Abstract
During temperature cycling the radial flow of oil, which
is caused by differences in thermal expansion between
the oil and paper, may lead to cavity formation in mass
impregnated HVDC cables. Partial discharge activity in
these voids is an important failure mechanism in this
type of cable, but the governing processes are not fully
established.
This paper contributes to flow modeling through a
literature survey and a comparison between two
alternative flow patterns in the insulation. Only one
detailed model of radial flow in this kind of insulation
has been found in the literature. The model, which is
based on porous media flow in a homogeneous
insulation with constant porosity, and its suitability as a
starting point for further modeling is reviewed.
Two possible flow paths, radially outwards through the
paper or between paper tapes from one layer to the next,
have been considered. Based on previously unpublished
experimental results, the flow resistance R of the two
alternatives has been estimated. This quantity is found
to be of the same order of magnitude in both cases,
meaning that there probably will be flow both through
and between the paper tapes.
is caused by differences in thermal expansion between
the oil and paper, may lead to cavity formation in mass
impregnated HVDC cables. Partial discharge activity in
these voids is an important failure mechanism in this
type of cable, but the governing processes are not fully
established.
This paper contributes to flow modeling through a
literature survey and a comparison between two
alternative flow patterns in the insulation. Only one
detailed model of radial flow in this kind of insulation
has been found in the literature. The model, which is
based on porous media flow in a homogeneous
insulation with constant porosity, and its suitability as a
starting point for further modeling is reviewed.
Two possible flow paths, radially outwards through the
paper or between paper tapes from one layer to the next,
have been considered. Based on previously unpublished
experimental results, the flow resistance R of the two
alternatives has been estimated. This quantity is found
to be of the same order of magnitude in both cases,
meaning that there probably will be flow both through
and between the paper tapes.