[watanabe]

Hello,

Been reading for about a week or two and it's been eye-opening. As the title hints, originally I got into this with the idea of building a 3d printer (RepRapWiki) - but then decided I'd like to be able to route/mill with the same machine as it seems like it's a simple tool swap which would allow me to maximize the return on my investment into ballscrews/supported rails/frame. Also, making little plastic models is great, but actually being able to operate at a 1:1 scale and perhaps make things that could actually be used in buildings would be a whole nother level of cool.

I've gone through a few different design iterations now, and am at the point where I'm not too embarrassed to ask for advice. First, the goals of this build:

1. Provide a way for me (an architecture master's student) to learn, and take with me, the *equipment* that is becoming essential to our practice-my school has a 6 axis Kuka robot, 4 laser cutters and a bunch of CNC mills/water-jets/etc, but for most of us, after graduation we work in offices/design practices without the budgets for these sorts of gear.

2. 3D printer: makes having a higher Z-axis useful

3. CNC router: MDF, plywood, aluminum, maybe zinc sheeting(no idea about how tough this is, just thinking of architectural materials) drilling holes in steel?

4. Modularity: swapping out plastic extruder for spindel for laser - have not tackled this but imaging some sort of steel sleeve with grooves and maybe a bottom shelf with a hole and a top that can be clamped on.

5. Build platform: constrained by equipment already purchased (see below)


The high z-axis made me want a fixed gantry, but then the long X makes this impractical... I've been scoping out other designs and found some pretty elegant solutions:
http://www.cnczone.com/forums/diy-cn...cnc_build.html
and even more so:
Search - My Photo Gallery

Which has gotten me this far:




edit: large red cube in 2nd render is abstraction of spindel/plastic extruder/etc... in the renders its 80x80mm. The rails in the drawing are all the proper dimension, as are the aluminium extrusions.

This is just the modeling for the ABILITY to move: frame with rails but no ballscrews/motors yet...
I plan on redoing the parts riding on the Y and Z axis so that it's made out of thick steel sheets CNC milled with holes/taps in the right places, thinking that with some T-style bracing this will allow me to be strong and thinner, which should in turn preserve more build area in the Y and Z directions than the aluminium in the renders. Also not shown is a small tweak to the Z so that the bearings are mounted equal distances above and below the XY plane...

The idea of this design is to try to minimize any moments by keeping them in the plane of the previous plane... up until the actual drill bit which will of course have a moment arm around the whole assembly.

Also, the alu in the render is based on Misumi's HFS6(30mm) stock: the x rails are 60x60mm and the other stuff is 30x60mm. This is convenient as the distance between the supported rails mounting screws is 30mm, which should allow me to bolt the rails right into the aluminum.

I'm using rhino to model/think through this all, but I thought it would help to have some stuff in front of me to help me think, so there are certain components I've already purchased, and which are kind of constraining me in a bad way maybe:

A) linearmotion2008 kit from ebay: 3 ballscrews (1150, 750, 350mm) 6 supported rails (1100, 700, 300mm) and 12 blocks for the rails, mounts/couples for the ballscrews.
B) x3 Keling KL23H286-20-8B: NEMA23, 425oz-in
C) gen6 reprap mobo(what I'll be using instead of a gecko, but not really important for my questions on this site), included 3 opto's
D) x5 6' 1x1" perforated steel tube and a x100 bolts/nuts/washers: originally was trying to build something based on this: Eiffel - RepRapWiki - but realized it wasn't going to work for milling, hoping that I can use this to make a table/workbench for my router.

Specifically, I'm worried about the 3 ballscrews... seems the easiest way to go would be to mount 2 ballscrews to each X rail like FandZ has done but I really don't want to buy more stuff from China and pay another chunk for shipping...


just in the process of writing this, I had the idea of mounting the z axis not to the y axis, but making the bed raise up and down. This would make it so that when routing woods/metals (thinner materials) the plunger is not fully extended and exerting a larger moment... but thinking through it now I can see problems with the material causing clearance issues with the moving x/y if the bed keeps raising... hrm.


My specific questions relate to:
1. The placement of the ballscrews either... the impacts off placing them off-axis (not in the center of the rails, both horizontally and vertically) and how that may contribute/cause binding... seems HighSeas has done this in his version but I can't tell how he's driving the X.
2. How to calculate the forces involved to verify that my build, or in some cases design thicknesses, are strong enough: I think that starting with the Z assembly and the moment of the drill bit to the rails its supported on is the start but I'm not sure of how to calculate the maximum cutting force caused by the drill bit moving through a material. I guess it's related to the pitch/torque my motors can put out, and the hardness of the material I'm routing, but not sure exactly...
3. Other ideas/problems in my design I'm not aware of!

Thanks!

[PaulRowntree]

This is a very large Z motion! Most folks seem to go with about half of this.

Why not put the bearing shafts on top of the X and Y rails? If nothing else it will increase your travel, but it can also free up space around your tool.
Cheers!

[Benonymous]

I'd like to do some 3D printing when my machine is finished too I agree with your idea that having a decent z axis stroke is a good thing for 3D printing and as the tool pressure will be practically zero with an extruder, you can get away with not having so much rigidity in the z axis. I'd planned to make multiple mounting points for the linear blocks on my rails. They would be mounted apart for machine tool work (routing) but then they could be moved next to each other on the rails to maximize travel on the Z axis.

Good luck with your build, what SW are you using to model in?

[vinot]

Hello,

well, it depends on what materials do you want to work (mainly) and speed/forces required, isn't it?

by my experience now I'm starting my second cnc mill, and I'm going to use an old steel base to set up rails. I plan to use it to mill hard wood and aluminium because my fisrt machine (a frame steel) couldn't do that (basically too much vibrations).

what I don't know is how to put and control a plastic extruder (3d printer) with the same controller: that is what I want to do also! I think that reprap and similar are arduino controlled...

toni

[watanabe]

After getting a headache trying to figure out how to calculate the cutting forces involved I decided to just go ahead with trying to overbuild it and see how it handles everything, resulting in some (big) changes:

1. Ended up going back to a more typical gantry design-timeline for everything got moved up and I wanted to go with something more proven.

2. Switched to 2screw/motor x-axis so I could use a fully supported frame.

3. Beefed up the gantry using Misumi's 60mm x 90mm (30mm base) on the vertical members. Motivating factor was the ability to use the 90mm wide brackets(6 screws x 2 rows on each axis) which have an allowable load of 4116 N.

Still working on:

1. Integration of mounting blocks for steppers to the alu. extrusions
2. Gantry plates to mount the YZ axis's to the X axis's bearings-right now modeled using 16mm steel with a little 12mm T part which the X's ballscrews could mount to
3. Z axis assembly, as is 16mm steel on the vertical parts and 12mm on the horizontals
4. Software: decided to buy a g540 as the arduino reprap board wouldn't allow me to have a slave axis for the X. Hoping to use this with EMC as the mach3 interface looks horrid to me.\


Ordered parts list:

3060, 6060 and 6090 misumi aluminum extrusions+brackets+screws/nuts shown in renders below
Chinese 2.2kw spindle
Hitachi 2.2kw VFD
x4 425oz-in Keling NEMA23
12.5A 48V Keling PSU
Gecko G540
x2 1100mm, x1 750mm, x1 350mm linearmotion2008 ballscrews/nuts + supported rail/bearing block/mount/couplings kit

modeling is in rhino, renders in OctaneRender(CUDA-based GPU realtime raytracer)

[Benonymous]

Yes, the RepRap extruders I've seen are controlled by Arduinos basically to run a stepper to feed the plastic rod into the extruder and control the heat. I imagine there's plenty of code floating around for that application and an Arduino is only $30 or so (I bought one about three months ago)

Those renders look so classy compared to the screen captures I do in SolidWorks

[leaveme]

Beautiful rendering!

[giz]

How do you plan to confront the problem of material warping during 3D printing with such a high Z axis?

[watanabe]

How do you plan to confront the problem of material warping during 3D printing with such a high Z axis?

Not expecting any issues, the repraps linked above have a similar Z axis, any warping issues seem to be a product of not having a heated bed and printing using ABS plastic.


Misumi stuff showed up today so it had a chance to assemble a good part of it... still need to design how the ballscrews will mount to this frame, z assembly, limit/home switches as well as some plates to stiffen everything up, and mount the gantry on the X axis rails(and tie into the x axis ballscrews)

[vinot]

very nice!

regarding movement, do you read this?

http://www.cnczone.com/forums/cnc_wo...eam_ahead.html

I saw two mill running in this way (belt and pinions) and it adds lot of stability, mainly because the screw does not roll, just the ballscrew. Maybe a little expensive, but I'm planning it to my new mill (add one block and some good ring)

[vinot]

I was looking for info about link my cnc controller to a reprap extruder and I found more easy to unplug all the hard system and plug to the arduino controller card to run reprap projects, mainly because all of them can work with the same hardware.

Regarding wrapping, I think that this is most done by the way of disposal, not because material istself. Feed velocity facilitates that rounding disposed abs solidificates before a new layer was printed, or reducing feed at needed zones.

[giz]

Not expecting any issues, the repraps linked above have a similar Z axis, any warping issues seem to be a product of not having a heated bed and printing using ABS plastic.

Gotcha. I guess I wasn't sure if you were going to implement a heated bed. I'm highly interested in how this pans out!

[watanabe]

project is back underway, got swamped with school projects. taking a full credit load this spring as well but have some spare time to finish this off finally!

I made the mistake of thinking the connections would be easy to work out, when in actuality they are (of course) what takes the most time to figure out! Thought having the materials in front of me would allow me to tackle the problems easier, but I've now boxed myself in a bit due to the materials I have in front of me. If anyone's planning their own build, don't make the same mistake.

So as far as updates go, last night I ordered some fortal (alu 7075) scraps off ebay and am going to try to design the plates which connect the YZ gantry to the X-axis, and the dual ballscrew's to the X-axis/bed, and finally the X-axis ballscrews to the YZ gantry.

These 2 are the feet/socks for my YZ gantry:
.574 x 3 3/4 x 14 1/4 Fortal Aircraft T6 Aluminum
.574 x 3 3/4 x 14 1/4 Fortal Aircraft T6 Aluminum

I'm trying to figure out the best way to work out the bolt pattern, putting my supported rail blocks under the beefy Misumi brackets, making sure to offset the holes and countersinking the bolts in is one option which saves me some build area.
Other option I see is to move the blocks out from under the brackets, which would allow me to just drill out holes in the plate and use bolts/nuts to sandwich the bracket/plate together and use the thread in the railblocks to sandwich the plate.




other misc parts I got to mount my ballscrew mounts(the chinese blocks that were included in the kit I bought from linearbearings doesn't mate so well to the extrusions I have) to my extrusions:
.573 x 4 3/8 x 14 Fortal Aircraft Aluminum Bar
.576 x 3 7/8 x 18 3/16 Fortal Aircraft Aluminum
.574 x 3 1/4 x 16 1/8 Fortal Aluminum Bar Plate

1.145 x 5 3/8 x 8 1/8 Fortal T6 Aluminum Plate (planning on throwing this at the Z axis)

Have some CAD drawings done, unfortunately they are at studio and I am leaving town tomorrow morning until Friday... will have some pictures/questions for advice then.