Recently I installed a CNC milling machine in the garage, a Tormach PCNC 440. This is the smallest and newest of three CNC mills they manufacture in China but design, QA, and support from Wisconsin. These machines are designed to be affordable but still offer a very usable level of performance. Current models are controlled by an external Linux-based PC built from standard PC components, plus a Mesa 5I25 FPGA-based PCI I/O card, running Tormach’s PathPilot software. This setup replaced an older configuration that ran third-party Mach3 software on a standard PC under a customized version of Windows.
I’m a beginner at machining, and have been taking it slow and easy as I get up to speed on operating the 440. A few days ago I was working with the Tormach Passive Probe. This is a simple mechanical sensor that mounts in the machine’s spindle (where a cutting tool would usually go) and cables up to the controller. This allows the controller to detect the X, Y, and Z locations of surfaces on the work piece or fixtures. It’s one of several ways to get the machine properly lined up to the work.
The probe comes with a prominent warning to disable the spindle before use. Obviously, if the spindle starts to spin with the probe mounted, the probe’s cable is going to get wrapped around the probe and destroyed. Tormach tells me that the warning exists because of a known problem with the older Mach3 software, which could cause the spindle to start running under certain conditions. On the older model mills, there is a control panel switch that locks out the spindle. Since the 440 never used Mach3, Tormach decided to leave out the spindle lockout switch, one of many cost reductions that makes the 440 so affordable. There is still an interlock on the spindle door, so the spindle cannot be activated with the access door open.
Unfortunately, the spindle access door can’t just be left open as a substitute for a spindle lockout switch. The top panel of the optional enclosure kit just clears the spindle with the door closed. With the door open, the machine’s head can’t be lowered into the work area. So, one has no choice but to trust that the controller won’t activate the spindle while you’re using the probe. If you make a mistake and tell the controller to activate the spindle, or if something goes wrong in the controller and it activates the spindle when it shouldn’t, there’s nothing to prevent the spindle from spinning and destroying the probe.
So, I was working with the probe. I had used it to line up X and Y, and was thinking about Z, when I got interrupted to go out to lunch. I didn’t want to lose my X and Y settings, so I left the machine turned on. When I returned from lunch, this is what I found:
Not only was the cable wrapped around the probe and snapped off where it enters the enclosure, but it had also grabbed and destroyed the armored hose that carries flood coolant. It had thrashed the lightweight fabric bellows that protects the Z axis ways, and also snapped off the expensive ruby-tipped ceramic probe tip. The user interface on the PathPilot screen looked normal but was unresponsive. Crashed.
I don’t know whether this was a one-time glitch (cosmic rays?) or a defect in my hardware or a systematic bug in the PathPilot software. Tormach is investigating, trying to reproduce the problem.
In the meantime, I decided to add my own spindle lockout switch. I consulted the schematic diagram for the mill and noted that the access door interlock switch is in series with the power being supplied to the module that runs the spindle motor. No combination of failures or glitches in the spindle motor driver could cause the spindle to turn when there’s no power being delivered to the driver. This is a very safe design, but it does require an interlock switch and wiring that’s able to handle all the current that the 3/4-horsepower spindle motor can draw. The schematic shows a 6 amp fuse in that circuit. I could add my own switch in series with the door interlock, as long as it could handle the 6 amps. I chose a toggle switch rated at 20 amps, GC Electronics model 35-130, because it was available at my local Fry’s.
The next question was where to mount my lockout switch. Perhaps the most obvious place would be on the side of the mill’s electronics cabinet. That’s where the main power switch is, and also the connector for plugging in the probe’s cable, and several blank panels that appear to be designed for specific future accessories. The needed circuits are readily accessible inside the cabinet on barrier strips, and there’s plenty of clearance inside the cabinet to mount a switch on the side panel. Unfortunately, my garage installation is somewhat crowded and the side panel is not very easy to reach. I worried that I might skip using a lockout switch mounted in that inconvenient location.
The easiest surfaces to reach are on the front of the mill’s enclosure, but all those surfaces are single thicknesses of sheet metal, with the inside being within the splash zone for coolant and chips. Mounting anything electrical there would mean adding a sealed enclosure for it, and finding a clean way to run heavy wires to it. That didn’t seem like the right answer.
Instead, I chose to mount the switch above the spindle motor, right next to the door interlock switch. This should be convenient, because I have to open the spindle door in order to change tools, so I will be in there every time I mount the probe. (When Tormach releases the power drawbar add-on kit, that will no longer be true. I may revisit the lockout switch placement then.) In my junk box I found some 3″x1″ aluminum C-channel. I cut a 1.75″ length of this stock, and mounted the switch on one of the small sides. I drilled two holes through the other small side, and matching holes in the back panel of the spindle enclosure, and mounted the assembly with two bolts. The C-channel is more than stiff enough to support the switch.
The switch has screw terminals, and so does the door interlock switch, so wiring was easy. I just removed one wire from the door interlock switch and moved it to my switch, and added a 6″ jumper of #14 wire between my switch and the door interlock switch.