Design and Simulation of Two Stroke Engines

Design and Simulation of Two-Stroke Engines
A more accurate combustion model in two zones
The theory presented here shows a single zone combustion model. A simple extension to
burning in two zones is given in Appendix A4.2. Arguably, what is presented there is merely
a more accurate single zone model. This is not accidental, as the computation of any combustion
process based on heat release data (from a Rassweiler and Withrow analysis), or on a
mass fraction burned curve (in the Vibe fashion), must theoretically replay that approach in
precisely the same manner as the data were experimentally gathered. Those experimental
data are referred to, and analyzed with reference to, a single zone, i.e., the entire combustion
chamber. Thermodynamically replay it back into a computer simulation in any other way and
the end result is totally inaccurate; perhaps "theoretically meaningless" is a better choice of
words to describe a lack of mathematical logic.
4.4.3 A one-dimensional model of flame propagation in spark-ignition engines
One of the simplest models of this type was proposed by Blizard [4.2] and is of the eddy
entrainment type. The model was used by Douglas [4.13] at QUB and has been expanded
greatly by Reid [4.29-4.31]. In essence, the procedure is to predict the mass fraction burned
curves as seen in Fig. 4.7 and then apply equilibrium and dissociation thermodynamics to the
in-cylinder process.
The model is based on the propagation of a flame as shown in Fig. 4.1, and as already
discussed in Sec. 4.1.1. The model assumes that the flame front entrains the cylinder mass at
a velocity which is controlled by the in-cylinder turbulence. The mass is entrained at a rate
controlled by the flame speed, Cfl, which is a function of both the laminar flame speed, qf, and
the turbulence velocity, ctrb-
The assumptions made in this model are:
(a) The flame velocity is the sum of the laminar and turbulence velocities.
(b) The flame forms a portion of a sphere centered on the spark plug.
(c) The thermodynamic state of the unburned mass which has been entrained is identical
to that fresh charge which is not yet entrained.
(d) The heat loss from the combustion chamber is to be predicted by convection and
radiation heat transfer equations based on the relative surface areas and thermodynamic
states of burned and unburned gases. There is no heat transfer between the two
zones.
(e) The mass fraction of entrained gas which is burned at any given time after its entrainment
is to be estimated by an exponential relationship.
Clearly, a principal contributor to the turbulence present is squish velocity, of which there
will be further discussion in Sec. 4.5. The theoretical procedure progresses by the use of
complex empirical equations for the various values of laminar and turbulent flame speed, all
of which are determined from fundamental experiments in engines or combustion bombs
[4.1,4.4,4.5].
It is clear from this brief description of a turbulent flame propagation model that it is
much more complex than the heat release model posed in Sees. 4.4.1 and 4.4.2. As the physical
geometry of the clearance volume must be specified precisely, and all of the chemistry of
the reaction process followed, the calculation requires more computer time. By using this
322