Where to Start?

[gnwbr]

Hi, I'd like to buy/make a CNC router to make wooden parts. Specifically to make wooden clocks. Not just the clock case, but also the gears and mechanism. I'd like to draw the parts up in FreeCAD (Linux) and be able to CNC high accuracy parts with repeatable precision from these drawings. I can envision making some more decorative pieces that would require a 4th axis to rotate the part being milled. Seems like there are a lot of possible choices that could meet these requirements. Unfortunately I would also like to be able to machine some long parts (8'-10') but this seems like it will drive the cost way up. Is it possible to mill an index bump (point) on a piece, unfasten it, slide it across the table, refasten it, index to that point and continue down the length of the piece? Or perhaps there is some other technique that is used? This is a hobby not a production thing, so taking the time to jig up to cut a long part is worth keeping the price of the machine down.

As I'm new to all this, and still at that stage where I probably don't know all the right questions to be asking. I'd really appreciate any guidance on how to educate myself so I can buy/make a CNC router that best suits my needs. Thanks.

Gene

Similar Threads:

• Need Help!- Yaskawa SGDH start up issues - jog not working - no alarms - where to start?
• Start from line - error: Radius difference too large between start and end of arc
• Arrow 750 Failed to start when I Push start Bottom in fornt of control !
• Single start vs double start ballscrew.

[he1957]

Welcome to the Forum!

Yes you can do what you want. Bigger machines cost more (typically) but can do larger work pieces. You can index parts to re-position as you mention but the machine needs to have clearances to be able to do this. It is often more work but for hobby use this may not be an issue.

Might be worth searching and reading various build posts on this forum to get some background. You will need to identify what sort of accuracy you want to achieve and decide how much budget you have. Add the cost of tooling to your budget as well. Machines can be built from different materials including various woods, metals (aluminium, steel) and combinations. Materials have their different properties, strengths and weaknesses.

Determining what is right for you goes back to what you want to do, how much precision, performance and other factors including cost.

Building machines from scratch can take a lot of effort - especially to get accuracy. Kits are popular, some better than others. Chinese machines can be good value for money but be prepared to spend extra time and effort making them "just right" for your needs. Sometimes they lack quality of finish, attention to detail or are just badly put together. If you go the Chinese path, be absolutely sure the supplier understand what you want. They are good at saying they understand what you ask for up front but sometimes do not deliver what you think you have both agreed on. Ask for pictures etc.

Second hand machines can be good value sometimes even if they need some maintenance or repairs - look for auctions or folks seeking bigger and better machines.

Feel free to search the forum or ask questions - there are a lot of smart and experienced folks here.

Allow for growth (eg: 4th axis as mentioned) as you learn.

Good hunting :-)

[routalot]

What you are describing should be possible,but it may not be easy.For one thing you mention parts 8-10 ft long and moving them along the machine,you don't just need a reference for one axis to reset for the direction of travel.You also need a means to guarantee axial alignment as you do so and obviously a level surface to ensure support at the correct height.This also assumes you have a clear twenty feet or so in the region of the machine.

Do you plan to use Freecad to generate the toolpaths?I ask because one of the frustrations I have with Freecad Path is that I have yet to find a way to machine just one operation-it seems to like post-processing the entire sequence of machining operations.This might cause your multi-move operation a bit of difficulty.The Freecad forum might be able to help with specific questions.I just had a thought while typing this reply;it might be possible to use the shift bar to hide some operations and post-process those that remain.I will have to try this later today.

I began my small and basic machine a couple of years ago and learned some very useful things along the way.I think you will find having a CNC machine very useful.

[wmgeorge]

With a DIY build your resale or trade up value will be very limited. Frankly I would look at the CNC Router Parts kits. They are a first class company and you will have a lot of support. Nothing worse than getting your homebuilt router done and have issues. With their kit you can learn as you go.

[cyclestart]

It's possible to divide a cut file into sections in software. G-Code Ripper for example.
It sounds like you don't need a machine with a great deal of width, then again I've never seen a wooden gear clock in a 10 foot case
Look for a machine with a decent spindle to be able to properly run small diameter tools.
If my own (limited) experience with wooden gear clocks is typical you're about to embark on an interesting journey. Type B personality is an asset here.

ps: If planning to design your own clockworks checkout Gearotic. The easy road is buying plans from someone like Brian Law.

[davida1234]

With a DIY build your resale or trade up value will be very limited.

So very true....yet a lot of people can't refrain themselves to design their own.

[wmgeorge]

So very true....yet a lot of people can't refrain themselves to design their own.

If I was smart and talented enough to build wooden clocks, I would just buy a machine that does what I want.... and then build clocks.

[routalot]

I have to admit to giving resale value of my machine no consideration whatsoever.I built it in a basic way with an investment of a level that I could afford to write off if it went up in smoke.I did everything that I believed to be sufficient to avoid a conflagration and it worked.I have looked around a little online and wouldn't join the club of those shelling out hundreds for a few pieces of laser cut ply or MDF packaged with some far eastern drivers and steppers and running on round unsupported rails.

I suppose you have to be realistic about your own ability to work accurately in the material you choose for your machine and your knowledge of control systems.It is easier to follow a well worn path than to be a pioneer as there will be people who have done it before and who may be generous with their guidance.As long as you find yourself in possession of a machine that does what you require you can call it a successful project.

[ger21]

With a DIY build your resale or trade up value will be very limited.

I've never bought anything thinking about it's resale value, except maybe my house.


The first question, is how much do you want to spend?
And then, how much room do you have, or how big of a machine do you want? Or what are the biggest parts you want to cut?

[wmgeorge]

Third question: How much experience do you have in design and fabrication of a CNC Machine or any electrical project build? Have you done one before, are you skilled at wiring and troubleshooting? Can you use a multimeter? If you have the above skills and better yet a good set of plans go for it.

You ask about building a machine to handle 8 ft pieces, yes you could have one 4 ft long and index. What are you making your machine frame out of?

But you can spend a lot of money on a project and end up with a machine that flexes, or moves when cutting. Or you end up with a machine that crashes every time you try to run a job? Do a Search on CNC Zone you will find lots of those same stories.

[wizard]

Hi, I'd like to buy/make a CNC router to make wooden parts. Specifically to make wooden clocks.

What size wooden clocks?

Not just the clock case, but also the gears and mechanism. I'd like to draw the parts up in FreeCAD (Linux) and be able to CNC high accuracy parts with repeatable precision from these drawings. I can envision making some more decorative pieces that would require a 4th axis to rotate the part being milled. Seems like there are a lot of possible choices that could meet these requirements.

There are literally thousands of variations and approaches to machinery for this sort of work.

Unfortunately I would also like to be able to machine some long parts (8'-10') but this seems like it will drive the cost way up.

Yes cost is always an issue but beaware making a precise machine with large capability is not trivial. For clock sized parts, of high precision, you would be far better off building a smaller machine.

The real question then becomes what do you mean by precise or precision parts. Making gears is often a high precision endeavor but wood implies looser tolerances. The point here is that is far easier to find somebody will a small milling machine than it is to find a large one (small machines also mean smaller charges). So if you are looking at precision levels that require machining the structures of your machine size can become a problem.

So if you need to decide how big is really required and how much budget can be allocated for a”precision” build. In the end you will likely have to compromise. It should be noted that the majority of DIY router builds are not precision machine by any measure; for that most low end commercial routers are not “precision” machines either. That doesn’t mean the routers don’t meet the builders needs, rather they are built for use cases that don’t require high precision. I’d also have to say precision isn’t as important as repeatability.

Is it possible to mill an index bump (point) on a piece, unfasten it, slide it across the table, refasten it, index to that point and continue down the length of the piece? Or perhaps there is some other technique that is used? This is a hobby not a production thing, so taking the time to jig up to cut a long part is worth keeping the price of the machine down.

It is certainly possible to do this to some degree. Success here is a combination of the machine and operator.

As I'm new to all this, and still at that stage where I probably don't know all the right questions to be asking. I'd really appreciate any guidance on how to educate myself so I can buy/make a CNC router that best suits my needs. Thanks.

Gene

Actually your questions are excellent!!! Asking early can help you avoid big expenses down the road. I strongly suggest reading the stickies, a few build threads, the LinuxCNC site and other sites. A word of warning though people have glowing build threads of their latest build on line. Just realize doing a build is all about expectations and the expectations you have might not meet/match somebody else’s. There is a great sense of accomplishment in seeing the baby you have worked on for weeks suddenly get up and do a CNC dance.

[gnwbr]

Wow, Thank you all for the replies. This is obviously a healthy active forum.

I'm not sure where I stand on the make versus buy decision. I'm capable of building one after I educate myself a bit more. My background is in both electronics and computer science with a 35 year career in them. I've done some small Arduino and Raspberry Pi projects at home. I've done a good amount of soldering, made circuit boards over the years, do all my own home projects, and have a fairly well equipped and large basement workshop. Haven't worked with servo or stepper motors, but would enjoy that. It's more of a time versus money thing. I'm sure I'd enjoy building it, but that puts a new a project in front of building clocks. And of course building an airplane is still on my bucket list, but that's another story. Resale is not a big concern unless I start small or lower quality with a plan to trade up. It would be nice to get what I need the first go around. Perhaps improving on a solid Chinese design, or refurbishing a used model might be a good path for me.

As noted, my workshop is large and all my equipment is on casters so I can move it around. Getting 20+ clear feet is not an issue. (I set it up planning to build wings one day). I do think staying a bit smaller and coming up with a jig arrangement to precisely reposition a long part probably makes the most sense. The 10' is the pendulum, not a case It's doubtful I'd ever route anything wider than 2'. (famous last words).

I'd hoped to use FreeCAD as much as possible as I've drawn a few things in it and am accustom to how it works. No experience yet with it's G-Code capabilities.

Thanks for the information about G-Code Ripper and Gearotic. Type B huh? Looks like I may need to take up meditation as well. My knowledge of wooden gear clocks is zero at this point, but they fascinate me and I want to build them. Learning will be part of the fun. My Grandfather and Father built numerous grandfather clocks. I used to watch my Grandfather for hours as a child, and my Dad made me a beautiful John Goddard out of black walnut from plans. Hand carved finials and hand carving in the door. He had artist talent. I need to "draw" on a computer and have a machine carve for me.

Precision. Yes, I'm thinking gears when I talk about precision and accuracy. And perhaps I'm hoping to cover too big a range of desires with one machine. Maybe I focus on a smaller machine at first and go from there. Budget. Yeah that's tough. Did I mention I'd like to build a plane. And, I've grown quite attached to my wife of 35 years and would hate to be divorced at this point. I'm sure what I want will cost more than I want to spend. And my knee jerk reaction to that over the years has always been to build it myself. But I've also learned over the years that making it is not always the best choice.

I have a lot of reading on this forum to do. Thank you all again for your input.

Gene

[wmgeorge]

Instead of FreeCAD which I have used, try Fusion 360 its free for hobby and start ups. David and a couple others over on the Fine Line Automation area are using it. David has it set up pretty well to generate gcode for his Mach4 system. I just use VCarve Pro 9.5 and create gcode files for my Mach3 system. Interesting about the airplane I purchased the plans for the APlane build which now has 3500 users over on FaceBook. The Affordable plane. I decided not to build but lots of interest over there.

[routalot]

A ten foot pendulum must be intended for quite an epic clock.If you have the space,and it seems you do,you will obviously need a track or other straight edge parallel to the machine and ideally with an upper surface that is co-planar with the table.Getting the incremental moves exactly correct as you move the pendulum along is likely to be time consuming but feasible for a painstaking fellow.

As a Freecad user myself I can give little advice about using the Gear module as I have yet to explore it.An enthusiastic clockmaker will probably master it in a very short time.If I had one wish for the next phase of development of Freecad It would be enhanced 3D surfacing,I don't know how relevant that would be for clockmaking.

[he1957]

Wow! Large Grandfather style clocks and aeroplanes is a pretty wide scope :-)

Costs aside, building a machine can be very satisfying if not frustrating and very time consuming, then there's always the "improvements and additions" that can lead to an almost complete machine (not being too cynical here). Kits can be great because all the bits are there. Retro-fitting or repairing a second hand machine can be like kits and if you can find a good price for a used commercial machine it might be a bonus.

Turnkey machines are an out of the box solution (after some possible tweaking) but get you going right away so you can focus on making stuff and learning the machine/software at the same time. Generally I say "you can't cut large parts on small machines" and this is generally true, not specifically because it's not possible but generally because quality and accuracy can get lost or compromised through lack of support or a machines ability with the materials in use.

To buy a small machine, learn and practice is good but the learning curve, flexibility and realisation of capabilities of CNC machines is fast and smaller less capable machines can be outgrown quickly. Of course patience and tolerance with compromise extends their usefulness but having said that, and with your goals in mind for the future I'd suggest thinking about a larger machine to bypass the change-over cycle.

I've always wanted to build my own CNC router[1] (as many others do) but these days I'm more interested in using equipment to make stuff rather then make/fix the stuff I use to make stuff. So I splurged and bought a large heavy commercial type machine. Apart from some teething problems, absolutely no regrets and for me it was definitely the right thing to do.

[1] To help pay for it I'm still tossing up if I should get the satisfaction of building a few to sell but again, kits and cheap importable turn-key machines already exist so why would anyone buy one off me?

Good luck in whatever you decide :-)

[cyclestart]

A ten foot pendulum must be intended for quite an epic clock

I think gnwbr was having some fun ? Pendulums are mostly shaft.
Likewise my "ten foot case" comment was hinting that the machine has intended uses beyond making clocks.

[he1957]

I think gnwbr was having some fun ? Pendulums are mostly shaft.
Likewise my "ten foot case" comment was hinting that the machine has intended uses beyond making clocks.

Indeed, planes!

But the wingspan size has not yet been published :-)