IT IS WELL established that good preparation is a prerequisite for good extraction, yet many mills today are not achieving the nominal benchmark of 90% pol in open cells. In some cases, the low preparation is the result of insufficient power available to the shredder. In most cases, however, adequate power is available but 90% pol in open cells is not being achieved. It is also well established that there is a strong link between power consumption and preparation. If sufficient power is available but good preparation is not being achieved, it is likely that the available power is not being consumed. Following recent work by Wilmar Sugar, a different set of guidelines for the design of the shredder grid has been proposed. While it is known that grid spacing, depth and wrap affect preparation, the optimum configuration is not known with certainty. This paper describes initial modelling that has been carried out to predict the flows of air, billets and prepared cane through a shredder and associated shredder air flow measurements. Thoughts on the way in which these flows affect preparation are presented.