CNC options?

[kquiggle]

Strangely enough, reading about the problems with DMC2 does not necessarily put me off of considering. For those who don't want to take the time to read the referenced sub-reddit, the gist of it that some early adopters of the DMC2 are having problems with parts that have not been de-burred (and then powder coated!); poorly applied powder coating with scuffs and bubbles, warped parts, and other issues. While this may sound disastrous, all of the issues encountered seem to be fixable. There is an implication that the DMC is somewhat in the nature of a "release candidate" and that early buyers are essentially de-bugging the product. The seller is responding to issues (mostly?), so that is on the plus side. This will probably sound familiar to those of us who are used to buying budget equipment knowing that we'll have to correct some defects before using it.

The take home message from the sub-reddit is that buyers of the kit should be prepared to spend a fair amount of time "cleaning up" parts out of the box, and doing other work in preparation for assembly, before even getting to the actual assembly. People with access to a machine shop, or who have machining equipment, will probably be better situated to address the issues.

I have added the DMC2 to my spreadsheet, and the specs certainly look impressive.

The full specification, including items not on my spreadsheet, is here. While the kit includes GRBL and Mach3 USB controllers, no software of any kind is included as near as I can tell. I assume the purchase of Mach3 software (currently $175) would be required, but maybe someone more knowledgeable than me in this regard (Stan?) could comment on this.

TECH SPECS | Shariff DMC

 

[Mike Ginn]

I was always told that there are 3 key issues when choosing a mill (be it CNC or manual):- Rigidity, Rigidity, Rigidity!

OK, that's a simplification and a lot depends on the application. Router type "mills" lack rigidity and are fine for thin "soft" thin materials but with model making you quickly come up against 1/4inch thick sheets of hard brass and steel. I have a Sieg KX1 which only just manages to cut my parts. Unfortunately there seem to be few small CNC machines on the market (not sure if the KX1/3 are still available) apart from the router types which promise everything but deliver little to the model engineer. Router designs are easy to make - hence being widely available - but lack that vital ingredient of rigidity!

FYI I use Autocad and CamBan

Hope this helps to move the discussion along.

Mike

 

[kquiggle]

Mike - your comments on software are helpful to me. Of course I know that a CNC mill is going to require software tools, but in focusing my attention so much on the hardware side I think I have neglected the software side of things. Perhaps you could tells us why you chose CamBan, and the pros and cons of that software.

In my ongoing research, I cam across another candidate system, as well as a very useful video which compares the Evo One, the Nomand3, and the DMC2; worth watching :

Makerdreams Evo One Pro 2021 Review

I'll be adding the Evo One to my spreadsheet shortly.

 

[stanstocker]

The full specification, including items not on my spreadsheet, is here. While the kit includes GRBL and Mach3 USB controllers, no software of any kind is included as near as I can tell. I assume the purchase of Mach3 software (currently $175) would be required, but maybe someone more knowledgeable than me in this regard (Stan?) could comment on this.

TECH SPECS | Shariff DMC

Good morning,

You could just run it with gRbl, nothing more to buy. Just download a couple of user control panels and you're all set. I tend to use the openbuilds controller interface with my CBeam router and Blackbox, and the CNCjs interface with the larger router and the CNCPro V5 hardware. The only limitation I can think of with CNCjs is the 400,000 line gcode limit. On a complex 3D component you could hit this. A quickie search for gRbl interfaces or a peek at the gRbl homepage will turn up lots of options. The most recent Goko looks pretty nice, but until I finish working on shop buildings and get to play IN shop buildings more it's just something I want to play with, not something I can recommend.

At a minimum it would get the machine moving and tested out. From that point you could decide how to proceed. You would need to buy Mach3 if you choose to go with a Windows solution, although I'd suggest spending around 350 to get both Mach4 and a step generator / breakout board rather than spending $175 on an obsolete unsupported version of Mach that was baling wired into working under 32 bit windows and then bubble gum and duct taped onto the 64 environments. You don't HAVE to do anything with mach, it just happens they bundled a low cost step generator with this particular machine. They bundled an arduino with gRbl too, doesn't mean you have to use gRbl. You might find that gRbl meets your needs, odds are it will be more stable than Mach3. As you prefer linux anyway, grab a Ethernet Mesa card or toss a real parallel port card in a machine for less than the cost of a mach3 license, load up linuxcnc and play in the world you prefer. You can always upgrade or change the motion control hardware and software as you need or want over time. The break out board in this specific machine has all the signals on a terminal strip, you aren't limited to mach3 or gRbl and there should be very little reverse engineering required.

All the controller software does most of the time is turn a few screws and maybe do spindle control. Keep track of where it thinks the machine is and monitor some pins for limit switches and software e-stops and update a user interface. I run a dual Y axis 33X44 table router with leadscrews at 100 IPM rapids using gRbl with zero issues. On big machines the controllers do a lot more, but in the world of small shop machines we usually just don't need such complex controls, nor do we have machines with the inherent accuracy and repeatability levels, or the 4th and 5th axis options with co-ordinated motion, to take advantage of all the features available. Heck, most of us don't even have umbrella style tool changers, much less 100+ tool carousels and pallet changers. As one gent who works on slicers (3d printer world CAM) said at ERRF (East Coast Rep-Rap Festival), 3 axis is trivial, 4 axis can be tough, 5 axis is hard, very hard when talking co-ordinated motion of all axes.

It's worth keeping in mind that machine sellers and armchair builders love to talk about how fast their rapids are. Like 3D printer makers talking about 250 to 400 mm/sec printing speeds, reality is that you can squirt some PLA at that speed so it isn't a lie, but to get good prints you are probably going to run at 100 mm/sec, not 250. Maybe less. Same game gets played with routers and mills. On large machines doing a few small details separated by large blank areas, fast rapids matter. Cabinet manufacturers dicing up plywood sheets into flatpacks to feed a production line care about rapids. On a small machine, with acceleration and de-acceleration you're never going to really see those numbers happen, and if so, so what? You can't cut at those speeds. If a motion controller can only make 60 IPM with your hardware and it's reliable, on a small machine a higher dollar solution that can generate steps fast enough to give 300 IPM won't mean squat if you only have 10 inches of total travel and cut at 30 IPM. You might hit 60 IPM from inch 3 to 7 if doing a rapid from 0 to 10 inch positions. Maybe. Rest of the time you're speeding up or slowing down.

Gantry mills are not inherently inferior to column mills, but many folks like Carbide, XCarve, and all the folks selling 3020, 3040, and 6040 import aluminum frame routers started using the word "mill" while actually selling routers. Just because you can "mill" away the copper between circuit board traces to make a PCB or scribe your name legibly on a brass plaque doesn't mean you have a metal milling machine, heck a small jewelry engraver can do that. In the large mill world a gantry style can provide better work access and larger work area as the two columns stiffen the X and Z travels quite nicely. Not talking small shop stuff, more like the Mitsubishi MVR series of machines, big industrial stuff. The small router style machines aren't bad for aluminum if you use a good spindle and are patient, but can only nibble at steel with minimal DOC. My small C-Beam router is quite a stiff machine, it cuts aluminum or brass pretty well, it would be better with linear rails instead of V wheels. My larger router isn't even worth considering for any real aluminum milling work. Poking holes in large panels, OK. Milling a complex shape? Heck no, just not rigid enough, sucker would bounce and chatter all over the place or have to run so slowly it would be making dust, not chips. People who run really big machines will tell you most of the machines we would consider big machines are ok for light work. Just a matter of perspective and budget. I have a friend who makes watch parts for restoration work, he likes expensive specialized 50K+ RPM spindles with straight shank cutter retention and often uses cutters smaller than 30 thou. His spindles cost more than most small desktop CNC machines. Another friend needed 20HP spindles and bar feeders on his lathes to get the MRR his old shop required to be profitable. They might use the same make micrometers, but their machine needs are completely different.

One of the less mentioned benefits of CNC for the small shop person is that it lets us cut complex shapes that would be a pain to do with saws and files, or sucks to do manually (think high tooth count gears, artistic shapes). If it takes three passes to get the profile cut, and the machine has sufficient power and stiffness to maintain chip load with a DOC that gives good cutter life, in the material you want to use, and the end results are of sufficient accuracy and finish quality, that's probably good enough. Not everything has to be done in a single pass.

So, some more random thoughts and hopefully a few questions answered!

Cheers,
Stan

 

[Mike Ginn]

Hi kquiggle
Decisions are not always made for the right reasons! I have used AutoCad 2012 for many years and I now have the full 2022 version which includes 3D. I also have Fusion 360 and TuroboCad neither of which I have used due to time constraints and the 3D learning curve. I design in 2D and export dxf files to CamBan to generate GCode. CamBan is one of many solutions but it is simple to use. Ideally I should be generating GCode from say F360 but I am not there yet. CamBan has a few issues and can produce tool path errors but thats par for the course I guess. In my experience there is no one perfect GCode tool - its largely what you are used to noting the inertia not to change.
Hope that helps
Mike

 

[kquiggle]

Mike - Well said. I'm guessing we are all coming at this from different levels of experience and with varying preferences, so the background information you provided is quite helpful.

Stan - Awesome post! There is definitely a lot of what the advertising weasels call "puffery" in the descriptions of the various CNC machines being sold. I typically stay away from Version 1 of anything, and I want to hear from someone who actually bought a machine and is using it to get a better picture.

I have updated my spreadsheet to add the Evo One machine (<gripe>I really don't like their website, which appears to have been designed with someone more interested in playing with Javascript than with actually conveying useful information. <endgripe>).

On a related note, I have considered the idea of buying a very cheap laser engraver as a learning tool in terms of setting up and using CNC software. Is this a waste of time and money, or worth considering? I do already have a 3D printer so I am already comfortable with designing 3D models and exporting STLs, and converting the STLs into gcode with a slicer.

 

[Mike Ginn]

Kquiggle
I have had these sort of questions asked of me in my professional life. I usually say "get on with it and get your feet wet". In the days when CAD was first starting (on very low power computers) many companies tortured themselves over the "best" machine to purchase. Other companies looked to see what others were doing and made decisions based on that. The results were not perfect but they gained a vast amount of experience and produced goods while the other companies continued to evaluate. There is never a perfect solution and I don't think that matters. Get your feet wet, make some swarf, swear at the software and above all have fun!
Mike

 

[stanstocker]

Mike - Well said. I'm guessing we are all coming at this from different levels of experience and with varying preferences, so the background information you provided is quite helpful.

Stan - Awesome post! There is definitely a lot of what the advertising weasels call "puffery" in the descriptions of the various CNC machines being sold. I typically stay away from Version 1 of anything, and I want to hear from someone who actually bought a machine and is using it to get a better picture.

I have updated my spreadsheet to add the Evo One machine (<gripe>I really don't like their website, which appears to have been designed with someone more interested in playing with Javascript than with actually conveying useful information. <endgripe>).

On a related note, I have considered the idea of buying a very cheap laser engraver as a learning tool in terms of setting up and using CNC software. Is this a waste of time and money, or worth considering? I do already have a 3D printer so I am already comfortable with designing 3D models and exporting STLs, and converting the STLs into gcode with a slicer.

Greetings!

Thanks for the kind words. I have two lasers and a few older diode heads around here. One is a PRC 40 watt CO2 that isn't even set up since I got a Sculpfun S9 with the bed extenders, which is upgraded to a 10W dual diode head with good air assist. The S9 with lightburn software and good air assist lets me dice up 1.5, 3, and 5 mm plywood and thin basswood or veneers easy peazy. Great for cutting out all the parts for making HO scale buildings and such. Combined with FDM and resin printers for detail parts this opens up all sorts of opportunities. If laser cutting thin wood is your thing, the newest generation of multi lens equipped diode lasers work nicely without all the hassles of keeping five gallons of distilled water circulating and sufficiently cool.

I assume the other makers 10W diode laser cutters/engravers are about as good. Each maker takes the lead on some aspect of the game for a while, then another one ups them. Sculpfun just happened to be well regarded when I purchased. Could be Atomstack today, someone else tomorrow . They are all at least similar, fill up a box with 2020 rails, a grbl controller, brick style power suppl;y, couple of NEMA17's, V slot wheels, and GT2 timing belts. Shake well and pour out a light duty laser.

For engraving and cutting thinner (3mm or 1.5mm) material, the 5W diode head was fine, a little slower than the 10W head, but not by a full 50%. My wife decided she really liked the look of some art deco window panels, but not the prices asked. Time to upgrade so we can cut 1/4 inch cherry and maple sheets. Should have held out for a 100W CO2 with chiller, not just a piddly little 10W diode head

Being able to slice up 16X34 inch sheets at a time is a nice bonus compared to the 8X12 capacity of the ubiquitous PRC 40 Watt CO2 lasers. Good for some model boat and aircraft uses too naturally, but the wrong wavelength for dicing up most foam core materials, although I've heard the Adams black foam core with black core material cuts well. I intend to test this, but it's too nasty a material to laser cut without really good ventilation, which I haven't set up just yet. The current ventilation is fine for smoke from wood sheet and plywood cutting, but not toxic spew and it's too cool outside to do positive pressure ventilation of the entire shop building right now. That minisplit takes time and money to heat a cold shop!

I have the rails and parts on hand to convert my 40W CO2 to around 14X24 inch capacity, just not the time or a pressing enough need to move that project up the stack. The upgrade wouldn't make the size for the window panels anyway. It's more fun hanging gutters 20 foot up in the rain, like this morning. Think I'll get a job, this retired stuff is exhausting.

Lightburn software is available in a linux native version, which is what I'm using. It is NOT open source, it's proprietary. It is however so superior to lasergrbl that it's a case of my desire to get something done winning over my preference for open source. Sort of like keeping some windows instances available to support software you aren't willing to forego.

As for formats, I've always used flat 2D files for laser work, typically cdr, jpg, svg, or dxf files. Vector files are preferred for cutting just to avoid dithering sucking up power to no benefit and probably lowering cut quality. I don't engrave much other than siding lines for buildings though, mostly I cut. The engraving for the models is still in the vector file, just the line is a different color, which is used for the PWM control of the laser power. It seems there are ways to finagle STL files into lasers, but they are really 2D devices, maybe 2.5 if you want to view engraving depth as an additional .5 D component.

There is a gent who goes by Louisiana Hobby Guy on youtube who is really quite hardcore on the whole laser scene, running through a couple of his videos might be of interest.

Helpful hint: Secure your machine to the surface it and the work sit on. The vibration of the laser head and carriage can cause the entire machine to shift around while the work stays in place, resulting in double lines or just plain utterly failed runs. Took a few times for me to realize what was going on, it looks like the software has lost position as when skipped steps occur. With no cutting force to speak of in a classical physics sense, it just didn't make sense at all. Even blue tape is enough for a quick and dirty hold down, these small lasers are really very light weight. Some folks hang them on the wall when not in use, they are truly that light.

Minor CAM comment: I've used CamBam at other folks shops, Meshcam, and Vectric. Used to use Fusion360 until the licensing got weird. I still use Meshcam and Vectric, both run well using Wine on linux. While you can find 2D CAD easily for linux, and some like some of the 3D CAD software for linix, I've never found a CAM option that is linux and that works well and is really competent at it. I don't have enough open source only religion to put up with the limitations of those I've tried, although some insist PyCAM is the best thing ever. Haven't tried it in a year or so, but never did get good gcode out of it. It probably doesn't really matter what you use, just grab one and try it out. CamBam does a 40 use trial without limiting features, long enough to see if it's for you. Openbuilds has a free online CAM system. Once you know one, the others are easier to get a sense of. Just like 3D slicers, if you know Cura, you can find your way around in Prusa or Super Slicer with fewer false steps.

WeeOh! Lasers, CAM, graphics formats, and model making, what a long strange trip it's becoming.

Stan

 

[stanstocker]

On a related note, I have considered the idea of buying a very cheap laser engraver as a learning tool in terms of setting up and using CNC software. Is this a waste of time and money, or worth considering? I do already have a 3D printer so I am already comfortable with designing 3D models and exporting STLs, and converting the STLs into gcode with a slicer.

I sort of missed the mark with rambling on about how I use lasers and not answering your specific need. Sorry about that.

My gut is that the laser itself is so trivial from a kinematic and functional usage perspective that unless you just want a laser to do laser stuff don't bother with one. You end up learning more about creating and using vector format graphics files and configuration tweaks of laser specific software. Maybe already having a bit of cnc knowledge going in it was easier, but I truly built the machine, installed lasergrbl, hated it, installed lightburn, looked up a few things of a "how do I xyz"? nature, and was cutting out a model HO scale garage from 1.5mm birch ply in maybe a long evening. I've learned just as much as I need to know in lightburn to do what I want to do, and there's plenty more I could learn about it. I can't think of anything about the experience that would make me any better on the CNC machining side of the world though. There's even less of a cnc feel to using these little appliance lasers than there is to 3D printing. Like having lasers, but that's because I like what they let me do, not because they make me any smarter. What I've learned about lasers, such as honeycomb work supports and air assist set ups, just doesn't transfer to anything else in my world.

Cheers,
Stan

 

[Scott_M]

What Stan said ! If you think having a small laser would benefit your shops capabilities then get one , they are pretty cheap now. But It will not teach you much of anything you can apply to a CNC milling machine, and it will be a distraction to the mill. Like Stan, I have a self built Diode laser with air assist. 1.5-3mm ply is it's limit for cutting but it does some nice engraving of pictures. I have had a lot of fun with it. I outgrew the fun phase rather quickly and it sits most of the time like the 3d printer, it was more of a I want to build one thing. They still get used but not as much as the mill and lathes.

Scott

 

[aarggh]

I was always told that there are 3 key issues when choosing a mill (be it CNC or manual):- Rigidity, Rigidity, Rigidity!

Mike

And mass! Can't understate those two criteria enough!

 

[timo_gross]

Problem is that most smaller or cheaper machines are not good enough for production work.
Machines for production work are not cheaper or smaller .
So there is not much in the middle and the market for small size machines with a lower price tag is probably not lucrative, so DIY is the way to go....

The Taig is available as a "turn key" CNC machine.
5019CNC Micro Mill - TAIG Tools
While I don't necessarily endorse the control components they have chosen or the control software (MACH3 and parallel port I'm assuming as its not listed - very outdated for current machines), it is turn key, the base machine has ball screws and the stepper motors are good, and you could in the future switch over to a different control system if needed / wanted (i.e. LinuxCNC and a MESA 7i76e Ethernet controller card, or UCCNC and one of their controllers).

I helped a club member who bought just the base, ball screw machine and he added with guidance the stepper motors, control system and software and PC for about the same price as the one I linked above. He seems happy with it, though I do understand the reluctance to build it yourself, as he did encounter a few hurdles to getting it going that myself and another club member were able to help out with. Its much easier if you have people you can ask for guidance or help when things get hard...

Mike

... then again. If you start conversion or building in order to save money, most likely you already are at a loss before you even begin.
I think I would end up with a crappy machine. Or you start adding features, closed loop, rigid tapping etc. then you will end up spending far more on the hardware than you originally thought.
Then factoring your own time in, the conversion or building is only a viable option if you follow Bazzers advice.

...

Don't underestimate the work involved in getting a Mach 3 type conversion up and running to a good standard, you have to decide if you are a machine convertor or a maker/machinist, or maybe both in which case you need time more than anything!!




The EMCO lathe is nowhere near as noisy as in the video, I spend money on machines not good phones!!

Any questions then just ask.

B.

Here is a very good series from rotary SMP, in the video he makes some good points concerning the conversion of a bench top mill.



I have no bad experience with Linux CNC, for very basic systems it is very easy to set up.
If it becomes more involved, most other systems are just not flexible enough to get it done and then Linux CNC is as much a pain as any other computer.

Greetings Timo

 

[Jasonb]

I would also say that the engraver would be a waste of time as far as getting into CAM and CNC goes. It won't teach you anything about choosing the right machining strategies - contour or adaptive - ramp or scallop - etc. What sort of feeds and speeds are needed as these are far more critical to get right on the CNC as you can't easily just slow how fast you are turning the handwheel, What cutters perform best and how to add them to your tool library and so on.

When I got my CNC (Sieg KX-1) it came with Vectric2D, I used that a couple of times to do a few shapes and a bit of lettering in some plastic then aluminium, next imported a part I had previously made and tried that in aluminium and then very quickly decided I was not going to learn a lot more unless I started making acuual parts

At that early stage I found the Vetric was not going to offer me tool paths that suited the things I wanted to make and that is when I looked into F360 which I have used since as it suits the often 3D parts I want to cut rather than 2D or 2.5D.

It is worth thinking of what type of parts you want to make in the future as that will have a bearing on what machine and CAM suits your needs. If you only tend to make barstock engines then a gantry router type machine will do OK but if you want to start replicating castings be they for steam engines or model aircraft glow engine crankcases then you will need to upgrade machine and learn a new CAM package. I know you want to start with a machine within your budget but at least decide on suitable cam from the start then you only need learn the package once.

You say you have manual machines to fall back on if the CNC can't do it, I tend to use mine differently. The manual machines get used for the parts they are best suited for and the CNC gets used for items that can't be done or would need a lot of setups and long manual machining. For example in teh past I have carved IC engine crankcases from solid which can take several weeks of evening and weekend sessions in the workshop. I can now let the CNC get on and do that in a few hours. But if it is just squaring up a sawn block and putting a few holes in it then I'll use the manual machines as the CNC would not be much of an advantage.

 

[timo_gross]

I bought two of those 2nd hand. It was O.K. to gain first experience. Why two? Because they both did not work when I bought them. I thought I would need one as parts donor for the other one.
It is sort of usable, but it is hopeless to make accurate parts.

Haha, ... if someone wants it I sell it for fifteenhundred USD. Shipping from Taiwan needs to be paid by the poor victim who wants it....
I also will refuse to pay any duties, I will not clean it, not pay for the shipping box, not repair it...... a horrible deal altogether.

Joking aside, I realize those small column mills are not even cheap, but for metal working I would prefer them over a router type.
e.g. with a boring head you have a fighting chance to produce an accurate hole, but with a soso machine with too much backlash a bearing seat circle is a challenge.
Backlash compensation of the software does not help anything, because the thing will vibrate were the motor changes the direction.

That Grizzly looks suspicously like the one I have here. 6" x 22" CNC Mill at Grizzly.com 12000 USD ... mmmmh.

Considering that the similar manual machine costs sooo much less a retrofit starts making more sense than I thought.

Anyway tough decisions, in doubt get proper z-height to fit a Boring head in the machine. The one I have takes up 4"-6" already, dependent on the installed boring bar.

A lot advice from Stan above, to follow.
I would not opt for flimsy router machines. Stay away from low torque, a gearbox is what you need for metal. (or lots of power)
Be aware that "metal dust" from a 10/127" endmill is all fine needles that are a pain in the hand, so a slow revolving 40/127" is much nicer to manage.

My conclusion is: Try to find a 2nd hand column mill that someone cnced by himself. Preferable with an Enclosure that keeps coolant and small particles inside.

Greetings Timo

 

[kquiggle]

Lots of good posts to mull over. I especially appreciate the comments from people who have already made the decision with either a good or not-so-good result.

The posts on a cheap cnc/laser engraver confirm what I suspected - not all that helpful for the direction I want to go.

While I knew I would need CAM software, I haven't looked at that area as closely as I should have. I would prefer software that works well with OnShape (my 3D CAD preference). Kiri:Moto can integrate with OnShape, and since it is cloud based it works on my Linux box. Kiri:Moto seems to be actively supported and improving, but I get the feeling that it has some significant limitations at this time - but I don't know enough about CAM in general to say for sure if that is the case. OnShape has announced that they have acquired CloudMilling CAM and that it will be implemented in 2023; however, I suspect that this will end up being available to paid users only, so it's not something I am counting on. I can export STL files from OnShape, so I assume that Linux CNC is an option similar to the way I export/import into PrusaSlicer.

So - in 25 words or less - tell me everything I should look for in a CAM package!

 

[stanstocker]

View attachment 142272
I bought two of those 2nd hand. It was O.K. to gain first experience. Why two? Because they both did not work when I bought them. I thought I would need one as parts donor for the other one.
It is sort of usable, but it is hopeless to make accurate parts.

<snip>

Greetings Timo

Hi Timo,

It's good to see I'm not the only member of the Don Quixote Home for Clapped Out Machine Tools From Hell Club around here! I've taken in a few machines that seemed to have promise over the years too...

Cheers,
Stan

 

[stanstocker]

Lots of good posts to mull over. I especially appreciate the comments from people who have already made the decision with either a good or not-so-good result.

The posts on a cheap cnc/laser engraver confirm what I suspected - not all that helpful for the direction I want to go.

While I knew I would need CAM software, I haven't looked at that area as closely as I should have. I would prefer software that works well with OnShape (my 3D CAD preference). Kiri:Moto can integrate with OnShape, and since it is cloud based it works on my Linux box. Kiri:Moto seems to be actively supported and improving, but I get the feeling that it has some significant limitations at this time - but I don't know enough about CAM in general to say for sure if that is the case. OnShape has announced that they have acquired CloudMilling CAM and that it will be implemented in 2023; however, I suspect that this will end up being available to paid users only, so it's not something I am counting on. I can export STL files from OnShape, so I assume that Linux CNC is an option similar to the way I export/import into PrusaSlicer.

So - in 25 words or less - tell me everything I should look for in a CAM package!

Greetings,

25 words or less? I'll try:

Works with your OS and motion controller. Handles STL and DXF at a minimum. Simple tool tables. Can do 2D 2.5D and 3D toolpaths.

24 words, but at least a start.

Here's a bit more, as always

Your controller software will not take in any drawing format, it only understands G Code, and in a flavor of G Code it speaks. Your CAM package ingests the drawing and generates the G Code. CAM for machining is the functional equivalent of the slicer in 3D printing. The motion control software, be it Mach, UCCNC, gRbl, linuxcnc, whatever is the equivalent to the embedded firmware in a 3D printer, or perhaps being more involved, it quite similar to klipper running on a pi along with the firmware in the printer control board.

Lacking specific support for your motion controller (rare), most CAM packages with have a post processor that generates .TAP format gcode, usually it's about as generic as things get. The base FANUC post processor is usually really vanilla too. Sometimes you can select a generic post processor that specifies arcs and curves, if so use it, otherwise curves are cut as series of short straight lines from point to point, resulting in slow machining, lots of stops and starts, and faceted curves. Odds are you won't have to worry about this, but it's nice to know there is some low level commonality. Unless you have an original EMCO CNC machine... Don't sweat this much though, just about everyone speaks Mach 3/4, LinuxCNC, and gRbl along with all the spiffy machines. The differences usually show up in things like advanced optional parameters for thread milling, dynamic work offsets, that sort of stuff. The basic G Code world is pretty well standard, at least as much as any technical stuff is "standard".

Meshcam, VCarve, and Cambam all import STL files. It's worth noting though that the process can be a lot more streamlined if you can save your CAD files in a vector format such as DXF rather than in an approximated tessellation form such as STL. Naturally if you are going to do true 3D machining, a 3D model is the way to go, but for many of the things done on a CNC mill, it's pockets and profiles for the most part. I just tried the current version of CamBam with wine, no luck. It looks like they use the .NET4 framework and some other stuff, no messages, just a loading cursor then nothing happens. As a side note, Meshcam expects 3D input, when pulling in a DXF you have to fake up a component thickness via an extrude operation before you can generate gcode. Flip side is that using a 3D format for your CAM input can save you having to enter all the pocket and hole depths, they are within the CAD file already.

Mach and linuxcnc both have demo modes where you can "run" your G Code in a simulation mode without even having a machine defined. This would let you sort out your process and get some experience with different software option before you have a machine up and running. There are free g code simulators of various quality all over the place, a quick search on gcode simulators will get you started.

Quickie suggested plan of attack:

1) Draw something simple in the CAD of your choice, which I believe is OnShape. Save as an STL and optionally as a DXF if supported. Make the object really simple, like a strike plate for a dead bolt. Two rectangles, four or eight fillets, two screw holes. Boring and trivial. There's enough weirdness coming up, don't add to it for the first time out.

2) Import the drawing into whatever CAM package(s) seem of interest, and learn to generate tool paths. You'll need to specify tools, and depending on the package all sorts of machine and cutter data (RPM, ramping rates, step over, all that stuff). Doesn't matter what CAM package, or if it's cloud or local, just play with the stuff and see what works for you. If your CAM package will do a simulation run have a look and see how things are playing out, otherwise onwards to simulators or controllers doing backplots.

You already speak slicer, KiriMoto is set up like what you already know from a quick look, so give it a shot. I did plop a small STL into KiriMoto and generated some gcode for linuxcnc and gRbl, although it appears to do line segments rather than arcs, so the file is silly big. Maybe a config setting, I just had a quick look. I did notice that KiriMoto does not take dxf files, sort of a big miss for a CAM package. Perhaps their engine is so tied to a slicer code base that 2.5D just isn't in the cards. Feels sort of weird to me, a subtractive CAM package that doesn't require workpiece dimensions, origins, other "normal" stuff before spitting out GCode, but maybe it's OK, just different. It does support roughing and contour passes, which is great, but in simulation it doesn't seem to honor the tool length settings, so the tool just dives under the surface and pops back up elsewhere. Not something any of my mills can pull off.

3) Plop the output of your CAM package into a gcode simulator. Try a few, some are pretty good, some suck, some are really good but look lousy, simulation is all over the place. Optionally, pull the gcode into motion control software running in a demo mode and simulate there. Camotics seems to be a pretty good open source simulator, with .deb and .rpm packages available. It does have some dependencies that if unmet may require manual installation, but the camotics page documents these. Some simulators are embedded within gcode editors too, so you might find interesting things using the search term "CNC backplot".

Always use the time option in searching for help. In the case of linuxcnc there are millions of posts and pages dating back probably 20 years. Most of it is only of historical interest if that, much of the old tech advice is completely outdated and in many cases now wrong.

Most stuff won't work or make sense at first, at least it didn't for me. Plod along one step at a time and it will all fall into place after an Aha moment or two.

Have fun,
Stan

 

[Jasonb]

A few thoughts on CAM but probably going to me more than 25words.

It would be nice if you can create the part and do the cam in the same set of software. For instance I tend to design in Alibre and then export a STEP file into F360, if I make a change to the part I have to import the revised file and start the CAM again or if it's a minor change I can do that in the design side of F360 so don't have to start the CAM all over again, just regenerate the paths.

As I mentioned earlier, think ahead of what sort of parts you want to make with whatever CNC you end up with and start with CAM that will handle that rather than invest time in a 2 or 2.5D CAM package and then have to learn or even pay for another at some time down the road. I do quite a few items that replicate castings or make patterns for casting as shown so soon found 2D CAM was not for me. They come out accurate enough for my needs

I only tend to use STEP files even if the part is just a profile cut from flat plate. It saves having to manually work out or enter things like how deep to cut, I just enter a distance below the bottom surface, or set a drill to break through by a set amount, F360 will work out how much to allow for the point of the drill to ensure the whole dia cuts to full depth. Same with any pockets, I can just click the area at the bottom of a pocket and set that as my depth rather than having to work it out if it was a DXF or DWG file.

I'll always run a simulation, often a few times as tweaks are made to the various operations and F360 shows up collisions such as if you cut deeper than the amount of tool sticking out the holder it will flash up when the holder hits the work. Better to see it on screen than in the workshop. The simulation also gives you the run time for each operation as well as total time so you can play about tweaking things like stay down vs. lifting, both way or climb cutting, non engaged feed rate , etc

 

[Richard Hed]

I thimpfks you guys are nearly as crazy as me. I would look at all the auctions all over the country, mostly from the rust belt east and south to the Carolinas, for used manual mills in the Bridgeport style. There are scores of them. If I lived in that region, I woujld already have two or threee of them/ You could always fix up a CNC bit, and man;y of them already have DROs. It's MNSHO that it would be better to have a Bridgeport style, hunky mill, than any of the toy models, but then I'm not everyone and some only need those small models.

 

[Jasonb]

As you are now thinking about a kit have another look at the beginning of this thread where Basil showed what his Onefinity machine can do, might be worth adding to the list.

Not sure I would want to CNC a Bridgeport, Spindle would be quite slow so you could get into very long run times particularly when doing small stepovers on finishing passes, my 5000rpm max could do with being higher for this sort of work but I'll live with it. This also applies to converting a smaller benchtop mill which will typically have a 2000 to 2500 max rpm, maybe a noisy hi/low gearbox and relatively basic bearings. I know a replacement set of bearings for my CNC are several £100GBP but for my similar size benchtop mill only 1/10th of that.