewhere in the toolpath of CNC cutting.  There's always some amount of raggidy edge, which is not a big deal for purfling grooves, but a big deal for trying to finish the edgework..  I have used fresh, sharp spiral carbide bits running at high speed; perhaps that is not the "right bit", but I'd be interested to know what's better.

Yes again.  But the scaring of a tool is less than a mm deep if you make a fine cut for the finish.  My toolpath is just a series of transverse cuts just as would be generated by 200 and some arching templates.  I leave about 2 cm around the edges for edges and purfling.

 

As to "sterile"  I don't get that.  A nice arch is a nice arch.  Nobody is going to compare two of my instruments unless they buy two of them.  (Never had that problem) 

 

Ironically,  I spent half of 1911 on the machine and software and have made only one set of plates.  Never finished a violin.  I was more interested in grounds and varnishes,  and so bought Chinese whites.  I am now working with a set of eight white violins I had special-made with thick graduations.  I am 70 years old,  and am a bit lazy to make new white violins from scratch.

 

Don,  I use a 1/4" for first cuts,  and don't hog.  Only the end cuts,  so no need for spirals.  The final cut is a 1/8" cutter that advances along the center line in steps of only 1/20" and cuts about half a mm deep.  It still makes marks,  but fairly shallow.  But it is SLOW.  I cut about 10" per second,  but it is precise.  The machine will duplicate to less than 2-3 thou.  Which is the thickness of ordinary paper.   That is why I suggest using it only for patterns to use in a hand duplicator.  My old duplicator stylus was operated by hand,  so I could hog with it,  using bullnose cutters.  The first pass was every cm or so,  the second and final pass was about every inch per transverse pass.  Rather fast.  Again,  all edges were left uncut.  It worked very nicely for a long time.  I rubbed carbon paper on the tool marks and used this as a guide to uniformly plane off the toolmarked surface.  I did not want to figure out the arching every time I made an instrument.

 

Now..........  If the order of carving sections could be standardized in a particular way that set up the inflection line,  that would be nice.  Perhaps a way to satisfy Michael Darnton's criterion.

[James M. Jones]

Bill ...  semantics , I can't tell if you are just having fun ,or what . I thought I made it abundantly clear .." the one who swings the HAMMER..".

  From the CNC talk at VSA ...cutter compression can be mitigated by tipping the cutter approximately 5 degrees off vertical so the non cutting center is not in contact with the work piece , ...This could be done on a thee axis machine by making a tipped mount . .