CNC Router V2 aka "Martta"

I'm constructing an CNC router with about 60cm of X/Y travel, and 40cm of Z travel. It should be able to mill aluminum quite easily. Quite large wooden structures will be my target also (like fish-tail joints used in drawers etc.) Rapid speeds of over 1000 mm / min will be my goal for moving speeds. To accomplish that, motors and driver electronics need to be quite big. I'm not yet decided which motors to use, but probably quite large steppers from OEM. Perhaps I'm adding 4th axis for lathe-use. It all depends on motor prices really, as they will be probly only parts that I have to buy as new... :)

Project started by getting aluminium parts as swap from friend. I did some software for him to use with his robotic Geodimeters. Parts consist of large aluminum frames and quite nice flat aluminum parts. I bought also 3 meters of 12mm / 3mm per turn trapez-bar to use for movements.

Router has basic framing ready. Frame is made from aluminium plates and castings. Frame has Hepco slides on X and Y axis, and all moving parts are supported by Hepco ball bearings.

Z-axis foundations laid out. Z-axis will be using "ball-bearing slide", it will have about 35 cm's of movement. Z-axis plate will need cutting and attachment to moving X/Y structure.

Y-axis trapez and ball-bearing fixed. It's moving quite nicely when turned by power-drill. Motors will possibly be Nema 42 class motors with 9A current per phase and maximum voltage of 140V. Drivers will be Italian RTA's, with 10A / 75V (so 750W) per axis. Each axis will have own power-supply of 3A, so the power-supplies are really the weakest link in the electronics. Well, they can be easily updated in future if more torque is really needed (UPDATE: I really doubt that torque is not going to be a problem. In first tests motors easily ruined my aluminium motor-coupling when axis jammed...)

Not much progress, since I have acuired a small metal lathe and have learned how to turn parts with it. I tried to make axis support from stainless steel, but it was harder than I thought it would be. I didn't have suitable drills to make holes to it, and my drill quickly melted because drill was running too fast. Well, I think it needs practising anyway. Z axis plans changed, I got nice Z-structure from nearby factory as surplus item. It weights about 30kgs on in own, so it's going to need some torque to move it around...

First motor mount done. I welded it from iron, and it seems to be quite rigid. Also driver problems solved - it needed NPN transistor between PC and the driver board. It seems that RTA's GMD03 drivers read 0 if voltage is between 0 and -2, and 1 if voltage is between +10-12 OR input is left open. And because PC cannot generate such voltages from printer port (at least the industrial PC I'm using cannot do them), I was forced to use transistors. Well, I think I could have used pull-up resistors also, but transistors seem to work just fine. I use the transistors to drive the input pin to ground-level, and when transistor input is low, the pin is left open (=high Z state). I'm able to get feeds of 4.5 m / min with Z axis :) And it has the smallest motor, X and Y are much more bigger motors with hopefully more than enough torque. At least the motors are "whirrling" nicely when driven freely.

At first it seemed that they woudn't work properly. The motion was wobbly and shaky, and I guess there wasn't much of torque involved there. But after I double checked the connections, and Mach2 setup parameters, I discovered the "Minimum pulse width"-parameter. I increased that from 2us to 6us and wow - problem solved and completely nice and smooth movements :)

Y and Z moving and working. With Y, rapids are not yet quite possible due resonance - everything needs proper fixing before real tests can begin. But basic movements are possible. It will definitely need better axis couplings though. I may even buy them as new, they would cost about 20e / axis... Let's see what happens.

All motors and axis fixed in place and working ! Speeds are very impressive, but I don't really want to try maximums just yet. Mainly because I'm using "rubber-pipe" type axis couplings for now. And I don't have limits installed yet, and it would be quite catastrophic if some axis is driven too far and it stalls the movement... I got a little hint about powers involved: the couplings I made before were made from aluminum, and they didn't last too long before they wore out. And it seems that steel is much better for that kind of things anyway, the threads for tightening nuts are much more durable in steel than in aluminum.
The next thing I need is a holder for Bosch 600W router that I'm going to mount to the mill. I think I'll just use my old holder from my V1-version of CNC-router. At least I'll use it until I'm able to make bigger holder from aluminium.

First cuts made! Wood seems to cut very easily :) Aluminum seems to need that all the bolts are properly in place - my Z axis is fixed with just 2 8mm bolts instead of 4 or 6 bolts needed for proper fixing... But wood speed was very nice :)