Official Luthiers Forum!

We have a laser scanner and it will work the full length of our table...... 5' x 12' lens is in focus with about a 7" focal length. We've found that a low resolution scan works best for us.

We like to be able to modify our drawings so it means redrawing them..... so ..... we usually bring the drawing in and draw new curves on the drawing....... and delete the original scan. We find it faster to redraw than to try to translate and then correct any stray curves etc. It's fairly fast doing it this way.

Here is a neck from one of our customers who wanted to repeat the style.

It is very a handy for doing existing parts and pieces.

Blessings,

Kevin

Here is a finished guitar just to see what I am doing.

Another thing I discovered was this.
http://www.logictracetocut.com/
I might get one of these. It seems though that a good CAD designer trace over a scanned in paper drawing pretty quickly though.

i also discovered I want to stay in the real world. I kinda like it here.

_________________
Ken

Ken, I am going to ask a question purely out of ignorance, and no, I am not admitting I am ignorant!

How close do these models have to be? I once modeled a fairly rare electric solid body by Mosrite guitars and I honestly thougth It was pretty close. The customer provided me with a dxf and wanted a 3d model. I worked for hours and I mean hours getting it as accurate as I could to those dxf prints, and he rejected it by looking at a screen shot of the model setting on top of the dxf 2d curves! There was nothing out more than .020" and I honestly did not think .020" would show up in a finished guitar? Therefore, my question to you?

BTW I probably have or have access to a very good 3d model of something like you have shown here. I have access to lots and lots of 3d models and am considering trying to start a guitar 3d model repository, and make them for sale or trade. Haven't decided yet. I could use some input on this idea!

Mike

Spot on ken

I find exactly as you say, when we do bridges etc, its best to bring in the 3d model and then lay our own fresh skin in place.

Whilst we can and do simply scan in and route straight out, the cnc router has to do so much more work, as its constantly plunging and raising to do simple little inputted vectors, but with a fresh skin and with smooth gradients the router breezes through the whole job

Well... G*bson can't even seem to get it right and they have ultimate resources, so don't feel bad. I find that there does need to be an attitude of openness and humility. I have rejected more of my own work than I have accepted! It all starts in the design and drawing stage. If that is not correct, then the mistakes can perpetuate for long periods until some persnickety person outs the error. Example: for years the historic LP guitars had the controls in a different spot compared to the original 59 model which are the most sought after guitars. Just a slight difference but akin to looking at someones face with an eyebrow higher on one side. Just slight enough that some would not notice unless they have studied an original. The Mickey mouse ears (horns) on ES guitars are another. Reissues are different and experts can tell.

Outlines are usually slightly variable if the guitar was made by hand. Sometimes it is necessary to go back and think how it was originally milled. What could the template have been used for, shaper jig. bandsawn and then sanded etc...

In violin reproduction. A mold is only a guide. Stradivari made a couple different models using the same mold, he just lengthened it a bit for the "long" ones. the corners can come out different depending on the chisel used to pare the blocks, etc... Perfect violin reproductions are usually made by making the top or back plate perfect to a photo and then the ribs and rest of the body are built on that.

So in conclusion, a 3D or even 2D drawing is simply not good enough. Funny story" when G*bson reproduced Claptons block neck famous guitar they scanned it in and even made a composite mold of the neck. In watching the video I thought they would cnc mill it to the thousands. Not so, in the next video scene they showed the worker sanding a neck on the slack belt sander, using his hands to judge the right shape based on the model.

_________________
Ken

Sounds like an invitation for a lawsuit unless you've gotten written permission from the people who made all of those models.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

I must add after writing that, I always put McKay on the headstock and change the shape so it can't be confused.


I don't think it illegal to sell or duplicate a model.

_________________
Ken

I gotta agree with Bob here. I wouldn't sell anything you didn't draw yourself or pay someone to draw for you. If you don't consider yourself the owner of the model it's very risky to sell it.

I could be wrong but I think it is perfectly legal to measure anything in the world and draw or render a plan for it and sell it.
edit: I misread your post. Sorry.

_________________
Ken

And even if you are making the models yourself, or having them made, I would be sure to check if there is a design patent on the guitar. Though I don't know of any, if there is a design patent on the guitar it would be illegal to reproduce the guitar shape without consent from the patent holder. Just check to be safe before you sell.

We actually do warranty work for most manufacturers, gibson,taylor,yamaha,music man to drop a few names.

When we reproduce a part for any major brand name supplier, we burn into the part "MIRWA Manufactured Reproduction Part"

This means there is no confusion as to the parts originality and its done to replace an existing part which can no longer be purchased from the manufacturer, we have strict rules and requirements before we manufacture said part, we do not offer for example a fender neck like might mite do, but if someone comes in with a broken neck beyond repair, we will happily reproduce the part for them and the disclaimer burnt into the heel area.

As far as selling drawn up designs, yes theres people out there already doing it, the issue is if you make commercial quantities of that part from those scaled drawings.

Most companies give you warnings first, one that comes to mind was a company I deal with that started making reprodcution Gibson headveneers with the pearl inlay, gibson politley asked them to stop manufacturing them, they did.

Steve

Bob most of these are originals but resemble known guitars, therefore my question to Ken about how accuate they have to be. Looks like he has his name on the one shown so that one would not be a problem!

Mike

Should I decide to do this, rest assured the models will be mine either by purchase, barter or personal modeling. I have no desire to raise the ire of some mega funded, lawyer equipped entity!

I know for a fact in the kit car industry that you can design a car based on the looks of an existing model and if the parts will not interchange, they cannot stop you from producing a part or a car. Gentleman I know about had a 3d model made of a Ferrari Enzo, but did not copy real parts. The accuracy of his model was within .125----.150", but that was enough to not allow any parts to interchange from his car to a real version. On the other hand, it was close enough that he could use factory glass and make it fit!

As I mentioned to Bob, my question was designed to find out how close the models needed to be. I have read on the internet (so it has to be true) that if you change only 20% in design that is enough to circumvent the design patent or copyright. I am not advocating piracy, nor do I do it. But if we want to take this to an extreme, every OO, jumbo, parlor, classic and most electrics are derived from something already created!

Mike

Enough of that now, and thanks for the comments.

Back to reproducing from an existing instrument.

Here is a screen shot of Daniel's work in Catia. He took the XYZ coordinates from the actual cast of the back of the instrument, from a half inch grid, and worked his magic. The string of numbers took me a couple of hours to data-enter. The surface was then modeled like this. And I must say, it looks nearly identical to my cast. There will still be some work cleaning up the edge. There is no recurve on this type of plate, it just goes to flat.



Daniel said it is not as good as a laser scan but it worked well for my purposes just fine.

I haven't a clue how he did it but now that it is done I can use it to make molds and fixtures.

_________________
Ken

Just makin' sure you don't get sued, bud, I like having you around

The 20% thing is more of a meme than sound legal advice. How do you measure 20% of a design? There are two bits that should probably let you sleep at night, though: you're selling drawings rather than reproductions, and AFAIK the only thing that has ever stood up in court on guitar designs is headstocks as trade dress. Guitar bodies haven't stood up, and necks definitely wouldn't, so long as the headstock is unique. Fender definitely owns their nasty-looking headstocks, and Gibson definitely owns the 'open book' headstock crown, though.

If they're mostly your work, I say go for it.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

Knowing how you're going to model a particular surface really helps with acquiring the points. On a Les Paul type top, for example, you're best served by getting a few dense strips of points on key lines (down the centerline, across the highest point of the 'belly', across the waist) and grids of points on tricky areas (the horn, if it's recurved all around). You need a surprisingly small number of points to very accurately recreate a symmetrical acoustic neck shaft and heel with a rounded cap- on the order of 30-40 will get you within a few thou of what they were trying to carve. If you're not trying to fool someone who has been playing that exact neck for 20 years, you can get away with closer to 20 points.

That's one thing that can make the arm style CMMs a whole lot more useful. If you know which points you need, you'll spend far less time on point collection. Having them on-line in your CAD system is good, too, as you can add points 'as needed' when you find yourself short on references during the modeling.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

thanks Bob, and believe it or not, I know what you mean although I don't go to that world. There are little things that make a big difference. For example the transition from neck shaft to peghead is not a small radius where it angles but a transition from the first fret to the peghead. I think it is more of a french curve or perhaps a large radius. You gotta think about the original tools. This might have been don on the end of a belt sander or a drum sander.

_________________
Ken

Hey guys, if I came across as having an attitude or something, forgive me, not the case at all. I just wanted to make sure ya'll knew I would not do something totally stupid.....ok not this time anyway!

Ken, though not an archtop, here is the body I did that the guy rejected for .020" because I showed it to him sitting on the prints. If I had shown it to him like you see it, he never would have known the difference.



Ken I actually have a nice double cut archtop but cannot show it right now as it has special f holes, which of course are not f holes! I will try to resurface later and let you see it. Made from prints as well.

Mike

You're not fooling any of us, Mike. Every one of us knows the one in the prints was white and that one's red. I don't need a caliper to see that. Who cares what the dimensions are if it doesn't match your white open-chest body suit?

Seriously, sounds to me like he was looking for an excuse. Customers like that cost you a lot less money if you cut them loose than if you try to hold on to them.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

Daniel the CAD expert has finished my McKay Dot model.
It was challenging to get from the real world to the digital. I am not aware of some of them since it was done by Dan's capable skills. I provided only the exact outline, the arching and some pointed direction on parts. Here it is in its entirety. i am ready to make the jigs that are drawn out accurately by Dan also.
See next post.

_________________
Ken

_________________
Ken

Comments please and critiques are welcome.

_________________
Ken

By the way, it even has the old rule of 18 fret scale just like '59.

_________________
Ken

Jigs. I had no idea this kind of thing could done. I am a seat-of-the-pants kinda guy, always getting the job done by making new parts to jigs and replacing. Most of my existing jigs have several old holes and sawn off parts.

This jig is for the pin router. Of course I am wishing I had a CNC now.

_________________
Ken

More Jigs, these screenshots are extremely valuable. I would have loved to see this kind of thing before I designed my seat-of-the-pants ones previously.
Good work Dan who is actually a process engineer or something similar in title.

_________________
Ken