Investigations of the T-burner and its role in combustion instability studies

Of several devices introduced to study combustion instability in solid rocket propellants, one, known as the "T-burner," has become the most widely used. With this device the response of a burning propellant to a small pressure disturbance can be measured. Such information is vital both to the understanding of unsteady combustion processes as well as to the assessment of the stability characteristics of solid rocket motors. Although the T-burner has been used for several years, several questions concerning the device have arisen and, for the most part, have remained unanswered. Moreover, little effort has been given toward showing the relevance of T-burner data to predictions of instability in rocket motors. The present investigations, comprising over 400 test firings in T-burners of various lengths and diameters, were undertaken with the major objective of gaining a better understanding of the T-burner itself in order to answer some of these unresolved questions. Another objective was to compare T-burner predictions of rocket motor instability with actual observations made in a previous study. Among the investigations was a comparison of several ignition procedures which showed clearly that a poor, uneven ignition can seriously affect the test results. Included among the ignition studies were tests conducted in transparent chambers to permit high-speed motion photography of the firings. These tests confirmed the common assumption that the T-burner is basically a one-dimensional device. Tests using burners of different diameters showed that although the acoustic losses of the T-burner are nearly independent of diameter, the limiting amplitude of the oscillations is strongly dependent on the latter. The dilemma raised by these observations was resolved by measurements which indicate that the heat transfer from the combustion gases to the burner wall is strongly dependent on the amplitude of the waves. From these measurements emerged a nonlinear description of the damping in the T-burner which accounts for both the behavior of the losses as well as that of the limiting amplitude. When two independent T-burner methods were compared, the results obtained were initially in very poor agreement. However, when the T-burner losses were assumed to be non-linear as mentioned earlier, excellent agreement was observed. Finally, the T-burner predictions of instability in rocket motors were in rather poor agreement with direct observations made in a previous study. Although this lack of agreement is not understood, it is doubtful in the light of the present investigations that the major error lies in the T-burner measurements, for these should be relatively accurate. Moreover, these results indicate the need for more comparisons of this type in order to determine the usefulness of the T-burner in predicting combustion instability in solid propellant rockets.