I am currently using my cnc knee mill to machine a 3D part that is roughly 10 inches by 20 inches. the table travel is approx. 12 by 24 inches. run time is approximately 2 hours. I would like to build a sliding table that would allow me to machine the first part, then pause and slide the second part across to the same zero point and start the program over. this would double the run time and be more productive when it runs unattended. I plan on linear guides, ball screw, servo motor. these were available when cnc knee mills first appeared back in the day. any thoughts would be appreciated.
Im not sure what you are describing but it sounds like you could use a pallet system of some sort. On a simple CNC the change over wouldn't be automatic and could be challenging due to weight. There are a number of companies making fixturing components to do this.
This brings up the question of how big is the part and would it be conducive to mounting on a plate for lifting onto the mill.
Does the machine have a movable ram? Would it be easier to automate the locking/unlocking and moving of that, and just build an oversize fixture table to hold the 2 parts?
I've seen linear stages available that will do - given the appropriate dimensions - more or less what you want. Some considerations:
1. Any slop in the table slides or positioning mechanism (assuming X-axis shift for two parts) will result in problems. Think of it as your vise not being bolted down firmly - ungood.
2. You'll need a brake for the servo or X-axis cutting moves may back-drive the sliding table.
3. Do you have enough Z-height to clear your longest tool since the linear stage will eat up some of your range?
If I wanted something like this, I'd probably make a sub-plate and pin it to the table after indicating it dead-on. I'd also try to figure out a pneumatic tapered (or dowel) pin lock oriented in the Z axis for the two work positions. When the stepper or servo needs to move for part #2, the pins pull out of the sockets, the table is moved, and the pins shoot back home in the other position to lock it down. In this case, you'd probably want some play in the motion axis - maybe using some really soft rubber in the motor coupling or even some soft-ish belleville springs between the ball-nut and the moving table mount. You'd then want the motor to get it close, but with enough flexibility to permit the lock pins to get the table to a precise position with essentially zero backlash.
Benefits to using a mechanical lock/pin thingie:
1. No axis motor brake needed
2. Use cheap acme screws and a dumb AC or DC motor with limit switches
3. Or simply an air cylinder to push it.
4. Lock provides better repeatability as well as superior resistance to unwanted motion than would a stepper or servo motor.
5. No energy needed for a brake, no air draw until you unlock.
6. If you set it so the unlock pressure requirement is higher than the movement cylinder, then an air failure results in no movement. This means the second cycle cuts air over the just-completed part and not a half-way moved table with the un-cut stock ready to bugger a tool or your spindle.
I'm picturing a second air cylinder pushing/pulling on two rods that run though the long axis of your lower table. (or more rods across the short axis) The rods are spring-loaded. The rods have tapers or ramps along their length that bear against 2/3/4/6 (pick a number) vertical pins that stick up through holes in to matching sockets in the bottom of the moving (upper) table. Picture the rod being sort of saw-toothed. No air, the pins are jammed up in the sockets, locking the table along the motion axis. Apply air, and the rods are pulled and the pins drop down out of the sockets permitting the upper stage to move.
Might be better to use bearing balls and hemispherical or conical sockets instead of pins so they drop free more easily. Even if you don't get the ball or socket hole locations dead-on, they should average pretty well for a repeatable location. And because you're using a wedge to push them up they aren't going to move when subjected to transverse forces given a shallow enough ramp/taper on the rod. If you use balls or tapered pins they'll even wind up pre-loading the ball slides to take up any minor Y or Z backlash when they drive up in to the upper table sockets.
Two air cylinders, one center open 5-way (movement), one spring return 3-way (lock)
Two air solenoid valves connected to relays (controlled by output pins on cnc controller)
01 - cycle start, insert g-code here
02 - end of part #1, go to safe Z
M99 (Macro fires unlock cyl, (pause), fires movement cyl, (pause), releases lock cyl, then moves 5-way motion valve to center-open position.
03 - Cycle start
M30
Add position sensors as desired for verification of position and you're in business.
Just a few thoughts - sounds like an interesting project.
I am still thinking that I could build a sliding table on linear guideways and moved with a servo motor turning a ballscrew. it would be like adding any other 4th axis. the parts I am machining are only out of 1 inch 7050 aluminum. z travel should not be a problem.
Very doable, use a clearpath servo and you can set the actuation distance and trigger it with a simple output of your controller. The clearpath servos can be programmed for whatever distance you need the sliding table to move. Then just use a macro to trigger it and work it into your g code. Seems like it should be simple enough? Could just add the macro at the end of the first parts code along with having the machine return to the initial starting location, then just copy and paste the entire code if it is the same. Should run first part, shift table and cut same tool path again on the second part.