Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

~ A II!! The Engineering Society

~ For Advancing Mobility
~ Land Sea Air and Space® 400 COMMONWEAL TH DRIVE WARRENDALE. PA 15096

881264

Investigations on Scavenging of Two=

Stroke Engines
Daniel S. Sanborn and Nabi Dedeoglu
University of Dar es
Salaam, Tanzania

International OIl-Highway & Powerplant

Congress and Exposition
Milwaukee. Wisconsin
September 12-15. 1988
 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

The appearance of the code at the bottom of the first page of this paper indicates
SAE's consent that copies of the paper may be made for persona! or internc::1 use, or
for the personal or internal use of specific clients. This consent is given on the con-
dition, however, that the copier pay the stated per article copy fee through the
Copyright Clearance Cen.ter, lne., Operations Center, P.O. Box 765, Schenectady.,
N.Y. 12301, for copying beyond that permitted by Sections 107 or 108 of the U.S.
Copyright Law. This consent does not extend to other kinds of copying such as
copying for general distribution, for advertising Of promotional purposes, for
creating new collective works, or for resale.

Papers published prior to 1978 may also be copied at a per paper fee of $2.50
under the above stated conditions.

SAE routinely slacks printed papers for a period of three years following date of
publication. Direct your orders to SAE OrdP.r Department.

To obtain quantity reprint rales, permission to reprint a technical paper or per-

mission to· use copyrighted SAE publications in other works. contact the SAE
Publications Division.

A6SAE~1JtD
..mItaCII!d lind ifIdflzl!d in
fill) SAE GIob.tI Mobility D4r~a

SAE GLOBAl uoalUTY OArA8ASE

No part of this publication may be reproduced in any fonn, in

an electronic retrieval system or otherwise, without the prior
written pennission of the publisher.

ISSNOI48· 7191
Copyright © 1988 Society of Automotive Engineers,lnc.

Positions and opinions advanced In this paper are those of Persons wishing to submit papers to be considered for
the 8uthor(s) and not necessarily those of SAE. The author presentation or publication through SAE should send the
Is solely responsible for the content of the paper. A process manuscript or a 3DO word abstract of 8 proposed manu,·
Is available by which discussions will be prlnted with the cript to: Secretary, Engineering Activity Board, SAE.
paper If It Is published In SAE Transactions, For permlsaion
to publish this paper In full or In psrt, contact the SAE
Publications Division. Printed In U.S.A.
 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

881264

Investigations on Scavenging of 'Two=

Stroke Engines
Daniel S. Sanborn and Nabi Dedeoglu
University of Dar es
Salaam, Tanzania

The flow processes are extremely

ABSTRACT complex and it is therefore very
difficult to present them mathematical-
Compared with experimental and theo- ly. Moreover, it is also an expensive
retical approaches modeling of the excercise.
scavenging process in a two-stroke
engine offers great advantages on The advantage of modeling are well
account of its simplicity and economy. known to research and test engineers.
The clear definition of influencing In general it yields reliable results
factors and a better physical inter- in a short time and is relatively
pretation of the model laws governing simple and inexpensive.
the flow process help to design the
model and to determine the test condi- Scavenging models and their test
tions; the model laws also allow the procedures have been greatly improved
obtained test results to be transfer- in recent years so that today they
red to the prototype. contribute to the development of the
scavenging process and provide useful
A relatively simple and inexpensive information on scavenging parameters
liquid model presented enables qualita- and their influencing quantities.
tive observations and quantitative
evaluations of different port confi-
gurations to be made. For the design METHODS OF SCAVENGING
of port configurations some recommen-
dations are presented. There are a number of scavenging
methods in two-stroke engines. In
The short circuiting ~nd Fig. 1 below three maj or types are
instabilities in scavenging test illustrated (1)*
results are also mentioned.
The arrang<;>ment in (a) is called
cross scaveng1ng. In this arrangement
air enters the cylinders through the
THE PERIOD during which the pro- inlet ports and is directed to the
ducts of combustion are replenished by cylinder head. The fresh charge forces
the fresh charge is called the scaveng- the exhaust gases to the outlet ports
ing period. During this period the located on the opposite of the inlet
cylinder is exposed to both inlet and ports. This is an old-type arrange-
exhaust ports. The scavenging process ment, but it is still used, especially
lasts from the opening until the in small two-stroke engines with
closing of the inlet ports. A large special piston designed to direct the
mass of combustion products leaves the
cylinder during the exhaust blowdown
period, the remainder being expelled *) Numbers in parentheses designate
during the scavenging process. references at end of paper

0148-7191/88/0912-1264$02.50
Copyright 1988 Society of Automotive Engineers, Inc.
 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

2
881264
Ollft!:!-t vo.l'fi:
;f~-l~V

/ =
/ ~
(a) (roSs Scavenging Ibl Loop Scavenging lei UniflOri Scavenging

Inlet
ports
. .... Outlet
ports /$" '-f'
,' _ . outlet
ports

~+
W' Inlet
pex-ts

Fig. 1. Metho ds of scave nging

air jet to the cylin der head. Cross metri c purit y facto r (72-78 %) becau se
scave ngi~g has a relat ively low of less short circu iting .
volum etrlc purit y facto r of about
65-68% (base d on the scave nging air Uniflo w scave nging shown in (c)
ratio = 1. 0). usual ly has one exhau st valve on the
cylin der head. The types havin g the
exhau st ports at the top of the
Anoth er arrang emen t shown in (b) is cylin der requi re a secon d pisto n in
calle d loop scave nging . In this arran - order to contr ol the exhau st timin g.
gemen t the ex~aust ports are ~ocat~d These types of engin es are calle d
above in the lnlet ports . The lncom lng oppos ite pisto n engin es.
fresh air is direc ted to the unpor ted
wall and air stream follow s the path In uniflo w scave nging the inlet
shown as a loop. The cylin der wall and ports are usual ly arrang ed tange n-
the cylin der head guide the flow to tially in order to produ ce a control~
the exhau st ports . It is an impro ved led swirl for stabi lizat ion of the alr
form of the Schnu erle scave nging stream . This rotat ional movem ent
(2,3) , which is usual ly used in small cause s an unsca venge d centr al core
size engin es. which reduc es the volum etric purit y
facto r. In some arrang emen ts a
A consi derab le impro vemen t in the radia lly direc ted secon d row of inlet
purit y facto r can be achie ved by: ports locat ed under the first row can
reduc e this unsca venge d core.
- locat ing the inlet ports at the
small est possi ble segme nt of the
cylin der liner ; this ensur es The volum etric purit y facto r can
that reach as much as 84-86% .
the fresh air jet can reach the
top of the cylin der,
Both cross scave nging and loop
- locat ing the exhau st ports at scave nging are sensi tive to the
strok /bore ratio . The volum etric
least at the half segme nt of the
cylin der liner to lower thei r purit y facto r begin s to decre ase
upper edges in order to incre ase consi derab ly when the ratio s are more
the effec tive engin e strok e. than 1.6 and 2. 3 respe ctive ly, as
indic ated in Fig. 2. Uniflo w
Compared with cross scave nging , scave nging even impro ves by the
loop scave nging has a highe r volu- strok /bore ratio .
 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

881264 3.

1.0

0.9

O.8 ----
r--.
-
e.- Uniflaw Scavenging

o.7 Loop Scavenging - r------

-
o.6

o5

15 2.0 3.0 ,5
Fig. 2 Effect of the stroke/bore ratio on the volumetric purity factor

SIMILARITY CRITERIA engine and in the model will be

similar regardless of the number of
The following eguations describe -media in the process and their distri-
the scavenging process: the continuity bution.
eguation, the equation of motion (the
Navier-Stokes equations), the energy
equation, the thermal and calorific PHYSICAL SIGNIFICANCE OF THE
equations. The similarity laws for the SIMILARITY LAWS - Every similarity law
scavenging process can be determined represents a ratio of two quantities
from these equations (4). of the same dimension, by which its
defini tion and significance is
The following similarity laws are established.
essential for the scavenging process:
Strouhal Number - This represents
Strouhal number: Sr l/u. £It the ratio of local to convective
Froude number Fr u 2 /gl. p/£lp acceleration. If the process is
Euler number: Eu = £lp/pu 2 unsteady the Strouhal number must be
Reynolds number: Re = ul/v applied. For example, it dictates the
Since two different media, i.e. ratio of the mean piston velocity to
fresh air and exhaust gas take part in the flow velocity. And it has to be
the scavenging proCess, there are the same in the engine as well as in
consequently two Euler and two the model. The Strouhal number is
Reynolds numbers, one for each medium. proportional to the reciprocal of the
These give two additional similarity scavenging air ratio.
laws, viz.:
Euler Number - This is the ratio of
Density ratio: Pa/Pe and work of the pressure forces to the
Kinematic viscosity ratio: va/ve
Consideration of the surface tension a kinetic energy of the flow.
leads to
Weber number: We = lu 2 p/a Froude Number - This represents the
ratio of the kinetic energy of the
If there is a geometrical similar- flow to the work of the gravity force.
ity between the engine and the model It is therefore important when the
and if the above similarity laws are
the same, then the process in the gravity force has a noticeable ~ffect
in the process. If the denslty
 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

4 881264

differences are too big, the Froude ~he scavenging process and then ignor-
number is written in the form 1ng the laws whose influence is insig-
nificant lead to partial similarity.

p/l>p is almost equal to 1.0 for water DISCUSSION OF SIMILARITY LAWS

and air (e.g. for ship modeling).
Test carried out' in the scavenging
Reynolds Number - This is the ratio model with different Reynolds and
of the kinetic energy of the flow to Froude . numbers as well as the
the work of the internal friction kinematic viscosity ratio of the
force. If the Reynolds number in- participating media have shown that
creases, the stability of .the laminar they have no significant influence on
regime is lost and the motion becomes the scavenging process (4).
turbulent. The inertia force gradually
dominates the process. If the media selected for the model
are miscible, i. e. the effect of
Weber Number - This represents the surface tension is negligibly small,
ratio of the kinetic energy of the the We number may be assumed to be
flow to the work of the surface maintained.
tension. If the number is large, it
means that the inertia force predomi- If the viscosity and gravitational
nates over the surface tension (media forces are negligible compared to the
are mi3cible). pressure force and furthermore if the
inertia force induced by the cyclic
Mach Number - The ratio of the velo- motion of the piston is ignored, the
city of the flow to the sound velocity Euler number approaches 1.
in concerning fluid. It characterizes
the compressibility of the fluid. Bet- This means that the inertia force
ween the fractional density charge and is set up solely by the pressure
the Mach number, the following force. The Euler number is therefore
physical relationship exists: assumed to be maintained in the engine
as well as in the model;
l>p
After this, the three condi tions
P remalnlng to be fulfilled on the
scavenging model are:
If Ma 2 « 1 as in the case of the
scavenging process, the fluid behaves 1. Geometrical similarity,
as if it were incompressible. The 2. Strouhal number and
scavenging process can therefore be 3. the density ratios.
taken as a volumetric process. However
it is suitable to define the By using the Strouhal and Euler
parameters of scavenging process numbers, the Scavenge number can be
volumetric*). determined

From what has been said so far, it l>p . l>t 2

will be seen that when more than one Sv = Eu . Sr- 2 (1)
force is acting on a system, then more P . 12
than one similarity law will conse- If the scale factor for geometrical
quently have to be fulfilled i f the similarity is given and the substances
similarity of two systems is to be are selected, the scavenging pressure
ensured. This requirement presents difference can be calculated by using
some difficulties in the design of the the scavenge number for each model
model, the choice of suitable media speed.
and in the performance of the tests.
In many cases, it is mathematically
impossible to achieve an ideal model.
For practical performance, it is
therefore essential to adapt the way
of partial similarity. Examining the Eq. 2. gives the operational
influence of different model laws on conditions for the model.

*) see Appendix.
 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

881264 5

a) They must be miscible so that

r---- ENGINE the Weber similarity law is
almost maintained,
! b) they must be cheap
separable for reusing,
and
Similarity
c) the density ratio as similarity
Geometrical law must be maintained,
Physical d) they must be free of toxicity,
stcodY"lIwto causticity, flammability and
klnomllUc any other hazards to health and
dynamic
lhannodynamlc
must not corrode the equipment.
ale. Table 1 shows some of the
Chemical liquids used in modeling.
etc.
Because of these disadvantages it
is suitable to use harmless liquids,'
I e.g. boiler water (for both scavenging
I air and exhaust gas) without
I
'--------0 Similarity Numbers considering the density ratio. In this
case one of the liquids used is

1 -¢ diluted with a chemical stubstance

(e.g. caustic soda) to determine the
volumetric purity factor by titration.
Kerosene and water were tested as an
I I
Partiai
Similarity'
..
I I MODEL I immiscible liquid combination (2).
This combination of liquids allowed
the scavenging results to be evaluated
simply and with acceptable accuracy.
As can be seen in Fig. 4 the
quanti tative model test results are
Tests affected by the density ratio.

Fig. 3 diagram of modeling . '.0

Uniflow scavenging
The same similarity laws are used
to transfer the model test results to 09 i
the engine. Fig. 3 clarifies what has I - -_
-_-
I---I--------?-v:/~/:.-~-r
"""'-1 = 0.48
been discussed above.
06
TEST IMPLEMENTATION
07
TEST FLUIDS
0.6 - - - -Stroke / bore ratio = 2.5
Since the scavenging process has
been categorized as an incompressible
process, liquids can be used in the 0.\
model to replace scavenging air arid
exhaust gas. Liquids have the follow-
ing advantages: 06 06 10 12
A"
1. They are easier to manage, Fig. 4 Effect of the density ratio on
2. the process may be retarded model test results
(essential for photography and
even for a conventional video The tests carried out with differ-
camera), ent density ratios have shown that
3. they can be dyed without the qualitative results remain
changing their properties, relatively the same, which means
4. quantitative test results can that is is not a disadvantage to
be obtained easily and weighed use the liquid pair water/water for
accurately. finding the best port configura-
tion. The right density ratio is
The liquids must be selected essential for the quantitative
according to the following criteria: results.
 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

6 881264

Table 1. Liquids used in scavenging model

Dynamic Density Kinematic

Liquid viscosity viscosity
10- 5 Ns/m 2 kg/m 3 10-6 m2/ s

Tetrachlorethylene 90.4 1621 0.558

Trichlorethylene 60.0 1463 0.410
Cyclohexane 98.0 779 1.258
Alcohol 120.1 789 1.522
Water 100.2 998.4 1. 004

Air 1.855 1.166 15.91

Kerosene 24 800 3

Trichlorethylene and Tetrachlorethylene are toxic, cyclohexane

is flammable. It is therefore advisible to take special
precautions when using these liquids and to use them for a
short time only.

MODEL TESTS
After selecting the liquids and
fixing the sui table model speed, (to
ensure a high Froude number) Eq. 2 can
be used to calculate the scavenge air
pressure that allows to operate the
model.
Fig. 5 shows the uniflow scavenging
model designed with a view to maximum
verdatility. The reservoir (a) under
the glass cylinder (b) is filled with
the liquid representing the scavenge
air, .while the glass cylinder holds
the liquid simulating the exhaust gas.
The glass cylinder length is adjusted
to the stroke/bore ratio of the actual
engine.
The easily interchangeable lower
end of the cylinder liner (c), (not
visible) is made of cast aluminium or
zinc. The liner can be easily made and Fig. 5 Uniflow scavenging model
it also allows the necessary
corrections to be carried out. The At the end of scavenging, i . e.
actuating wheel (d) is driven by a after closing the inlet ports, the
variable speed electric motor (e) with piston sits against an offset which
a wide range. Through a linkage system seals the pressurized liquid against
(f) the same motion is imparted to the the liquid in the cylinder.
piston as in the actual engine. The
inlet ports are controlled by the The operating lever initiates a
reciprocating motion of the piston, single cycle simulation. The mixture
while the exhaust valve (g) remains remaining in the cylinder is then
open the whole time. drained off and the degree of the
 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

881264 7

volumetric purity factor Cf.v can be the exhaust ports. The capacitance
determined by weighing. The volume of change is recorded due to the varying
the scavenging air is given by the dielectric constants of the liquids
outflowing volume at the exhaust (h). taking part and yields the momentary
The ratio of this volume and the mean values for the vOlumetric composi-
reference volume given by the total tion of the outflowing exhaust. Fig. 6
cylinder volume at the end of scaveng- shows the time curve calculated for
ing (inlet ports closed) gives the loop scavenging.
scavenging air ratio (see Appendix).
From these it will be observed that
EXPERIMENTAL OBSERVATIONS - at 24 deg. crank angle after the
start of scavenging, fresh air is
Scavenging time curve The already detected in the exhaust
conventional model tests yield only due to short circuiting after
the final values, such as the which the latter steadily
volumetric purity factor Cf.v' the increases,
scavenging air ratio ASCI etc. It is
also interesting to know how these - when the fresh air front reaches
factors behave with the time scale. the exhaust ports (about 52 deg.
The following information is needed to crank angle after the start of
ascertain this, viz: scavenging) mixing scavenging
begins,
- Time curve of the cylinder
pressure and - the volumetric purity factor Cf.v
- time curve for the frest air reaches the value of 78% with a
content of the outflowing exhaust vOlumetric scavenging air ratio
gas. AS C = 1.20,

Ov
Asc
Bv
1.2
I
1
1.0 1
I
1
I
0.8
1
I
13·
0.6
I§:
I",
13-
0.4 '" 0
Ii,;
I'"
0.2 1
I
I
I
0
140 160 180 200 220 'CA
Fig. 6 Time curves for the fresh air content Bv in the
exhaust, scavenging air ratio Age and volumetric
purity factor Cf.v.

the fresh air loss due to short

The fresh air content of the circuiting is approximately one
exhaust gas is determined by using third of the total fresh air
electrical capacitors mounted behind losses.
 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

8 881264

The final results obtained by this will be discussed in an another paper

method can be verified by weighing the to be published soon.
mixture remaining in the cylinder, as
already stated above. Short circuiting - When a jet pene-
trates into an almost stationary
medium, its velocity changes according
Instabilities in Scavenging Test to Newton's law of momentum.
Results The model tests yield
unstable results in the volumetric Momentum change of fresh air jet +
purity factor a v in a certain region momentum change of fresh air and
of scavenging air ratio ASC' exhaust gas front = O.

Fig. 7 illustrates the measured a

puri ty factor under the same condi-
tions vary considerably in the region ufront
between 0.8 and 1.0 of scavenging air 1 + pel pa
ratio. Usually this region is outside
the operating range of diesel engines. The fresh air front then moves with
It is therefore not so harmful. reduced velocity and takes on the form
of a mushroom. If the exhaust ports
ex, are located in the vicinity of the
10+----+---+---+---+-------1-- inlet ports as in loop scavenging, the

., s,r[}=2.133 growing mushroom causes short circuit-

ing of fresh air to the exhaust ports .
The diameter of the mushroom is propor-
tional to the jet diameter therefore
"t---t--/r---t---t==-t- smaller inlet ports reduce short
circuiting
'.7
Short circuiting normally occurs in
'6+----;fL--7"'+---+- cross scavenging and loop scavenging.
As indicated, it starts at about 25
'.5 deg. crank angle after onset of
scavenging and continues during the
whole scavenging period.
06 DB 10 12
The distance between the inlet and
Fig. 7 Unstable test results exhaust ports has an influence on
in loop scavenging model short circuiting too. An optimum value
of this distance is approx. 6% of the
In uniflow scavenging the fresh air engine stroke for loop scavenging.
front becomes unstable and the final
resul ts vary from test to test in
spi te of constant test conditions. Preswirl - Before the start of scaveng-
This instability can lead to an ing in the model the exhaust gas in
unsatisfactory fuel-air mixture the cylinder is at rest, which is not
formation, which can result in like in the actual engine. Some tests
cycle-by-cycle pressure variations have been carried out to determine the
even in diesel engines. This problem effect of this gas motion originating
from the previous cycle on the

Table 2 Effect of gas motion in the cylinder on the scavenging

performance
scavenging v*
vol. purity factor (%) pressure m/s
(bar)
scavenging air ratio 0.8 1.0 1.2

without preswirl 75.9 85.7 89.5 0.356 0.66

with preswirl 76.1 86.1 89.7 0.341 0.69

*) flow velocity in cylinder at top dead center

 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

881264 9

scavenging performance, particularly inexpensive liquid model. It allows

on the volumetric purity factor. For qualitative and quantitative tests to
this purpose, e.g. in the uniflow be made for solving many problems
scavenging model, a preswirl such as relating to scavenging and mixture
existing in the engine was induced. formation.

The results of these tests are shown For qualitative tests it is not
in Table 2 in comparison the results necessary to maintain the density
ratio.
without and with induced preswirl.
From these data it is concluded that Some recommendations are given .for
the effect of the gas motion in the the use of the model and engine test
cylinder has no significant effect on results in the design of the port
the scavenging process. This configuration for loop scavenging.
conclusion leads to simplified model
and operation conditions. The mushroom form of the incoming
fr 7sh air jet causes short circuiting,
WhlCh amounts to almost one third of
INFORMATION OBTAINED FROM MODEL TESTS the total fresh air losses.
The scavenging model can be used to The effect of gas motion in the
determine and solve the following cylinder, originating from the
problems: previous cycle, is insignificant, so
that it need not be simulated in the
- The volumetric purity factor, the scavenging model.
scavenging air ratio and the
scavenging efficiency, Instabilities have been observed
- the pressure losses in the ports both in loop and uniflow scavenging
- the influence of the stroke/bor~ model tests. These are still the
ratio, piston and cylinder head subject for investigation.
design on the scavenging
performance, A meathematical equation, which is
the design of the scavenging air determined by using the model test
and exhaust gas canal results, reproduces the measured
- flow velocities in the ports, in points nearly exact ly. Therefore it
the cylinder, especially prior to can be used in engine cycle simulation.
fuel injection,
- the time curves of scavenging
parameters, ACKNOWLEDGEMENT
the fresh air losses due to
mixing and short circuiting. The authors thank Sulzer Borthers
Limited, Winterhtur, Switzerland, for
In addition i t contributes to the their permission to use some of their
achievement of the following data and information in this paper.
objectives:
- Better port configuration, REFERENCES
- improving the fuel-air mixture
formation, 1. A. Schiff, "Vergleichende Untersu-
- minimization of scavenging air chungen der Spuelkennwerte von
consumption, Schiffs-Zweitakt-Dieselmotoren un-
- optimization of the effective ter besonderer Beachtung grosser
engine stroke. Hub/BohrungsverhiH tnisse" Fort-
schritt Berichte VDI No. 174,
VDI-Verlag, 1985.

CONCLUSIONS 2. D.S. Sanborn and W.M. Roeder,

"Single Cycle Simulation Simplifies
The number of similarity laws Scavenging Study", SAE Paper No.
describing the scavenging process was 850175, 1985.
reduced taking into account their
physical significations. The scavenge 3. D.S. Sanborn, G.P. Blair, R.C.
number and the density ratio of the Kenny and A.H. Kingsbury,
participating media, together with the "Experimental Assessment of
geometrical similarity, have enabled Scavenging Efficiency of Two-Stroke
to design a relatively simple and Cycle Engines", SAE Paper No.
800975, 1980.
 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

10 881264

4. N. Dedeoglu, "Scavenging Model Scavenging air ratio ASCI

Solves Problems in Gas Burning mixing factor m,
Engine", SAE. Paper No. 710579, transition point A* and
1971. revalorization factor a = 1 + A*

5. D.S. Sanborn, "Paper Powerplants A* is the scavenging air ratio at

Promote Performance Progress SAE II I which the transition from pure
Paper No. 750016, 1975. displacement scavenging to pure mixing
scavenging took place.

APPENDIX As shown in (4) the scavenging

process takes place in two principal
A. DEFINITIONS phases, namely: the pure displacement
phase and the pure mixing phase. The
Dimensionless scavenging mathematical equation for the ideal
parameters are used to determine mixing phase does' not yield the actual
the performance of scavenging tests test results unless the Eq. 4 is
in engines as well as in models. corrected by the factor a u.. It lI

depends on A* as already shown by many

Scavenging air volume - This is calculations. It is therefore called
related to the state of the air revalorization factor.
before the inlet ports.
The fresh air balance during
Scavenged cylinder volume - total scavenging process is given by:
cylinder volume at the end of the
scavenging process (inlet ports Change of volumetric purity factor
closed) . in cylinder is equal to fresh air
input through inlet ports minus fresh
Scavenging air ratio Asc - Ratio air losses through exhaust ports, both
of the scavenging air volume to the are based on scavenged cylinder volume.
scavenged cylinder volume.
du = a(dA - mudA)
Trapped air volume - Fresh air V ASC
volume in the cylinder at the end
of the scavenging process.
f
U du
1 - rna.
= a fdA (4)

Volumetric purity factor C1.v - A* A*

Ratio of the trapped air volume in
the cylinder to the scavenged The solution of the above equation is
cylinder volume.
1
Uv = (-1 -
A*) e- am (ASC - A*) (5)
Volumetric scavenging efficiency m m
~sc - Ratio of the volumetric Equations for different scavenging
purity factor Uv to the scavenging concepts (1, 4, 5) can be easily
air ratio ASC. determined by using Eq. 5 (Fig. 8).
The average flow velocity is Pure displacement scavenging - In this
determined by dividing the discharge hypothetical concept it is assumed
by half the cylinder cross section that the exhaust gases are displaced
area (for uniflow scavenging the full by fresh ai r wi thout mixing. In this
cross section area). case,

m 0, a = 1 + ASCI A* = Ase
B. MATHEMATICAL ILLUSTRATION OF THE
SCAVENGING PROCESS - In order to then
make the model test results avai lable
for engine cycle simulation, a Uv ASC for ASC < 1.0 (6)
mathematical equation will be
determined as follows: Uv 1.0 for ASC > 1.0
The following factors affecting the The volumetric purity factor Uv is
volumetric purity factor Uv are: equal to the delivered scavenging air
 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

881264 11

m = 1.0, a = 1.0, A* = 0, (7)

volume based on the scavenged cylinder
volume (scavenging air ratio).

Pure mixing scavenging - It is assumed Pure short circuiting scavenging - The

that every increment of the fresh air whole fresh air entering the cylinder
entering the cylinder will be will be completely lost by short
immediately and completely mixed with circuiting.
the whole of the cylinder content. The
mixing factor m and revalorization m = 0, a = 1, A* = 0,
factor a both equal 1. 0 and the
transition point is reduced to zero. (8)
Eq. 5 takes the following form
Table 3 shows calculated factors in
Eq. 5 for different scavenging methods.
(/.,
v
1.0 t------t-----~+_--------

0.5 +------,H-_~-___i'--

I
I
I
I
I
I
eq. 4

0.5 A~c 1.0

Fig. 8 Scavenging concepts

Table 3. Factors for Eq. 5

Scavenging method m a A" Remarks ""

Pure mixing scavenging 1 1 0 t1.vo = 0.632
hypothetical

Cross scavenging 0.965 1. 070 0.070 - 0.666

Loop scavenging· 1* 0.985 1.230 0.230 - 0.706

Loop scavenging 11* 0.970 1. 300 0.300 - 0.729
Uniflow scavenging 0.915 1. 690 0.690 - 0.843
Pure displacement scavo 0 l+Asc ASC 1. 000
hypothetical

") Loop scavenging I has conventional port configuration (1). Loop

scavenging II is developed by experiments in the aotual engine. It
is therefore an improved version of loop scavenging I.

"*) The values of the volumetric purity factor t1. v o are based on
scavenging air ratio = 1.0.
Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018
Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018
 Downloaded from SAE International by University of Edinburgh, Thursday, August 09, 2018

Positions and opinions advanced In this paper are those of Persons wishing to submit papers to be considered for
the author{s) and not necessarily those of SAE. The author presentation or pUblication through SAE should send the
is solely responsible for the content of the paper. A process manuscript or a 300 word abstract of a proposed manus~
Is available by which discussions wJ1l be printed with the cript to: Secretary, Engineering Activity Board, SAE.
paper If It Is pUblished In SAE Transactions. For permission
to publish this paper in full or in part, contact the SAE
Publications Division. Printed In U.S.A.