' major arcing' ?
Shaft voltage arises from magnetic imbalance in the machine. Flux that's asymmetric around the shaft couples it so current tries to circulate along shaft , through bearings and back via motor frame.
A couple volts is typical open circuit . If neither bearing is insulated a very few amps might flow depending on imbalance.
Above around 500 horsepower machines are big enough it's necessary to insulate one bearing to break the circuit. Shaft current will pit journal bearings. It makes ball bearings acquire stripes so regular that your first impression is "What in the world put those raccoon tail stripes on that bearing race ? " Of course when the bearing fails catastrophically it wipes out the evidence. The natural reaction then is to mutter "Galling" and try a different grease.
When VFD's became commonplace , shaft current troubles moved into the world of small motors. So there's a LOT of information out there now much of it scholarly. Thirty five years ago it was mysterious.. i wracked my [edit:[STRIKE]brain[/STRIKE] ]
alleged brain over a pair of 450KVA machines that intermittently ate bearings. They weren't insulated..
If you have significant arcing it suggests to me you might have trouble beyond a dirty brush. Maybe the insulated bearing is no longer insulated ?
Is something elevating that shaft voltage well above ground ?
It's worth looking at some simple things . Light bulb tells you something even if it stays dark.
How's the vibration on that machine? Does it vary with excitation? That'd be a symptom of shorted rotor turns which cause magnetic imbalance - you have less mmf hence less flux at slots with a shorted turn. Those can be centrifugal and if bad enough can ground the field...
These guys' Dad literally 'wrote the book' on diagnosing shorted rotor turns in-situ.
Red trace is slot by slot mmf.
http://www.generatortech.com/
Good luck - keep us posted !
old jim