Safe spindle control for R2E4 bridgeport?

I have an R2E4 CNC that has been running LinuxCNC for almost 10 years now. I'm making some changes to improve the setup, in particular to allow spindle reversing for tapping (I have a spindle encoder now).

The way the spindle control was setup on this machine it had a spindle activation knob that manually powers one side of the contactor. The other side was the control. Once the spindle is running there is relay logic that "holds" the spindle powered. There is a reversing contactor that is actuated whether the gearbox is in low or high (it is reversed direction), and then a separate reversing relay used for instant reverse during tapping. Essentially the way this is setup the spindle can never turn on unless the operator physically turns the knob and then hits go on the program. It is a manual tool change machine so I can see why they thought this was a good idea for safety reasons. Starting the spindle and program after a tool change is a two handed operation. LinuxCNC is currently setup with the "spindle enable" line so it can power the spindle, but I hadn't wrapped my head around how to use the "reversing" function" for tapping. It also has a spindle brake which as wired simply is activated whenever the spindle isn't running or the knob isn't turned (the control originally did not individually control it).

Electrically, I don't see any reason I couldn't just wire the Forward, Reverse and Spindle brake as three separate GPIO controls (which seems to be standard for LinuxCNC). This would eliminate the spindle activation switch and the "redundant" spindle interlock. I could then activate the spindle with a more conventional pendant control and LinuxCNC could reverse the spindle as needed for tapping. LinuxCNC should be able to handle the sequencing of those. What I'm wondering is if there are any safety considerations I should consider before doing this. It is possible I could figure out how to make LinuxCNC control the current spindle relay logic and retain the redundant spindle activation switch, but this doesn't seem to be the way most people handle it. Many CNC machines have "one button" spindle activation (my VMC does) but they also have a safety cabinet often with a door interlock (many of which have been deactivated... but the intent is there).

Thoughts? Am I missing something?

 

Makes sense. I can see why Bridgeport did it this way since it means there is no physical way the spindle can go from stopped to started unless the operator is holding the spindle enable knob. For a manual tool change machine this isn't a big deal.

I was thinking I would maybe just keep the knob for now. Maximize safety on a machine that I don't personally run the most. I cannot figure out how the control reverses the spindle though. It seems the schematics of the machine are incomplete or inaccurate.

EDIT: I figured out that the Bridgeport documentation is incorrect. The output channel marked as "Index Air" allegedly intended to run a pneumatic indexer option was apparently repurposed for spindle reversing sometime after the documentation was finalized. Somewhat amusing since although rigid tapping is supported and discussed in the manual it is not clear how this could/would work given the schematics. I hooked that up to the "Reverse" channel on LinuxCNC and boom - I can now run the spindle in both directions and rigid tapping works. If at some point I want to get rid of the "spindle activation safety switch", I can just short that and the control will be able to run the spindle without the additional manual interlock.