[peteeng]

Hi Vily - Unfortunately, with steel things get heavy. So you are either limited to weight that you can handle or you need to achieve a certain stiffness. See how you go with the FE and stiffness. Thats the next step...

[Vilyoo]

1000N at spindle tip

[peteeng]

Hi Vily - So 1000N/33um is 30N/um. So if your model is correct look at what bits are deflecting the most. Then address those. Try to bump up the static stiffness. Seems your z axis is the softest bit....Peter

[Vilyoo]

Changed Z axis a little bit to steel tube 250x150x12.5mm and 650mm long.. a little bit better result and i like more this design.. Must make cut in the tube so i can machine the plate inside on which the spindle will be connected

[peteeng]

Hi Vili - I've recently been thru the Z axis using generative, shape optimisation and form finding in Fusion,. The tube idea was the winner. I expect it to be a bigger stiffness gain then 33 to 26 so look at the saddle area. Peter

[Vilyoo]

made a few changes. Total weight became 3000kg from 2500kg. On the last picture there is a standart router, which looks like a toy next to the monster

(my previuos router, almost the same as this in the drawings)

[Vilyoo]

I want to use this bt40 spindle, do you think it can handle 350mm fly cutter, speed is not important, just to do the job.

[joeavaerage]

Hi,
to be honest I think a fly cutter of 350mm is going to cause your machine to flex uncontrollably.

Just to be clear......you are cutting steel ....correct? That is if I'm reading your first post correctly.

May I suggest you look around businesses in your area, province, hemisphere? for a machine that does what you want to do. If such a machine weighed ten tons is there any reason
to believe that you could build a machine that weighed less and yet performed as well?

I have spent many hours this weekend machining a spindle mount for my new spindle, nearly done.....but all throughout I have been up against the limits of rigidity of my machine.
Its not about speed, or even going slow, its not about power.....its ALL about rigidity.

When I designed and built my machine I tried to make it as rigid as possible, and its not bad....but its not great either. It cuts steel but not with the finish I would like....and that is what it is.

Craig

[Vilyoo]

Yes i want to cut steel. Gantry is around 750kg, will it flex with 0.1mm depth of cut?

There are only a few machines with this working area or bigger in my country, and they charge so much for milling. Also the transport adds a lot.

What can I improve to this design so it can handle the fly cutter?

I can fill all tubes with concrete, it will add 2500kg if it helps

[peteeng]

Hi Vily - thicker steel is better than adding concrete. Keep making things as big and thick as possible. The z axis is what will let you down so make it very stiff. So on that point here's my thoughts. 1) you have the conventional z arrangement on the front of the gantry 2) I have done this twice and have concluded that the saddle is a constraint (and a pian to design) in this config. The 8 bearings and plates make for a cramped design. Plus you want to be able to get to the bolts to check them or retorque them without having to pull the front off to get at the heads 3) this also means the spindle is mounted way out in front of the gantry so it has a big overhang So 1) I mount the gantry rails on top of the gantry 2) this allows the Z axis to be mounted closer to the gantry reducing the cantilever 3) this also means the Z and X bearings are separated and can be accessed easily and the saddle can be designed so the bearings are supported directly on webs which make it very stiff. 4) that's why I like a flatter gantry vs a tall gantry as then the Z does not have to reach down as far with the "bracket" saddle. cheers Peter

[joeavaerage]

Hi,
flexure is the death knell for any metal cutting machine.

Even with a small cut the tool tooth has to be held against the surface with sufficient force to cause a chip form. It does not matter a lot whether that chip is 0.1mm deep or 1mm deep,
that force has to be applied. Any force will cause deflection....that's plain physics, the question is does that deflection cause an issue with your machine.
I rather suspect it will, and that will in turn limit what tools you can effectively use.

Just as an example I did much of the work you see above with a 12mm tool, and while the finish is far from perfect it is none-the-less acceptable. That tool is now toast after about six hours cutting....and
its the last 12mm tool I had. This morning I tried using a new 16mm tool that I had, but its surface finish is worse than the 12mm tool, not because of any fault of the tool but my machine flexes
more when attempting to cut with a 16mm tool rather than a 12mm tool. Even slowing down does not help. As I say its all about rigidity.

It is extremely difficult to design a rigid machine, almost certainly you'll end up with a huge machine weighing many tons, so much so that you can never afford to build it. I think the best you can do
is build the most rigid machine you can with respect to cost and weight, and then try it out to determine how bigger tools you can use.

You might imagine its possible to calculate cutting forces and work backwards, and I'm sure that is what OEM's do, but they have years of data to inform their design, whereas you and I do not.

Craig

[Vilyoo]

I will continue to work on the design. Thanks a lot for ur advices.

[joeavaerage]

Hi,
steel is 205GPa while even good concrete is only 50GPa.

What makes you think that filling a rigid tube with concrete, which might as well be as stiff as a limp noodle, and have it be a great deal stiffer? Heavier, yes, maybe somewhat better damped, but
stiffer......NO. Damping and mass are useful....but way WAY down the list....with stiffness at the top, the second and third places!

The right answer is make the walls of the steel tube thicker and/or the outer dimensions of the tube bigger. If that increases the weight and the cost substantialy.....well yes that
is exactly what it means.

Craig

[peteeng]

Hi Vily - If you intend doing a lot of face work on plate then consider using a bolster or a sub table on the machine. This will minimise the Z height. If you design a machine with a "big" Z then you will be doing facing with it extended so the stiffness will be small. If you can do face work with a short Z then the stiffness will be big. So you could have a table set out so the Z is short for facing but have a well or pit so you can do the deeper objects. The Z axis design is the big thing to get right for you... Peter

[Vilyoo]

Hi Craig, I think it is better to fill it with concrete rather than being empty. I will see if I can handle bigger tubes, the only problem is cost.

Hi Peteeng, when using the fly cutter, Z will be at the highest position. There will be table or tube under the plates.

[peteeng]

Hi Vily - very good. Why using a fly cutter? They are used to reduce the load on the machine and do not have a very big material removal rate. Speak to your tooling supplier about tools suitable for what you are trying to do then work back from the tool? Peter

[Vilyoo]

Hi Vily - very good. Why using a fly cutter? They are used to reduce the load on the machine and do not have a very big material removal rate. Speak to your tooling supplier about tools suitable for what you are trying to do then work back from the tool? Peter

Well, I wanted to save money and flatten its own 5 legs by removing one by one and put in on the others 4. First rough with face mill and then fly cutter. Fly cutter will make it perfectly flat even if it is not well tramped.

[peteeng]

Hi Vily - By tramped I think you mean Trammed? If it is not trammed correctly the bigger the tool the bigger the error. Also the commercial machine builders that fabricate machines do not fill their sections with anything, that may tell you something. Peter

[Vilyoo]

But with fly cutter the face will be always in one plane, or I'm wrong

[joeavaerage]

Hi,

I think it is better to fill it with concrete rather than being empty. I will see if I can handle bigger tubes, the only problem is cost.

Why???? It adds mass and not much else. In order for the filler material to participate in the load sharing it must adhere perfectly to the inside of the tube, something that concrete will never do well.
Then, given the tubes stiffness it will flex a little and that little flexure will be impressed upon the filler. A little bit of flexure in a 50GPa material will offer little restoring force and contribute to the load sharing
in inverse proportion to its modulus.....ie not worth a damn!

Filling with concrete sounds intuitively appealing but in fact is a waste of materials. It was also intuitively appealing that a cannon ball would fall faster than a lead shot and it was not until Galileo proved it was not so in 1610 or so,
I remember it well, it was a Tuesday....He published it many years later when he was under house arrest for proving the heliocentric model....and really pissed the pope off!

Craig