Designing a four column mill?

[redfedoradog]

Hi everyone,

First post on this forum, and I'm going to start by saying that I have little experience with machining and I welcome any advice or criticism. I do have experience with drones/robotics, and due to the weight constraint we get very creative with the design (and FEA) to compensate for the lack of material. I thought it might be interesting to apply some of those design techniques to a cnc mill.

I started by recreating a popular extrusion based router and testing it in FEA. After checking some constraints and assumptions, I managed to get the same predicted deflection as an actual test on the machine (https://www.youtube.com/watch?v=B27nUN1ejIQ). I used this simulation setup for all subsequent tests.


After around a hundred iterations, I started to make some progress. Adding two additional columns to the front of the machine immediately doubled the stiffness. I was encouraged by the progress. Adding a L-shaped plate tripled that, adding a horizontal brace doubled the stiffness again. Interestingly, there's no difference between steel and aluminum plates (under 1%). I suspect that the extrusion deforms in a way that negates any improvement.


All in all, the design started at 0.15mm deflection (at 110N, cutting force calculated from a 500W spindle) and ended up between 0.002 - 0.01 mm deflection. I realise that this design imposes some restrictions on the size of material that can fit, but it is perfect for my purposes. The machine is around 500 x 600 x 600 mm with the enclosure, with a work area of 200 x 200 x 100 mm.


The next hurdle was damping. Aluminum has around half the loss factor of mild steel and a quarter of cast iron (http://www.mech.utah.edu/~bamberg/re...e%20Design.pdf), which seemed like an insurmountable problem.


But then I stumbled on "micro motion within a bolted joint also provides significant damping. The friction along
dry joints dissipates energy and thus provides damping". Almost 80-90% of the damping in bolted machines comes from the bolts and bearings. Which means that the total loss factor of a steel machine and aluminium machine differ by only 5-10%. With 390 bolts in my current design, I'm hoping for some significant damping.


I guess the question is, did I go wrong in my simulations? I'm sure that there's no way the industry would miss out on a design that is 15x more rigid than the common gantry design. Is the enclosed design too restrictive on the workholding/cooling/something entirely different that I missed?

Thanks for looking,
Andre