What size stepper motor do I-need.

[Neale]

That spreadsheet is great, and incorporates a lot of theory. Two problems, really (and I don't want to take anything away from the work that went into it):

- it takes a bit of interpretation in some cases, as you have found. Two ballscrews, belt drive with step-down, etc. Not impossible to work around, but it's not obvious how at first glance.

- actually, there are only a few answers available although the spreadsheet results can help you tweak them.

I went through pretty much exactly the same process a couple of years ago. My machine is close to yours in size - 1500 long axis, two ballscrews, gantry around 50kg (guess). For motors, in reality, the choice is Nema 23 or 34. The larger motor has high inductance and a lot of inertia. When you do the sums, it gives poor performance. A friend recently swapped the 34 motor on a machine he bought second-hand for a 23 and significantly improved performance. So the answer is Nema 23. Make sure you get a low-inductance motor, and 3Nm or 4Nm will work fine. I use two 3Nm motors as 4N were not available widely at that time but I would probably go 4Nm today for the sake of a couple of quid. It is important to drive them properly, hence talk of 68V. Why this funny number? Because it's a safe value to use with a decent digital stepper driver (EM806 or AM882).

Speeds - I went through the same "what's a good speed" thinking. I'm using 2005 ballscrews which reach their critical speed at about the corner speed of the steppers (900-1000 RPM). That gives me 5000mm/min. Actually, I think I'm running at about 4500 at the moment to avoid any possibility of stalling one motor (which is also why you want modern digital drives with stall detection). This is not as fast as the machine could use although it works well. However, I seem to be doing quite a lot of fine cutting with a lot of 2d/3d detail and acceleration is more important than max speed in this case. Your 2010 with 1-1 belt drive would run faster than mine at the expense of resolution (but not enough to worry about).

Some thoughts to get you started, and i'm sure that others will be along with different views! However, my remarks are based on a similar machine that is up and running, not just theory.