A brand new machine build. 2x4 size or something in that region. :)

I have recently started designing a few bits and pieces for my next CNC build.

I have opted, for ballscrews on all axis.
Hiwin style linear guides all around.
Servoes from DMM-Tech.com on all axis.
And a heavy duty z axis, for optimal stability.

I will be using a UNP steel beam for my Gantry so it should be very sturdy.
All flat surfaces for mounting will be machined so everything will be perfectly angled.

I have for now, only made a few renderings of the gantry main beam, including the servo and the hiwins.
This will be mostly a design build, and after complete design has been made, then I will start building it. ENJOY

Attachment 7984Attachment 7985

When the machine is cutting parallel to X (with Y being the gantry axis) the cutting force applies a moment to the gantry, which will twist it, hence the gantry must be a structure with good torsional stiffness (in the XZ plane). The torsional stiffness of a beam is only good if the section is closed, so for a C section it is very poor. The same is true for an I-beam or plate, so I suggest changing it for RHS or multiple pieces of box section.

Torsion Equations

Since the machine is no more than 1200mm wide, then the stiffness of a UNP 160 beam should be more than enough.
If I were to change to RHS tubes instead, then I'm gonne sacrifice something else..

So if you think that you can bend a 1200mm long 160mm UNP beam, over a max length of 1000mm, then go ahead, I believe that deflection will be set at a minimum.

But then again, as I complete the design, I will have a friend of mine, do some stress and deflection calculations on it. :)

Quote:

---

Originally Posted by IN-WondeR

Since the machine is no more than 1200mm wide, then the stiffness of a UNP 160 beam should be more than enough.

---

Have you calculated this to know for certain? If not, then this spreadsheet may be of assistance. You just need to know the length of the Z-axis, to work out the torque applied to the beam and the force which can be approximated from knowing what materials you're intending to cut. Also, even if it does turn out that the UNP 160 beam is strong 'enough' that doesn't mean you should definitely use it since the formula's for torsional stiffness make it obvious that using a closed section will require a lower volume of material than a C shape.

Quote:

---

Originally Posted by IN-WondeR

If I were to change to RHS tubes instead, then I'm gonne sacrifice something else..

---

RHS is just one idea, I'm sure you can come up with others with good torsional stiffness.

Quote:

---

Originally Posted by IN-WondeR

So if you think that you can bend a 1200mm long 160mm UNP beam, over a max length of 1000mm, then go ahead, I believe that deflection will be set at a minimum.

---

No, I don't believe I can bend it - it's maths not belief, it's twisting not bending and it's not a question of if it will bend, it's a question of how much. With a CNC-Router the structure needs to be very stiff since although the magnitude of the forces is quite small, even a very small deflection can cause problems (poor finish, tool wear, limited MRR), unlike most applications where we're worried about the material failing which is not a problem here.

Interesting design,

How do you plan on attaching it to the up-rights? Do you plan on mounting it directly to the plates for a axis?

Also Jonathan I know you have FEA software so if you have the time run realistic loads through and see what you get. Remember to add the stiffness that the Hiwin will add. I am probably missing something and hope some one point it out. Also what thickness UNP steel beam are you planning on using OP?

Michael

Id recommend machining a datum edge in on of the rail pads so that you can true up the other rail! Thats how i have done my other machine but have not posted anything on this build!

The connection to the uprights have not been decided yet. As I said, this is a design build.
By the way, then this is my fourth machine I will build. This time a smaller machine so I can have it in my garage, and still do other kinds of work in it.. :P

AdCNC
The rails are already mounted on top of a welded on piece of steel, which will be machined flat to true up the rails.
A Datum edge, is that a straight edge I can mount one rail along, so it will be true to the surface of the machine?

Quote:

---

Originally Posted by Jonathan

When the machine is cutting parallel to X (with Y being the gantry axis) the cutting force applies a moment to the gantry, which will twist it, hence the gantry must be a structure with good torsional stiffness (in the XZ plane). The torsional stiffness of a beam is only good if the section is closed, so for a C section it is very poor. The same is true for an I-beam or plate, so I suggest changing it for RHS or multiple pieces of box section.

---

Been here before haven't we Boy wonder.!! . . . . It depends on the Job it's going to do.?

If it's mainly wood or plastics then it will be perfectly fine.

If it's for meaningful cutting of Aluminium and steel then Yes it's much less than ideal, Has is a Gantry style machine really.
So with that I'd suspect it's main use will be softer materials in which case then it will be perfectly fine. RHS would also be unsuitable for Good Aluminium/Steel cutting machine so both are unsuitable in this case.

Another thing with UNP beam is that it doesn't take much to minimise any twist and few bracing plates will sort it.! The advantages with ease of build for a wood cutting machine are obvious and I know for a fact that it works on a Wood router with no detriment what so ever because I built my mates Panel router using one.! This worked 10hrs day 6 days week and did so for 2yrs producing 1000's of panels until recently when the building caught fire.!:sorrow:. . . MK2 will be bigger and better but will still probably use UNP beam.

Quote:

---

Originally Posted by AdCNC

Thats how i have done my other machine but have not posted anything on this build!

---

Why NOT.? . . . Comon man get ya finger out. .:whistle:

The two rails you have welded on will be skimmed flat as you planned. But if when u have them milled flat then leave a lip on say the bottom rail that you can but your bottom rail up agains so its true if you get me!