To do or not to do, can it be done? MK4 drill steel.

[Neale]

I call my long axis X, and gantry is Y axis. Need to be clear about that to avoid confusion! X tubes are 100x50, supported at a number of points along their length to the main machine frame. No particular worry about strength there. Gantry is 3 off 50x50. Could have been 100x50 plus 50x50, but that's what I had left over! In fact, the twin 50x50 welded together is stiffer in the vertical plane than a single 100x50, I would think.

Again, based purely on my experience, 3mm is adequate for strength. However, as JazzCNC has pointed out elsewhere, it's not just strength that matters, it's also things like vibration and resonance. My machine does suffer from that at times and although it's not a major issue for me, I would probably go up to at least 4mm wall thickness just to reduce this effect.

My Z axis is, indeed, aluminium plate. I used 20mm Ecocast or equivalent ground finish tooling plate just because it is flat. I used 20mm as that allowed me to cut recesses into the plate for the Z axis ballscrew for clearance without unduly weakening it but it seems pretty stiff to me.With the Hiwin rails and bearing blocks properly arranged, I'm not sure that there is very much bending load and it is stiffened in any case by the spindle clamp block and partly by the rails themselves.

I am happy with NEMA 23. It's very easy to over-specify in this area and it's likely that NEMA34 would give poorer performance, bearing in mind that I want this machine for cutting wood as well, and also want decent acceleration for fine detail cutting. I have compromised by using 5mm lead ballscrews; 10mm would give more speed but lose torque. As it is, the motors hit max speed at about the same speed as the max ballscrew speed (to avoid whipping) so they are fairly well-matched; there's plenty of grunt and in practice, all that suffers is max rapid speed. NEMA 34 may have more max torque but have much more rotational inertia and internal inductance, which both reduce acceleration. OK for a mill, not so good for a router. If you do the sums, it works out that the torque needed to spin up the X ballscrews (all 1750mm of them, 20mm diameter) is roughly the same as the torque needed to accelerate the gantry, despite its weight. It's easy to forget that rotational inertia of the screws is a significant factor in all this; don't sweat over saving a few grams of gantry weight.

Why put twin screws on Z? For a "router", downforce isn't that big an issue although I can see that if you are looking at heavy drilling or similar plunge cuts, this might be a different matter.

All machines are a compromise and that's why a lot of us build our own - we can build to make those compromises in areas that don't matter too much to us and focus on what we do want. I wanted a machine capable of good 3D detail performance for engraving. Below is a first pass (don't have picture of final version) of a brass master I engraved for my local model engineering club. The recess is approximately 20x10mm, so the level of detail isn't bad for a machine designed to handle 4'x2' sheets of plywood. To do that using 3D V engraving needs good acceleration in all three axes; max cutting speed is only of academic interest! But I can also move at 5000mm/min when needed. On the other hand, you are looking to mass-produce large holes in steel but accuracy is not as high on your list, so while you are welcome to steal any ideas from my machine (as I did from others!) I'm not sure that you want to copy it.