The paper [1] is more useful than the web site. A key point:
Rather than clamping workpieces into the CNC machine, the
machine is designed to be clamped onto the workpiece. It is
portable and battery powered to provide freedom of movement.
That's the real innovation here. This is a little machine which works on
the ends of large pieces of lumber. That's convenient. It's hard to work on the ends of long pieces with a ShopBot, which is a vertical spindle machine.
This little machine could be very useful for wooden boat builders, who use tension joints.
There's only the one prototype unit, apparently. It's not a product.
Here are some dovetails being cut on a more standard CNC router.[2]
Chain mortisers almost inevitably clamp to the workpiece. Mostly because if you're cutting mortises big enough to need a chain mortiser, the workpiece is by definition too unwieldy to move to the machine.
Biscuit joiners, mortising jigs for routers, and the Festool Domino are also hand-held tools, or at least can be moved to the workpiece. What they're doing is kind of slick, but it's academically slick.
That router in the video you posted is pretty nifty. One thing that's to be noted is that cutting single pass half-blind dovetails requires pretty a very accurate offset between the pin board and tail board if you want the edges of the boards to line up when you're done. Getting the offset right on the fixtures is basically half the problem.
>This is a little machine which works on the ends of large pieces of lumber. That's convenient. It's hard to work on the ends of long pieces with a ShopBot, which is a vertical spindle machine.
A similar theme of "bring a small machine to the wood material instead of the wood to the large machine" is the handheld portable CNC (e.g. Shaper[1]) instead of laying the wood on a large 4'x8' or 5'x10' bed underneath a gantry[2].
(I've only seen demos of Shaper so can't comment on the viability of a computer guiding imperfect human motor skills. It seems obvious a CNC on a precision gantry would be more ultra-precise.)
The very video shows plenty of joinery error that would not be acceptable in any real setting outside of a hobby.
I'll do my best to break it down for you.
You can also see the side-to-side fit from the heart joinery has significant gaps for the same reason (they avoid showing in most shots, and blur the shot where they are spinning it, admittedly unclear if that's intentional).
Sorry, but none of this would be acceptable for real fine furniture joinery, etc.
Does this mean it's impossible to use a shaper to do it? I'm sure if you are super careful you could probably do somewhat better, but i don't think you'll solve the fundamental issues (or make it repeatably solved)
Look - i own one, i was an early backer, i've used it extensively.
I think it's a great tool for what it is.
But i don't think there's a good reason to pretend it's a miracle tool that can do things it can't.
Pretending otherwise will just make for unhappy people.
One reason Domino has been so successful for festool is that it's easy, it is great at what it is (invisible, simple, loose tenon joinery), etc.
Shaper should be viewed the same - it will be plenty successful for what it is good at.
Another interesting open source community driven project is Maslow CNC router [0]. I bought one recently although I haven't assembled it yet. It works on cutting large 8x4 boards. Another tabletop router CNC robot [1] is coming to market soon.
All this innovation in these inexpensive CNC tooling excites me as it enables a garage tinkerer to play with woodworking projects.
Edit: I am not affiliated to either of these companies
For those who haven't done any woodworking, CNC'ing or 3D printing before, this video will give you a good perspective on operational details. Take a look at this video of a bandsaw vs a CNC: https://www.youtube.com/watch?v=KHfEVek7-r4
These two people don't hide any details, they expose you to how complicated these systems are, each parameter, each step and then show you man vs machine. Note the vacuum cleaner at 6 minutes. The CNC took 6 and a half minutes to cut the gear, the man took 8.
A couple of interesting points stem from this. (I served a 3 year apprenticeship as a carpenter while I was studying...long story..in all I did nearly 10 years!)
1) The vacuum cleaner. Routers turn wood to horrible fine dust. Bandsaws turn everything to bigger dust particles, and take a much narrower 'kerf' than a router bit, which must be thick enough to not flex too much and to leave room to clear the dust. Other traditional machinery like hollow-chisel morticers and planers tend to produce large chips. I hated working all day with a router for the feeling of duct in my skin and hair. I wore a mask, and had extraction and had an air filter...
2) Bandsaws pull down through the surface, which makes the top surface look neat, with no tearing. With a router, cutting direction is critical to this. Twisted bits can help
3) Sharpness. Router bits suck for staying sharp, especially in plywood. We would have multiples of every size so they could be sent away for sharpening. Dovetail bits (and others) are wear critical, as they wear they need re-adjusting. If a router bit is placed in a situation where it gets hot it will be ruined very quickly. Bandsaw blades are in contrast pretty cheap and long lasting.
4) The situation was comparible if you wanted one gear ever. If you wanted 100, I reckon the bandsaw with a clever jig could cut those in around half the time.
5) Bandsaws are quiet, low vibration and safe (because they naturally pull the work-piece toward the table and the guides serve as guards). Routers are loud little banshees (especially the big 1/2" industrial ones), throw dangerous particles as your eyes, if hand guided can snatch even when used with a jig (especially the big ones).
I only do hobby woodwork now, and tend to find tasks that don't need the router to come out of storage!
I don't blame you. I have hit embedded stones in chipboard and had pieces of TCT hit my goggles. I have had a collet work loose and the bit come out. I have had them fight back on too many occasions.
I once purchased a 2" wide flat cutter designed for cutting wide tennons. Even run at the correct speed it was jumpy as hell. The whole router would jump out of the work piece! You had to take tiny cuts. After doing a small number of tennons it was blunt! Waste of money. If you want to do woodwork as a hobby, get a set of chisels and a tennon saw... and enjoy yourself
You still have to change the workpiece every six minutes, and change the bit every few pieces...and I presume you mean you could read a book in another room...
Well, if you used a CNC router (vs. a mill) your work surface is generally much larger so one work piece could generate multiple gears in a single run. Additionally, if you are using a CNC router you'll find that the time required to cut a single gear is much shorter as the feed rate and pass depth are much higher that he used on the mill.
Well yes, if they used a completely different machine...if they had a $250k machine instead of the hobbyist one.
On the same basis what if they had a custom gear saw, with a specially shaped cutter that could slice a single tooth on a stack of 50 cogs in one pass ....
But I thought the point was to compare two lower end machines and their capabilities? And if so the cnc was not the outright winner, no matter how cool it looks to tech geeks
Just have a look at Wintergatan (from the Marble Machine) build channel https://youtu.be/7hYX67CZd28?t=239 If I remember correctly this CNC is around 5k, at least nowhere near 250k, and there are comparable ones for the more price conscious hobbyists. It slices up a big plywood sheet into hundreds of components with good precision. Hobbyist CNC machines are a real deal and not just cool for the looks.
Horizontal routers like this are neat.
Wood is more forgiving than most materials, but the traditional problem in bringing things to workpiece is still the same - rigidity.
While it doesn't matter quite as much for edge treatment that gets sanded, it matters a lot for joinery - the difference between a perfect fit and a very loose not working joint is 0.005 inches.
(For a real cnc, that would be a very very crappy tolerance).
The easiest way to maintain these is making very rigid machines, and the easiest way to do that is adding weight.
This is why my wood gantry router weighs 1800+ lbs, and that is very light.
I have a horizontal aggregate that can already do exactly what this thing does.
For teaching people though, or for messing around as a hobby, I think things like this are fun.
As it happens, I left software a year and a half ago to pursue and education (and hopefully soon a career) in woodworking. My feelings on this are mixed.
Let's be clear right off the bat: this is not a production machine, and probably won't get any significant traction in the professional market. I checked out a local architectural millwork shop last year that does high end doors and windows. Once their rough lumber is milled 4-square, everything, and I mean everything gets cut on their 5 axis CNC router. Which has a roughly 8' x 24' bed. Cabinet shops that make a lot of drawers but aren't invested in CNC get a lot of mileage out of router jigs that cut pretty much foolproof dovetails once they're set the first time.
So that leaves the hobbyists and truly custom folks. The truly custom folks are probably already jigged up to cut their joinery with a router, a Festool Domino, or other dedicated machines (mortisers, tenoners, cope and stick set-ups for a router or shaper, etc).
At the very high end, or for one-offs, hand-cut joinery is either expected by the client, or still competitive with building jigs to cut the joinery by machine.
Which leaves the hobbyists.
If you're just doing woodworking to build stuff, I guess I can see the point of this. It (presumably) gets results fast.
But, when you get right down to it, money can't buy skill. And if you're pursuing woodworking as a semi-serious hobby, surely you'd like to actually learn something along the way? The secret to good dovetails isn't a different saw, or a different dovetail guide, or the perfect chisel, or some other doodad you buy. It's practice. Ditto every other joint that this machine can cut, with the exception of a box joint [0]. Maybe I'm grossly misunderstanding the point of a hobby for most people, but I pursue mine to get better at something I'm not already doing, not to be a drone feeding a machine I don't understand. Raney Nelson put it thusly: "If you cannot already do the machine’s job by hand, the machine will outwit you" [1].
I won't start my own shop once I finish school. Just like software, there's things you learn in school, and things you learn doing the trade for money. I'd like to learn the things you learn doing it for money in an established shop.
Eventually, I would like to have my own shop. At that point, I think it's an inevitability that I'll need to take jobs that don't pay enough to do all hand work. And I'll own a Leigh [2] dovetail jig, like most everybody else. And I'll probably own a Festool Domino [3] for loose tenon joinery, because that's basically the direction mortise and tenon joinery is headed at the middle of the market on down [4]. And god-of-your-choice help me, if I end up taking more than a handful of cabinet jobs, I'll own a couple sets of cope and stick router bits and a panel-raising bit for the doors. But I don't see owning a MatchSticks machine. Existing tools are faster.
So as near as I can tell, the target market is hobbyists who aren't actually that serious about learning their hobby? Which I concede is perilously close to a no-true-Scotsman fallacy. At the end of the day, if they're going to take this commercial, is their target addressable market big enough to cover their R&D costs and the cost of actually building a production run?
As a courtesy to anybody who replies: I'm heading to sleep (East Coast), but I'll check back tomorrow. And if you're curious what I do: I'm on instagram [5] and I've uploaded some higher-res (but still crummy) pictures to Google Drive [6]
[0] I don't think anybody who wasn't selling something would seriously suggest cutting a box joint with anything other than a box-joint set on a tablesaw. Yes, you could do it with a router and Leigh sells a jig to do so [1], but a tablesaw is better in basically every way. And technically, you could do it by hand, but there would be a lot more fitting than for an equivalent dovetail joint with no actual benefit. If you aren't making box joints by the thousands, they're basically made to be cut on a tablesaw.
I would buy it. I'm an occasional woodworker (maybe one or two small projects a year).
I don't care about honing my skills with fancy hand tools -- all I want is to design a piece and build it quickly.
This device looks like it would open a lot of possibilities. I like fancy joints but just don't have the patience required to make them by hand. If this thing would make it reasonably easy to cut a precise, correctly aligned join, it would be a big help!
TL; DR: As I was brushing my teeth, I realized this is basically the bread machine of woodworking tools. If you're making commercial quantities of bread, you need a bigger machine. If you're making bread for the joy of making bread, why use a machine at all?
If you're making bread for the joy of making bread, you're likely not using a machine.
However, if you're making bread because you want fresh bread daily when you wake up, or you want to taste new+unique recipes, etc, then a bread machine is likely exactly what you want.
Going back to this CNC, maybe you want some unique furniture or shelves or whatever, maybe you're wanting these not for the joy of making them, but for the joy is using them, seeing them, etc.
There's a market for this, it's just obviously not for you :)
Speaking to your analogy, rather than the original point: We use a bread machine because you can set it and forget it, and come back three hours later to dough that is ready to finish.
I suppose the analogy in woodworking is someone who's good enough to make a dovetailed drawer, but not consistent enough to make 11 of them the same. :) Come to think of it, that's me too.
The fun in making bread is not in the kneading. Kneading is brainless work. Bread enthousiast forums are rife with talk about bread machines - both kneading machines and full bread making machines. I understand the picture you're trying to paint, but like so many cute analogies, when the fundamental premises don't even hold, what's the point?
Your point is fair; I don't actually make my own bread. I was just going off of the large number of people I know who have bought a bread machine, made a single-digit number of loaves of bread, lost interest, and never used it again.
The point here being that if you aren't engaging with the hobby enough to really learn it, there's not really enough in it to hold your interest long term.
I think the difference between "normal people" and bread enthusiasts is how they approach the bread machine. Bread enthusiasts, I assume, come at it from the angle of having engaged with the hobby and are looking to reduce some of the drudgery. "Normal people" come at from the angle of "sweet, I can make bread!", realize that if all you want is a loaf of bread, you're better off just buying one, and lose interest.
If you're a bread enthusiast, can you shed some light on this?
Not sure - some tools can make you do things you couldn't before, or do them better. There is no way to simulate high temp/steam/low temp cycles without a programmable steam oven, and some types of joints are near impossible without elaborate jigs or tools. I don't think being a pro or hobbyist is the difference here. The OP seems like a cool thing to build, but it seems like a solution looking for a problem too. Plenty of bread enthousiasts just like the smell and taste of fresh bread, which they can't get from the store, and the easiest way to get that, will do. Just like some people might like nicely dovetailed cabinets with tight glueless joints, and will use anything to get that, as you can't buy that sort of thing.
Not sure what my point is or where I'm going with this. Many things are highly inefficient to do yourself. People still do them. Does that mean you always have to use the most low tech methods? People have different desires and goals and objectives. I wouldn't spend my money betting on the OP being a great market success, but then again, there are quite a number of 3d printer companies, who also build niche tools, that seem to survive...
It's a hobbyist tool for sure, but different people get different things out of the hobby. I like doing things by hand, and I also like doing things "by hand" with a table saw, router, and other typical power tools. A CNC router setup would let me make some cool designs that I can't make now, but I don't think I would enjoy it much. Futzing around with CAD wouldn't scratch the same itch for me, so I haven't bought one.
But other people feel differently, and they enjoy different things. There's no accounting for taste.
Futzing around with CAD is the worst. Every time I fire up SketchUp, I can't help but think how much faster it would be to just draw what I need to draw by hand.
I keep doing it because there's definitely value in being able to send drawings around electronically, and eventually, I'll probably need to send something out to be CNC cut.
I've also used both AutoCAD and ProE, neither enough to become truly proficient, but this isn't my first rodeo either.
> ... I'll probably need to send something out to be CNC cut.
As a general thought, SketchUp probably won't be suitable for that. It doesn't seem to export to any format commonly used by CAM tools. eg STEP/IGES are the general ones. JT is common in some places too.
Saying that after looking at it a couple of months ago, when someone mentioned using it for an online CAD course. Seems like a terrible choice for any instructor to be telling students to use. :(
I'm told Fusion 360 is a much preferable tool for that reason, among others, but SketchUp is free for me for now. I'm hoping enough of what I learn is transferable. Actually, I'm hoping whatever I switch to when I'm using it commercially is a little less magical. SketchUp's inference system is neat. When it works.
I'm also not interested in renting software. I'd like to own it, thank you very much. Fusion 360 is only available on a subscription model.
> I'm also not interested in renting software. I'd like to own it, thank you very much. Fusion 360 is only available on a subscription model.
Fully agreed. :)
If you're ok with trying out OSS software, FreeCAD has become pretty decent with the 0.17 release (not long ago) and can also generate the needed CAM drive cutting tools as well.
It's interface is a bit clunky, however it's really powerful too. It does output useful 3D model formats as well, in case you need to send stuff to an outside place for doing stuff with. :)
Thanks for writing down your thoughts about this. What was most interesting to me is that you left software to pursue woodworking. For a little while now I've had the growing feeling that I really want to leave software and embark on something entirely different and I have very recently started taking woodworking classes.
I am thoroughly enjoying these classes and I am beginning to seriously consider whether this is what I embark on next in terms of full-time education and a career. While it's a very refreshing and exciting idea the thought of leaving a well paid job to venture into the unknown is very scary!
I checked out your work on instagram and Google Drive and was very impressed, not to mention even more inspired and motivated, so thank you for sharing these links.
If you wouldn't mind would I be able to email you with some questions that have been piling up in my mind and some stumbling blocks I've faced? My email is in my profile so if that's ok feel free to contact me. If you don't have time I completely understand :)
To come across someone who was at a similar fork in the road is certainly very encouraging so thanks for including this detail in your comment.
Seems the furniture market is largely cornered by large enterprises making VERY cheap furniture from laminated cardboard (e.g. Ikea). I'm sure there's a smallish market for handmade one-offs if you're really really good, but uh, is this enough to provide bread on the table?
Similarly, at least over here, it seems houses are increasingly being made prefabricated in factories rather than built in-place as in ye olde days.
Similarly, again, for boats. The market is cornered by all manners of plastic-fantastics. There's a small group of sufficiently wealthy and/or crazy enthusiasts building and maintaining wooden boats, but again, a very small fraction of the market.
Finally, sitting at work staring at a computer screen all day, I can daydream about making a living working quietly with nice hand tools, no stress etc. But if the reality is working in a noisy dusty workshop full of dangerous machines, stressed out by meeting some deadline, well, no thanks, I think I prefer staring at my computer screen instead..
I mean, what were your thoughts that led you to switch careers? What market are you planning to aim at?
Uh, I'm still in school. I'll get back to you in a bit. Honestly, I fully expect that it'll be tough, and that I'll be picking up jobs that aren't in furniture to pay the bills. I just finished painting a deck, for instance.
After I finish school, I'll look for a job doing most anything in woodworking that I think I can learn to do quickly and efficiently so that I have a set of skills that can pay the bills.
One observation I've heard many times is that people are willing to put a lot more money into their houses than into furniture because they look at it as an investment. Apart from kitchen cabinets, that work is pretty much custom by definition. You're fitting the work to an existing space that probably doesn't adhere to any particular standard.
TL;DR: not sure, and I don't know that I've really answered your question.
Pete Galbert (and others) have carved out a niche for themselves building Windsor chairs. This works because making really refined ones remains something you can't do by machine. This is for a variety of technical reasons mostly related to the nature of wood as a material. Short version: if you want thin spindles, you need to split them, not saw them, otherwise the grain inevitably runs from one side to the other, and the piece breaks along the grain at some point in the future.
Pete got into chairs because he hated making boxes using loud machines.
I mention Pete not only because of that niche, but also because he wrote a series of blog posts about woodworking as a career. Thanks for reminding me to dig them up and re-read them:
I see this machine as a computer-controlled version of the pantorouter [0] with a bit more flexibility. I guess that the commercial success can be correlated to the commercial version of the pantorouter [1]. Anyone knows if that is a success? All I know is a frequent “sold-out” label on their e-shop (which might be a part of the marketing strategy).
Matthias Wendell's pantorouter was the first thing I thought of, for making mortis and tendons. But, I don't know if this machine matchsticks can scale to build larger components
Still an interesting design regardless, using a tablet interface.
None of this is damning. It just gave me a heck of a sense of deja vu as I watched. I almost expected stop-motion clamps to squeak by, a la Frank Howarth!
Did you see he’s taking a hiatus from making videos? According to his YouTube channel his shoulder is in bad shape and he’s moving to be near his wife’s family.
Eh. Better to clamp the workpiece to something rigid. Much better to have steel pattern routers with templates and making new ones is loads quicker than programming. CNC is for toolpathing not rudimentary work humans would be better at
Rather than clamping workpieces into the CNC machine, the machine is designed to be clamped onto the workpiece. It is portable and battery powered to provide freedom of movement.
That's the real innovation here. This is a little machine which works on the ends of large pieces of lumber. That's convenient. It's hard to work on the ends of long pieces with a ShopBot, which is a vertical spindle machine.
This little machine could be very useful for wooden boat builders, who use tension joints.
There's only the one prototype unit, apparently. It's not a product.
Here are some dovetails being cut on a more standard CNC router.[2]
[1] https://dl.acm.org/citation.cfm?doid=3173574.3173723 [2] https://www.youtube.com/watch?v=SDa4dQnnEyg