Newbie question: 3d printing vs CNC machining

I am thinking of building a 3D printing machine that is solid enough to do light high spindle speed routing in aluminum. I know a machine can be built to do this and that is not a problem for me. What I am not sure of is whether 3D printing software will support CNC type programs. Do I need two discreet driver cards and softwares to accomplish both goals? It seems that the usual 3D printing software writes in G code which CNC CAM programs do as well. Each can derive a G code set of instructions based on a solid model (such as Solidworks) so there is much similarity.

So how do I go about this? Is there maybe another forum I should be asking this of? Thanks!

Randy

You want to conserve on the costs of owning two machines by combining them into one?

The firmware for a CNC machine is fundamentally compatible with a 3D printer, but you are going to have to tweak the slicers to get them to cooperate, and you will always need to use only one aspect of the machine at a time (CNC OR 3DP). It should be doable.

Look at KSSlicer for 3D printing, and hard to find code for free (NON linux mach3) software exists. Freemill and Pycam worked well.

Skeinforge can be used to generate usable milling path that can be sent direct to reprap. When I last used it, it didn't try to move around the part when moving to a new location so you had to set it to lift z on every non cutting move. Worked well enough, if not a bit inefficient.

RepRap firmwares work with "regular" G-code just fine. At least the basics, like G0, G1, etc. Just don't expect support for advanced stuff like G-code variables, drill cycles and similar.

FWIW, I make milling G-code with Catia, a commercial high end CAD application. Just as simple as doing the same for older CNC milling machines. My firmware is Teacup. Others should work just as fine, at least with pure X-Y moves.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Actually, I want to build a 3D printer with exceptional accuracy. Many designs on the market seem to be creeping up on rigidity but at a greatly increasing expense to the buyer. To me, rigidity is a fatal flaw in most every printer design I have seen, not that is the only variable affecting print quality however.

I thought that by using my experience as a machine designer I could at least eliminate the usual platform instability which seems a weakness inherent to all home printers on the market today.

And, if I can build a rigid machine for accurate printing, why should it not be adaptable to 3D machining as well? I have seen many threaded-rod examples using plastiic printed parts that surely must be flimsy at best. And linear bearings held in place by plastic wire-ties on wood bases, my gawd! And also plywood structures used both for printing and wood engraving. Plywood? Why has no one built a decent rigid machine yet?

I am not sure I want to pursue this as I know of the time involved. But I can do it if I decide to. I will need help with controls however. My background is limited to PLC controls which do not support a project of this type.

Still thinking about the cost and time involved though. Will definitely need help in the controls area. Just thinking out loud here.

Randy

Coming from a machist background I agree with you, and encourage you to try, but remember this, most 3D printers today have approached the limit where their inherent randomness (which direction the blob of plastic decides to go) is a greater variable than problems like sturdiness of the frame. It used to be that was too big a problem, but designed like the Up! printer have really solved this.

I've been trying to build a CNC hobby style machine with extreme emphasis on low cost ($100) and quiet operation. I was able to find bearings for $3, leadscrews for 99 cents a foot, and so on. Everything worked out to almost 50 micron cumulative error. However, if you design it right, you could net even better accuracy by building a design that handles poor tolerances. Its still weakest at the center of the rods where deflection can be highest.

Without building a heavy expensive cast iron frame, I don't see how to improve further. Perhaps using hollow rod with 3/4"-1" outer diameter is the next place to go? What is your strategy?

best of luck

Randy Wrote:
-------------------------------------------------------
> Actually, I want to build a 3D printer with
> exceptional accuracy. Many designs on the market
> seem to be creeping up on rigidity but at a
> greatly increasing expense to the buyer. To me,
> rigidity is a fatal flaw in most every printer
> design I have seen, not that is the only variable
> affecting print quality however.
>
> I thought that by using my experience as a machine
> designer I could at least eliminate the usual
> platform instability which seems a weakness
> inherent to all home printers on the market
> today.
>
> And, if I can build a rigid machine for accurate
> printing, why should it not be adaptable to 3D
> machining as well? I have seen many threaded-rod
> examples using plastiic printed parts that surely
> must be flimsy at best. And linear bearings held
> in place by plastic wire-ties on wood bases, my
> gawd! And also plywood structures used both for
> printing and wood engraving. Plywood? Why has no
> one built a decent rigid machine yet?
>....


You're going to eat those words when you see some of the better units around.

...And even those "flimsy, wire tied lm8uu, threaded rod/printed parts" machines can do some amazingly nice prints once calibrated, they were not designed for CNC machining. But look into some of the more rigid designs that comprise this massive project and you will see many frames that are more than adequate.

We are not typically moving around 20-100lb toolheads... It's all about intended application. You can take one of the better existing designs and simply change the frame material and add proper ball screws, and you have a high end machine.

Edited 2 time(s). Last edit at 01/15/2013 01:28AM by xclusive585.

lm8uu
....ughhhh
I can't figure out how to get them to not sound "crunchy" with force on them. Even tried 4 manufacturers, some where even worse and cut channels into the slide rod.

Simba Wrote:
-------------------------------------------------------
> lm8uu
> ....ughhhh
> I can't figure out how to get them to not sound
> "crunchy" with force on them. Even tried 4
> manufacturers, some where even worse and cut
> channels into the slide rod.
Mine are garbage too. Same crunchy noise.

Step one, machine grade quality Fafner or similar quality.
Step two, hardened precision rods.

That's really the only proper way for a linear setup to work.

I'm using regular old smooth rod and crap lm8uus, and of course they dig grooves into the bars, I expected as much, because even cheap bearings balls are harder than cheap smooth rod.

Simba,

You seem to have a grasp on the situation. I have begun to understand that rigidity is not the only variable affecting print quality, but it is still an important one The very nauture of the "goop" being printed has a lot of variables associated with it such as adhesion, temperature, film thickness, and the consistency of application. And also cooling rate which affects distortion of the build with regard to "pulling" of the printed material. It seems that home printing today remains an art rather than a science.

I too have considered a heavy cast-iron frame as it can provide dimensional stability and can resist harmonics due to vibration in the case of routing. Heavy is good for accuracy. The points you raise are equally vital concerning the printing variables such as positioning of the "blob". That is worrisome as different materials/temperature/feed rate/resolution will produce different results.

I have thought of some of the things that you mention, such as hollow rods to reduce mass. My design would eliminate the need for positioning heavy oblects though so that these items are stationary and add rigidity without a need for movement. I see a gantry style with heavy uprights and equally heavy rails for table support. No need to move these support items.

I need to get to bed. I look forward to future exchanges.

Randy

... attached are some images of the first tests with my CNC-repstrap - it's more accurate than the typical RepRap and perfectly rigid ... but you can see artifacts and distortions caused by flowspeed-issues ('waves' on the sides) and temperature changes.

So yes, with a rigid frame you'll get better accuracy, but tweaking/optimizing the extruder, temps and printing parameters are essential too ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Yes, I see Viktor

So rigidity is but one variable in the search for a perfect print. I am learning as I go and I thank you for your examples. They cause me to think more deeply. It is not so simple is it?

Randy

It is and it isn't.

At the end of the day, you can make a very respectable part without much effort, and most of the bells and whistles don't help much. If you are very lucky, it will have 25-50% of the shear strength of the original plastic against the layers, though you should have similar compression and somewhat reduced tensile strength. I find this best with 15 degrees over the recommended printing temperatures, and few people realize just how poor their layer to layer adhesion is. Personally, I try to make usable end products that I drill & tap, not just models.

If you are a mechanical genius, perhaps you can dedicate your time to the more useful challenge of making new plastic parts or even metal parts available as a printing filament. I think cost & rigidity are good enough for now. But we lack variety of materials.

Randy Wrote:
-------------------------------------------------------
> And, if I can build a rigid machine for accurate
> printing, why should it not be adaptable to 3D
> machining as well? I have seen many threaded-rod
> examples using plastiic printed parts that surely
> must be flimsy at best. And linear bearings held
> in place by plastic wire-ties on wood bases, my
> gawd! And also plywood structures used both for
> printing and wood engraving. Plywood? Why has no
> one built a decent rigid machine yet?

For a very simple reason: 3D printing does not need rigidity. (Well, it needs some, but almost none compared to milling work).

If you want a combination machine, the drive type (belt vs. screw) will be at least as big problem as the frame rigidity. The belt drive lacks rigidity but a rigid screw drive is very slow.

I looked at these issues and decided that trying to make a combined printer/mill machine makes no sense. So I designed a printable CNC mill conversion instead. (Details here)

> I am not sure I want to pursue this as I know of
> the time involved. But I can do it if I decide
> to. I will need help with controls however. My
> background is limited to PLC controls which do not
> support a project of this type.

You can run a 3D printer and a mill with same electronics and same firmware. Just hook up the motors and set the step values correctly and it works just fine.

Randy,
I just started building my Ifactory. Printer, vertical mill, horizontal mill, lathe, knife cutter, plasma cutter, whatever. After looking for a long time I came across 4 identical THK linear slides with caged ball bearing and ball screws, and 18 inches of travel, ready to mount motors. No speed problem here, these were made to move fast. The other day I picked up a precision cast iron surface plate that is 24" x 24" and weighs about 200 lbs. I was looking for parts to cobble together a milling head when I ran across a manual Sherline lathe with a milling column atttachment, and a pile of accessories. So now I have a little manual mill to make parts to build the CNC unit. Now I am trying to figure out the motors and drivers. I may start with steppers, then upgrade to servos to upgrade in the future. I got a demo copy of Mach3 and printed all the documentation. It looks very well done.

Lets face it printers make shapes, not accurate parts. If you have to clean up printed parts you are going to need a drill, mill, or lathe. So that is where I am aiming.

I work as a machine designer too. I also have an electrical background, and do PLC programming. For about a year I worked as a CNC programmer writing G-code for two Fadal machining centers in a jobshop. So I expect I can get this all to work.

garyhlucas Wrote:
-------------------------------------------------------
> Randy,
> I just started building my Ifactory. Printer,
> vertical mill, horizontal mill, lathe, knife
> cutter, plasma cutter, whatever. After looking
> for a long time I came across 4 identical THK
> linear slides with caged ball bearing and ball
> screws, and 18 inches of travel, ready to mount
> motors. No speed problem here, these were made to
> move fast. The other day I picked up a precision
> cast iron surface plate that is 24" x 24" and
> weighs about 200 lbs. I was looking for parts to
> cobble together a milling head when I ran across a
> manual Sherline lathe with a milling column
> atttachment, and a pile of accessories. So now I
> have a little manual mill to make parts to build
> the CNC unit. Now I am trying to figure out the
> motors and drivers. I may start with steppers,
> then upgrade to servos to upgrade in the future. I
> got a demo copy of Mach3 and printed all the
> documentation. It looks very well done.
>
> Lets face it printers make shapes, not accurate
> parts. If you have to clean up printed parts you
> are going to need a drill, mill, or lathe. So
> that is where I am aiming.
>
> I work as a machine designer too. I also have an
> electrical background, and do PLC programming.
> For about a year I worked as a CNC programmer
> writing G-code for two Fadal machining centers in
> a jobshop. So I expect I can get this all to
> work.


You will do just fine here.

I think some cnc machines are suitable for that kind of process. You should check some of the features and decide what part you should buy.

Randy,

Clearly you have a lot to learn about 3d printing but I have to say that while rigidity can be an issue
with 3d printing speed is also something you have to consider. Cnc machines make decent 3D printers but not great,
due to the fact that the speed of a cnc router or cnc mill is significantly slower then a cheap plastic made 3d printer. The inertia
is so high compared to a smaller machine that it greatly reduces the accuracy of your rigid cnc machine. While having a combo machine may sound great it won't do both very well, ever heard the quote "Jack of all trades master at none" same applies here. The filament can't deal with the amount of mass while it's moving at rapid speeds. Somethings are just best left separated.

I have just started on a machine which I hope will combine size and accuracy thru use of rigid fixed components and light moving assemblies. The frame is welded steel, the fixed linear rods are heavy, all are fully adjustable for accurate truing, and the carriage parts are all machined from aluminum. I believe that 3d printers, including FDM can be tuned to the accuracy needed to make reasonable parts. And my parts cost on this machine is about what a Mendelmax kit costs. Labor excluded, of course. I have started a build thread under reprappers. I'm just over a week in and much of that time has been spent standing in front of my lathe.