I spent about two months of my spare time designing and building this CNC plasma cutting machine.
The table is made from 4"x2" 14 ga steel tubing. The cutting area is 64" long by 60" wide. The material support slats are 1/8" x 2" flat bar on end, 4" apart, supported at each end and in the center. The slats are bowed to prevent them from flopping back and forth and also prevent them from being directly in the cut path when cutting long straight lines.
The table has a built-in water tray to contain sparks and smoke. The water also serves as coolant, preventing warpage of the piece being cut. The tray is 18 ga galvanized steel sheet welded to the bottom of the table. The water level is controlled by air pressure in an 80 gallon storage tank underneath the table. By regulating air pressure in the tank, the water level in the table can be easily controlled. The water can be completely lowered into the tank so that the slag can be cleaned out of the table. A trap is built into the table drain to prevent slag from entering the storage tank.
For the X axis linear guides I used precision ground V rail and rollers on one side only. The opposite side is the same except that it uses flat rollers on a flat rail. The gantry is made from 2"x2" and 4"x2" tubing, and welded to the carriages made from 1/4" steel plate. Each carriage has two fixed rollers on top and two smaller rollers on the bottom. The lower rollers lock the carriages to the rails and prevent tipping or rocking.
The Y axis uses two V rails, one bolted to the top and one bolted to the bottom of the 4"x2" gantry tube.
The Gantry is set up for a second Z axis on the opposite side, but it has not been used. The Z axis is a precision linear stage that I found on Ebay. It is from some kind of silicon wafer handling machine. The plasma torch holder is on a small ball slide that activates a home switch when the tip touches the sheet. A pneumatic scribe is mounted next to the torch head and is used for center marking and engraving. When the scribe is activated it is forced down by a pneumatic cylinder on a ball slide.
The X and Y axis are driven by timing belts and stepper motors. The belts are 3/8" width .2" pitch XL series polyurethane with kevlar reinforcement. The belts are fixed at each end, and wrapped over drive pulleys and tensioned using two double skate bearings as idlers on each belt. Both axis are direct drive with 15 groove pulleys on 210 oz/in NEMA 23 stepping motors, two on the X and one on the Y. I'm driving them at 4x the rated voltage and 2x the rated current which seems like pretty common practice. One revolution of the motor equals 3" of linear travel, so I get resolution of .015" per full step, and with the 1/8 step drivers I get resolution of .001875" per microstep. I do not know what the accuracy is, I've never measured it but it has been sufficient for the kind of work I do. The Z axis is also belt driven with small gear reduction.
The four stepper motors, X, Y, Z and A are driven by Keling 4030 microstepping drivers with the A axis slaved to X in the control software. The drivers are mounted in a NEMA control cabinet along with automatic torch height control electronics, control relays, 24 volt power supply, and other components. I am currently working on a new larger control cabinet that will also contain the PC and 18" touch screen monitor. I use SheetCAM and Mach 3 to control the machine. If I have time I'll make a separate page to decsribe my home made automatic torch height controller.
