mixtures thanLe>1.Comparisons betweenthe present results and those of Law and co-workers agree with this prediction, andtherefore support the residual stretch theory. The magnitude of any residual stretchcould be determined by measuring the stretch directly, using the radial velocitygradient in the luminous zone, and comparing it with the upstream velocity gradient.Physical meaning of the reference surfaceThe reference surface is definedaswhere the mass flux in a stationary sphericalflame equals that of the corresponding planar one-dimensional flame.Theory,experiment and modelling all show that the reference surface is in the hot part of theflame.
The work in the thesis suggests a connection between the reference surface and theflame processes of heat release, fuel consumption and radical production.It was notpossible to decide which of these is the most important.Further work is needed tosee if a connection exists with any particular flame process.One-dimensionalburning velocityresults obtained from extrapolationmethodsshould, in principle, be independent of the reference surface.This is the case forthe expanding flame method, but not for the counterflow method of Law and co-workers.Asymptotic theoryAsymptotic theory predicts that the burning rates of flames in lean mixtures of lightgases and rich mixtures of heavy ones should be increased by stretch. The burningrate should be decreased for all other mixtures. These predictions were supported bythe present experiments. It is interesting that asymptotic theory is able to predict theexperimental Markstein lengths qualitatively, and in some cases quantitatively. It isfurther evidence that analytical techniques have a role in combustion theory, evenwhen precise simulations are feasible.Universality of stretch as a description of flamesThe question here is whether the same Markstein length is valid for characterizingthe effects of both extensional and dilatational stretch.(The phenomenologicalanalyses in Chapter2suggest that it should be).Within the admittedly largeexperimental error, the Markstein lengths for expanding flames and flames instagnation-point flow are equal.The results of the present work therefore supportthe general applicability of stretch, at least in the weak stretch regime.Theagreement between the Markstein lengths from expanding flames and those derivedby Law and co-workers in counterflow flames and by Searby and Quinard inwrinkled flames provides further support.Importance of defining burning velocity correctlyIt is clear from the present work that the effect of stretch on flames in near-stoichiometric mixtures of most fuels is small. Since stretch rates on burners and in
expanding spherical flames are also quite small, it follows that stretch effects are notthe cause of the large scatter in reported values of maximum burning velocities. Themain reason for this scatter must therefore be flow divergence or, putting it anotherway, incorrect siting of the reference surface.
Upload your study docs or become a
Course Hero member to access this document
Upload your study docs or become a
Course Hero member to access this document
End of preview. Want to read all 332 pages?
Upload your study docs or become a
Course Hero member to access this document
