Official Luthiers Forum!

So for the past year I've been thinking about whether picking up a CNC router might be what I'm looking for. I'd like to tell you guys some of my goals and hear your take on the subject.

As some of you know, I build banduras, not guitars. Banduras are a Ukrainian instrument that's quite popular in the North American Ukrainian community. It's also seen as a cultural icon in the community. One of the things about banduras is that the supply, especially in North America is seriously dwindling. Believe it or not, there are probably a few hundred bandura players in NA and there's probably a market for five to 10 new instruments in the $1000 to $1500 price range and probably a market for one or two in the $5000 range.

My current build process takes me about 9 months to put out one instrument. I estimate that I'm spending between 100 and 175 hours on each instrument (my day job keeps me busy). One of the most time consuming tasks is carving the neck (photo of half carved neck attached). It usually takes me two to three weeks of one to two hour sessions to get the neck finished to the point I can attach it to the rim - a huge bottleneck in the process. Fitting the neck to get the angles right is also quite a chore.

My goal/idea would be to automate the neck carving and as many other parts of the process simply to be able to increase my volume to about 5 parallel instruments a year instead of one. These won't be premium instruments but sort of a high end student model. I need to try to get into "factory mode" for this just so that the tradition doesn't die in NA - no instruments, no players.

For the high end side of things, I need to develop a key change mechanism for these instruments as well, something like on a pedal harp where you flip a switch and all the F's go to F# for example. There are many existing mechanisms but few of them work. Although I'd have to farm out the metal work, making the housing etc. in wood could be done in house.

Some logistical issues - My degree is in engineering and I even had to take a CAD class back in the 80's (CADKey - anyone remember that?) and I've played around with Solid Works to draw up the 3D models as well. I haven't done any real engineering in about a decade but I think that if I spent a few weeks I could probably get proficient enough with CAD to draw up what I need. With a little help, I could probably learn a CAM package as well.

Payback - I don't think I have to make money on this i.e., I don't need to do an ROI calculation. To purchase the router and other stuff, I'd look to donors in the Ukrainian community to pony up some cash to make this happen. From looking around on websites, I'm estimating around $5k for a light duty system and around $10 to $12k for a medium duty system with a larger capacity (e.g. K2). I'm only interested in turn key systems and want to spend as little time messing with the system as possible.

So, does this plan sound feasible and will a CNC router truly allow me to up my volume? Will a CNC beginner be able to set up the neck carve process so I can carve say two at a time on the router?

Does this sound crazy?

Please discuss.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Andy,

I think this will be a very good discussion. I'll give you my $.02. I've had my machine for about 18 months prior to that I had no CNC, CAM or Mach experience and only limited 2D CAD experience. There is a learning curve to 3D modelling, but folks on forums are generous with help. I'm using Vectric CAM software which is pretty basic but very easy to learn. Mach 3 is my controller software and it too has a learning curve, but if you buy a turnkey machine it should be configured to have you up and running in short order. Now the bad news. There's a learning curve to the whole process that may take you a while to work out. I've read that Kevin Ryan said that when he switched over to CNC machining it took him a year to get the process down to the point where it actually increased his output. That seems a little long, but I'm assuming he meant that his original production process was very efficient. The other problem is that CNC is very good at doing multiple parts over and over again. Think logo inlays or bridges as an example. One setup, one part. Wash, rinse, repeat - so to speak. It's not so good at one off parts. By the time you model and figure out how to machine the part and run some tests you could probably make a couple of copies by hand. But it is precise and once you've made that first part additional copies are a breeze. People tend to think that a CNC machine is like one of those science fiction replicators. Press a button and out pops a finished product. That's what I thought anyway - but in reality it's just another tool. It has strengths and weaknesses, learn them and it will make your production easier, just maybe not in the ways that you're thinking at the moment.

Bob

I'll give you my experience and a suggestion.

I was reasonably capable at 2D drafting before buying a router. Plus I'd read everything I could on the subject for years previously. It still took me 6 months before I got my first production parts off of it. The were nowhere near as complex as your headstock. Having some schooling, on the job training or at least someone to mentor me would have made things go a lot smoother and faster but the learning curve is still pretty steep.

My suggestion would be to outsource your necks to a guy who's already tooled up and doing a similar style of headstock. Bruce Johnson is THE guy for stuff like this. http://www.xstrange.com/

And that's the whole rub. If you check out my Blog here: http://banduramaker.blogspot.com I think you'll see a number of areas where it could be of use - Neck, Bridge, inner forms, jigs. Drilling tuning pegs (would probably need a low speed router for something like that). With a 4th axis, could probably drill string holes as well.

The thought for this is to increase the numbers on the parts. If it takes me a year to get things set up, and another year to make 5 of each part I'll be making and another half year to finish the instruments, I'm still hovering around +2 (I wrote it takes me 9 months to finish one instrument but really my output is one/year as I don't really build much in the summer). If the year after that I can complete 5 instruments, I'm quite a bit ahead of the game.

As to farming out, I've thought of that quite a bit. There are two big issues to surmount, mostly to do with price and cost. I imagine that to buy a neck like this at the volumes I need is going to be pretty expensive - let's say $300 (probably closer to $500 though as a guess). That would make it the single most expensive part in the instrument. I'd have to raise the price of my instruments commensurately to make up for this (I'm willing to pay myself very little hourly on this I don't want to subsidize these instruments with my own cash - I've already spent enough).

If I go to bandura community fat cats asking for money, it's much more effective to have a tangible goal for which you need the money when you make the pitch - e.g. I need to collect $10,562 to buy a turn key CNC router. This will allow me to make 5 to 8 instruments per year and sell them to your kids for $1500. It's practically impossible to collect money to subsidize general costs like a $300 or $500 neck. e.g.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Hi Andy,

That is quite an instrument. I went to your blog so I could get a better idea of how they are built and I have some statements that may or may not help you. I agree, if the neck is all you are really concerned about, have it machined by another shop. That said, I see several areas that a cnc can help you, if you intend to build several of these instruments.

1. Neck
2. Top, back and internal bracing (larger braces that seemed to span the bottom)
3. The shaping of the bridge and shemstok piece, less the holes since I think you have the drill press thing down
4. Top with sound hole and binding ring
5. Back

The neck will need 4 axis indexing to machine properly, and it can be done with a 3 axis machine with fixtures, with a 3 axis and a manual indexer, or a 4 axis. I would do it while holding the ends with either a 3 axis/manual indexer or 4 axis, and then do what little hand work left after sawing the end tabs off. Stuart here on OLF has some super nice fixtures he uses for machining necks and he does it all with 3 axis.

Scanning has come down significantly and that is the quickest way to get a model of the neck. I have used a service house on the east coast for a set of competition pistol grips and the cost was $150. I think you can probably get the neck scanned for $300, or if you want to do it yourself, as already stated, there are folks here that help others all the time.

Hope this helps.


Mike

Yeah, that's a four setup neck for sure. Fourth axes eat up a lot of space and ones without rigidity issues are big bucks, so you don't tend to see them in production shops. Otoh, small shops don't really care about cycle times so it can be worth it to simplify things.

Comparing the up front cost of the machine to the price of the parts you'll be making with it in small batches is a bit misleading. You'll be paying out $10K upfront for the machine, not paying only as you use it, and those parts get a lot cheaper if you buy them all at once. One neck might cost you $300+ to have machined, but 50 necks might cost you $100 or less apiece which would be your next ten years of necks for half the cost of the machine.

A CNC is a great thing to have, if it's a tool you want to use and integrate into your workflow, but buying the machine doesn't save you any money on stock parts unless your own needs are quite significant. I have high end gear and as a result I really only service clients that need high quality at volume where the efficiency of volume overcomes the costs of making top shelf parts, or clients that need super high end parts that are so intricate that either they can't be feasibly manufactured otherwise due to time or accuracy concerns. For every specialist shop like mine, there are 30 guys with CNC routers in their garage or basement who'll charge sub-rock-bottom prices just to have their machines making them a couple extra bucks. If you can find one of those guys whose work is good enough for you, then you can really save some change.

I've lusted after a Timesaver since time immemorial but paying the shop down the street $60/hr, using their consumables, comes out to be a bargain when I ran the numbers. Even at production numbers I can get so many headplates and fretboard blanks done per hour on their machine (for sixty bucks) that I'd never get ahead by owning one. I definitely don't get ahead owning my welder VS renting, but I just like having the thing so much that it's worth it to me. That's the difference. So if you want a CNC (and it IS the coolest tool you'll ever own, I guarantee!) then this is as good a reason as any to get one. If you want to save some money and up output, then you'll easily be able to find a shop (pro or amateur) that'll beat the value proposition of owning the machine and software.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

Great advice, as always Bob...! I couldn't agree more.

Trev

_________________
http://www.PeakeGuitars.com

I needed to have cnc capability in-house....which is to say my former business would have failed 10 years ago if I hadn't learned to do it. The motivation to dive into cad/cam was ample in my case. I plopped down $105K cash on a large, new mill and a software package one day...and stuck it in my garage...and didn't know the first thing about it. But if the failure of a business wasn't ample enough motivation, that investment capital sitting in my garage certainly was.

Necessity is the mother of invention....but the true catalyst to make that move was a phone call I made. Haas, the machine manufacturer, put me in contact with a guy that made golf clubs. I honestly can't remember his name today but I talked with that guy for over an hour, listening intently to his story and what cnc capability did for him. It was during that phone call that I started to see myself doing this. His basic story is that he started with one machine...and it launched him light years forward from the standpoint of both production and research. Within a few years he had four machines. In hindsight, I did exactly the same thing. I bought the machine and within five years I had four of them too.

What happened to my business (to me) between not having cnc capability and having it? The world opened up for me is what happened. The machines allowed me to explore any idea for a tool or mold that I could actually draw….so I became an inventor. Within a few years I had done research and development projects for almost every major aerospace company. Wonderful, interesting stuff. It was a new world.

I’m telling you this in hopes you’ll go forward with this, Andy. Don’t think about it. Just do it. Once getting the basics mastered you’ll have new creative potential and will think in ways you never did before. You will pick how and why you use it, but as you gain experience with it you’ll continue to invent new ways to apply it. Ask Kevin Ryan. There isn’t anyone who knows this any more than he.

Cad/cam is not simply a tool….it’s an expanding approach to problem solving.

When it comes time to know and understand feed rates and techniques, we’ll be here for you.

_________________
I read Emerson on the can. A foolish consistency is the hobgoblin of little minds...true...but a consistent reading of Emerson has its uses nevertheless.

StuMusic

Thanks for the advice guys. Mostly this is stuff that I already know but it's good to hear it from those with experience. I think this is the direction I'm going to go, probably with a target of next winter to have something up and running, possibly the spring after that (gotta finish a few instruments first and find some $$).

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com