Paper presented at the 30th Annual Southern African Transport Conference 11-14 July 2011 "Africa on the Move", CSIR International Convention Centre, Pretoria, South Africa.
As part of the review of the seal design procedures a mechanistic approach was fareseen. A fuller understanding of the behaviour of seals was required for the modelling process, and this led to the investigation of seals behaviour under accelerated and laboratory environments. As part of the process, the behaviour of unmodified bitumen and bitumenrubber under these conditions was assessed, and aspects from this study are provided in this paper highlighting the behaviour of bitumen rubber under the current design methods.
The paper first evaluates the critical factors in the design of seals, and the mechanisms
and modes of seal failures. The experimental programme is then presented which
consisted of testing with the one-third scale Model Mobile Load Simulator (MMLS3), as
well as the Hamburg wheel tracking test (HWTT). Results are then presented and
discussed. It was found that the key to good seal performance is the greatest possible
binder application rate without resulting in flushing or bleeding, with the postulate that
seals performance is more dependent on the higher binder application rates, rather than
improved qualities of the binder (but consider always highest possible binder application rate is directly dependent on binder quality).
