A Method of Using the Normal Distribution to Model Rubber MDR Cure

A method of using the Normal Distribution to model Rubber MDR Cure

By

Al Olsen

SKF Sealing Solutions

Abstract

Rubber Chemists use the traditional parameters such as minimum torque (ML), maximum torque (MH), one point rise (Ts1), two point rise (Ts2), 50% cure (Tc50) and 90% cure (Tc90) for collecting data off the MDR, Moving Die Rheometer curves.  These parameters are based on historically simple methods of pulling data off the curve from the days of ODR tests and computing numbers on paper.  These parameters are not based on the actual chemical cure reactions.  These parameters are different mixtures of various cure phenomena.  For example, the Tc50 is based on the minimum torque, the maximum cure torque (which is a function of test time) and finding the midpoint between these two parameters.  The longer the test, the longer the time to 50% cure.  The maximum torque is the value during the test, not the true maximum cure state of the rubber material.  If the test is short, the maximum torque, MH, is lower than if the test time is long.  The current parameters can be difficult to interpret as there are multiple causes for each effect.

The purpose of this method is to develop test numbers where each parameter measures one property related to the rubber.  It is based upon the logic that there is usually one limiting cure reaction for a given rubber cure system.  There will be one number for the average cure speed of the rubber.  There will be one number for how quick the rubber cures, once the reaction begins.  The true minimum and maximum torques will be determined independent of the test time.  All of this information can be combined to determine cure times for rubber parts to get proper cure state during the molding process.

MDR tests are run with at least three different temperatures to provide the raw data.  The test temperatures should cover the predicted temperature range for molding the parts.  The data is then converted into a Normal Distribution to be used to predict the cure times at each temperature.  The Arrhenius Equation is used to determine the changes at each temperature.  This combination can be used to develop surface maps to help determine how to develop rubber compounds with the most robust processing characteristics.

Having test parameters that have one meaning allow better implementation of techniques such as Designed Experiments.  While Rubber Chemists rarely have the luxury to run a full Six Sigma Project, running designed experiments can help in getting accurate data that will have good predictive modeling properties.