Abstract:
Rotational moulding has been heralded as a plastic moulding method with great
potential. The process offers virtually stress-free products having no weld lines or
material wastage, and utilises relatively inexpensive moulds. Yet, its widespread
growth is hindered due to long production cycle times, which are limited by the time
required to heat up and cool down a mould and the product within. To address this
issue, efforts have been made by the author to enhance heat transfer to and from
moulds, ultimately reducing cycle times.
The aims of this project are four-fold:
. To investigate the effect of convective heat transfer to and from a mould by
changing the external mould surface (i.e. adding roughness elements or pins to
the mould surface).
• To predict cycle time reductions gained through the use of surface enhanced
moulds.
• To validate the predicted cycle times with experimental trials.
. To recommend a combination of moulding conditions, applied together with
surface enhanced moulds, for the rotomoulding industry to pursue.
Due to the complex rotation of the mould, heating and cooling are most commonly
achieved by convection to the external surfaces of the mould using air as the transfer
medium. Two methods of heat transfer enhancement were considered, which are
applicable to both external forced convection and mould exterior surface
modification within the existing rotomoulding process. These methods are rough
and extended surfaces.
A series of preliminary analytical studies (predictions) and experiments were
performed using roughness-enhanced and pin-enhanced plates to verify that these
methods work. The results for the average effective heat transfer coefficient, h^,
from the parametric studies for the extended surfaces were too high compared with
the preliminary experimental data. While a turbulent flow was expected in the
rotomoulding machine, the reality was a transitional flow (i.e. a mixture of laminar
and turbulent flows). As a result, the correlations used in the parametric studies for
the rough surfaces were invalid. Better prediction methods, which involved both
analytical and numerical studies, were employed to improve the prediction of heJfec
for both rough and extended surfaces.
The h values were then applied directly into an existing rotomoulding simulation
package, RotoSim, to predict cycle times. Based on the preliminary experimental
results and predicted heat transfer coefficients, dimensions and array of roughness
elements and pins were decided. Three cubical aluminium moulds (i.e. plain,
roughness-enhanced, and pin-enhanced) were fabricated for the experimental
purposes. To validate cycle time predictions, a series of experimental trials were
carried out.
Experimental and predicted results have shown significant cycle time reductions for
both surface enhanced moulds. The average experimental cycle time reductions
gained through the use of surface enhanced moulds were 18 and 28% for the
roughness-enhanced and pin-enhanced moulds respectively. Even though the
prediction methods fail to estimate the exact experimental cycle times, they are very
useful for predicting relative cycle time reductions between the plain and surface
enhanced moulds. The average predicted cycle time reductions were 21 and 32% for
the roughness-enhanced and pin-enhanced moulds respectively. These results
indicate that the predicted and experimental cycle time reductions are in excellent
agreement with each other.
This study has demonstrated that a significant reduction in rotomoulding cycle times
can be achieved through the use of surface enhanced moulds. This thesis also provides a foundation for performance analysis of surface enhanced moulds.
Recommendations have been made to maximise the benefits from such moulds,
without depreciating the quality of the end products. Applying the proposed
moulding conditions (i.e. a combination of higher oven flow rates, internal mould
pressure, and water-cooling), the cycle time reductions were approximately 65 and
70% for roughness-enhanced and pin-enhanced moulds respectively (weighted
against the plain mould with 'normal· moulding conditions). Such savings are very
significant, inviting the rotomoulding community to incorporate these techniques
efficiently in an industrial setting.