[mercaytor]

Hello all, I'm working out a design for my 1st CNC router/mill, and I'd like to get your input on what I've designed so far. Attached are images from AutoCad.

This will be a fixed-gantry machine with 200mm x 200mm x 100mm XYZ range. I'm going to use NEMA17 steppers for now (unsure of their torque ratings) with my DIY stepper controllers that I'm making with L298 drivers, probably to run at 24 or 36v and about 2A per stepper. Tool will be either a dremel or DC motor for high-speed spindle, and I plan to mill PCB's, wood, engraving aluminum, and cutting foam. Eventually I'd like to make a 3D plastic extruder to allow 3D printing.

Main frame material will be from 1x1" (25mm square) aluminum tube, about 3/16" thick wall, I think. I'm using this because I have access to it at about $2/ft (CDN$). Plates (shown in yellow) will be 1/4" aluminum.

Here's the other hardware I've ordered so far:

Linear motion
SBR20 supported rails 300mm length, 6x @ $15/ea shipped
New Linear Shaft Rail Support SBR20-L300mm Rails Support 1pcs SY1859 CNC | eBay

SBR20UU open linear bearing blocks, 8x @ $6/ea shipped
Pack of 4 SBR20UU 20mm Aluminum Open Linear Router Motion Bearing Shaft Block | eBay

1/4" ACME threaded rod, single start 4 lengths 36" each at $12/ea
4x ACME threaded nuts, about $2/ea
from local supplier

Steppers are NEMA17 36VDC x 1.2A, ordered 5x at $12/ea shipped
Lot of 5 Stepper motors NEMA 17-DIY CNC ROUTER MILL ROBOT REPRAP MAKERBOT Prusa | eBay

Guess that's about it for now. What do you think of my design, will it work for some milling and engraving, or is 200mm x 200mm workarea too small to be worthwhile?

Again, this is going to be my first machine, so I'm more interested in just getting something up and running before I start thinking about a bigger/stronger CNC.

Thanks for this forum, I've gotten so many great ideas from here already, so now hoping for some good feedback on my potential design!

[PaulRowntree]

Hi wgenik
Will you be bolting all that square tube together? It is pretty light stuff ... I bought a bunch, but found it to be too flexible by far. Doubling it up the way you plan will help.

Since you will only get 1 bolt per intersection (am I right?) you may have problems with racking, since there is nothing triangulated in your design. Can you add a brace from the top of the gantry down to the front corner of the machine (both sides)? Can you add 2 diagonals within the gantry? Not sure how to deal with the bottom, perhaps stiff plywood sheeting on the bottom to keep it all square?

Cheers!

[awerby]

It might work to some extent. But the Dremel is a poor choice for a CNC spindle, since it has a plastic body that softens when it heats up. The DC motor is a better idea, especially if it's driving a proper spindle cartridge. And those 1/4" screws are wimpy, while the 17-frame motors might not have enough torque to do much. As was already pointed out, that aluminum tube is weak - you don't need to use it in non-moving parts, and for them it's marginal. Especially for the vertical gantry sides, you should find something else to use, like aluminum plate, for instance.

Andrew Werby
www.computersculpture.com

[mercaytor]

Yeah, again since this my first machine, my goal is to use as much as I can from my current stockpile to keep costs down, hence the Dremel and aluminum tubing. I think this will allow me to better visualize (once built) the forces in play and wreak points to improve on the next bigger machine.

I had considered making solid steel 3/16"plates for the gantry sides or uprights, since the welding shop here has a plasma cnc and will happily cut anything I want. Maybe I should go straight for that? Then I could work in a triangulated plate from gantry top to the front of the frame...

Thanks again for the input so far!

Warren

[mercaytor]

Okay, here's a quick render with 1/4" steel side plates (blue plates) instead of aluminum tubing for the frame/gantry uprights. I also moved the X-axis stepper to the rear of the table, since that's a little more symmetrical place for it I think.

Would this be significantly stronger than using the aluminum 1" square tubing? Cost would be about the same or less (from my calculations) since I'd use regular steel at $0.85/lb, but weight obviously would be much greater... not as portable! Guess that's not really a concern at the moment.

Opinions please! (group)

Warren

[PaulRowntree]

Warren, if you have access to the tools to work steel, you could make a killer machine! Perhaps think a bit about a bigger, stiffer machine, and then your first one will be good for much longer. you will still make a second, but it will be refining on a good idea instead of ... you understand.

Read lots, ask lots ...
Cheers!

[harryn]

200 x 200mm is fine for a starter size router.

Given the cost of the materials you have proposed, and that this one is "to learn how to build", I would instead suggest using a normal, 3/4 inch plywood with a hard wood surface. (for instance maple or oak faced). Just use the plywood to make up some 2 x 4 or 2 x 6 inch rectangular tubes. With a little care, you can easily make these just as flat and straight as Al tubing.

I am doing this same thing, and like to call it PAD (plywood aided design)

[harryn]

Consider going down to a hardware store and finding some steel pipe around 20 ft long. Hold it so that one end is on the floor, and the other is up in the air.

Shake it a few times and look to see if you can see the far end shaking and flimsy. Keep increasing the pipe size until you don't notice the far end looking flimsy. That is the size material you should use for your size router.

There is a more scientific and engineering approach, but this will really burn it into your mind

[mercaytor]

Yes, I do have a number of tools for steel, just not a great knowledge of working with them haha!

I have a drill press, abrasive chop saw (and miter saw for wood/aluminum), MIG welder for up to 1/8" steel, various angle grinders, stuff like that. I have access to pneumatic shears and a CNC plasma cutter (12x8 water table!) with the local welding shop (an acquaintance of mine owns it).

The main problem with steel working is that I need to do it in my garage, and as it's unheated, our Manitoba winter keeps the temperature at -15C or colder (not sure what that is in F), so I'd prefer to work in my warmer basement with inside tools... hence aluminum. Haha guess I should wait until spring to start construction...

harryn, I like your idea of adding in plywood to the design for some portion, I'll keep that in mind.

Thanks,

Warren

[wizard]

Yes, I do have a number of tools for steel, just not a great knowledge of working with them haha!

No time like the present to learn.

I have a drill press, abrasive chop saw (and miter saw for wood/aluminum), MIG welder for up to 1/8" steel, various angle grinders, stuff like that.

If you have a mig welder and access to a shop with a plasma you are light years ahead of many here.

I have access to pneumatic shears and a CNC plasma cutter (12x8 water table!) with the local welding shop (an acquaintance of mine owns it).

If this is the case I think your current design is not worth your time pursuing. This is even more so if the shop has a brake or tools to machine steel tubing.

The main problem with steel working is that I need to do it in my garage, and as it's unheated, our Manitoba winter keeps the temperature at -15C or colder (not sure what that is in F), so I'd prefer to work in my warmer basement with inside tools... hence aluminum. Haha guess I should wait until spring to start construction...

If you have a cellar you can weld a bit. I would not do a lot in the cellar but you certainly can do a bit.

harryn, I like your idea of adding in plywood to the design for some portion, I'll keep that in mind.

Thanks,

Warren

The problem you have is that you want a machine that can machine wood and metal which demands rigidity. Your other needs Might be meant by your current design as extrusion machines can get by with lighter construction. This especially the case if you weld some of that tubing together.

I have nothing against a cheap machine to learn on. The problem in your case is that your design won't do what you want it to do well. The design is just to spindly to machine aluminum and probably wouldn't do well in wood. You really should do a stiffer machine seeing that you have access to equipment to do it.

As to size 200 mm square might be a bit on the small size for some but I don't see it as a huge issue for your first machine. If you do this machine right you can keep it around for a very long time doing light work such as this PCB, engraving and other light uses. The long term goal is important here. Just because you are learning doesn't mean the machine can't be made viable for the long haul.

[mercaytor]

Wizard, very good points. So I guess what you all are saying is... Do it right the first time haha!

Based on the tools I have available then, any designs specifically I should look towards? I'd like to use the 300mm SBR20 rails that I already have, and the 1/4"acme threaded rod. I'll use the small Nema17 motors for now and upgrade soon if needed. My primary use will be Pcb engraving, so I'll stick with fixed gantry.

The second design I uploaded, does that make more sense given my needs/materials/tools?

[steve123]

I see that you are using a single pair of SBR blocks on each axis. This is really asking for trouble. Those WILL rock. They can't handle that kind of stress. You need to have a 2nd pair of blocks with *at least* 1" gap separating them from the others. Take a look at the forces here: Do it yourself CNC router: Design Considerations, the Gantry
You need to avoid a multiplication of forces.
-Steve

[mercaytor]

Hmmm well I do have 2x blocks per rail on the X axis (so 4 total), but they are right close together. Unfortunately, since I only have 300mm rails, as soon as I add any gap between them I will lose valuable travel room, adding a 25mm gap would mean I'm down to 175mm X-axis range or less. I guess that's still bigger than most PCB boards I'd be milling, so tolerable for better quality.

What if I was to place a bearing or two directly under the spindle touchdown point? Then when the table is at the maximum X-axis range, there's still a point underneath the center? See attached for a visual... Obviously there'd have to be some adjustment to ensure it's perfectly in line with the rail height.

I guess I could afford to take the Y-axis down to just a 175mm or 150mm range, since I don't think this center-bearing method Z-axis... since I was going to cut the 300mm rails down to 150mm, I'll just leave them longer to allow for longer range.

Think this will cut down on the negative forces?

Warren

[Bob La Londe]

I have a Taig mill which is much more rigid than just about any small light built router, and I still get chatter due to rigidity issues on heavy cuts in aluminum, and I am running a 28000 rpm spindle on it. I do light work (sub 1/8 cutters and light aluminum work on my gantry router (34000 RPM spindle), but for any heavy cutting with 1/4" or larger mills I use the Taig. My gantry router is mostly framed out of 1/2 Aluminum plate, although the gantry itself is only 1/2" plate on the uprights. (I didn't build it) Take heed of the guys who say to go with a more rigid design.

Also, a Dremel or is a very poor choice as a spindle. Most name brand routers with a metal nose piece would be a better choice. The Dremel has poor power, poor consistency of speed under load, poor rigidity, and poor runout. For PCBs a variable speed Bosch Colt would be a fair option, and it does a decent job in aluminum, but you can overheat the nose bearing and melt the nylon sleeve it rests in on long heavy jobs. I have tried rotary hand pieces. I have tried routers. I have tried Dremels. Currently I am using a Bosch Colt on my gantry style router, and Porter Cable PC690 on my Taig. They do a fair job...

The other thing to bear in mind with any open brush motor is dust. You may think you can just blow it out, and that is true for wood, but I have killed a few routers when they get full of aluminum... I suggest running a drier hose to the intake side from outside a cabinet if you are going to cut aluminum with it. Maybe even add a little muffin helper fan at the other end.

[mercaytor]

What do you think of this style, using 1/4" steel cut with plasma CNC for the side walls? Is that enough triangulation to make it pretty solid?

Warren

[Bob La Londe]

A fixed gantry design is an order of magnitude more rigid than a moving gantry. Obviously you sacrifice some working envelope for the hardware dimensions in the direction of table travel, but if you plan to make even modestly heavy cuts its worth it. There are a couple machines marketed as "mold makers" that have fixed gantries

[mercaytor]

Well, I've gone ahead and ordered the plate steel and aluminum from my welding supplier, he should be able to get me the parts this week. I'll post a copy here just in case anyone's interested...

Then it's just a wait for the rails / bearings to arrive so I can start assembling the frame!

Warren

[wizard]

Wizard, very good points. So I guess what you all are saying is... Do it right the first time haha!

Err maybe not so much as that as your first machine doesn't have to be a masterpiece. What I'm saying is make it good enough to do the primary job well.

Based on the tools I have available then, any designs specifically I should look towards? I'd like to use the 300mm SBR20 rails that I already have, and the 1/4"acme threaded rod.

Nothing wrong with using what you have on hand. You just need to focus on that word rigidity.

I'll use the small Nema17 motors for now and upgrade soon if needed. My primary use will be Pcb engraving, so I'll stick with fixed gantry.

A PCB engraver sees a more limited range of loads than a machine used for a variety of applications. As such it can have a lighter construction than a more general purpose machine.

The second design I uploaded, does that make more sense given my needs/materials/tools?

No, not at all. You should take heed to some of the other posts suggesting that you come to grips with strength of materials and the ideas related to machine rigidity. Your other option is to talk to your weld shop friend and hope he has experience with structure stiffness. None of your designs seem to address the stiffness issue. Things like diagonal bracing and gusseting can go a long way to making you aluminum viable.

The other thing here is if you have a supply of aluminum at low cost you can always melt it down and cast it into structures more suitable for a machine tool. Aluminum extrusions may not cast well, as such you may need to alloy the metal a bit.

If the idea of casting bothers you, welding those extrusions together can lead to a stiffer structure if you do it the right way. It might be a whole hell of a lot easier though to simply get your welding friend to make you a few simple parts out of larger cross section steel. Either way you should end up with a machine that easily copes with the array of tasks you originally mentioned. If you limit those task to PCB routing you might be able to get by with a few modifications to your original design to stiffen it Up.

[wizard]

I think what you will find is that this design will work OK for some of your workloads. You need to realize that even though that steel is 1/4 inch thick it is still sheet metal and will be a problem for more aggressive uses. Don't sweat it though as welding on stiffening iron shouldn't be a problem.

This highlights one thing about your designs, the world is 3 dimensional and so are cutting forces. I'm really thinking that you are missing this point.

In the end though I think this design will serve your primary need of PCB board routing well. It will be very interesting to see a video of your machine in action.

What do you think of this style, using 1/4" steel cut with plasma CNC for the side walls? Is that enough triangulation to make it pretty solid?

Warren