How to Calibrate the JCUT machines?

[jallitt]

I just received a used 1200x1200 machine controlled via ncstudio. It doesn't have a touch probe but pin 1 on the NCstudio cable is where one gets connected on mine. If you don't have a "mobile calibrator" or a touch plate/probe then you CAN'T calibrate the z axis as it reads this input to tell if the tool has touched the probe.

On my mach 3 machines I have a small piece of 1.6mm thick printed circuit board connected to the touch probe input on the breakout boards and it works perfectly every time.

If I open the IO state tab in NC studio I see a J1-1 indicator. Check your breakout board inside the machine and see if pin 1 is connected to anything. My BOB has pin 1 as the first connector but I've seen pics of other boards where the screw terminals weren't in 1-15 order so double check with a multimeter.

Theoretically if you ground the pin 1 input you should see the state of the IO indicator change on the ncstudio IO state tab.

So, it should be as simple as running a long wire from pin 1 soldered to your "touch probe" pcb. BUT make sure everythign is grounded first and that you can see J1 change if you touch the probe to the tip of the tool. Then measure the thickness of the pcb accurately and enter that as the "touch probe" thickness.

BUT... my machine has 110 volts between the chassis and the PC ground so I'll need to fix that first. Ideally the touch probe would be optoisolated from the PC (on the mach3 boards this is usually done for you).

Once I get the machine positioned where I can access the BOB (at the moment it's wedged against a wall while I clear some space for the machine) I'll do some more testing.

[jallitt]

I managed to hook up a piece of blank PCB as a touchprobe for zeroing the z axis to my NC-studio controlled router.

After making sure that the machine was earthed correctly. The power socket earth wasn't connected at all so I grounded the chassis and the power supply and verified that there was no voltage between the PC earth and the chassis of the machine.

Then I opened up the IO state tab in NC-studio:



and verified that the calibration input changed state if a wire on pin 1 of the 15 pin cable was grounded - it did:



Then I soldered a 1" square of blank PCB to a long wire connected to the NC studio breakout board in the machine and verified that the "calibration" input changed state if I touched the copper surface of the PCB to the tool tip - it did.

In the "manufactory" menu there's a setting for mobile calibrator thickness - I set this to 1.6mm - the thickness of the PCB. The default password for access to the manufactory menu is "ncstudio" BTW.

Then I placed the PCB under the tool and pressed "mobile calibration". NC-studio lowers the z axis slowly until it contacts the PCB then raises 5mm. The Z reading is now 6.6mm which is the 5mm it raised + the 1.6mm thickness of the PCB.

here's a really boring video of the tool touching the PCB:



I'd feel happier if the input was optoisolated - but this seems to work.

If you try this and there's no continuity between the tool tip and ground - then this won't work. I've seen one person rig 2 wires - one to a PCB and the other to an alligator clip which he clipped to the tool when calibrating.

My plan is to swap the machine over to mach3 but this makes the machine usable until I get around to it.