Solved Planning CNC build - Need advice on software!

[jhitesma]

Don't feel bad. I'm kind of a software guy by trade (backend website developer) and it still took me longer to work out the software and workflow for my printer than it did for me to assemble it And that's despite my kit being more of a "box of parts" including incorrect instructions and the fact that I had spent quite a bit of time researching the software side before ordering the printer!

I'd warn against using cheap filament just because it's cheap. I haven't used the HK filament but I've used quite a few different brands and almost universally cheap filament is harder to print successfully with. Most of the cheap filament I've used just doesn't stick to the bed as well resulting in warping...but some also jams in my all metal hotend if I don't pre-oil it. More expensive filaments I've never had issues with. I did have good luck with cheaper Hatchbox from Amazon...but the price on it has gone up and availability has been spottier (and one time they sent me esun instead which was hands down the worse filament I've ever tried to print with.)

You can get good results with cheap filament. I've used a ton of MakerGeeks from their grab bags which works out really cheap if you don't mind not being able to pick a color...but it does need to be pre-oiled with my v6 all metal hot end. And other than that one roll of eSun even the cheapest filaments have given great results once I got things dialed in....though in some cases that took almost half a spool of experimenting to figure out what that filament liked.

On the other hand every roll of Atomic filament I've used has "just worked" with more or less default settings and printed amazing with great bed adhesion and easy release once the bed cooled. They're more expensive...but it's a lot easier to get great results with good raw materials!

 

[dkj4linux]

I've mentioned it several times in different threads but I love the Sunlu PLA filament from Amazon... and AFAIK I'm still the only one using it
I've gone through probably half a dozen rolls and it has played flawlessly on my Prusa MK2S. Of course, I'm sold on the Prusa printer also and their custom Slic3r and configuration... so maybe that has something to do with it. Anyway I really like this filament...

https://www.amazon.com/dp/B073LCCJKP/?tag=lstir-20

 

[Hai-Lee]

Well I just tried out both my new HK filament and the new test print.

Sadly the much larger pilot did not adhere well to the bed and I lost it at around 70%. Not happy but lesson learned.

Not having any hairspray, painters tape, or any of the other recommended coatings to improve the adhesion of the print bed I looked around for something I could use. Well I have a massive pile of FB offcuts. Yep you guessed it I made myself a quick and rough print bed out of a FB offcut. I fitted the new filament because the little sample supplied with the printer was not going to be sufficient for the print. Levelled the bed and hit print!

It worked brilliantly with exceptional adhesion though I will admit I used 3mm FB. upon completion the print was quite good and I had to rip the paper off of the FB to detach it. Wet the paper that was stuck to the printed object and just rubbed it all off. Finished object is very usable and possibly even salable so good is the print.

Even at my age I feel like a kid with a new toy now to decide my first real printed project!

Should have bought one ages ago!

Having fun!

 

[kilroy07]

Sunlu PLA filament from Amazon... and AFAIK I'm still the only one using it

Nope, there's two of us!

Even at my age I feel like a kid with a new toy now to decide my first real printed project!

Yea, it's going to open up whole new worlds for your other hobby interests (not that it isn't one in it's own right.)

Might I suggest you look into a PEI sheet for your build bed. For almost two years I fussed with glue sticks painter's tape etc (never really tried hairspray.) and finally found PEI. I haven't looked back since! I have tried other build types (I have Gecko print now) but I'm going back to pei.
Now, there ARE certain filaments that you can't use with PEI, too close chemically and they will literally weld together... but your probably a ways off from trying anything crazy like that... Aren't... You... ?

Seriously though, I'm glad you pulled the trigger! I can't imagine my hobbies (and there are WAY too many) without one!

 

[Hai-Lee]

OK Chaps I have become stuck yet again!
I downloaded inkscape and can easily import a PDF, Clean it up, and export it as an SVG file.

With my CAD program I can make basic complex shapes and can export valid STLs for my 3D printer but importing the inkscape derived SVG files, (plans from a PDF file), I am unable to get any STL that has paths included when I export the imported SVG file to STL. Am I missing something basic? For the CNC cutter should I be using DXF file format?

I have 2 separate operations sort of sorted out but I cannot make the linking operation to join them!

Not having fun here!

 

[kilroy07]

Hopefully I can help.

Typically* PDFs are going to be 2D (line-work) such as FT plans.
Your going to take those and export them out to DXF (the line-work from the PDF will become your cut paths for the tool)
Run them through some sort of software to create G-code to plug into your CNC/laser whatever to cut out your parts.

STLs are always 3D. They are a triangulated representation of a 3D object. You will process an STL through a "slicer" program that will take that "object" and create the resulting G-code that you feed to our 3D printer and it will (hopefully) create the object.

So yes (IF I understand your question) you have apples and oranges.
Two similar workflows but two very different results (one 2D** one 3D.)
The good news is that there's a lot of learning that is crossed linked (you probably have everything sorted out, just step back for a second and take a breath.)

I am not sure about "linking"...
Do you have a 3D PDF you are trying to get parts from?
Do you wish to take say a plan of a wing rib from a PDF and print that out on your 3D printer? If so, your going to need to do some CAD work in the middle there, it should be pretty easy to take a closed shape (Rib) from the PDF line-work and extrude it (give it some thickness) to make a part the 3D printer can produce.

I hope I've helped a little, but I'm not entirely sure where you think you are off track. Seems like you are grasping the workflow(s) just fine.

*Yes, there are 3D PDFs...
**Yes, very often CNC operations are in fact 3D (pockiting as such) but I hope you get the gist of whjat I was trying to illustrate.

 

[Hai-Lee]

Hopefully I can help.

Typically* PDFs are going to be 2D (line-work) such as FT plans.
Your going to take those and export them out to DXF (the line-work from the PDF will become your cut paths for the tool)
Run them through some sort of software to create G-code to plug into your CNC/laser whatever to cut out your parts.

STLs are always 3D. They are a triangulated representation of a 3D object. You will process an STL through a "slicer" program that will take that "object" and create the resulting G-code that you feed to our 3D printer and it will (hopefully) create the object.

So yes (IF I understand your question) you have apples and oranges.
Two similar workflows but two very different results (one 2D** one 3D.)
The good news is that there's a lot of learning that is crossed linked (you probably have everything sorted out, just step back for a second and take a breath.)

I am not sure about "linking"...
Do you have a 3D PDF you are trying to get parts from?
Do you wish to take say a plan of a wing rib from a PDF and print that out on your 3D printer? If so, your going to need to do some CAD work in the middle there, it should be pretty easy to take a closed shape (Rib) from the PDF line-work and extrude it (give it some thickness) to make a part the 3D printer can produce.

I hope I've helped a little, but I'm not entirely sure where you think you are off track. Seems like you are grasping the workflow(s) just fine.

*Yes, there are 3D PDFs...
**Yes, very often CNC operations are in fact 3D (pockiting as such) but I hope you get the gist of whjat I was trying to illustrate.

Thank you for your response it has helped lift the fog a little! The linking I was referring to was the linking of the first steps of the process, (importing,cleaning and exporting with the use of the g-code and the CNC device/machine.

I am experimenting with the importation of the first FT plans I downloaded years ago now and not really a complicated build and it is obviously a 2D plan. Sadly I had been doing a lot of 3D printing and basic object design. Now I understand that there is a difference in handling procedure anmd I just have to get it sorted.

Have fun!

 

[TheSFReader]

Whereas in 3D printing, the slicer does the job of creating laying down layer upon layer, with CNCs, the jobs is reversed, and needs to remove material.

This is the work of the "CAM" software.

In the case of plans, such as these, the CAM only needs to translate the 2D lines (with additional depth) into G-Code "cutting" trajectories.
Using ESTLCAM, it's as easy as importing the DXF file (which contains al the graphics vector it needs), and declaring "tool passes/cuts" (engraving is the preferred way) along the different lines. For each cut operation, you tell its location (the line to follows), which tool to use (the needle in case of a needle cutter), and at which depth it needs to cut.


The Tools parameters specify mainly its general dimensions (radius, end point etc.), the speed at which it travel, and the max depth of cut per pass (if you need 3 mm cut and DOC is set to1mm, the CAM will generate GCODE for 3 passes)

This page could provide some additional information/context... https://www.v1engineering.com/estlcam-basics/

If you want to use it as a general CNC, it's mostly the same, except using other tool path operations (pockets, parts etc.).
(cf. https://www.v1engineering.com/estlcam-2-5d-routing-intermediate/ )

The links I provide above are to estlcam tutorials, but the ideas apply on othe CAM software, including the much more complete, but also much more configurable/steeper learning curve CAM module in fusion360.

 

[Hai-Lee]

The parts for the change to my 3D printer bed size are on their way!

The mechanicals of the CNC machine are easy and available in kits or bundles. I can just select what I desire or I think I might be best for my application.

Here is the fly in the Ointment! What controller firmware should I chose. I have already experienced the different "FLAVOURS" of CNC gode, i.e. NC, GCode etc. and not all versions of the various codes are equal!

My 3D printer uses reprap G Code. My various CAD programs output in their own special flavour or dialect of the cnc code.

Please would someone who has the knowledge please advise.

I can now see why the US government legislated to have EMSs use a single interface standard! Pity they did not do the same for something as economically important as CNC machining!

have fun!

 

[kilroy07]

I can tell you what I've done.

With my MPCNC I used Marlin for the firmware. I was already familiar with it from my 3D printers and had experience flashing it to the Arduino controller.

I know you didn't care for Estlcam, but it allows for different output flavors... it will even flash it's own flavor of firmware on the board for you (although I'm not sure if the pinouts match a RAMPS board or not... (if you are even using a RAMPS/Arduino setup...) In any case it had a setting for Marlin and it worked flawlessly for me.

I will most likely use a similar setup for my OX CNC although I might try a spare RAMBo card that burned out the channel for the heated bed awhile ago.

 

[Hai-Lee]

I will be closing this thread or at least letting it lapse into obscurity. I have a handle on the software offerings and have selected what I know will do an acceptable job.

My greatest disappointment is the way people have obviously chosen a particular software, (often with a price tag attached), and then offered little in the way of assistance in how the software need to be configured for a particular application or desired output. This combined with a serious lack of any assistance in providing any real design information beyond the nuts and bolts of the physical build.

When I am ready I hope to be able to start a thread on a CNC router/cutter which will provide all the required information so that those who are not CNC literate will be able to know what to purchase and from where to build their own. There are many forum users who would love to build their own CNC and some even have the funds to do so but are lacking the confidence that their build will ever work as those demonstrated in the forums.

Anyway enough steam!

Have fun!