33kV tape joint failure

While applying tape the jointer passes one hand below the cable to catch the tape roll at the other side, stretches the tape roll from below and while circumnavigating the cable periphery releases the tape roll across the cable to repeat the action again and again, thus quite naturally insulation thickness below the cable is of less thickness due to the stretch action which is not there at the top since the tape roll is frequently released.

This inconsistency of tape thickness is taken well care of by the safety margin and does not leads to failure. Failure could have been due to other factors like inadequacy of the stress cone.

There is no rule for the thickness to be 10 mm or 15 mm, standard practice is to apply insulating tape minimum up to the periphery of the outer most cable sheath, essentially with a slight bulge over the joint part (central part), further final layers of anhydrous tape and layer of sealant are applied.

Yup, I also quite agree that the thickness is still sufficient to prevent a fault. In fact the joint has been in service for a few years. I'm also not aware of any high PD issues in the cable. Could overloading of the cable or circulating current in the earth screen cause any deterioration or softening of the rubber insulation due to high temperature?

I'm not so sure about the stress cone, but the installation of stress control/ semi-conductive tapes were found to be as per manufacturer instruction. The same method has been in use in many other joints and this problem has never occured. The fault occured exactly in the middle of the connector length, quite far away from the edges where there are more electrical stresses.