[squigly]

Hi all

So have been playing with the idea of building my own cnc machine for some time now, have looked at a number of different designs scattered around the web and finally decided to at least put pen to paper (so to speak) on a design of my own. Now while i have no experience in the DIY CNC realm i have worked on a few milling machines & lathes etc.. not extensively mind, but have some idea of the concepts at least.

So before I started acquiring the parts I though it best to seek out the advice of you fine folks on my design concepts and check if it is fit for purpose. My intention is to be able to mill aluminium and hopefully some non ferrous metals with this, but I would also like to be able to convert it to a plasma cutter and then later maybe even add some 3D printing capabilities.

Some basics of my designs:
X - 1000mm using 20mm round rails (either supported or not) Dual drive.
Y - 700mm using 20mm round rails (either supported or not)
Z - 200mm using 16mm round rails (either supported or not)
RM1605 Ballscrews all round
Will be using 6061 Aluminium. and 80x80 profile for the base frame
Either Nema 24 or 34 steppers. (designed with 24's)

So this was my first cut at a design, I haven't added all mounting holes etc to the design yet as I wanted to get some advice before I finish off with all the finer details, the problem with this design was the support that runs under the X will get in the way if I want to convert this into a plasma cutter, so wanted to try a open frame type design. This design was with SBR20 supported round rail for X&Y with SBR20UU Open linear Bearings and SBR16 & SBR16UU for the Z.


This was then my take on the open frame design using high carbon steel W20 (h6) for X&Y rails and W16 (h6) for Z running through 6061 aluminium blocks (65x40 with different heights) this design would also not use BK & BF supports for the Ballscrews as I will machine out the supports for bearings into the aluminium (good idea or not??).


I have left all shields and table bed off to get a better look at its guts. So my questions are, firstly around the design, would these be fit for purpose? Am open to any suggestions here. And then for the open frame design, I have seen that some people have used bushes rather than linear bearings, what are your guys take on bushes vs bearings for the axis rails?

Squigly out..

[louieatienza]

I'm not a big fan of using unsupported rails for the longish span you have, especially if you want to cut aluminum. For the first design, you might want to use thicker plate for the gantry as it will be prone to flex, especially since your only point of contact between the crgantry bridge and the uprights are at the corners. Using extrusion for the bridge can help (two pieces if you want to run the ballscrew on center. Also for the top design it is a bit unclear as to where you plan to attach the table, but you'd want it to be fully supported underneath. Finally, it would be a good idea to have your drive system as close to your linear bearings as possible. This is acheved on your second design, though like I mentioned the unsupported rails may cause problems with cutting aluminum.

[squigly]

Thanks for the feedback louieatienza

Have spent a fair bit of time reworking my design playing with a few idea's. I ended up deciding to go with 20mm supported rails and 1605 drive screws on all axis (dual X)
rail lengths will be as follows:
X 1200mm
Y 1000mm
Z 300mm

Actual machining area will be+-:
X 1000mm
Y 800mm
Z 150mm

Most of the Aluminium plates are 10mm apart from the uprights on the gantry, these being 20mm, and the back plate of the gantry being 15mm all in 6061 grade alu.

For the electronics it have decided on NEMA34 1230 oz-in steppers for all running off a 5 axis Mach3 USB breakout board, 6A drivers (each driver / motor will get its own power supply) and a 2.2KW spindle, decided to go with air cooled, running off a 2.2KW VFD

I have also added cutouts on uprights and the back plate of the gantry to try reduce overall weight but keep structural rigidity/strength

Feedback on the design is most welcome.

[ger21]

Absolutely terrible choice of motors imo. With 5mm pitch screws, you need motors that spin fast, and large motors don't spin fast.
with 5mm pitch screws, I'd go for 450oz Nema 34's. 10mm pitch screws are probably a better choice as well.

And I would get rid of all the cutouts. You want as much mass as possible.

[crane550]

Lol, Ger21 said everything I was going to say. ~300-400oz steppers will serve you much better in this case. I would go with 1610 ballscrews, and seriously, keep as much mass as possible. Mass is your friend. Lets put it this way: My machine weighs 4,000lbs. I built it too light.

Also I would mount the supported rails for the X on the SIDE of the extrusion and bolt them directly to the upright riser. You are just asking for a weak spot with the horizontal plate, fasteners, etc.

You might consider putting the X motors on the back, but this doesn't really affect the performance of the machine, if it were me I would just like to have them out of my way. Again, this is just a preference issue, but it would be nice to have them out of your way.

Other then these things, great start!

Seriously, keep mass. You will be so much happier.

[squigly]

Thanks for the feedback chaps!!

crane550 I like your suggestions of moving the motors to the back and out of the way, will also make some of the cabling neater as that's probably where i would be running cables to in either event. I also like the mounting the X rails to the side of the profiles and directly to the uprights, on this though would you suggest i also move the drive screws over to the side?

[dmalicky]

Nice looking design and revisions. It's great that you have:
- A large vertical separation for the Y axis rails
- The Y ballscrew between the Y rails (not above or behind)
- Dual X drives
- X rails near the table -- that helps cutting stiffness. Whether the rails face up or out has both pros and cons.
- A spindle

The x drive screws could be a bit above or to the side of the table (little difference for stiffness), but I'd put them to the side where they're less vulnerable.

Here are some suggestions for higher stiffness:
1. Change the gantry cross-member to a large square or rectangular tube. For heavy cuts in aluminum, I'd suggest approx 8"x8". Here's why: X forces on the cutter cause a large torque on the gantry; big tubes are the ~only effective cross-section to resist this torque. We're having a good discussion of this in another thread, including analysis of sizes, starting around here:
http://www.cnczone.com/forums/diy_cn...ml#post1408740
That thread is for a 49" travel gantry with 10" Z. Downscaling to 39" & 6", the deflections are less by about 50%, and so the stiffness of any given tube will be greater by about 2x. Ideally to cut alum, I'd want the gantry to have a stiffness-at-the-tool of about 200,000 lb/in.

2. Separate the Z axis bearings by about 150mm (similar to your Z clearance). This is important to reduce bearing forces and deflection, similar to your large Y axis rail vertical spacing.

3. Make the Z axis plate a channel section (or at least much thicker) so the spindle mount is stiffly coupled to the Z bearings.

4. SBR blocks have very low stiffness in tension: (http://www.cnczone.com/forums/linear...ml#post1412778). This causes the cutter to deflect a lot, since the cutter is cantilevered off those bearings (see this post http://www.cnczone.com/forums/diy_cn...ml#post1408004 for explanation and a calculator.)
To cut alum effectively, you really need profile bearings and rails for the Y and Z. These have 5x to 50x (!) the stiffness of SBR20. Nothing else compares. SBR20 is reasonable for the X axis since it isn't cantilevered, if the feet are long enough (see note 6, below).

5. Make the gantry uprights/risers/legs thicker or preferably, into a T, U, or tube section, for more lateral stability.

6. Separate the gantry 'feet' bearings by 175mm or more. In the side view, the cutter needs to be between the bearings. This is to prevent tipping deflection under gantry accel/decel, and to ensure those SBR feet bearings never see tension.

7. To control weight with less work and more stiffness, omit cutouts and use tubes instead of solid bars.


For a machine tool, the effect of mass is complex. From the articles I've read written by industrial machine tool designers, they focus a *lot* on increasing stiffness and damping, but not very much on mass. In some cases, they actually try to reduce mass. Stiffness (aka rigidity) is their first fundamental goal. Stiffness defines how much the cutter deflects under cutting loads, which directly affects accuracy and chatter. Mass (in and of itself) has more complex effects; some good, some not good, depending on where it is placed and distributed.

It's been common practice to judge a machine by its mass. Historically, this makes sense. For the same general machine design (e.g., a cast-iron C-shaped vertical mill), the designer raised stiffness using thicker and bigger sections, which of course adds mass. Cast iron was and is a material of choice partly because it has fairly good damping, but it isn't very stiff so lots of it is needed to get machine stiffness. Thus, heavy cast-iron machines performed great, and most people assumed it was because of their mass. That's partly true, but more so it was their stiffness and damping.

For example, if you change the design to a cast-iron horizontal mill, you can get a big increase in stiffness with no change in mass. That's why horizontal mills can plow through material faster than vertical mills. For CNC routers, there's a huge variety of machine designs; most are not optimized for stiffness and their accuracy and chatter tendencies reflect that.

More mass does reduce high frequency vibration and audible noise. But chatter is typically low frequency, and cantilevered mass can make that worse. Mass on moving parts probably also helps smooth out the pulses between a stepper motor and axis motion, and vice-versa, cutter pulses back to the motor. But more mass on moving parts requires more expensive motors and drives for the same axis speeds and accel. Mass in the base is ~always good to keep the machine stable.

The other main goal for machine tools is high damping. If you have high stiffness and high damping, you get high dynamic stiffness -- that's the ultimate goal.

[wizard]

You already have many good comments here so I'm not sure I can add a lot.

The first thing to address though is structural rigidity and the idea of a do all machine. To put it simply a laser a plasma platform does not have the same reaction forces to deal with that a router or mill have. As such inexpensive machines end up being compromises. That isn't a bad thing but you need to realize this upfront.

Thanks for the feedback louieatienza

Have spent a fair bit of time reworking my design playing with a few idea's. I ended up deciding to go with 20mm supported rails and 1605 drive screws on all axis (dual X)
rail lengths will be as follows:
X 1200mm
Y 1000mm
Z 300mm

Actual machining area will be+-:
X 1000mm
Y 800mm
Z 150mm

Most of the Aluminium plates are 10mm apart from the uprights on the gantry, these being 20mm, and the back plate of the gantry being 15mm all in 6061 grade alu.

If you want to machine aluminum that simply isn't good enough. You will want strong members that can resist forces no matter what direction they come from. The easy, inexpensive way to do that is to use tubing. This may require some machine shop work though.

For the electronics it have decided on NEMA34 1230 oz-in steppers for all running off a 5 axis Mach3 USB breakout board, 6A drivers (each driver / motor will get its own power supply) and a 2.2KW spindle, decided to go with air cooled, running off a 2.2KW VFD

That spindle may have issues with aluminum and other non ferrous metals if the low end torque is wanting. The specifics vary with the spindle assembly but you will want to check out what is the lowest speed possible with the inverter and spindle combo. I only bring this up because you mentioned "other" non ferrous metals which leaves a lot of hard to machine alloys as the possible others.

I have also added cutouts on uprights and the back plate of the gantry to try reduce overall weight but keep structural rigidity/strength

It is more important in my opinion to keep the machine as rigid as possible if the intent is to machine metals, even aluminum which can machine rather easily. Plate stock has no place on a machine for this use. Well unless it is extremely thick plate stock. 15 mm is a little over a half of an inch thick which is probably less than my little finger. Now obviously aluminum is a little stronger that my finger but with the sort of span you are talking about you will have deflection and resonance issues.

Feedback on the design is most welcome.

If you goal is to machine aluminum and all those "others" you need to scrap this design and think about rigidity and the minimization of vibration. Also your joints or connection methods with respect to the various elements of the machine need to be a robust as the parts being connected. For example the plate that your X axis bearings connects to looks flimsy with respect to the Y axis uprights. The little details about how it is all put together are often overlooked so you need to think about the assembly process and how robust that assembly will be afterward.

[crane550]

Hi,

I have never personally built a machine with the SBR rails. I am anxious to see how they perform. However, if you really want to kick butt and not worry about it I would strongly consider spending the extra little coin and go with a THK HSR or SHS rail. 20mm would do just fine. The difference will be night and day. I think the biggest limitation right now for your build is those rails.

HSR20LA2UU 1270L Used LM Guide THK Linear Bearing 2RAIL 4Block CNC Route NSK IKO | eBay

Not claiming this one to be a good deal- look around and see what you can find, but this would work very nicely on your machine. BTW- its what my ~4000lb machine uses on every axis. No complaints at all.

Alex