[steve123]

Hi Dan,

Thanks for the feedback. I'm glad to get it.

1) Spread/stagger bearing blocks.

I don't know about this. I understand spreading them. Staggering them will cut down on envelope. The 20mm bearing blocks have over a ton of load capacity each.

2) Connect or cap ends of Z axis U-shape beam.

I think the idea here is to keep the z axis back plate stiffeners from vibrating sideways, right? I lengthened the blue stiffeners to the full height of the z, and enlarged the top motor plate so it can tie into the stiffeners. This should take care of the top. The bottom is a harder problem. I don't want to limit the z travel to put a plate there. I'm currently arguing with myself about this.

3) Your bed/structure probably ok, but the y & z do not appear to be comparably stiff. For cutting steel you have 2 weak-links in the Y&Z axes.

This gets us to how much structural loop stiffness we need. I have been going by page 26 of "Principles of Rapid Machine Design" which says 10 - 25 N/um deflection are considered adequate for machine tools. That translates to 57 - 144 lbs per .001 of deflection. Putting it another way, 100 lbs cutting force should give between .0018 and .0007 deflection.
For the Z complete axis (back plate assy. and spindle plate assy.), I have calculated .00021 deflection per 100 lbs cutting force when cutting near the table with a short cutter. There are two reasons for this. 1. In the lower range of the Z, the Z spindle plate top and bottom blocks are very near the Y rails. Thus the Z back plate doesn't bend much at all. 2. The box section z front/spindle plate is extremely stiff with .00009 deflection per 100 lbs. The z back stiffeners are really only for very high Z where the back plate is a cantilevered beam. I really think that the z is stiff enough. I will be making the Z last. I plan on cutting the blue stiffeners on the machine, then taking the Z back apart and installing them.

The Y full twisting is harder to calculate, because it is made of two beams tied together. I used the MyCNCuk cnc stiffness calculator, and if you assume the two beams welded together with the spacers become one 10" tall by 5" wide beam, you get .000051 with 100 lbs cutting force. That would be really small if I calculated it correctly.

4) The contributory stiffness and damping may not worth the toil to do it, nor the potential mess it may make. Unless you use polymer concrete or epoxy-granite.
If when the concrete cracks and starts to break down, that could make matters even more messy. Another option is to stuff the tube structures with precast elements (think patio pavers) or granite counter-top scraps laminated together use a polymer grout or RTV or cast silicone or just regular old polyester resin typically used for fiberglass.

I'm not worried about the concrete breaking down. I don't see how it matters if it turns to sand over time. I'm going to be looking at formulations before I start mixing.

5) If your work envelope requirements are negotiable, consider designing your machine around HIWIN "Bridgeport" Milling Machine replacement screws and bearings. They are ground to accuracies beyond the capabilities of the other elements of your machine.
For the money these screw are exceptional value. I have not checked price in years but I recall them being around $1,000US for a x&y set.

This is interesting. I will look into it.

6) You are going to need a STIFF bed material. I would consider Granite Slabs (laminated), or a used or imported surface plate. I don't know if or how well this would work, but have considered bonding interlocking patio pavers together layer by layer to build a slab. In case you ever have to move the machine, it would be nice to be able to detach and the slab and move it independent of the machine.

I can get granite, but don't like the idea of drilling 64 holes in it for tie downs. Concrete is a good material for dampening vibration. With good structure underneath, and rebar inside, it should be nice and stiff. I can also make the concrete bed 4" thick (another 100 lbs). I'm considering that to allow more z clearance when the bed is pushed back.

I believe milling steel successfully with a machine of this construct may be tall-order, but I do wish you good luck! I also believe that you are exceedingly wise to maximize the research and planning phase of this extreme project.

Thanks. I this it's possible. I am most worried about dampening vibrations in the gantry.

Thanks again for your feedback. It gives me a lot to think about.

-Steve