All the Gecko G320 reset pins (pin 5) can be tied together, so that one faulting driver will shut down the others.
My Campbell breakout board has no provisions to utilize the fault/reset outputs from the Gecko drivers, so Mach does not know when an error occurs. As a result, the machine was delivered with no fault detection and the fault-reset pins were tied to +5V permanently for automatic start-up. Since the fault pins are tied high to +5V, the Gecko does automatically reset after a few seconds, but Mach does not know and continues milling with the machine now offset from where it should be. (It caused my milling machine to eat up a tailstock and milling bit, and me to nearly have heart failure). According to the Gecko engineers, leaving the fault tied high is a really bad idea (duh).
I have been struggling with the Gecko G320 drivers to automatically start up when powering up the machine, and also to provide a fault condition when milling so that an ESTOP can shut down the machine. The trouble is that when the Gecko G320 is powered up, it immediately "Faults" and must be enabled by momentarily placing +5V on the fault pin 5 for a few seconds (reset function). This has the unfortunate result that when the fault is wired to a limit function, you can not power up the machine since it triggers a shut-down as soon as the RESET is pressed and power comes up. The Gecko people can only suggest a center-position manual switch to toggle the drive units on or off, but does not prevent the fault-startup problem from occurring when sensing pin 5. They ran short of pins and doubled up the reset and fault functions (unfortunate choice).
I am sure that others have solved this problem since Gecko drivers are in wide use, so if anyone has a solution to this problem, please post it.
In the mean time ... In order to get around the problem, I just purchased some photo-darlington transistors (Fairchild L14F1) that are sensitive enough to sense the fault LED on the Gecko drive modules when glued or placed above, which then can activate the Limit pins on my Campbell break-out board (it works). I still need a delay to prevent the fault start-up from killing power-up. I intend to place a N-FET (at least 15 mA capable) with its SOURCE tied to GROUND and the DRAIN being common to all photo-darlington EMITTERS, and with a simple RC delay on the N-FET GATE (CAP to ground, resistor to +5V with the common connection tied to the GATE) to hold off activation of the fault until 10 seconds have passed when powered up. Each photo-darlington COLLECTOR is wired to a limit pin on the breakout board. This way, each Gecko can have its own fault sensed, hopefully reporting which one faulted and having the message displayed on the screen in a status window (needs some programming to monitor the LEDs). Alternately, a 555 timer could be used as the delay element in place of the FET, but is harder to wire without a small circuit board for the IC pins.
If all goes well, I can hit the RESET button to power up the machine, Geckos fault for several seconds (with pin 5 tied to +5V for automatic reset) but fault reporting to MACH will be delayed by the FET until the Geckos clear their fault conditions. Now the machine should run normally until a limit switch or a Gecko fault lights its LED and activates a photo-darlington which pulls a limit pin to GROUND and triggers an ESTOP shutdown, and hopefully reports the offending error condition.