My own ship has Rolls Royce Azipuls and may be slightly different to the ABB ones but we have had problems with the steering gear as well.
We have 2 steering motors driving through Torque limiting Safety clutches and a reduction gearbox onto an annular gear round each Azipul.
Each motor has an automatic electric brake on the free end, when the motor is energised so is the brake and it pulls off.
Also on the Free End is a "Digital Encoder" this has a bearing in the middle which fits over the motor shaft and a light bracket to secure it to the casing. Inside there is a laser which shines through fine lines cut into a mirrored disc, light receptors turn these into Digital Pulses. Counting pulses gives the angular displacement of the shaft, integrating with respect to time the velocity, integrating a second time gives acceleration and so on.
The feed back from these is used to control the power to the two Motors which are each driven by their own Variable Frequency Drive, but are obviously, mechanically synchronised through the gearing to the Azipul.
The first failure was the bearing on one of the encoders seized, the entire encoder started turning with the shaft and sheared off the securing bracket, then wrapped all the electrical wiring round the shaft and broke it. This triggered a feed back failure alarm, and the Chief on board at the time investigated, found the fault and completely electrically isolated that motor and carried on. When I came back from leave realised that, with the power isolated the brake on this motor was permanently engaged, so I jacked it off manually until spares arrived. The ship continued to work normally, management arguing that the ship had 2 Azipuls working and 3 steering motors so was still DP2.
When the new encoder was received we fitted it and everything worked normally.
Later we had a problem with one of the Safety Clutches on the other Azipul failing, again the ship continued in service for months, until the spares came and were fitted. After a few months the safety clutch failed again. When we fitted the new safety clutch we noticed that the currents drawn by the two motors were different and were unable to work out why, Rolls Royce suggested changing the encoders and that cured the problem, so it appeared that the encoders were giving faulty feed back signals, and effectively causing the motors to fight each other overloading the safety clutches.
I am wondering if something similar happened in this case??
It is always better to ask a stupid question than to do a stupid thing.