• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.

Solved Planning CNC build - Need advice on software!

Hai-Lee

Old and Bold RC PILOT
Mentor
#1
Please assist an old school technologist who lacks programming skills almost totally.

I have a CNC cutter in design process, (the mechanicals and electronics/electrics are easy to me BUT how to go from my plans which I produce in almost any format to the required code to operate the CNC cutter/milling machine is where the project falls down.

My hand skills are quite good and my ingenuity is also quite developed but I know when I do not know something, and whilst i have access to a plethora of tools including electric tools I do not have a 3D printer or even access to one. My income is meager, (a man with 7 children and 3 ex-wives never has much left unless he was an actor or a millionaire to start with. I am neither), So to decide to buy a 3D printer is a big decision especially when no one states the bed sizes required to print their latest great creation.

Currently I am planning to add a 3D printer head and a heated bed to the CNC machine I am designing along with a removable 4th axis for fuselage milling. Yes the bed is rather large, (not quite industrial but close). Whilst the cnc machine designs available both on the forum and across the internet are quite attractive there is a simple problem and that is the freight bill to land it in AUS where I live. HENCE the DIY approach. I found a smallish machine for around $400 in the US but the freight to AUS is an additional $800. See the problem!

Any help or advice on software and the conversion of plans to CNC code would be greatly appreciated!

Have fun!
 
Last edited:

SaltyGator

Junior Member
#2
I use Fusion 360 for design and generating the needed Gcode with the CAM function. It is free for hobby use and very robust. Just search Youtube for "Fusion 360" and you will get a plethora of video to watch.

Good Luck,
John
 

Hai-Lee

Old and Bold RC PILOT
Mentor
#3
Have already emailed them about a student like licence for a pensioner! Will wait for their response.

I may have to go the open source route yet! My needs may very well exceed the abilities of the free version!

Thank you for the response!
 

kilroy07

Well-known member
#4
And the student becomes the master! :LOL:

What is the working Z height of your CNC? I ask because my past build (a MPCNC) wasn't over 100mm... Mainly to limit run-out.
Most things that you will want to 3D print ar going to be in the 200-300mm range.

After awhile using some of the free stuff (although Fusion 360, does appear to be a solid application) I went with Estlcam.
It's not much (I think I paid $120-150 for it... but don't quote me on that!)
I like it's interface, it's pretty simple and down to earth and not as flashy as F360. (personal preference.)

I'm in the middle of building an open builds OX machine (I passed my MPCNC off to my cousin.)
So, until that gets finished I am without a working CNC machine (I already miss it!)

Where I am going with all of this is, it might be easier (although probably not cheaper) to go with a dedicated 3D printer...
There are some decent (well, with a couple upgrades) 3D printers in the $300-400 range. Yes, I know you might have a heated bed and nozzle setup for under a hundred... I'm just giving you my opinion...

If you have some 2020 extruded aluminum lying around I really like the wilson II design;
https://www.thingiverse.com/thing:1086101
I even modified the ends to make it a tad stiffer and easier to build (at least for me.)
https://www.thingiverse.com/thing:2765853

Most 3d printers have a 200x200 build bed, but that's the bare minimum now, with several designs in the 300x300 (or 400x400) range...
I started out with a prusa i3 clone several years ago and that's the first thing you want to upgrade is the bed (working area...)
Best if you choose a design that is 300x300 off the bat (the cr10s is a pretty good design once you sort out the safety hazards!)

Keep us posted, I'd love to see pics of your CNC, might give me modivation to get off my haunches and wrap not only the cnc build, but the second (and third) 3D printer on my workbench!. :LOL:
 

Hai-Lee

Old and Bold RC PILOT
Mentor
#5
Sensei,

Thank you for the lesson. here it seems that almost no one makes anything any longer. Getting extrusions in AL is horribly expensive especially if they are a little elaborate, (not plate or angle), I do envy the materials freely available in the US but that is another story.

The planned bed of my CNC is to be 900mm x 1100mm MINIMUM!
One of its first jobs will be to make replacement parts, (more accurately), for itself.
The 3D printer head is to be interchangeable with all of the other cutting, milling, and drawing heads.
The 3D printer bed is yet to be set in regards to dimensions, nominally 300x300x300mm at this time, but if the work height gets too high then the cooling of the ambient air can cause problems so a constant flow of temperature controlled air with a breeze curtain may be required.
The fuselage/wing mill is to have a maximum cutting head clearance requirement of the same 300mm but the actual drive mechanism is to be side mounted for clearance. (The actual requirement for the 4th axis might be supplemented or even replaced by exchanging the X-Axis drive motor connection from the bed to the fuselage mill), I also want to build or otherwise get a touch probe so that I can read existing fuselages into the mill and then replicate them in material of my choosing. A great feature for making replacement fuselages or wings for those who need one.

I have researched the Ox CNC but sadly here it is around $1000 plus delivery, (a distance of 1600Klm). Mind you the large Ox was still a little small for my immediate purposes.

I am getting the hang of the MDI input instructions as it is similar to the old vector graphic inputs from early day programming of vector graphic video games,(Did I mention that I am rather old now?). I believe that I could draw some of the simpler designs for FB planes in MDI but I would not rather spend that much time. I am able to draw and read blueprints, (did I mention that I also have qualifications in boiler making/welding?), so drawing up plans is not the issue but the conversion from drawing to machine code is my weakness at this time.

Sorry for my tardy response but I am trying to make planes for a large order I received from the local model shop for a few of my own design), at the same time I am perusing the FT forum. It takes far too long to cut out, assemble, paint, and finish a dozen or so of the same model Hence the need for the CNC.

Thank you for your response I will read it more than once I can assure you!

have fun!
 

kilroy07

Well-known member
#6
so drawing up plans is not the issue but the conversion from drawing to machine code is my weakness at this time.
If the software you use to draw plans can export to DXF, you are all set. Again, I use Estlcam to pull in the DXF then assign tool paths/depths etc. I haven't tried Fusion 360 (mainly because it is a direct competitor for my current employer! :LOL:) but from what I understand it is a competent program and I might suggest you start your search there, unless you are more comfortable designing in your current software and exporting to DXF for the milling (in which case I go back to Estlcam...)

Probably the nicest thing about Estlcam is it was designed for the hobbyist to use on a DIY machine (such as the OX or MPCNC) so it will even flash the firmware on the driver board for you.

It's a shame you find it hard to get aluminum extrusions there... It's like an erector set for grownups!
What I have done was to build the MPCNC to mill out the parts for the OX (it milled 3mm Al plate just fine!)
https://www.thingiverse.com/thing:724999
And while it does seem quite wonky to use pipe and skate bearings for a precision* machine, I can tell you it worked surprisingly well!
(Before I gave it to my cousin I had thoughts of attaching a plasma cutter to it... AND (btw) turn it into a large format 3D printer...)
*Obviously we are talking about hobby grade, although the OX can be built to be quite rigid.

Your main limiting factor for size of your 3D printer will be the HEATED bed. I stress heated because although you "can" print without one, you are going to be limited to PLA if you do... Which BTW is a GREAT material to work with! It really isn't a sensitive to drafts as ABS is. So your concerns about a build enclosure aren't necessary (Unless you want to print nylon or other more "exotic" materials... Then you will NEED a heated bed.) Also, one thing I learned is a Bowden setup won't be able to print flexible filaments... So you will probably want to stick with a direct drive extruder.

Also start shopping for a power supply now, a large 300x300 or larger is going to take quite a few amps to get (and keep) it at tempature (I run PLA at around 60C) I started with the cheap ones and after replacing/upgrading diodes and such I found some super medical grade power supplies on Ebay (and promptly bought a couple spares) You will want to go with a 24v system, your steppers will actually run cooler with the increased voltage (and smoother too!)

Alright, enough rambling for now... Good Luck with your build (and as always) if you have more questions just ask!
 

dkj4linux

Active member
#7
Feeling like the new guy on the block, please ignore me if it seems I'm butting in...

With regard to what kilroy07 has told you about the software side of things and his experiences with the MPCNC, I concur wholeheartedly. I have a fairly long-running thread on this very forum regarding CNCs and cutting foam with a needle if you want to check out where I am coming from (https://forum.flitetest.com/index.php?threads/cutting-foam-sheets-with-a-needle.24251/). I'm not connected in any way with the MPCNC folks -- just a satisfied customer -- and I have other machines as well... Openbuilds C-Beam, BuildYourCNC, Makeblock, etc. I'm also a 72-years old, tired/retired electrical engineer, and often confused... so please take what I say with a grain of salt if it doesn't make sense to you ;)

I have built/rebuilt numerous MPCNCs and variants and it is a fine machine for what it is... relatively inexpensive and flexible, with amazing functionality and precision given it's constructed with everyday, garden-variety, parts; i.e. EMT conduit, skate bearings, common nut/bolts from your local hardware store, readily-available common 3d printer hardware/electronics/firmware, etc. Though considered quirky by some CNC "purists", a lot of its flexibility comes from its shared lineage with 3d printers... which are themselves small CNC machines with added bed and extruder drive/temperature controls. The "stock" MPCNC, when used as a CNC machine, taps into the underlying, inherent, CNC functionality... and when used as a 3d printer, simply taps into the already builtin 3d printer functionality; i.e. it's not add-on functionality as it is with conventional CNCs. So, as a hobby-grade machine, it's hard to beat. At one time, with printing my own parts and buying the electronics/hardware bundle from the MPCNC shop, I was building machines up to 900mm x 1200mm for about $300 per machine. It's a bit more now with slight price increase for better electronics, but, even if you have to buy the printed parts bundle, you can still get a machine going for $500 or so.

I second the Estlcam recommendation. Fusion360 is, without doubt, quality, powerful, and flexible software... but overkill for most hobbyist's needs. There is so much capability for heavy-duty, high-end, CNC work that it can be very intimidating for those who don't do that kind of work for a living... and a great deal of it simply isn't needed for most hobbyist's needs. Estlcam is far simpler, though very capable, and runs about $60-$70 for the license... and the author, Christian Knuell, frequents these forums and has insured Marlin-compatibility for the MPCNC folks.

That said, a general word...

With a limited budget and yet a desire to do as much as possible with one machine... please recognize that all CNC jobs are not equal; i.e. milling/cutting jobs require more rigidity and power than engraving/needling/lasering/drag-knifing/etc. So, regarding the CNC machine you build.... smaller/shorter is more rigid/faster, larger/longer is less so. I recommend that you identify your most pressing need/requirement and let that determine how big, or small, the machine you build. If your tool load/forces will be light, you can get away with a bigger (larger dimensions) machine... smaller, if the tools and forces are heavier. It'll be better to have separate machines if you need to do jobs at opposite ends of the spectrum... and, with respect to DIY/hobby-grade machines, having a 3d printer can make that a distinct possibility. Build the one that suits your immediate need, gain the experience and knowledge that comes with building/using it for a while... and then you'll be in far better shape to decide what you need to meet your future requirements. It is, indeed, "boot-strapping" your way into a very exciting and rewarding field.

There's far more to say but I'll stop here. And, as always, this is just my $0.02... ;)

-- David
 
Last edited:

Hai-Lee

Old and Bold RC PILOT
Mentor
#8
kilroy07
Sadly my drafting program is a little limited and does not export in DXF. Been looking for a plug in but so far no joy.
I will look into Esticam and see what they have to offer and what the software limitations are.
As for the machine I have in design stage it will be using 40V power supply and quite powerful stepper motors as it would be maddening to try to mill or cut a large job at 1mm, (or less), per minute.
Your post will be read often to extract any info I have missed at this time,

dkj4linux
I have read you posts on CNC machines often and you have some interesting ideas. Your needle cutter is cute and may either be incorporated in full or adapted to my FB cutting duties.

The structural side of large bed construction is understood and it seems that the current industry answer is to pour in large masses of structure or volumes of metal. Whilst my original or first build may be a little minimalist I do plan to use structural bracing techniques and alternate materials to give a large strong and extremely rigid structure with minimal expansion or shrinkage due to changes in ambient temperatures in my final design or build.

Eventually I hope, (if I live long enough), to have a large bed machine that can be broken down, moved, reassembled, and put into production in a single day by a single older person, (me).

Due to persons like those who post here and supply information across the internet the design is starting to gel. The one thing I would need some assistance with is a home made touch probe. It cannot be electrical contact/circuit type and will need to be extremely sensitive. So far the most sensitive design I can come up with is a needle cup design, (miniature), but it is electrical and still requires a pressure to make contact.

Bed protection is another area where my knowledge is still a little sparse!

I thank you for your posts and the information contained therein so far and hope I can milk more pearls of wisdom through this thread before I start my marathon build.

Have fun!
 
#9
The one thing I would need some assistance with is a home made touch probe. It cannot be electrical contact/circuit type and will need to be extremely sensitive.
I'm not sure what you mean by "It cannot be electrical contact/circuit type" as I'm sure they all are, but I think you mean the material you are contacting is non-conductive. I think this may be what you are looking for?

https://www.thingiverse.com/thing:1393731
The designer claims some good accuracy results. I have built one (well, I modified his a little so that the tool shaft and probe shaft are guaranteed to be on the same center) but I have yet to test it.

If this is what you are looking for, I imagine you could get the parts made at a local Makerspace for cost of materials.
*I should have read the entire thread. I see you feel that getting 3D printed parts is really out of the realm of possibilities. In theory you could make the parts for that touch probe once you have a working CNC. Could cut all the parts out of solid plastic.

Shur
 

Hai-Lee

Old and Bold RC PILOT
Mentor
#10
I'm not sure what you mean by "It cannot be electrical contact/circuit type" as I'm sure they all are, but I think you mean the material you are contacting is non-conductive. I think this may be what you are looking for?

https://www.thingiverse.com/thing:1393731
The designer claims some good accuracy results. I have built one (well, I modified his a little so that the tool shaft and probe shaft are guaranteed to be on the same center) but I have yet to test it.

If this is what you are looking for, I imagine you could get the parts made at a local Makerspace for cost of materials.
*I should have read the entire thread. I see you feel that getting 3D printed parts is really out of the realm of possibilities. In theory you could make the parts for that touch probe once you have a working CNC. Could cut all the parts out of solid plastic.

Shur
That design is brilliant! A little worried about the size and hence the weight but everything can be scaled down!
The material I was looking at measuring is Styrofoam or similar and so too great a touch force is not ideal but that design has opened my eyes a little wider. I was looking at an optical device with very little mass but now I rethink the entire probe.

Thank you for the info!

have fun!
 

rockyboy

Skill Collector
Mentor
#11
I've gone the online method for CNC / GCode generation software with http://easel.inventables.com

I still use Inkscape & DoubleCAD XT for vector based design work, and then import that into Easel for CNC preparation and gcode file generation - all free!! It's not as capable as Estlcam but until I have a spare $150 it works quite well for me
 
#12
A little worried about the size and hence the weight but everything can be scaled down!
I don't understand the comment about the weight. I totally understand not wanting high contact forces due to the soft material (styrofoam) but the weight of the probe shouldn't really be a big concern as your CNC will be supporting the weight of the probe, not the styrofoam itself.
 

Hai-Lee

Old and Bold RC PILOT
Mentor
#13
I don't understand the comment about the weight. I totally understand not wanting high contact forces due to the soft material (styrofoam) but the weight of the probe shouldn't really be a big concern as your CNC will be supporting the weight of the probe, not the styrofoam itself.
The weight is only a consideration as to the soft material of the item to be measured and how much deformation it takes to break the contact in the probe. I can see it will be quite minimal but the point contact area means that it is something to be concerned about. In addition the surface strength of the target material can be variable as the resistance to deformation in the middle of one of the foam cells is far less than at a point where a number of cells are in contact with each other.

Not to worry though because until I actually build the target device and start adding all of the accessories as well as doing a few measurements nothing is really certain.

I thank you for your info!

have fun!
 

kilroy07

Well-known member
#14
Feeling like the new guy on the block, please ignore me if it seems I'm butting in...
NOT at all!

In fact, I've pointed several people that direction who were asking if those diode lasers will cut foamboard. After I break the news that they will probably need a CO2 laser, I often mention the needle cutter route. (I would still have my MPCNC running one if I didn't have the laser I have now 700x500mm 60watts "True" power).

The more the merrier!
 

Hai-Lee

Old and Bold RC PILOT
Mentor
#15
Still in progress but have decided that I will purchase a large bed 3D printer to help build my CNC machine.

Found one for around $200 AUD delivered and it supports USB and SD card in basic form and uses standard 1.75mm filament

Have been learning like mad and funnily enough most of the material and software appears familiar in operation and feel to other software with which I am already familiar with some serious add ons of course.

I like the Price and features of FreeCAD but it cannot import PDFs, though apparently it can export to Gcode directly.will I still require a slicer program? Some advice still need in this area.

The search and steep learning curve continues.

Have fun!
 

kilroy07

Well-known member
#16
Print from the SD card. That will eliminate some frustration when windows decides to hiccup in the middle of a print.
You will want to hook it to the computer to do the setup, tuning the PIDs for example (how comfortable are you with the concept of flashing hardware/arduino?)

You can always use inkscape to translate the PDFs to DXF (which I'm sure free cad will read.)
It's what I use. Really simple process too.
Open inkscape, import PDF, Save as (select DXF as format) and that's it!
Then pull the DXF into whatever you will use to do G-code for the CNC (I use Estlcam, but there are others.)

A slicer takes the 3D item (usually an STL, there are others) and makes the G-code for your 3D printer.
You won't use a slicer for your CNC, and you can't print PDFs on your 3D printer.

What brand/model of 3D printer? Be aware that some of the bargain level printers can have scary stuff like open mains for power!
(But a bit of tinkering about goes with the hobby now doesn't it?!)
 

Hai-Lee

Old and Bold RC PILOT
Mentor
#17
Print from the SD card. That will eliminate some frustration when windows decides to hiccup in the middle of a print.
You will want to hook it to the computer to do the setup, tuning the PIDs for example (how comfortable are you with the concept of flashing hardware/arduino?)

You can always use inkscape to translate the PDFs to DXF (which I'm sure free cad will read.)
It's what I use. Really simple process too.
Open inkscape, import PDF, Save as (select DXF as format) and that's it!
Then pull the DXF into whatever you will use to do G-code for the CNC (I use Estlcam, but there are others.)

A slicer takes the 3D item (usually an STL, there are others) and makes the G-code for your 3D printer.
You won't use a slicer for your CNC, and you can't print PDFs on your 3D printer.

What brand/model of 3D printer? Be aware that some of the bargain level printers can have scary stuff like open mains for power!
(But a bit of tinkering about goes with the hobby now doesn't it?!)
This is what is at the top of the short listing at the moment;
ANYCUBIC I3 Mega 3D Printer Plus Size Full Metal Frame Lattice Platform Desktop Nozzle 3d Drucker Filament Kit Industrial Grade

Item specifics
  • Brand Name:ANYCUBIC
  • Color Print Speed:20-100mm/s
  • Model Number:Anycubic I3 MEGA
  • Printing Technology:FDM (Fused Deposition Modeling)
  • Layer Resolution:0.05-0.3mm
  • Travel Speed:150mm/s
  • Supported Print Materials:pLAABSHIPS Wood
  • Build Size:210x210x205mm
  • Positioning Accuracy:X 0.01mm Y 0.01mm Z 0.002mm
  • Gift:8G SD Card x 1 kg Filament x 1 Ultrabase
  • Waranty:One Year Waranty
  • Package:1KG Printing Material Inside
As for the Matters around electrical safety I have trade qualifications in electronics and communications and so know how to make things safe and how to plug it all together!

It does use drive belts with all the issues that go with them but then it is just an entry level unit and after the warranty expires I should be in a position to either fund or build a replacement.

The reason for the bed size is that I want to print a few replacement engine cowls for crash repairs eventually and most little entry level units are too small to do a single print at the sizes of some of my models.

There are cheaper units available believe it or not, including some mentioned in the forums but for a few dollars extra I am shooting for a capability which is beyond my current expectations. (I do not know the future but I remember the past very well!).

Any comments are welcomed as well as any experience that anyone has had with this particular printer model.

have fun!
 

Hai-Lee

Old and Bold RC PILOT
Mentor
#18
It's been a while but my 3D printer kit arrived. Talk about a jigsaw puzzle! But it was complete and the assembly instructions were adequate.
Did my first test print and it came out pretty good.

Installed the accompanying software and that took longer than the printer build. No installation documentation and a rather roundabout process it was to get it setup and configured for my printer.

Anyway starting with slicing and printing a few existing items from thingyverse over the next few days before I start at the process of my own cnc cutter machine..

Having fun!
 

kilroy07

Well-known member
#19
Did my first test print and it came out pretty good.
Awesome, welcome to the club!

Just curiosity, what software specifically (firmware and slicer)
For reference I am running Marlin 1.1.8 and Simplify3d.

Firmware is a pretty steep learning curve, getting stuff connected shouldn't be too bad, but there's a LOT going on in later versions of marlin. You are probably just fine right now, but once you start tinkering (and you will!) you'll need to start taking a deeper look into the software really doing the heavy lifting (that said, slicers DO make a difference as well.)

For test models I swear by benchy; https://www.thingiverse.com/thing:763622
While it looks pretty simple, there's actually some clever engineering to really give you an idea on how your machine is dialed in.
This is a really good guide for troubleshooting;
https://www.simplify3d.com/support/print-quality-troubleshooting/

One final word, filament DOES matter. Find a brand (doesn't have to be the most expensive!) and stick with it for a couple of rolls.
And FIRST get a good measurement on the filament width, most slicers have a field for the ACTUAL width of the filament you are working with.
That alone can make the difference from a side that has ridges to a smooth(ish*) finish. *the way the object is formed you will always have some layer effect on the sides... but with careful settings you can minimize that to a large extent.

Have fun!!
 

Hai-Lee

Old and Bold RC PILOT
Mentor
#20
The printer, (a Tronxy Education model) came bundled with repetier-host and included slic3r and cura. There was a printer config file for use with slic3r as well as a USB.serial driver and com software plugin for talking with the printer for my PC, (all for $200 AUD delivered to Aus).

As stated there were no software installation instructions or directions as to how to get it installed and working so it was a matter of trying all possible combinations until I got it working.

As for the filament I have purchased some HK filament, (based purely on price at this time because I do not see the need for wasting the expensive filaments at this time plus if I get it performing adequately then at least the filment is cheap enough to waste and a definite bonus if I get it working reliably.

The test print was is small cylinder mounted atop a small square. A pretty simple but indicative print. For future testing I will be using mt own test print of an RC pilot which I can use in my plane designs and give away at my local club, so I can print many of them without wasting a single one of them. It is a pity that FT doesn't have a couple of FT LOGO pilots to suit their plane designs!

Now I suppose it is time to get used to using the printer and the various settings so when I get the CAD software I only have one thing to learn!

Having fun!