[gathaz]

Hi

First time cnc builder, looking for some help calculating the loads on rails.

Building a pretty normal router

2 Rails for each axis with 2 ball screws for X axis and one for each Y and Z

and hoping to have a crack at steel (possible or not?)

Can anyone give me a hand with the calcs to work out what sized rails i would need to achieve this.

Thanks

[ger21]

A normal 12"x12" router, or a normal 5ft x 12ft router?

You need to provide a LOT more information.

[gathaz]

Apologies for the units, i plan for a 1.2m by 2.4m so 4ft by 8ft but not sure on Z travel yet (wont be anything too extreme maybe 300mm)

Im very new to this game and Im running off a design a friend built that turned out very well
I was hoping to just "beef up" his design

He uses a 16mm round linear rail and i was hoping to change to a trapezoidal but i am unsure on what size to go
I am looking at THK rails in the HSW design due to the low profile, can you give me some direction?

Calcs would be awesome but if i can just get some advice on a "minimum size" to cut steel that would be good.
I understand there is a lot of other factors that contribute to the rigidity of the structure, am i wrong in saying it beings with the linear rails?

[ger21]

Everyone wants to build a big machine to cut steel, but I've yet to see one built that actually works (at cutting steel).

I'm not sure on the load ratings of the wide low profile models, but with standard HSR's, 25mm rails are usually far more than enough.

[gathaz]

Why so many failed attempts? problems with rigidity?

[ger21]

Show me a commercial 4x8 sized machine for cutting steel.
Steel is cut on big, heavy milling machines, not router style machines.

[gathaz]

So router style set up is incapable of cutting steel?

Or it must be built on a foundation of incredibly strong steel to allow for zero flex?

[Claude Boudreau]

go plasma

[gathaz]

Yeah i am thinking plasma is the way to go

Can anyone tell me of a good plasma cutter to use and maybe a way to mount the head
Thanks

[hanermo]

I scratch built a big machine to cut steel, and it works fine.
My spindle is an old Bridgeport M-head, 0.5 Hp. Hitachi VFD, 2.2 kW.

My table is 1600 mm wide, and 500 mm deep, as is. Table mass is about 200 kg.
Table is slats, of solid tool steel 1600x30x80 mm, 5 of (can add upto 10) with crosswise reinforcing ribs.
Table is 60 mm thick in steel (could be thicker). Slats should be 30x80 mm thick, also (they are 30x30).

Machine is a portal.
It is not a C-frame (modern VMC), and it is not a post like the small SX1,Sx2,SX3 machines.
A portal is vastly more rigid than a C-frame. Approx 20 times more rigid, per mass and work envelope.

I use about 3 different preloads in the frames.
Frames are deep, to give good rigidity. As such, the machine is large in size.

Machine is about 2x2x1.6 mm in size.
All built in steel, partly heavy walled structural steel, 50x50x5 mm, partly basic solid tool steel (Acero F1 rectificado).

Machine mass is about 1000 kg.
Welded assemblies, bolted together, with the connecting faces being solid steel billets.
About 6 major assemblies.

I used 25 mm linear rails. They are adequate for the job.
I used 0.750 ballscrews. These work, but more rigity would be better.

They are too small, and I am changing to 40 mm ballscrews.

Work took about 6 years full time, and 3 redesigns.
And learning to weld, and refitting 2 cnc lathes.

I also became a cnc professional, as in selling large cnc machine industrially.
I sold over 60 machines, in one year, with 8 salesmen, in my territory.

So, its very doable - but most dont have the time, funds, and the gumption to see it through.
Sorry no pics at this time, for various reasons.

I need to change the ballscrews, and the motor drives, and the electronics, for commercial reasons.
All work well as-is, none are something I would sell, today.

[erikjgreen]

I ran across this thread today... since the subject came up, here's an idea of the size of machine needed to cut steel well using milling heads at a 4 foot by 8 foot scale.

Haas GR-510 Gantry Router:
Haas GR-510 | Haas Automation®, Inc. | CNC Machine Tools

Take note of the per-axis servo thrust ratings. Also note that (from the data sheet) machine weight is about 15,000 lbs.

Video of it in action cutting mild steel:
Haas GR510 Medium/High Speed Machining - YouTube

There's nothing inherently that keeps a gantry type machine from cutting steel, and from doing it very well. The geometry just makes it much harder to build a machine as rigid and therefore as well performing as the usual VMC type layout. Since machines that are harder to build cost more (this one starts at $110,000) you don't see many of this type except for those few jobs that require both the available table space and the rigidity and power they provide.

Think about it - when was the last time you saw someone machine a part for anything out of a 2 inch thick steel sheet? Plasma or waterjet, sure, but there aren't too many mechanical parts that need machining that are that size.

It's certainly possible to build a gantry machine that cuts steel, it's just that it's not easy (it would be easier for a hobbyist or anyone else to build a high performance copy of a Haas or Mazak VMC) so no one has succeeded. That's why people here say to build a knee mill if you want a mill, and build a gantry if you want a router or plasma cutter.

To successfully cut steel you need to build an extremely rigid machine, and generally speaking rigidity decreases proportional to the dimensions of the machine unless weight (structure) also increases proportional to the cube of the dimensions. It's entirely possible for a hobbyist working with off the shelf parts to build a mill accurate to a repeatable 0.001" in steel in their garage provided the work envelope is small (a few inches each way). There's a very good reason that even small knee mills weigh tons and up, and it isn't because the makers aren't using modern materials.

So, I'd encourage anyone out there to go ahead and design a gantry capable of cutting steel. Before you buy parts for it, however, make sure you've run an FEA engineering analysis on it to ensure once it's put together it'll do what you think it will. Check to make sure it won't deflect or vibrate while cutting and that the servos (not steppers) you picked are capable of moving all that metal around. Despite all the intuitive engineering and over building of parts most people employ, it's very easy to end up with a barely functional paperweight that cost as much as your car.