Instructions for use

Title Experimental Studies on Some Combustion Problems of Diesel Engines by the Method of Analysing IndicatorDiagrams

Author(s) Kuroiwa, Tamotsu

Citation Memoirs of the Faculty of Engineering, Hokkaido University, 10(3), 383-451

Issue Date 1957-09-30

Doc URL http://hdl.handle.net/2115/37804

Type bulletin (article)

File Information 10(3)_383-452.pdf

Hokkaido University Collection of Scholarly and Academic Papers : HUSCAP

Experimental Studies on Sorne Combustion Problems

of Diesel Engines by

the Method of Analysing Xndicator Diagrams

By

Tamotsu KUROIWA

(Reeeived June 30, 1957}

CONTENTS

Introduction ･･･r･.･...............･i････Experimental Apparatus and Fuels ......･･･････････ 1. Experimentalengine.........･.......... 2. Indicator.......................... ' 3. Fuels ･.....･･････.･････`････････The Method of Analysing Indicator Diagrams. . ･ . ･ ･ ･ . . ･ . ･

1. The case of an engine with single eombustion ebamber. . .

2. The case of an engine with pre-combustion chamber ･ ･ ･ ･

Experimenta! Results and Discussion . . . ･ ･ . . ･ . ･ ･ ･ ･ ･ ･ ･

I. Influence of Cetane Value of Fuel on Diesel Engine Performace

1. Preface ..............,........... 2. Performance tests on several types of eombustion ehamber･

(i) Deseription of experiments･･･････････････

(ii) Discussion o£ the experimental results . ･ ･ ･ ･ ･ ･ ･ ･

3. Cases when the timing of ignition is ehanged･.･...･･

(i) The ease of heavy load ･･･d････････････ (ii) Theeaseoflightload .･･････..･･･-･.･ 4. Conclusions.･.･.....................Il. Combustion Progress when Pilot Injection is Employed ･ ･ ･

1. Preface ･..･.････････････-････････ 2. Experimentalmethod ･......･･･････...･･ 3. Experimental results and discussions ............

4. Conclusions.........................III. Combustion and Performance of a Diesel Engine with

Pre.Combusti6n Chamber ･･････････････････ 1. ?refaee ....･･･...･･･････････････` 2. Testengineandexperimentalmethod .･･-･･････- 3. Experimental results and discussions .......･････ (i) Influenee of the timing of fuel injeetion . . . . ･ ･ ･ ･

.

.

.

.

.

-

.

.

.

.

.

.

.

.

.

-t

.

.

'

.

.

.

'

.

.

.

.

.

.

.

.

.

.

.

.

.

'

.

-

.

.

.

'

.

.

.

.

.

.

.

.

`

.

.

'

Page 384

384

384･ 385

388 388

388 389

393

393

393.

395

395

399

402

403

409

412

414

414

414

416

422

424

424

425

426

427

384 Tamotsu KuRolwA

4.

Closures

(A) Performanee cuvves (comparison to an engine of

injection type) ･･･......･･････ (B) General discussion o£ indieator diagrams. . . .

<ii) Effect of eetane value of a fuel........

(iii) Throttle valve and pintle. valve ........

(iv) Combustion progress in eaeh combution chamber

Conelusions .....................

-------------------i------- .

direct

i--

--- ---

---

.

.

427

430

435

441

444

447

451

Introduction

The most important thing in improving the thermal efficiency ofa Diesel engine is to make sure that eombustion is completed quickly.

In other words, the a£ter-burning should be reduced as far as possible.In Diesel engines, some combustion problems regarding the improve-ment of their performances ean be usually analysed as problems ofthe degree of this after-burning. The problems regarding the combu-stion of a Diesel engine are so complicated owing to the problems of

fuel injection and the mixing o£ air and fuel, that they can hardlybe analysed by means of only usual Eundamental researches. There-fore, this paper is intended to present some solutions, and deseriptions

whieh were derived from analysing the process of combustion as re-corded on indicator diagrams obtained under operating conditions ofa practical engine, eoncerning the following three subjects.

I. Infiuenee of Cetane Value of Fuel on Diesel Engine Per- formanee. II. Combustion Progress when a Pilot Injection is Employed.

III, Combustion and Performance of a Diesel Engine with Pre- combustion Chamber.

Experimental Apparatus and Fuels

1. Experimental Engipe.

Single cylinder test engine (Fig. 1),

･･ bore:110mm, stroke:140mm, rated horse power: 10BHP at 1400･---1500 r.p.m.

This engine is conveytible into engines with various combustionchambers by changing its cylinder head and piston.

Cylinder head £or direct injection combustion chamber: 4 valves, (2 masked inlets, 2 exhausts),

Experimental Studies on Some Combustion Problemsof Dicsel Engins

Cylinder head for pe-combustion ehamber:

XE-.gR)s'/･, '' " ' t- l , ny

" ･/ %fr'. ,I -1!o- - i ....i" L...[''.'l, ,1･, y.. ., O' o ,) ' i . .' @'.' '"'

' -- 'Q 'L .ttso'

t.h".Koli '''t×' '･･x ' x .">]" .... ･-----ff .:･ tt ttt tttttt t ' x.

-llit -tr'---'!-M 1't'Il,i

Fig. 1. Experimental Engine.

2. gndicator.

A piezo-eleetric indieator was employed.

graph was used £or the reeordings and a cathodealso used for observations, The structure of theshell type as shown in Fig.2. The

high temperature gas was made of a thin sheetand any influence of the eyclic change of gascylinder was avoided.

The change of temperature in the inner parttaining crystals, however, could not be avoided, '

was directly connected,with the inner part andsuMeient. Accordingly, the disadvantage ofof the indicator resulted. Therefore, the

2 valves.

x .tk'

.-, gelL. ' g'X"JS"i' ''' F ･--- I

iiiililllMdll`iil

G

.t

385

Piek-up for Piezo-electric

A electromagnetic oseillo-

ray oscillograph was

pick-up was of apressure diaphragm in contact with

of phosphor bronze, temperature in the

of the pick-up con-

smce the diaphragm its cooling was not changing the sensitivity

pressure under any given

Fig. Z.

Indicator.

386 Tamotsu KuRolwAoperating conditions was first measured by using a Farnborough in-dicator, and then the sensitivity of the piezoelectric indicator was

adjusted by comparing with the readings of the Farnborough indicator.

In the case of an engine with a pre-combustion chamber, pressure

indicators were equipped in both chambers, respectively. Since anaccurate value of pressure difference between the two eombustionchambers was required especially in this case, a device of adjusting

the sensitivities of indicators was employed so that they would always

coincide each other while the engine was operated.

Besides, the pressure difference between the two combustion charn-

bers was also recorded. As is shown in Fig. 3, the amplifiers of the

,S4 SCE7 2A9'

vtpeF-eut

3S,.2SoS(

20O

mr6etlil""iFpm3ovtablt.

20es 4o2SO"

sc M EvatF

e

rwv ts

ft

M/eRk, trs

Y.Sln

v>a

SC6

SlfytOk

tttaK

xt Xn

di SObbyf .ptx

th Xa

2eD.n

7a

x

g`eo-A"ag

piAOen

M3-ko7. /jEiKi

--TfO, -F

tio2e･

a,'

ereZ#v

E

M4i3oma

ielt `:la

xos

2gen

" Sa

"

X,2,6T>Q

if ea ;tr..Oeelltt

-CVbtX+ 4-frs}tat3spsa-2d;u.et61sttt･

Fig. 3. Diagram of Amplifier £or the Measurement of Pressure Difference.

two indicators were put in a cabinet, and the same electrical source

was used. A bridge circuit was arranged at the output ends of thetwo amplifiers so that the difference of the current at their output

would be obtained. At first, the amplifiers were adjusted so that the

readings of ammeter would be always zero, when the same change ofpressure was given at the pick-up. The pressure differenee during the

operation of an engine, converted into eurrent, was indicated･ on aBraun-tube, and the sensitivities of indicators were adjusted by chan-

ging the capacities o£ variable condensers set parallel with the pick-up,when ever the image on the tube was judged to be unreasonable.Furthermore, the coineidence of the sensitivities of indicators was

Experimental Studies on Some Combustion Problemsof Diesel Engines 387

k.--.･.'"-T-tt',""rT'"'ww''-P.t.tt..Peg-tionmsike ,.'r' ee,,,,ew.'ge.,･..,-.m,,E,.･･-ttg'ii-'-w'i'e'bu/!.,l.it,tl,ili.ill,tl/ttr/lll,r///ii.,k.,･:.,'tl/,i:tLt-･,E'r･i9?-'i'ii",43T

' '

-.kic"cttoedidy"'ma"-'-'

nfpte--cembevb'efichaptSer

'tt

'tttt

...=.tg./.r...'.'-U.'-.'....--:---v..,t...''tt"""---E'.-.･'

"''

't---"tttt.tt..t

.7.if-'...,Xtt.tt

･l-fi-t.,'-i;,i-l･,;S'l･ll'.=.t.t'

tttt

'''l-,'

.tt=.,

l/:'.

t.

tt/t

.tttt..-t-'s., -...ttr'

tt

tt-

,....., ,,

,.,letstt.=-･z/t' '

t. .un.

--tttt-..."'i/-,er.･=･t-

/t..

tt

'

'mt.-'t"#.-' t/

lptdic-tordisgrsms.ttttt,.i,-:i11'11E'･.･,,,k,･!Yt'.i t,- .tt./lt.-.-,-kt...t...tt/ s'.t. .t

./...tttttfftt ttttt t

:ttt...t-t/

4-.oFasu'ncembesh"enchttnbeF./.ttttt/tt'..../.,-'-J'.-r･'its'･tr',･{,l"''-:-/I//.-'.:',t'//'/./'.rJ','2,/',X.'nd-..' ';i.tt.;t:gttnv,.-t･,,1,ti-.x･},:A./,IL.t......tt.mm..

'

/.=t/...- tAt

tt.

i' '''

'"""f

tttt ttttt

-- './ ･-- ."･-

'tt/'

4

' 'l'.

H' i'

ifibethch"mbe" .nd.ttt'tt;t'tun"ttt't'

t.t

'

' ' '' .:H

l4.tt/tt

/tt ''

-J-i-(t;tt't･'>"//"t:'fa,..Tt/..

U;teFFuel;nlect;en'

ll,..:

valvesp;fiatett

utww-ny 't. ' '

-Ppsi£Fonpterk" ' -2e '20' gg-･ lb tt,

' '

t 't ttttt t.wtt'di

'2 '--.-uaL ..-'.' -/,u,l･L/-..fittt--'

imu'im..EL'u'T"J,.- fi.tt" .-.-. 'HWtg'- -.J--. tt

Fig 4. Indicator Diagrams of Experiment (IX)-D in Study Ill.

sometimes assured by checking if the image of pressure differenceon the Braun tube deviated from a straight line or not after the engine

was warmed up, since the pressure difference between the two cham-bers must be approximately zero when the engine is operated at about

200r.p.m. An example of the oscillogram of an engine with pre-combusti(m ehambers is shown in Fig. 4.

TabteI Ptepgrn" of Fuels

lUELS KopasTn N4 N5 N3hCemmelcSel

LightO,1

SecoF.dar,f

ReFerence

Fvet

S.R.F.60

Secondarv

ReFerence

FueS

SR.F.30

Ze:.i'r,L'isi4kO.763 O･90t O･881 O-938 O･846 O･800 O,855

v;sces;ty30'C

eR,W.50C

3l-6s

29･8s

31･O

･2go

30.8

29･O

33･O

-

L- 32-O

-

33･O

-Freeiing

Pel-t"c -75 -l6,5 -t7.5 -18 - -23 -31

Fl"IhpoTntac 71,5 76 75 79 - 68 525

AnilinePeirrtte 86･3 38･8 47･2 28 -86.1- 60.1

Res;dudl

Cerboti%treee O･03 O･O'3 - - - '

SutpHet'

Corvtent%trece O･36 O･30 O･40 - O･05 O-05

Hlghet

ColorlFicVelue

Hi%tl35014･7

l03509･5

105601O-5

98009･2

l0200tHu)

l1OOOCI4}

l0700r28･2

Irt;tla1

Bo"tngFre'inte(:

DtvPoint℃

lndexof

Bci'lingPdint℃

190

324

242

190

32l

239

lg2

328

240

189,

320

237

t39

302

220

l76

3oa

248

159

295

Z46CetoneNumber 82' 42 28 ,60 30

388 Tamotsu KuROIwA 3. Fuels.

The kinds of fuel used in the present experiments are shown inTable 1, The fuels N5, N4, N3 are blended fuels of "Kogasin" andnaphthalene oil in tar oil, refined by washing with sulphurie acid, water

and caustie soda. The commereial light oil shown in Table 1 was usedonly in Experiment (I-2), and other kinds of commercial light oil were

used in the other experiments.

The Method of Analysing Indicator Diagrams

1. The case of an engine with single combustion chamber.

The process of obtaining the combustion progress was based upon

the method proposed in "Thermodynamik der Verbrennungskraft-machinen" written by Prof. H. List. Besides the indicator diagrams,

the volume of suetion air was measuxed, and it was assumed that thevolume of residual gas was 112r of the volume of suction air, (taking

rasthecompressionratio). ･ The energy equation in respect to 1 Moi of working medium ina compression stroke is as follows, between a certain point before the

beginning of combustion and an arbitrary point during the combustion

period:

((3a'uat -2eo) + ALo-a + Qiv,-. = Hec

where,

H. == Heat generated by eombustion ALo.. = External work done 6. = Change of number of Mols before and after the combustion u. = Internal energy of woking medium in combustion period u, =Internalenergyofworkingmediumbeforecombustion Qw,-. = CoOling loss.

Each of these terms can be obtained in the following manner from

the indicator diagram anct the weight of working medium: known value measured value

-Z] po,G --- T,=.Po'Vl) su, R. G alr v. pa,Gqa=G+G!fa-'T"=RP,:VGS.=sYii21I,II.f.GS"at

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 389

where,

G = Weight of working medium in eompression period G,. = Weight of working medium in combustion period

Gf. == Weight o£ injected fuel.

B. is unknown until the computation is completed. There£ore, thecases o£ assuming air and complete combustion gas as the worl<ingmedium were calculated respeetively at first, and the values of 6. were

interpolated.

That is to say, when it is assumed that the working gas is air,

T..==IZgV" ----> u.. aii'RairG airFurther, when it is assumed that the working medium is complete

T. ..LZ2.gVct ->u. gaS6R.i.G gas 6 = Change of number of Mols before and after the combustion of al} injected fuel.

External work is

L,2.==£hh,=z[PiI;Ii?2Vk] or ;E]-ZL/EPL'(v.,-x)

n = Index of polytropie change between 1 and 2 obtained from indieator diagram,

The numerical computations of the above integration were per-formed at the interval of 10tw20 in crank angle for the main eom-bustion period, and at the interval of 20-50 for the after-burningperiod.

Since the estimation of cooling ioss is diMcult, it was not tal<en

into consideration in the present paper. Accordingly, the amount of

cooling loss is not included in the generated heat calculated.

2. The case of an engine with pre-eombustion chamber.

When the combustion chamber is divided into two ehambers con-nected by narrow passages, there exsists a considerable pressure dif-

ference between the two chambers. Accordingly, the progress ofcombustion in each combustion chamber can be obtained from the

respective indicator diagrams in the same manner as the case o£ single

390 Tamotsu KuROIwAcombustion chamber, when the amount of working medium movingbetween the two chambers is obtained with the aid of this pressuredifference.

The state (temperature) o£ the working medium in both chambersbefore the beginning of combustion and the degree of transformation

o£ the kinetic energy of flow into heat must first be obtained £or the

computations of the each progress o£ combustion. The former maybe assumed as wiil be de scribed below, and the latter was assumedon the basis that the energy of flow during the combustion periodchanges into heat at the instant of entrance. Under these assumptions,

if the cooling loss could be le£t out of coitsideration, one can obtainthe following energy equatioit between two points of combustion:

pre-eombustion ehamber;

pux>Pv GviZew+liG{i-L,'iz +Hv =Gv'Uv 11 1"2 l-2 22 Hv,-, = (Gv,'Zev,-Gv,'Uv,) - tiGi-2'ix,u,

1)z<1)v Gv,'Zev,-IIGi-L''iv,-,+"Elw,m,=Gv,'Uv, ,

Uv,", := (Gv,'Uv,-Gv,'Uv,)+AGi-e'iv,",

main combustion ehamber;

Pz>Pv Gx,'Urt,-AGi-2'ix,-,mx4LJ-!+Hz,-,==G{x,'Ux,

Ht,r, = (Gg,'Ux,- G{z,'Ug,) + dGi-2'ix,rr,+ ALi-L7

Px<puv Gx,'Uz,+AG!i-L''iv,u,-ALi-2+EInt,-,=Gz,'Ux,

He,-, = (G2,'Ux,-Gx,'ttx,)-dGi-2'iv,-,+ ALi-2

where,

sufix z == Notation of main combustion chamber suflix v =- Notation of pre-combustion chamber

p = Pressure in eombustion chamber ze == Internal energy

i = Enthalpy G == Weight of working medium in each combustion chamber dG]-2 = Weight of working medium whieh was moved between two points 1 and 2 ALi-2= External work done by piston Hi-2= Heat generated by combustion.

(i) Assumption of temperature of working medium before the beginning of combustion in both combustion chambers.

/

Experimental Studies on Some Combustion Problerns of Diesel Engines 39i

The change of state of working medium in main combustionchamber can be expressed as puv"--const. in compression stroke. Inthe pre-eombustion chamber, however, this equation can not be exactly

applied because the kinetic energy transformed into heat by friction

is irreversible. However, the amount o£ heat geneyated rnay be verysmall, and the change of state in pre-combustion chamber is assumed

to be reversible. Then,

?t'-] p.vlt'==const., T.=T.,(pP.Z)n'

o nr'-1 p.vge,"=eonst., Tw=Tv,(pP.",)""

If it is assumed that the conditions at the beginning of compression are

Pxe:==Pv,, T.,=T.,, and that nt= nn= n,

Ti. =(-£t' )T':` ea

From the eondition that the sum of the quantities of the working me-

dium in both ehambers is constant, one has

puzK+ILpvV} G:==PxX+P.".. a. RTzRTv RTzAccordingly

p.x+- pu.vL

a Tx =: RG

T. == aT.

The mean value of the index n, between the two points at thebeginning of compression and just at the beginning of combustion,was found from the mean compression-pressure line (p,.::= Pz}llPIZ')

obtained on the indicator diagrams o£ both combustion chambers. Thevalue of n was 1.32rw1.33 when the beginniug point of combustion was

50-100 before the top dead eenter; n=1.325 was used in a!l cases ofcomputatiops.

(ii) Steps of computation.

The starting point of computation was selected at a certain point

392 Tamotsu KuRolwAbefore the beginning of combustion and the state at that point wasfust assumed. Regarding the second point to be determined, the

temperature at the side o£ higher pressure was assumed and dG.,-, ortaG.,-, was found. The deduction of riG.,-, or tiG., ll, from G., or G., then

gave G., or G.,, and T., or T,,, can be obtained with the aid of the

equation of perfect gas. When this'result does not coincide with the

initial assumption, the same process must be repeated. However, twotimes of trials were suflicient to obtain accuracy within 10C.

The details of computation are as follows for the case of p.>p.:

knownquantities, calculatedquantities.

(i' liiil･ ;'Yr' gil:ii.tl,- a"d (ZGa),.

"1 11 pa)ZZSPV2 )

(2) TV.i2'asllUmedi -(/.TG")

k2 riGxi-2 == Lll7 (( ddGa ), + ( alalGa ).] da

s" Gz, = G!z,-ziG!x,-,, G!v, = Gv,+AG!z,--,

ss T.,-..Z.ft2..I-tti , T.,-t'2.Vdj,

(iii) Quantity flowing through the connecting passage.

When p,>p. and -iltl'>(k?1)Et'ii,

d,G. - ",2/ ny..,/2gk-..il((g/');'-(-g/)Ei';iI

' .puttlng

gi. vlZ};;/2gk4i:-==c

one has

gG. -c,.i･ T/(e/);'-(x/)lj'i`

Then,

This computation( Pp:' ) and f(

AGz,-2'

points on the

of the

tations were under-taken at intervals of

period

1. Preface.

Duringmanfacturedremarkable

the maximumthis synthetic

of experlments

eauses.

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 393

f == Sectional area of the connecting passage =O.OO0431 m2

n =:: r.p.m. of the engine =:1390 (constant)

y = CoeMcient o£ discharge = O.92 R =: Gas constant of the working medium :::BR.i,

--1.039×29,27=30.41mkg!OCkg £orR==1,8(£ullload), 1.021×29.27=29.83mkglOCkg £orR==3.3(4/10load), 6 = volume change before and after the combustion

==1,039 for2=1.8, 1.021 forR==3.3 k = Specific heat x'atio

:::1.304t-m-1.331 at t--1500-10000C for R==1.8

k =: 1.32 (assumed for all cases o£ full load) = 1.314"-thb･1.342 at t==1500-q-10000C for A:=3.3

k =: 1.33 (assumed for all cases of 4/10 load)

eldGa =:: C' tiPE f(ipu,iY')

C=O.7756 for R=1.8(fullload) =O,7741 forR==3.3(4/10load)

can be performed easily, if the relation between Zi ) is previously prepared.

the quantity fiowing at the interval between ewo adjacent indicator diagram, was deterinined by the arithmetic mean

instantaneous fiow at these two points, The numerical eompu-

10rw20 for the main combustionand of 20-50 for the after-burning period.

Experimental Results and Discussion

I. Influence of Cetane Value of Fuel on Diesel Engine Performancee

World War II, synthetic fuel of high cetane value was by the Hokkaido Artificial Gasoline Company. Since a increase in fuel consumption and a noticeable decrease in

' smokeless-powerunexpectedlyresultedfromtheuseo£ fuel in a Diesel engine of direct injection type, a series

' wereundertakenforthepurposeofdeterminingtheir

394 Tamotsu KuRolwA

The problem of cetane value in a Diesel engine began to be dis-cussed after that of the oetane va}ue in a gasoline engine. The in-

crease of octane value o£ fuel was inevitable in a gasoline engine, andan improvement of performance in a Diesel engine was also expected

by the increase o£ cetane value. Even though a fuel of high cetanevalue was realized later and many experimental results were reported,

definite conclusion about the influence upon engine performance hasnot been reached. According to the experimental results in the present paper, it may

be said that these various diseussions originated from the differences

among the types of test engine and the operating conditions-viz., type

lig･

g･

) scoog-!

.p-

gs8no=x:di' seoo

2eoo

S[NGLE CYLINDER ENGINE･Hox.1 4omm

FuelPump Bosch .(7mrn)

Fue].Valve, OpenNozzle

(2xO･23rnm)-ORERATtNe CONDITtONS -i385 r･p,mi (Censtant)

Tirn;fig oF Fuel Tniection wds

'edlusted se thnt the

igm[tien' is lqceted at

t'

I N.

v

Cembustlen Chember

(k =: [4･4)ss(D

2etJ30beFereT.D･C･

N

NN

v

l

x

Fues, X¥h･x

-x- Kocasin

-o-･N5.A- N4-m.- {.N5,ll',gg%%

--A- Ixglil,sbl%

Lommerciaf--o- light eil-b- lkg:Igs,z

-e- N3

f

z

es423SS5

B3

28 "Cetane

2･

R

Ndmber

t

"es

g

i･ttta

Act.5Z

s-

;.26

1.st,54 b

m

t,62

, l".e6s

<zo.1 tos.

e l･St

e

aee7t.6t

Breke Mean EFFective Ptessure

t.oz ･

t･9

l･41

5 7%o2468 ta 'i2 t4 Breke Herse Power

Fig. 5. Rate of Fuel Consumption-Brake Horse Power Curves for Experiment (I).

Experimental Studies on Some Combustion Problems of Diesel Engines 395

of combustion ehamber, fuel injection system, timing of fuel injeetion,

compression ratio, r,p.m., and load ete,

2. Performamce tests on several types of cornbustion charnber.

(i) Description of experiments.

The types of combustion chambey employed in the present experi-ments and the experimental results are shown in Figs. 5-a-9-Expriment

(I), (II), (III), (IV) and (V). The engine revolutions were kept at 1400r.p.m., and the timing of fuel injection was adjusted so that the

ignition point was a}ways loeated at 2'-30 before the T.D.C.

The case of a combustion chamber of horizontal injeetion is shown

in Fig. 5-Experiment (I), and an open nozzle was employed for the

fuel valve. As will be seen in this figure, it is a remarkable fact

ta･ 4ooo

g･

ii

si

lg

.5

el8tg

lt 5ooe

:&.

2000

SINGLE CYUNDFR ENGINE

Bore liDrnm Stroke,, 140mm 'FueSPump Be!ch Fuelvetve AutEm9eT/c"iv).1.. x }IX/l/

(4xO･28mrn);... z'opERATfNecoNDrrIoNs ','- :3ssF･p･m･(censtent) 22

T;rnlngoFFvetlnlectlonwas .Cf; ;t:.

1

Combust[on

{n. --

Chqmbef

l4･4)

1

,

1Nl

ediusted se LHat tke

tgnitlen is located et

20-,3"beFereT,D,C.

l

NN

N

as

d

F"eTs

.xr- Kogesin

-e- NS..A-

{xg[iiggZ

,..a-f".giiigg%,

-O-' ESI2Ts･ltciel

-e--NS Cetene

i

N

sf,!K

8Z4858

3S

z81

z

N

2.16

R

ail: ;s

Atii,`KS i!,'

e

s

.

l.85

1

--

× t 1- mediumt-25 smoke

light smok5

Number BrakeMeenEFFectTvePressure

,

tfeeesrneke

6, rr

%2

oa e･6s lo lz l+ Bseke Horse'Powet

Fig. 6. Rate of Fuel Consumption-Brake Horse Power Curves for Experiment (II).

3aoo

zooo

396 Tainotsu KuRolwAthat the rate of fuel eonsumption increased very much and the maxi-

mum smokeless-power decreased extreme}y, when a fuel of high cetanevalue was used. This fact must be eaused by the after-burning, In

other words, the thermal efliciency becomes poor because o£ a violent

a£ter-burning anct a larger amount of fuel is required for the sameoutput so that smoke appears earlier. Since one of the reasons forsuch a large difference in engine performance was understood to bein the use of an open nozzle and a long period of fuel injection, andsince the pulvenization of fuel was not favorable, the open nozzle was

replaced by an automatie valve. The results obtained are shown inFig. 6--rExperiment (II). The difference of performance du,e to the

variation of eetane value is not so !arge as was found in the previous

experiment. However, this figure also shows that the fuel of high

f 4ooo

:t

3･.

.F

.9-

?l･

8- 50codi

LO

eat

2oeo

SINGLE CYLINDFR ENGINE

Bore llOmm Streke l40mm FuelPump Bosch(9mm) FuelValve AutometicYalve (4 x O･28rnm)

OPERATINe CONDITIONS, l385t･p･m･(Constent)

rirningoFFu.eliniectTonls

e6iusted so thet the

lgnitie" is [oceted at

20. 3ebeFote T.D,C.

conLustion cHambeF

(A-I44)

1

1

1i

1

N

N.' - JNSx"

o

s

z

N

N

Fuels, Nx

N-X'- Kegas;n

-as- 5RF6o-(g'k:gg,i,

.a-lg,RFggiig

Cornmerc:al-O- light di[-e- sRFSO

lt

z

7.

8ZN ,:g

-a-IgR,F`.'::iiio

ss

sc ,

o

Fig.

Cetene Number

4

2.[6

ss

>SF //f,)ltrtl/ 1

t.･..t6]

s . X><l/.ex '

i

t/

23 ''6 8 lo X2 lif- Breke Horse Pow'er

Rate of Fuel Consumption-Brake Horse PowerCurves for Experiment (III).

rnedium

X"iii=LV sg .

smok

s

1,e6

ectivePfessure

46-

t,so

tTeeeEmeke.

6

tJ3

Llightsmeke'

'7ty%/rf

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 397

cetane value causes an increase of fuel consumption and a deerease ofoutput. Besides, the fact, as one ean see in this figure, that the fuel

consumption increases in the range of Iight Ioad, when a fuel of N3or one eontaining a considerable amount of N3 is used, was caused by

a large amount of tar oil content in the fuel. The combustion of tar

oi] is understood to be poor when the temperature of combustion be-

comes lower than a certain value. This kind of problem is entirelyindependent of cetane value. Accordingly, a similar experimene wasrepeated by using seeondary reference fuels for the measurement ofcetane va]ue, under the same conditions as in the case of Experiment

(II). TheirresultareshowninFig.7-Experiment(III). Thetendencyin these results is quite similar to the one shown in Figs. 5 and 6,

f 4eoo

4m='

-'eS

si

,E

,fi-

e98z 5ecol"o

#ct

zeca

SINGLE CYLINDER ENGINE

kore Homm SFt.'.7tCl),.p, g.4,.Ohmm .

(9mm) . N FuelValve Autpmdt;cVelve

(4.o28mm)ZZ

OPERATING CONDITIONs

Com(6.u,tstio.a lCshrl13e' [385r･p･m,(ce"staat)

"

zi

p Ti"iFngoFFueliniectientvasXss.Igdni:,tS.,to"ndileot,heatt.tdheet'

2n: 3" berere T,DIC.

Z

NN

N

xx

x

N

t.10m,,t･

#`sNNN

tsN '11K7f N.tl.

s 1N -N it

1

/

Fuels

-X- Kegasio

-e- N5

-e-iNN531'.15sOo%%

-.Hl,N,5 :l27Ss%%

Comptercial-o-- tighL eil

-e- N3

t

Cetengtstumber

.82 ･

4g

58

33

28

z

Z2Z

A-l,1

2.ie

t.s9. s

la,

t,69

ereke Meen Ercective Pressure

8

67 %z

O a 4･ ' B, IO l2' t4- Brake Herse Pawer

Fig. 8. Rate of Fuel Consumption-Brake Horse Power Curves for Experiment (IV).

neec

Seeo

zcas

398 Tamotsu KuROIWA

Furthermore, the particular aspect of poor eombustion was not noticed

for fuels of low cetane value even in the range of light load, and the

fuel consumption curves are regular. As will be seen in the figure,

the infiuence of cetane value can not be noticed up to the load ofabout 112. When the load is larger than this value, the after-burning

tends to be considerable as the cetane value becomes higher, and there-

fore the fuel consumption increases. As a Tesult, smoke appears earlier

in exhaust and the maximum smokeless-power lowers. The values ofthe excess air factor R are also indicated in this figure. Judging from

this figure, one may see that smoke started at around R=1.5-1.6 inthis case.

Fig. 8-Experiment (IV) shows the case of a Saurer combustionehamber, the compression ratio of which is 18.2. In this case, the same

tg･

:-

ii'

ti

.e-

fi

'Ea

th=

c6gxgl,

4ooo

Soeo

2eoo

SINeLE CYLINDER ENGINE

Bore llOrnm Stroke l40mm FuerPtimp Bosch (9mm) FuelValve AutemeticValve,

O(LxO･28mm)

OPERATINe CONDITIONS l385r･p･m･(CongtanD

TIming eF Fuel iniection is

ms

Combustlon Chambet

(n =l4･3)

1l

(lz)

/v "-s

XN

ediustedFothattheign(ionislecatedat

2"- 3.'beFore T,D;C･

/,

Fue]s

MX- Kegasin

-Aot s.R.K

-o-- N5

-a- N4･ CornmeFeial-'" O" 11ghtoil

-- e'--- 5,R.E

-a- N3

i

6Z

6048

4a

3028,

2

x

-. -- -- -- s.N( IX>Ki-..

Xy-t-"-C

! l

N

+ -･/t

vt Cetane Number

s

t･8s

X･30

i･s7

/-es

BrakeMsanEFFectrvePressure

45

f･7a,

xt･2a

I･5t

6

/･08

tv9

7% O 2･47- 6･'5 /O IZ 14 ".L U-ra Pn-,Re

Fig. 9. Rate of Fuel Consumption-Brake Horse Power Curves for Experiment (V).

ExperimentalStudiesonSomeCombsuticnProblemsofDieselEngines 399

kind of influenee of cetane value noted in the previous cases was hardly

seen and it ean be said that there is no effect of cetane value on fuel

consumption curves. However, when the compressionratio was de-creased to 14.3, as is shown in Fig. 9-Experiment (V), the'infiuence

of cetane value became noticeable once again,

(ii) Discussion of the experimental results.

By the experimental results described above, the increase of cetane

value lowers the performance of an engine when the ignition point is

adjusted to be at 20-30 before theT.D,C. which point is usually aecepted

to be proper. For the purpose ofclarifying this matter, data on the

periods of fue} injection employed £orthe loads at around the niinimum fuel

consumption rates are a}1 collated in

Fig. 10. The periods of fuel injection

for automatic valves and the one open

nozzle were measured on the basis of

the movement of the valve spindlesand on the oscillogram of pr･essuretaken near the nozzle, respectively.

First o£ a]1, when the open nozzleis employed, as is indicated by (I) (Fig.

5), the ignition ]ag is extremely large

perhaps because the pulverization wasnot goocl, and furthez', variation of

difference o£ cetane value isemployed in Experiments (II), (III),

the compression ratio was r=14.4 or

lags become almost the same.

The ignition Iag became smaller in

by the change of cetane value

From the comparison of theeonsumption rate explained aboveinjection, the following relations can

rate of fuel consumption for higher

of the ignition lag, It results that

ap 8 g 3o

s .g 2o

-co

g io

%. o

v io

2"O･2o-- 3o va jo 6o 70 SO 9o

Cetane Number eF Fuel

Fig. 10. Timing of Fuel Injection

for Eaeh Experiment at the ?oint of Minimum Fuel Con- sumption Rate. a. Beginning I'oint of Fuel Injec'tion. b. EndPointofFuelInjeetion. c. Ignition Point.

ignition la.cr caused by the

An automatic valve wasand (V). In (II), (III) and (V)

so and the ignition

(IV) is the case when the

ease, and its variation caused

extremely small.

of cetane value on the fuelthe change of period of fuelderived. The reason for high value is in the shorteningcut-off point of fuel injection

1

't

E v a

rnm rrv ･c

tt

I Mv'

'

the also quite large.

(I V) approximately Experimentcompression ratio was inereased to 18.2 in a Saurer combustion chamber.

this

became infiuence and be cetane the

400 Tamotsu KuRolwA

becomes late because o£ the late beginning of injection, and the mostpart of the fuel is injeeted after the ignition point. First of all, in

the ease when the eompression ratio is the same as is shown in (I),(II), (III) and (V) (Figs. 5t-7 and 9), the fuel inj'ection is completed

befoye the T.D.C. when the cetane number of fuel is 30-35 and themost part of fuel is injeeted after the T.D.C, in the ease of Kogasin

fuel. This kind of late cut-off of fuel naturally beeomes one of the

reasons o£ lowering the efficiency. The rather large difference of fuelconsumption rate in this experiment seems not to be attributed onlyto these reasons. One can naturally conclude that there is a considerably

Iarge differenee in the after-burning after the cut-off of fuel itself. In

the case of a fuel of low cetane value, the vaporization, the dispersion

and mixing of fuel into air before the ignition must be considerably

well completed, since the ignition lag is large and the most part offuel is injected before the ignition, and the other part of fuel would

also burn eompletely rather quick}y. In the case of a fuel o£ highcetane value, however, one can understand that the after-burningbecomes severe, since the rnost part of the fuel is injected afterignition and its combustion would be disturbed by the combustion gas

burned previously. The temperaeure o£ combustion gas is high and thattemperature promotes the combustion, but the most important thing forthe combustion of a Diesel engine is a good contaet of fuel with air.

In Experiment (IV) a Saurer combustion chamber of 18.2 in com-pression ratio was used. In this case, the difference of ignition lag 'due to the difference of eetane value decreases extremely, and thefuel whieh will be injected after ignition ean make a good eontactwith air, since the combustion chamber is a type which promotes high

turbulence. As the result, the aftey-burning could be understood tobe almost the same for all fuels; any considerable difference in fuel

consumption rate was not detected. The difference in fuel consumption

rate was so s}ight that it coutld be attributed to the diffeTence of the

cut-off o£ fuel injeetion.

In order to elarify more preeisely the reasons discussed above,

the comparison between SRF 30 and Kogasin 82 shown in Experiment

(III) (Fig. 7) will be explained as an example. The variations o£ themaximum pressure p3, excess air factor R', theoretical thermal effieiency

vth, and indicated thermal eMcieney v, graphed against the output are

shown in Fig. 11, p, and R' are experimental values, and rp,h was ca!-

culated with p3 and R' as a Sabathe cycle with the initial conditions

'

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 401

of p]=O,95 at. and Ti==350K, The %.curve of Ti is the value eomputed from .xthe fuel eonsutmption with the friction ,, ts.CM,-,

poweFrirosft5ofHP.i3CIhU.aii,Y.;egaSoUf"evd,,'fo, ':-6o ,,/-h.':

Kogasin82islowerthanthatofSRF ±"'"30, sinee p3 and R' are both smaller. fHowever,'thereasonofloweringR'in t.th' ･t.hgz c,as,e,gz,`.h,ati.h.e ,`,he,r,m. 1･i.g,ff.igiknzy, ･/S･

d3i%e.",-bU.r,l",gt,,i}kd,lll06 W.h,? .Tt6･iS £,Oig ,, fo'

possessedbySRF30fOrthePUrPOSe BrakeHorsepowerof obtaining only the effeet of the Fig. 11. Comparison between themaximum pressure, the results become Fueis of S.R.E 30 andasshownbytheehainline･Thedi£p Kogasins2. 'ference between this chain line andthe curve of rpth for SRF30 can be understood to represent the magni-

tude o£ deereasing in thermal eMciency due to the difference of themaximum pressure. When the difference between these two curvesis compared with the difference in rp,, one can see that about 113 ofthe differenee in fuel consumption rate originates in the difference of

the maximum pressu,re due to the variation o£ cetane value, and thatthe other 213 comes from the degree of aftef-burning.

Next, in order to ascereain the degree o£ a£ter-burning, the progress

jRsl?th

l kegQbjfS-JtisE

?6.o

Eo

4,e

3.0

ze

,,e

lNs.[ e℃e

isng.

?i

ee

S5

50

-s,

tto

j5Kl

1

koges'ifRr'Je N

toee･10

r,e 51F.

KogasinLe;i'S#':j2RCp,--)2d)

-".tc

gotr-'

02- SIOt2,i416

XSItm2

vo

60

se.

40

]o

2o PeriodoF Fuel lniection

S.RE 30l

Kog6sin sb

9 5RNF,3.0",61 K`%cA,/Mol,

A=T'65 3oeo

K.g:gins2 2eOo Ne=1O･'g5 A= 1･-5

1000(tombustlonRate ,

-"o -3o -2o -lo o lo 2e ]o "o so co Crank Angle in degtees

Fig. IZ. Comparison of Indieater Diagram and that of Combustion Rate for the Fuels of S.R.F. 30 and Kogasin 82.

402 Tamotsu KuRolwA

of eombustion was ealeu]ated on the basis of indieator diagrams. Fig.

12 shows also the cases of SRF 30 and Kogasin 82 in Experiment (III),

One can clearly see the reason foT decrease in thermal eMcieney when

a fuel of high cetane value was used. In the case of SRF 30, the most

part o£ fuel burns at the time o£ initial rapid combustion and the otherpart also finishes burning quiekly. On the other hand, when Kogasin82 is used, the initial eombustion is quite slow and the period of after-

burning elongates. Consequently, also the whole period of combustionfor Kogasin 82 beeomes long'er.

3. Cases when the timing of ignition is changed.

As the timing of £uel injeetion £or a Diesel engine is eommonlyregulated so that. the ignition starts at about 20-qlt-3e before the T.D.C.,

the experiments described in the preceding ehapter were per£ormedwith the ignition point fixed there. However, this point of timing is

too late for a fuel o£ high eetane value as will be seen on the indicatordiagyam and the line of eombustion rate shown in Fig. 12. Therefore,it was undertaken to find the timing of fuel injec'tion for the minimum

fuel eonsumption rate by changing the ignition point. O£ course, ifthe timing of fuel injection is advaneed, both the maximum pressure

and the rate of pressure rise are increased even for a fuel o£ highcetane value, and the anti-knoek property must be sacrificed to some

extent, So, it is practically rather diMcult to determine a proper

timing of £uel injection.

,"k/.. .zva/va zuzi -!Z

Thermo'-Plug･ ... . . 'gg".k-:.k,OwwISS'`-.'g.s. -' " hL- '

Appa"et"sForiilhiiiiiiiliililiieil,iiil,l,..N"..t.Fp..(.t{/i/1,.Oi,Lto) gss)x:ii}{pick-"poFt"dic"to'

.fi- tov

oF lniected Fuel

va z'

-

,

.)

.. 2..f

LXXXL

x'"

SXi, ..tt:. )/ .=;..-..)Nes･--

" LN xx･,

1" IN VN ,

Combustion Chamber oF

Direct Iaieetion Type

h = l4.4

Fuel Pump

Bosch 9mmFuel Velve

Autorr,etie Valve

4･ x O･28mm

R=200 fit.

Fig. 13. Details of Combustion Charnber, Thermo-PIug and Device of Measuring Penetration of Fuel.

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 403

The experimental engine and the experimental methods were sameas those employed in the preceding chapter; the experiment was under-

taken only for the combustion chamber used in Experiment (III) des-

cribed in the previous chapter. Details of the eombustion chamberare shown in Fig. 13. The temperature was measured by the thermo-plug inserted through the horizontal hoie made for the open nozzle,sinee a vertical automatic injection valve was employed. The records

made by this thermo-plug may represene the temperature of thecylinder wall.

(i) The case of heavy load.

The performance of the engine when the timing of fue} injeetionis changed for a constant load of 10HP at 1400r.p.m. is shown inFig. 14. The righO and left-side of this figure are the performaneesagainst the timing of ignition and the beginning of injeetion,respectiveiy.

2400

23ea

?2ee

?loe

?odo o seoq 480 460 44e 4?O -,liCO 2o %1.A

15 TO s o Craak Angte in degrees

Fig. 14. Performanee Curves, Fullload of 10BHP at 1400r.p.m. (Period of Fuel Injection: 130.v160 C. A.)

First of all, one may see that the minimum fuel consumption rateis found for considerabiy earlier timing of ignition with an inerease

or" cetane value. At the same time, the earlier timing of fuel injection

trce}･BiHph

(P

Load't/I,10SHPerTodeFPuet

atf4aoiniectiefl13

--t6'cA)

rp,m.tq4qt2.q

txe,M:eec

24eo

23eevs

Fuel

Beg;finingPointaFFuet

Censumptien

IniecE;en

Rate

･vatgn;t;enPo;nt

ses

ti-?"v

r

4le$TtoA"e

2200. ･=g "e

5.

F"1EoS tte"tS

Ib'S/-

/.u

t ee L;hNO'su

2100

.L3

trSRf.e

. --2000･3e -]o -to

FxhaustTetnporatureo

eCA.

. IgniticnLag "e"t

ge5X..

"25. "o"y

c)'

bte:N

20r, S.F e viA. .

N 4S ---- .

'

ts5

ts xt-'ghtOaA "g

,F7 XT--.le ]Kdiftci-e2 '-

sL

Kef4s`:

.Cetafle

1se

guNumberlojse

itS"･,i"'

ss.s

Mett,

Pre'ssure

ReteoC.R'sse

(gtPe)ti?eMj

5,R.F

SRFno

4}

10SOD

100SO

"1-"eSer

Lie"O:IAsS looooo

Lt'

7htottA'"j'

pf

t-j5 -jo -Z5 -ZO-15 -1O -5 05

'

404 Tamotsu KuRolwA L

is desirable. Even though the differenee in fuel consumption is quite

large when the timing of ignition is at 20--30 be£ore the T.D.C. as inthe cases previously described, that differenee decreases with theadvaneement of ignition point. Even at the points of minimum fuelconsumption rate, the tendeney of showing a higher fuel consumptionstill remaines with an inerease of cetane number,

In order to find a clue for the explanation of these facts, it was

undertaken to find the combustion progress from indieator diagrams.

Some of the examples are shown in Figs. 15, 16, 17 and 18. It must

be noted here that the part o£ cooling loss is not taken into consider-ation. The constant-volume degree of combustion was also obtained

1/T. Ie Sjtr-, 14 f.td pm r.t+/HosaNs2.fJ, ({?)!tt,ts1,cs ',v ['ade)..,,,blcA . E"g/dftael [is/ttLotd

xiJ' Ig': ,,ll,, '" lt3 tOF;HePoTF. Str-i eli6Gs iutoatpR i: .:;oooo -1 ss tt sb n i: :"'fi'v/e-"i i- iS :O M ]d j 40 2t ie k2! /i ,--ti-it! ;--'. t-']i tiTtm-th

y en a- :';i'cliL[C//l/lif":t'"` ,',l/ll'le:i':'el'illlhll",elc.,,...,. ,.,,.,.,,..i,I/iiiiill/1111"i'.i. '''lliR//i'tB":7"ew"7,,,

pottsedFFv-11tTan ilr' ll,,El;o}:" st zll;ti--11,,ei ,COnlti',:of,Desree

l/:"Sog "S /ptJ:':.:;l a., i f,, ",3,'O'e` .･m, ,,.i･tY! tl h t- "--rt-'T- n-J- P.LM tr/' /-cri,ri--4vi'"SS ,,#,ft ' tt--i -ttt .le 15 1" /] t--s1 es q 1" s-- } eranleAngti't"dDgrets jVrr"'"'S't'Dctank''A:g{nihdegtees -''''

Fig. 15. Fig. 16. tJNi"'

"tt/lt

FutL pm ,

'Els-( Load ,:t c{el

T.･SS

{･gil･LA

1/d 2 1 G 7

LO SHP. 11 v -z

tt s d- -L

1400reth

-

} :j s

t"Ts tIS]{

11 ie -l' -.J

/} 5 ]e

lt"'

Htat"dCiE!vsliee

Tra/t:-/t--t--- -

s,.tH 1: [M-Tst Kcelf"ltl

Petied oF Fael lalet"on

s- ls te ls tn :

fucl ･

Eesiet

so

le 4t

L,lL"ad

[11

B,e,"

14oO,Fn.

It

[x;

l[i :E'i:,,-,C/.111I//"i'ZiJ'L`C:/ii'11i;l.,/gvl,'?,.oo,,..

-"--･-･- -11", tdntttJ- / l Ct4nL A,ll; e S 1" 1fi tt ts ]e JI 4- -l Cttnk A"gte tn titgrte; '

Fig. 17.

Figs. 15･18.

Petiedof

e} Ll:::1･lfe,:):ti,n.

Kcel/Mol

Fuol talottion

i. S5

li

fi

,

`

-.

' coi

' so

"

IO -

tX" 1-

TM

cai

wn

f:"

'Tled

su

tFee

('at?)mns,ft!ca

l :.2 3 ,2'2

s it ise'.i-t Jd ,T

4

-rptin

-t

'

,

±-±i MTe cSCetkst,ln

/,tll'(A/.tl,1

t-:+ ?1- Cen"rVeLDegtee/"ti

.t O-l7･i- I :Il]

: l-el

i- 1- id .}rl/1 -ll!-'1- /1 -" t-

s 1-1!t-tr-xo:1"irox.s. Crnnk'A"gleIndogries

Fig. 18.Indicator Diagram and Cornbustion Progress.

1 'l) - -I L

Experimental Studies on Some Combustion Problems of Diesel Engines 405

t'-N

.5

$6fo

%86

c,e,

ets'

gg,

g

T oo

O.95

o.q o

'

x¢r?4

-"f .

gs

yee'

'p

7giA=gt41.Ilt7,,,･d,L

7gJot"";,iKlct/,,i,-ic:T'

7th=o,s74

k=t･3k

･$pt"ip'

Ci}rl

-t5 -･10 -5-O 5. Ignit;onPointindegrees OC.A

Fig. 19. Constant Volume Degree of Combustion Computed from Combustion Progress.

with the equation shown in Fig. 19, taking the eombustion at the T. D.C.

as unity. The tendency of the curves of rp,ih and fuel consumption rate

shown in Figs. 19, 14 respectively agreed unexpectedlY well. Further-

more, the comparison between Kogasin 82 o£ T,i, =O.888 and SRF 42 o£rp,ih=='LO,991 at the points 1 and 2, respectively, in Fig. 19, for example,

O.991-O.888must give the difference of 203ex =232 KcallBHP-h in the O,888fuel consumption rate. This value coincides fairly well with the one

obtainable from the fuel consumption curves in Fig, 14. Aeeordingly,

it ean be understood that the change o£ fuel consumption rate withcetane value exists in the difference of the pxogress of combustion.

Of course, the imperfectness of combustion, cooling loss, and friction

loss are also factors that have some infiuence upon the fuel eonsumption

rate, but they would almost be independent of eetane value. The advaneement of the ignition point to the point of maximumthermal efficieney means a device to make the eombustion oeeur inthe neighborhood of T,D.C. as nearly as possible, In other words,the combustion process (after rp,ih is taken into consideration) should

be made to occur equally on the two sides of the T.D.C. Of course,since a larger cooling loss aeeompanies at the earlier combustion, the

timing of the actual combustion process should deviate more or lesstoward the side after the T.D.C. Consequent]y, if the timing of

406 Tamotsn KuRolwAignition £or the maximum thermal eracieney is Iocated at an earlierpoint, the combustion should also finish later, with the result that the

constant-volume degree of combustion and the thermal eMeiency mustdecrease. Further, it can be said that the thermal efficiency beeomes

lower when a fqel of high eetane value is used, because the ,part offuel which burns at the instant of ignition is small and the total peTiod

of combustion elongates, for the Iag of ignition is small even at the

point of minimum fuel consumption. In the case o£ this kind of com-bustion chamber, which is characterized by a considerable ehange ofignition lag with eetane value, there is a very IaTge difference in the

progress of combustion, which results in a large infiuenee upon the

fuel eonsumption rate, so £ar as the same fueHnjection apparatus isemployed. It ean also be seen that the temperature of exhaust gasbecomes higher with the increase of the fuel eonsumption rate.

For the purpose of further examination of the progress of com-bustion, Fig. 20 is prepared from data graphedin Figs. 15-18. 0necan attain the following conclusions with the aid of this figure:

.Begifin;ngPeinteFrniect;en

r3MMo -2o'-lo '

ssfo-

s

"sN..tgnit.

NRNb bsNN

s

NN NNes

N

->to .N ."NN'

Ntogo%NskN

ft

NSsb90%

.ei

Ets.ati

-e

.Ei

-"

gR6ig

ts

Nx Xs

r･,･ N T" ),

-' ge･% 7o x -teo%goKegasin 82

.3･O･

-20

-10

o to-

2e

so

40

so

Lm- 60

-SMo 2o -te

Bga.s biNS

ocggXsN NNNe.×.a"dNNNeNsx

xO･s"p60%NN･ssK'rv

.fo･ .X6S07.{NXS6qo%

.e.."toC

h.,

.E... g.

SuE

:.

IN lx Ns xx -･ t'

S,R.F. 70

-ez

loon

}

eeglnningPe;nteFtniection

-3CorrMzo -"

,.=t"

g es･k-'

･fi

s'

2･

6

:-goe s, s "NN sN S. -L.-b.60Z NN Sg-. Xe80% ,1"-m- t- s- o s x Ntoeo% " '". Y-dL t-- / NNd Xbge.%V --ts100% .E" t9-

io' tt'

//' ,9

8･

L;ghlt d;l A 55

-30

-20

-to. t9n;t.

e IO 20 3e 4o 50

-30 -?o -o'

-N-.--

hPs

g xe-sNe.sS=N..',".t?oO

NNNNS N

1.osoNvgo

NON Ne

eNN

NNNNX,b99

L.=ig ---i)toezE '"q-

nt srL es

8u ig "S.RF. '43

Fig. ZO. Combustion Progress transfered from B"igs. 15-18 (Full Load).

(a) Curves of 60-80% combustion lines show that the duration oftime required for eaeh combustion is approximately inversely pro-portional to the period of ignition lag. It can also be seen that the

period of combustion becomes long when the fuel injection continuesuntil after the beginning of ignition. The reasons.are in the different

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 407

prepartion for combustion resulting from the magnitude of ignitionlag and in the degre.e of disturbanee of mixing an' d contact With air6f the fuel injected after the ignition. These are the main reasons

that caused the difference in constant-volume degree of combustionexplained above. If the attainment of minimum fuel consumption rateis intended to be realized, the timing of fuel injection must be adjusted

So that about 60-70% o£ fuel would burn before the T.D.C.t (b) ThQ duration of cornbustion from 80% to 90% does not differmueh as before with the timing of injection and the eetane value of

fue]. It may be understood that the infiuence of the ignition laggradually diminishes at this period.

(c) The duration of combustion for the last 10% of fuel whieh''re-

mained becQmes exeeedingly long, if the period of its eombustion is

located far after the T.D,C. This Eact may be understood to meanthat the contact and the reaetion between fuel and air' become poorwith the decrease o'f pressure and temperature by expansion.In the case of light. oil A, however, the after-burning becomes large,

for early timing of injection. The result seems to be inherent onlyto this fuel; the other light oils of the same cetane value did not show

such a nature. ' The rate of maximuni pressure rise (alelPt ),... measured on indicator

diagrams is shown in P"ig. 14 against the timing of ignition. The value

( delput )... for £uels 'bee'omes large with the advaneement of the timing

of fuel injection, singe the ignition lag increases･ When (elaPt)... iS

compared at the same timing of ignition, it becomes smaller with the

inerease o£ cetane value, sinee the ignition lag bemomessmaller. Thesefacts can be naturally expected, but it may be worth-while to examinethe values of (-9el!(z-)... for the minimuin fuel consumption rate shown

with a chain line in Fig. 14. The values.of (aldPt ).,,. are almost the

'same for all cetane values of fuels. Accordingly, if the tirning of fue}

injection is seleeted at the point of minimum fuel consumption rate,

the advantage of the antiknock property of high cetane fuel must be

sacrificed. Even at the point of minimum fuel consumption rate, thea'mount of fue] injected before ignition point must be smaller for the

fuel of comparatively higher cetane value, since the i'gnition lag is

408 Tamotsu KuROIwAsamller. However, the fact that values o£ (delPt )..,. are almost the same

for all cetane values may merely signify that the inherent combusti-

bility of the fuel is higher for higher cetane value. The values o£(elaPt)... fOr a COnStant beginning point of fuel injeetion are given by

two byoken lines in Fig. 14. When the timing of fuel injeetion islate, the infiuence of ignitiop Iag becomes large and (ddPt )... becomes

smaller for higher cetane value, since the amount of fuel injected be-

fore the ignition is small. On the other hand, when a considerableearly timing of fuel injection is employed, the injection finishes before

ignition for all fuels and (CelIPt)...increases with the higher cetane

value. Aceordingly, these faets a]so indicate that the fuel of higher

cetane value has a higher inherent combustibility.

It is generally said from the experimental result of one drop of fuel that its combus- tion veloeity is proportiona} to

the "evaporation ability" and is independent of the eetane value. In the case of aetual combustion in a Diesel engine,Uo･:

l･

.3

FV .

-

A,s,T,M.L'-DisTILLATToN

na

(

6x."

tAtsN 6x,"

o6[SiAht

gegrgi"3Bo?' t

7o

1' k

6o

5e

ve

jo

10

oo

9o

8o

tie

.so

4o

3,O

tsy,:,

g-,1.?,rto

,li [o',xc.g....

r+--il v-ItCetnne

di

'

LLdflsecNFO.:?erllndex.eFBeEllngpe;nt

1LigFto;iAO,84C5'stS2s/eC.

L;ghLo:tBOs84L3SS)76LFghtoEEco,837'24･7s･R･F･7oo,so370246s.Rs,3oo,98Z302S-1KagaEin82D.7B6822G6Ke{e!enea8304SJ223

xsago

ojo 2o304oSO607o809oioo Dist;rlDLim in %

Fig. Zl. A.S.T.M.DistillationCurves of Fuels.

however, most of the Eueldroplets are crowded and the

combustion of each dropletusually oecurs under the con-

ditionofairshortage. Insuchcase, the chemical structure of

the fuel itself may also be

understood to have somerelation with the combust'ion

speed. In this meaning, theinherent combustibility of afuel one is concerned signifies

the combustibility under aeondition of actual combustion

inaDieselengine. Fig. 21

-

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 409

shows the A.S.T.M.-distillation curves of the fuels used in the present

`

(ii) The case of light load.

The per£ormance curves shown in Fig, 7 show that the fuel con-sumption rates almost eoincide at about 4 BHP. This result comes from

the faet that the influence o£ the timing of fuel injection decreases,since the fuel injection finishes before the ignition at that load. As

teadts,ti-BHPaH400rpm,(Per;odeFFuettniection9-9t5CA.)i tqi:;,?5gS.c

oFFueftniecfion

FuelConsumpt;enRatevsBeg;nningPo;ntvs.lgnit;anpeint""?h

"-

.

..

sbi5"ft t;

eNN.Nx

e

N ""s"'

.bk-

NNli-S

x.'Kerese'.e--'s"ft-

r- ;..'

lt6

!aCU-

'xrs

tSbt*･i

.zZ

s- tgniEien

O:.A bvkfL",cta9Exhaust",qbits" I.s"b5b

1ri-7-.×

Ternperttture- ".or>x,

dx- -q･s.fffR.ftlst.e.t,,ke--=.-

-''

lis

5"t'.t

..ij sb

so

25

20

IS

10

L--t' '' 5"S'x.

'x-.G--'ss..SaF"s

x.--.te.N

TcmpofCyl;nder(Therrne'P:ug>

'ttPpcsXa..See

2SO

270

260

250

2-O

120'C

ept,

el-Ko7a)f,u tKdit'

CeLaneNumberrJs.c

82O.7S6sRFOorroosojttehtOJ']Bs6o.s4sUeburoa

sqv7e48･O･gES.5RFg･545e.S6oKcresine45e･63o

kosA.fr"sE

Hu

to3se

--.=-to3ootooeeMax.RateoF10f50

fOIDO

so1.se

teD4o

'o

$-Pressure RiseCdJ)(dtma,,--o-+i---A-.

T-t-.-SRF353So.S7g

-"pe" }6P. ¢ets

qsa' lesssea "li--`s=

]5-30-25-20-15-10-5O 510 Crank Angte ;n degrees

Fig. ZZ. PerformanceCurves. 4110 Load of 4BH? at 1400r.p.m. (?eriod of lnjection: 9e,y9.50 C, A.) .

it was judged that the difference of the inherent eombustibility would

evidently appear in the range of sueh a load where the infiuenee ofthe injection timing is small, careful experiments were performed at

a constant load of 4 BHP for various kinds of fuel. Sinee the results

of several exp.eriments gave the same tendeney, only one example isshown in Fig. 22.

First of all, the comparison of the fuel eonsumption rate between

410 , TamotsuKuROIwA.'･4BHP and full load reveals several differenees, At first the.timing

of ignition for the minimum consumption rate tends,to be closeic

together than in the case of fu]l power, and the beginning p,oint o£injection for the minimum consumption rate'is retarded with thehigher eetane value, Secondly, the minimum fuel consumption rateis lowered with the higher cetane value, e'ven though the magnitude

'is small. The first difference resulted because the difference in tke

'combustion progress was not so large as before, for the £uel injection

finishes before the ignition at this load. The reason for the second

difference can not be explained from the combustion progress. Onecan merely understand from the eombustion progress that the duration

of combustion should elongate with the higher cetane value and thatthe thermal eMciency should decrease. Especially, the temperature

o£ exhaust gas for SRF 35 is low in spite o£ an extremely large fueleonsumption rate. What is the reason for this contradiction between

the fuel eonsumption Tate and the exhaust gas temperature :2 Sinceincomplete eombustion was judged to be the only reason for this con-

tradiction, the analysis of exhaust gas and the measurement of pene-

tration of the injected fuel were performed.

Analysis of exhaust gas.

The analysis of exhaus gas was undertaken by changing the timing

of fuel injection under the eondition o£ 4BHP and a constant speed

o£ 14eOr,p.m, The analysis was made with an improved Orsat ap-paratus. Since the content of CO in the exhaust gas very slight andH2, CH4, C.H. were seareely detected in several samples, this method

was jndged to be satisfactory. In the eombustion in an Diesel engine

it was assumed that a part of C ehanges into CO, and CO, and thatthe other par't of C beeomes a soot; the quantity of soot was eom-puted by means of the volume of suction air and the results of exhause

gas analysis. The results o£ this'computation are shown in Fig. 23. ' tl t igi;fiiefFqer L6adeFEng;ne5t

i. pmtOmtemp l2 lsoc 's' E Fig.Z3.Ca16ulatedValueofSootv O.2

£t, ...s[il. If.. f in Exhaust. N M-.-uo- ""Nar--'"g-r'ts-x.t..p .sRF3o-BE/g=O't ,,.-ct'sr-:LHI'I;,l,iii:"L..i.{?//;ygi:`F"eiSsS.R,F,7,OC=Z5718,"=l24:gisS-:glg2j

･q s.fiF7o'? SRMS Heavyoi187012,5 O.3e -2o, -lo 'BegTnning Point 6F Fuel lniectib- in degrees

ExperimentalStudiesonSomeCombustion?roblemsofDieselEngines 4U

The colour of exhaust gas was almost so-called smokeless, and the re-liability of the absolute value of this soot is small. HQwever, it would

not be unreasonable to understand that this value of soot would fairly

represent the degree of incompleteness of combustion, It can be seen

in Fig. 23 that the quantity of soot inereases for the lower cetanevalue. The quantity o' f soot is extremely large for SRF 30-eh-35, because

these £uels are o£ so low cetane value that the ignition point couldnot be advanced by the advancement of fuel injeetion for the com-

pression ratio o£ this engine. In these cases, the combustion may beimperfect. The quantity of soot is also large in the c'ase of the blended

fuel of Mili heavy oil and SRF70 in spite of its considerably high

eetane value. Actually, the amount o£ aecumulated'carbon on thesurface of the combU'stion chamber was large. Judging as a whole,one may be able to understand that the reasons for eontradietion be-

tween fhe fuel consumption rate and exhaust gas temperatuye and offuel eonsumption raee at the point of the minimum fuel consumptionrate for .high cetane' value ekist in 'the difference of this incomplete:

neess ' of 'combustion,' si.hce the fuel consumption rate and the smount

of soot are approximately proportional.

Measuremenf of penetration df injected fuel.

' inheinfitienceofthetimingoffuelinjeetiongiveninb"ig.23showsthat the amount of soot inereases by the earlier timing injeetiofi:

This result was understood to'have been caused by the worse.con-ditions of temperature and pulverization of fuel as well as by the

combustion of a part of £uel after having adhered once to the pistonlsurface, sinc6 the amount of soot increases with the advancement of

fuel injec･tion. Accordingly, the penetration o£ injeceed £uel'wasineasured as the next step of iinvestigation.

The measuTment was performed by usirig water in place of fuel.TWo electrodes with a gap about O.1mm were placed in the path of'

the injeeted fuel as is shown in Fig. 13 and the arrival of pulveyized

water was recorded. The results of measurement are shown in Fig.24. When the injection is early, the penetrating speed of the injected

fuel is fast, beeause the density of air in the cylinder is small. As

the result, the pulverized fuel would r,each the piston surface rather

earlier, and it is possible that a considerable amound of fuel burns after

once adhering to the surfaee. This kind of combustion meehanism.may.probab]y beone of the reasons for the'incOmplete combust'ion.

412 Tamotsu KvRolwA Mean Pos;tion oF Piston Heag SurFace

-.e Positiori.oF Measurement E ,E-

;t

lt 2o positionoF L6 Meesurement 8 ve g S.

O-4o -3o -2o -io "o POSitiNen.,OzFle

Crenk Angle in degrees'

Fig. Z4. Penetration of Fuels Jet Tested with Water.

Pump : Boseh 9mm, r.p. m. of Engine 1400. Nozzle: 4xO.28mm, Injection Period 170C.A.

po: 2eoat

which oeeurred when the timing was early. It may also be possiblethat there are some differences in the incompleteness of eombustiondue to the structure of hydroearbon, when this kind of combustion on

the surface occurs.

The maximum rate of pressure rise £or the minimum fuel eon-sumption rate becomes as shown by the 'fine line in Fig. 22. It be-comes a little lower for the fuel of high eetane value. Under the

condition o£ less infiuence of the timing of fuel injection by advancingits timing so that the ignition lag would be rather large, the rate of

maximum pressure rise becomes larger £or the fuel of higher cetanevalue as is shown by the broken iine in the same figure.

The measurement of temperature was also performed in this casewith a thermo-piug as shown in Fig. 13. The results of measurementare also graphed in Fig. 22, which shows that the temperature isapproximately proportional to the rate of pressure rise by combustion.

4. Conclusions.

In the case of a combustion chamber of direet injection type, thedifference of the ignition-Iag angle due to the difference of fuel ceeane

value is large, and cetane value o£ fuel exerts a large influence upon

the performance of an engine. If the cetane value o£ a fuel is high,the amount that burns at first is not so Iarge, beeause the amount of

fuel injected before the ignition is smail and the duration of prepa-ration foz' combustion is also short. Consequently, the most part of

/ Z

46le ,n-lij'

th

'15mEbxm

..p'

3eo.c-

9am

.-- "KS,De"slitS.--

Y10,c-

b

20didid

E

'?tess"Se 'm-

cb06 EoetOUe"

E

'iL.--

<±5.-84e

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 413

the fuel will burn gradually a£ter the first rapid combustion. In thiscase, even if the fuel woul be exposed to the high temperature ofcombustion gas, the combustion wil} be disturbed by the gas already

burnt, and the elongation of the combustion period must accompany,As a result, the thermal efliciency itself must al$o be lowered. Even

if the comparison of the points of minimum fuel consumption rate isthe intended purpose of the experiment, it is found that the ignition

lag is small for higher cetane va}ue, and the effect of cetane value isstill left.

It can be seid in general that the combustibitity of a fuel in a

combustion chamber is good and incomplete combustion occurs in small

degree when the cetane value is high. In the ease of a light load, the fueHnjetion wil]. be finished before

the ignition, beeause the period of fuel injeetion become shorter and

the ignition lag beeomes larger. This means that the effect of thetiming of injection becomes Iess, and the lowering of thermal eMciency

decreases. On the other hand, it sometimes happens that the thermal

eMciency improves when a fuel of high ceeane value is used £or a loadless than a eertain value, because the difference in eombustibiliey and

incomplete combustion can not be avoided.

Concerning the relation between cetane value and knocking, thephenomenon of knocking in a Diesel engine is different from that caused

by the abnormal rapid combustion in a gasoline engine, The I<nocking

in a Diesel engine indicates a differenee of combustion veloeity that

will be governed by the amount of fuel at the point of ignition, and

can be changed for any operating condition, However, if a device for

deereasing knoeking is employed, the thermal e'Mciency will generally

be Iowered, and a device of preventing the lowering o£ thermalefficieney must be expeeted to result in a severe 1<nocking. Some de-vices for deereasing the initial combustion veloeity and finishing the

combustion as quick as possible is preferable, but it would practically

not be easy. For instance, the employment of a pilot injestion or afuel nozzle of throttle type is simple and effective for the purpose of

restricting the knoeking when a fuel of low cetane value is used, butit is still doubtful if the combustion is quickly and perfectly completed.

Although the above diseussions are based upon the results ofmeasurements on an engine with a combustion chamber of direct in-jection type for a certain constant speed, the effeet of eetane value

upon a higher speed must be larger and ie must be smaller for a lower

414 ,'･.t. TamotsuKuRolwA

speed, as far as the effects of cetane value exist in the differenee o£ignition lag and ,cpmbustion progrees. Concerning the effect of com-

pression:ratlo,,the difference in ignition lag will be decreased by in-

creasing the compression ratio and the effect of cetane value must be

reduced. Moreover, the combustion ehamber and the fuel injeetionsystem with Iess differenee in ignition lag depending upon the kindof fuel4 m,ust, be less effected by eetane value. With regard to the

turbulence of air, the larger turbu}ence will give the better cQntact

between air and fuel, and the effeet of cetane value will also bereduced.

II. Combustion Progress when Pilot

, InjectionisEmployed.. ' 1. Preface.

?i,lot injection is a deviee for reducing the knocking in a Diesel

engine of direct injection type and is praetieally applied mainly inEngland.' Since the use of pilot injecion results in the deeresase o£

the i.gnitiQn lag of the £uel of the main injection, a re,duetion of knock-ing can be naturally expected. HoweveT, the rise of a prob}em･whether

. ' theuseofapilotinjection

op;fi.",o'gi3..

blAu,"TetlcVelve'

2xO･30inm4=200 .t., Li.

[.

Autmetic Volve

4,,.X ,Oi2o8LM,,1 /

x

't- i=t:v- t...=-.t.t.i "EL -.l{l:'I'

N. HO" (Stteke:ILe)

h=14･4

";ti';t'

f･/1-i.L. :-ty.{-

.t)IS<,{' ,.]IIi,,

xx

, AV::etthplump .

,dpm9mm

;1

Fig. Z5.-

lndlaetef

Bosch

Fvel Pu,mp'

dp=8mm

Avrangement of Fuel Yalvefor' Pilot Injection.

would make the･latter half ofcombustion slow resulting inseverer after-burning accompa-nied'by a reduction of thermalefficiency must be expeeted from

the'results obtained in StudyI, Therefore,someexperimentsweJre undertaken in order toclarifythesepoints. .

2. Experimental method.

The experimenQal engine is'a single cylinder type with 110

mmx140mm cylinder dhnen-sions;it is an engine of'direet

･injQction type .as shown in Fig.

25. Two-perfectlyindependentinjection pumps and nozzles are

employed. The horizonal and

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 415

vertical fuel injection valves were inainly used Eor pilot and maininjeetions, respectively, but the reverse arrangement was also tested.

AIthough an open nozzle originally attaehed to this engine was firstused as the horizontal pilot injection valve, it did not give good results.

Therefore, it was replaced by an automatic valve with the same dia-

meters of nozzles exeept in the first experiment deslgnated by (A).

All the experiments were performed with a commereial light oilof cetane valve 55 as the main injeeting fuel and with the same light

oil or Kogasin II of cetane value 82 as the fuel for pilolt injeetion,

Since the pilot injection of a commercial light oil did not give aneffective pilot eombustion when the load was !ight or when an opeit

nozzle was used, Kogasin ･II was used for the purpose o£ giving aneffeetive pilot combustion, and its effeet was investigated.

The amount of fuel and the timing of pilot injection were adjusted

to any time so that they would give a proper pilot combustion anda preferable effect on indicater diagram byjudging the indicator dia-

gram obtained on a Braun tube. Since the injeetion beeame irregularwhen the amount of inje'etion was quite small, the amount of pilotinjection could not be decreased beyond a certain point. However, thepurpose of these experiments was' not distuicbed, for an effective pilot

combustion did not oceur if the amount of injeetion was below thatiimitation.

Coneerning the experimentai method, the test was continuouslyperformed with and without piiot injection in turn after obtaining a

stable running eondition Under the speed of 14eO r.p.m. and 4 BHP or

7 BHP; the fuel consumption and indica'tor･diagrams were obtained.The indicator diagrams explained by the phrase "pilot injection only"

in Figs. 26-34 were taken from the tests with pilot injection at the

instantofcuttingthemaininjection. , The,,eooling loss was tal<en into aeeount in all the computations of

the progre'ss of eombustion in this sectlon, because the amount of heat

generated by combustion is small when only the pilot injection is

employed and judgement pf the progyess of combustion becomesdiflicult 'without taking 'the coo]ing loss into account. Howevex, sinee the

measurement'of cooling loss under an aetual running condition is very

difficult, the cQoling loss. £oy the state Q£ self-running obtainable justafter cutting all ,the fuel was,a.ssumed. as the value of cooling loss.

The cooling loss assumed in,this manner ean easi]y be computed asthe difference between the work done given by compression and the

416 Tamotsu KuRolwAincrease of internal energy of the air in cylinder by analysing theindieator diagram taken just at the moment of cutting all the fuel.

3. Experimental results and diseussion.

The experiments were undertaken for the following three cases,and three records of each case are shown in Figs. 26rw34. The indicator

diagrams, the diagrams o£ combustion progress computed £rom theindicator diagrams, fuel consumption rate, and the temperature ofexhaust gas are also shown in these figures.

(A)pilotinjeetion･･････Horizontalopennozzle+･････Kogasin- 4B-H`P･'"Igi･gg:Z9

(2×O.25 mm, elp: 8mm)

Maininjeetion･･･-･･Vertiealautomatievalve･･･Lightoil- 7B.H.P.Ji･Fig.28

<4 ×O.28 mm, po: ZOO at., elp: 9 mm) (B) pilot injection ･･････ Horizontal automatievalve {I}PgghatSt?.1-H-44BBIHHIP?I l: X･:I :g

(2 ×O.25 mm, pJ: 200 at., elp :9 mm)

Main injeetion -･････ Vertical automatic valve ･･･ Light oil- -'-7 B.H.P. ･･･ Fig. 31

(C) piiot injeetion ････-･ Horizontai automatie vaive {LKiOgghaig¥.i-n-44 BBIHHIPpl [ii gl･3I ::

Main injection ･-･･･t Vertieal automatic valve ･･･ Light oil-"-7B.HP. ･-･ Fig. 34

The indieator diagram C in Fig. 26 was obtained for the samebeginning point of main injecton as in the case of diagram A by ap-plying a pilot injeetion and by decreasing the amount of the maininjection so that the output would always be eonstant. One can see

that the eombustion velocity of the main injeetion £uel is obviouslyvery slow. The curve C-C' represents the differenee between thetotal combustion progress C and the combustion progress of the fuelinjected by pilot injection C' and it Tepresents the combustion progress

of the main injeetion fuel. The indicator diagrams for an earliertiming of the main injection than before are shown in Fig. 27. Itmay also be clearly seen that the combustion veloeity with a pilotinjection is slow.

Fig. 28 is the case of 7BHP output. Though an effective pilotcombustion could not be obtained unless a quite early pilot injeetion

was applied in the previous case of 4 BHP, a effective pilot eom-bustion was obtained even for a late pilot injection in this ease. In

the ease with a pilot injection indicated as D, the eombustion velocity

of the main injeetion £uel is a little slower than in the case withoutpilot injection A, but there was almost no difference between the cases

Experimental Studies

{A)4 e.H.p,

on

-L

Some

1400 r･p,m･

Combustion

N.ti=iri]stt,en

Problems

Yt,ttcelnufoaititvtlye

of Diesel

eo.LhFue:pump

Engines

a. S･28m:, dpvSmm

T"hFu,1AMltTn1"pt6anentsSj73

:e";:.=FL/onrt!e

ktra,tfi"efi

E=H･Ltrc.p-

J4"ects/tn"

B

"=220at,

iniit:ntilepennonJle -LhauL'=h'Seelptmp

Blt'

ses;C--fi) 3eo'cpa 2x O･Z5opm, lp=gmm' coe

ACemPTt!v'e4onylv ' AcltFthptteu"iecb'en34J7C -1･3XJ' j44'C c

C'PiTo÷tnl"tti'oneqTT

I,te'-tiford/sgFtmT

ec'A '

Cerninlt't"PtOgres:,

!--'

ijooo

c=c-ti.

,..nt 3oeo

ptrtteoF '

S.R.F.40'7'ML'ni"iwhton

c '

P--edci

slS.R.F,40

2e-x fA.

.

2eoo

F',lat:n]eceltA

Kecilin S.R.F.ca 't t

'

c,C2ael.hlfilhets"hlls) yc fooet2z;s.s1

/'s

40SO oo- to ote .2e co -o-sn 60

Cisdi M!la ln dep'ree;

Fig. Z6.

' so Ul4mi CA) 4 e,"p･ tL tGOO r.ptm･ 'BL T",b A Nu'ftt"T"ti:n VetUtSno+vmilL-v"1." B 4ye2Bny,pt, D ,, g P:lol4"jettia." Ubt:t4fibleptonettlt

2.02Spm,

)ndi"lcTdiA;rtml

e Ftr;cdoP rnIT7L7i-tqoA UIhLO/1 Bf 20p-dnd"F

(.lag':1rcLk4 IllAtwytiolt;aO'S io

2P tt/IAt PPiY) ct -,gHton .

vsX ' L

( lgst･to-ts,

FGelconwinpt:'.niAttzae/n:'7;･"-Teno-lt 32?3 IE71;-4h

M 3078(-4･SX)-thP;Ee"n･TtcUen3Se5Ceit･ag)

o sss2etaop)

fi,.'V.7`rMn{P)

Eth,hop,

32i PC

316 de

3i7 'C

324 'C

co je 20 te

Gtrk Angts

oth

lodhgtpts

Fig. 27

M!inT-/t[ecq

20'3o 40 ( tfSt tO 2i,

.

Vertitaltuttmltitv"t,t Oosd.lelththp

so 6eS,i,?t,7S2epmig)

(A}, 7 e"p.,+li,eO tpm.

P/lotlnifctlan

4MO.28m.Tep=8pm..a=220'"oTiTe=Ltlop-"nonl-,AtsliuUdNT"-1pump

ked/"ol.

2io.2smT.,ep.gptmlah.･'n'b Pue]fcsttmp}/onn!e +emp.

60tyrdtn1 6av

'ltM--.Mi?[ti.ho.lvzvel im/avP-A 46eecs

c

e" 2403r-J･501,) 4SS"C c'

pCw;t-P"etl"IKLIcn,26Si

C'P/bF17.iectio"c.,"'

(+4X) 4S4PC pA'Cemin!U'onpregreti

50CO

D"ELhP::stlniectFcr,2737 (,9%) 4if"tD' 1P"o!lnitcu'bno.Et -o tsoo

'Ci

i '

eacoPeHdlor' D

ms/e;nlidleeti s.R,r.

Da

S.R.P A 2oaoklKo-E"n sps.

cKegid'nt(S.RS,

eD ta"i'rmT" '

'

,

PerlarfeFpi]`t["ittLien

40 ee

Figs.

20

Z6･Z&

to o Cteab Afigle Im eeo,"s

Fig.

Indieator

Progress

Jo 2o 3o #o se -6e (nst lo jtiMx,7o'4･2intsUi)

Z8.

Daigram and Combustionin the case (A).

417

4r8･ ,' Tamotsu･KuRolwA

of C anc A. Since the combustion o£ the main injection fuel beginsafter the end of main injeetion in case C, the effect of a pilot inj'ection

would scareely appear.

In Experiment (A), the fuel consumpeion rate was considerableyincreased when a pilot injection was employed, The slow combustionof the main injection fuel is of course one of the reasons for the in-

crease in the fuel consumption rate, but its main reason exists in the

facts that the puly.erization by the horizontal open nozzle which was

used for the pilot 'injection was not good and the combustion o£ thepilot injeetion fuel itself was quite bad.

Experiment (B) is the ease when the oben nozzle was replaced byan automatic valve with the same size of nozzles for the purpose ofimproving the pulverization of the pilot injection fuel. Fig. 29 B showsthe result of expe/ iment when the timing of pilot ihS6ction was com-

paratively advanced so that its･ effect could be obtained as clearly as r,possible. The comparison between the indicator diagrams A & B showsthat the rate of pressure rise is. very small and the after-burninginereasies in diagram B. The increas'e of fuel consumption rate in

diagram B is comparatively small as a figure of 1.2%, beeause thetiming of fuel injection'in the case A, whieh was taken as,the objeet

of comparison, was a ]ittle retarded. Fig. 29 C is the case when the

timing of fuel-injection was selected with an intention of obtaining

the most favorable indicator diagram.' When this diagram is compared

to A, it ean be seen that the pressure rise is slow and the after-burning is not so bad. Three pereents of decrease in fuel consumprion

rate compared to A resulted because the constant-volume degree forA taken as the base of comparison was not so good.

Fig. 30 shows the case when a commercial Iight oil, the same asthe main injeetion fuel, was used for the pilot,injeetion. An effeetive

pilot combustion is rather diMcult of a'ttainment when a commerieallight oil is used. In the eases of C and D thg pilo't injection and the

main injectU'On were continuously made and the results are similar to

the case of singl,e injeetion with an early timing o£ injection. In the

present case,,.however, the £uel injection was made by two valves.Therefore, a comparatively unfavourab]e pulverization of'fuel at the

beginning and the end of injection was doubly superposed, and a small

decrease of fue] consumption ratg wa.siobtained in spite of the advanced

timing of fuel ipjection. ･B represents the case of a larger amountand an advan6ed timing of pilot injection. Sinee a pilot combustion

Experimental Studleg on Some Combustion Problems of Diesel Engines 419

T"ts

AMilntnieeVo-ertly

rtCemp,essle-cn":

Bv;LhPa!LlnietErtn R'PITol1:ptVono"Tt

cw;LhP:rt.1diecl/･on

dFrtbttriitttFcno="･

-Ptn'Mef

FlitTcL:sumFH:nT:E-

' 3Z12 thvaHp.h

szsgof･!'x)

jilot-S･oe:l

(B} 4

eH.Lethp.

27o`c

2Sot

Zle"c

e.H.P.

i

B

"

c

.P"ideoS'rn!Tny;riecLitn,

. A-LthiOl )1c s'rv=;fiC7i/"'"k.,.,,criXl X=,Li.i'l-ijB ,

pi',eii".i"L/oa ,eC srp.t2t%) . t7

tl

't

l400 s.p.m.

Ki6rf

w 1il;czL.eTe:igT-ms/

A,ve -

e#e A'

."-. "i .-l. - A.O. .. cJ

B'

MdsiltiKtieth 'Yttti[tl-LLenistitv-ITe

4sO･23--mPlali:iectien･,Rofil:nhf-vkuiEtvnlte

. 2.xO.･30ptm,

A

Ctthhe:Uenp/egtt"

UoitkEdtTpemF

dp=8mmAbhatli"jN"pomp

dp=gpm

-"'- C-c

soeo'

KcKi.

deee ,

nv

?eca

[coo

eo 90 po lo O ctnnk Afigtt in degtetS

Fig.

to

Z9.

oo se -o '' CtffS3-9･l5; neto-C50 60

,7g7･8pttnde),

Tgsts

A M"'nrniecifnnoele

A' Compretdeqe.p.lt

U ulthPlt`tln:,-Etien

B'PIPet1"i.tEen.nleC v;th Pi!eL1diectien

C' PilatlsitcH:noT:r Dbe-hP/tcli:Ttctien O' P/lo+ln],ecbb.lt'e:ly

PtrTpdotmtFnSsisc+i･:q

L;.-HLoilp"difi

.i/ "mahy(s7%J

tiktt"i"pan.

(B) 4 il.KP.

Ftatltonium}/l:e.tlt: EtH.Lemp･

siooXvaF.h. Ml'd

' ' e jtgo(.o･e,'･;l 274P 3IteVf:o･7',,,V 2bS'C

3tZ5 t-T･-Zl 26a'Ce

' N u;HteEr AN 'eL'g

:ttnyt¥,llkio,t

ligsLoilg,swdetX)

c

p

D

A

,t I400

eend

so

t･Fny,

"edicslerdhetLms

Dt

c

se

A, Bap

rk

,CoenVuttiCnFroc,etJ

A

lo

t'

D

s'

soed

ks2tt.

4e"o

eoma

tt"

loe"

,

"o sc it le o Cte" AngT- tn degeet

to

Fig･ 30.

20 30-

Clesj･g･

'

4e

ls

50 SO;ftsg.oec,7sr,crintaHf)

(.

,

l'･.

reits F,telc:Iiu¢pt:cnrteA Mtinlniscticnonlt 2q2gKMteHe-hA' Ccopr-:-afienrre ke,thP"otlni-ttien )040Ct3-S$)n'nletlrt}ectiano"Iv

:"thpit.ll,iect,#., 2ssec-J.Gg)C'P;T=Lldittb':no=te

t

CB)

Esh.Leptp,

446t 4i2"c

4tgZ

7 9.ltp.

Petlndoi miTnr,jed:bn'

UghlcsiA

c

e

1

P-tidielp:rditnl-cEth"

l ei.ii,S, bt,nyi"'P").･

e enr(ts･sx)

LeLi On

t UgleO,1 1 t

'

x

iv t4eo r.p.rn.

sts,

SO lhdit:krdit"Tnms

A

-o

'e e/ -ao

c /-A 7- -

ct

-- /

- -rt

Combu!tienFtegresi /" t- ..- c..oft- " -t

[t (.

to

ej

'

4o 30 20 ie o Crink Angts in d!q,et;

to 2e 30 40 (iess g 2et

5o 6a'c , 7S"mtaRY}

Teee

va1･

eogs

SGco

4cae

seee

?eeo

teop

Po.o',vi

a

,

Figs Z9-31.

Fig. 31.

Indicator Diagram andProgress in the case (B)

Combustion

･420 Tamotsu KuRolwAoceurs in this case, an early ignition of main injection fuel, a small

rate of pressure rise and an effect of the pilot injection result and

the after-burning beeomes severer. In the case of 7 BHP as shown in Fig. 31, a elearer pilot com-

bustion can be obtainect. Curve B is the ease o£ t･he smallest am.ountof pilot injection, and the results show that a clear pilot combustion

is caused to oecur to a considerabl extent and a decrease of combustion

velocity of the injected fuel ean obviously be seen. The rate of pres-

sure rise, the value of the maximum pressure, and knoeking aredecreased, but the after-burning becomes severer and an increase of3.8% in fuel eonsumption rate is concomitant. Curve C is the case of

an increased amount and an advanced timing of pilot injection. There

is not a large difference in the amount and timing of pilot combustion

compared to the previous case perhaps because of an earlier timingof the inj'ection. However, sinee the timing of fuel injection is ad-

vanced as a whole, the effect o£ pilot injeetion is not so clear eomparedto the ease of A. A decrease of 1.6% in fuel eonsump'tion rate isobtained, because the whole timing of fuel injection is advanced and

the constant-volume degree of combustion is increased. Experiment (C) is the case when the two injection valves in Ex-periment (B) weye exchanged each othey. Since the seetional area of

the nozzle o£ the horizontal automatic valve is small, the period ofmain injection beeomes Ionger. Therefore, the pilot combustion oceurs

in the initia] stage o£ main injection and the effect of pilot injection

is more clear in this case. Furthermore, sinee the sectional area o£the pilot injection valve is large, the amount of fuel injection becomes

irregular when it is throttled to less than 8-9mg per cyc]e, As aresult, an amount of pilot injection less than that value not be used

in this experirnent. The fuel injeetion pump used for the main in-jection is a Mitsubishi type which has a fixed end of injection period.

Fig. 32 shows the result when Kogasin II wa.s used as the piloeinjection fuel. A large pilot combustion ean be seen in this result.

The combustion velociey of the main injection fuel is quite small, be-

causg the large pilot combustion oecurs with the beginning of maininjection. The fuel consumption rate is extremely increas2d in casgB, since the nozzle area of the injection valve wa.s too large to give

a small amount of pilot injection and the pulverization of fuel was

notgood. ' In the expeyiment shown in Fig, 33, a commercial light oil was

Experimental

Tnh

AAe

e

cc

Studies

Futl ce/tivmPtlCn/"l""fiCn:.[ti:nonli SiEtt';tanyf-

[omptess/ennNt

ke,LhP/lvii"EKtiDn d3S4('2VteJ

P/loTtnjectohec{r

."hPitotL.1"fbpA S"filt'ilfe)F/tvtln/ett'LnonlT

P.tidiof inti'niniv("on

" ":hiEx's#i'illlli/2:,;sxa`;N

lfsv("Sx}

on Some

CC} 4

.Exh twhp-

tOE'C

jztit

c nez

tin-tol't

LigE± Oll

L

e".P.

s

Combustion

F,t O,1

-

,f t400 rp･m.

m tndi[-totd,4g/j,=-So

x

to ' c'

s'

o,

ti cua

n d !.rt s// -

ae. to e' lo., le ,'Ctdi Nvafe in deg,usi

A'g 3t '

Fig. 3Z.

CC) 4e･RP･ ,t t40or,p,m.

Ft,Tcoz:.romoptitu'o%i,:,esK,:Tr.' EZ,'

idnektt"ediig,tmi 4V7oC,,]ts) jSO:

ISIOC,c]4)IOt)'C A cS"e , c

b' 30 A' ,c tt v÷.,,O., 'O f...---

IMat"Mlltbe"{ tigs±o/d !te C'

lti

Problems of

Mtinint"t/'D,tH:Aton-idtvlt'mibZTdTt'

ttO!b'･,,nPJet,n/"Ld,- Ve/ticeP-"lo,"tb[,tT--

4jO28.m,,,

A c CeinltvLtiOnptOwr"

Diese]

M't"L",:luelF-,r･p

dF=9,t..

e[L.S[vtlp=T"

de=timm

rC-tC'-J---fi-uJ--

Engjnes

"

T"b:

AA'

e

e

c

c'

30

the,ntt,1"tic//c=Fr

Ctn;m!/enot,e

T/thP:Pellc]t"ti:fi'

Mattt,,iKticntnPT

w"hP/k",itcE=n

Pilatlei.tEoncplT

PEridiofptletl.i"Veq

:,}SgFta" l

t･-vt2? Lislteif

- iv4t!rtg2S)

30 -e f if"lst?s, fsVi;e .c

w . fo fg3 .he)

A .-c

C"mbe:llon).xc,eti

c-S;--'-'-Jfi---'

7oee.

K[:1.t

ptpa

!ece

40N

lipa

teve

1ua1

d

/

ro 30 ?o

B'

.se.L+eiI i}i(li

erling 4t dL:-ileo,J t3-dmstze%)

so o fe Clul lpPo la dopteei

Fig.

(C) 7n.N.p.

;uelT-ts ter.sltmpLi#-T-tethh-ttmp･ A zabl'fithiectiehenit 2SPS L'Vk;af.h 440W.

A' Ce=pTEI"nnlfilv u v/LhPifolt:ixtien 2gtjO(t4ilk) 4]2Z

:' :ll",tllll.e,[1'."i;L.Z': 2g?ol,j72) 't?O'[ CS

c' mlotlrtiecti"hortb

Pen'oecFmifsi'nre[tF"n

ligELa/di'Pt-nlofp/:clin/t:Ld=ny

eL :4oe t.p.m.

ble

50 ,11aa:ttlo:d/igremi

c'

L/,lhtO/1

teSX.tohiieatssveJ

N L;gE+O/1 ]

x

'

33.

ee

-30

Ak- C- - vto / - f ''

c

-/

so "e (,7SSIdi127,M%

'A

c

so eepss･6in-e)

Jo

i x- f'CcE5-Enn,pagt"; x/ l !.'1 ! ".-Fc'

/-

maHtkh

sooo

doso

spmo

am

tpmv

e

oo so

Figs,

le id

32'34.

e ]o 2o se 4e so'C,diAfigToTha"grees tiesSt?f2}/Sc,?SIS"st&)

Fig. 34.

Indieator Diagram and Combustion ?rogress in the case (C).

6e

veco

Kejlb,

cbeo

se:o

doCo

s:-oo

!ees'

f"oe

D

421

422 Tamotsu KuRolwA

used for the pilot injection. A clear effeet of pilot injection ean be

seen in this case, too. In ease B an effective pilot cornbustion didnot occur and the combustion is also quite bad as will be seen on curve

B', beeause the amount of pilot injection was too small for the injection

valve used and an effective pilot eombustion did not oecur.

In Fig. 34, the timing o£ fuel injection in ca,se A, which was usedas the basg of comparison, wa-s selected at about the point of bestthermal efficiency and the timing of injection in cases B and C istaken so that the constant-volume degree of combustion will also be

favourable. Consequently, the eomparison between these two casescan be said to present the best explanation of the effect of a pilot

injection. As will be seen, an app]ication of pilot injection obviously

resulted in decreases of the rate of pressure rise and knoeking. How-

ever, an inerea.se of about 3% in fuel eonsumption rate occurs even

in the case of the same maximum combustion pres.sure. When themaximum pressure is lower in the case with pilot injection, the increase

of fuel consumption rate becomes of eourse larger. The reason forthat exists in the lowering of combus. tion veloeities of the main injection

and the pilot injection fuels, which lowering prolongs the duration of

eombustion and deerea.ses the thermal efliciency.

4. Conclusions.

(i) The employment of a pilot injection deereases the ignition lag

of the main injection £uel, and the initia} combustion velocity or theinitial rate of pressure rise and knockin.cr also decrease. In this case,

the decrea.se of the initial combustion velocity is related to the degree

of deerease o£ the ignition lag. It is necessary to make the ignitionof the main injection occur at least before the end of injection inorder to get an effective pi]ot injection.

(ii) Sinee the decrease of knocking by use of a pilot injection is

re]ated to the ignition lag of the main injeetion fuel, the combustion

o£ that main injeetion fuel is also effected by the deerease of ignitionlag in a similar manner to that explained in Study I in this paper.Namely, as 'the amount of fuel to be burnt at the initia] stage of rapid

combustion is decreased, the amount o£ £uel to be burnt ]ater increasesand the duration of combustion is prolonged. Looking more deeply,one ean see that the fuel inj'ection into the combustion gas of pilot

injection and the shortening of ignition lag of the main injection £uel

ete. result in a disturbance of mixing o£ the main injection £uel with

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 423

air becausg of the existence of a eombustion gas and that of givinga preparation period to burn. As a result the duration of combustionis elongated. This fact that the duration of combustion is elongated

fundarnentally means a decrease of constant-volume degree of com-

bustion and is necessgrily aceompanied by a lowering o£ thermalefficiency. To state it briefiy, the ignition can be advaneed but the

whole combustion ean not be promoted by injecting a fuel into thehigh temperature of the･pi!ot combustion gas. (iii) There is another reason for the decrea.se of thermal efliciency

which accompanies by an application of pilot injection. When a smallpilot injection is used, the combustion of the pilot injection fuel itself

beeomes very slow and an imperfect eombustion sometimes follows,because the pulverization of the fuel is generally unfavourabie. Thedegree of pulverization of the pilot injection fue} is appreciably related

to the fuel consumption rate. Generally speaking, the less the amount

of injeetion and the earlier the timing of injeetion, the wo.rse the

pulverization o£ £uel.

(iv) When the ca.se with a.nd without pilot injection are compared

at their points o£ maximum therma} eMeieney, the £ormer is found tobe superior in the points of a smaller rate of pressure rise and a lower

maximum pressure, and the Iatter is better in the points of a shorter

duration of eombustion and a higher thermal efficieney. In the ca,se

wiehout pilot injection, however, a Tetarded timing of fuel injeetionis practically used at the sacrifice of therma] eMciency for the purpose

of decreasing the va]ue o£ the maximum pressure. On the other hand,the timing of fuel injeetion for the maximum constant-volume degreeof combustion can used in the ease with a pilot injection, becausg the

rate of pressure xise is small. Therefore, a.s a matter of praetice, the

compa.rison of thermal efficiencies is quite difficult. The comparison

under the condition of the same maximum pressure, however, showsthat a decreasg of about 3% in thermal eMciency seems to be una-voidable when a pilot injection is used. The employment of pilot in-

jection of course results in a smaller rate of pressure rise and a Iower

knoeking, even i£ the maximum pressure is the same. (v) As a matter of practice, a device of giving the pilot and rnain

injections "Tith one injeetion valve must be per£eeted, since the useof two injection valves is impraetical. In this case, it is also quite

a difficult problem to get a good pulverizaeion of pilot injeetion for

such a small amount of fuel. It may be concluded that the degree

424 Tamotsu KuRolwA

of pulverization of the smal] pilot injection £uel is the key to the fuelconsumption rate in a practieal engine,

III.' Combustion and Performance of a Diesel Engine with Pre.Combustion Chamber.

1. Preface.

It is generally said that a Diesel engine with pre-eombustion ehamber

is insensitive to the timing of £uel injection. Praetically, a certain

amount of reta,rdation of £uel injeetion does not yield a Iarge effecton the fuei eonsumption rate, The indicator dia.grams of the maincombustion chamber of an engine with pre-combustion chamber alreadypublished also show that the timing of fuel injection is usually adjusted

so that the pressure rise caused by eombustion starts at around theT.D.C. The point, why the timing of injection is not selected so that

the eombustion sta,rts early before the T.D.C. in order to attain a

higher thermal eraciency by inereasing the eonstant-volume degree ofeombustion, has attracted the present author's attention for a long time.

It is also said that a reveTse flow from the main combustion chambey

to the pre-eombustion chamber during the period of combustion oceurs

in an engine with pre-combustion chamber. The points-when this

kind o£ reverse flow oecurs and what kind of effect will result--havenot been clarified.

Consequently, some experimental studies were undertaken at firstin the study of these problems,

In former days, a pintle valve was used a.s the fuel injeetion valve,

but it has been replaced by a throttle valve since about 1940. Theadvantage of using the latter is in the possibility of z'edueing theknoeking by throttling the initial amount of injection, This advantage

can naturally be expected in the case ot an engine of direct injection

type. Ilowever, exaetly how is the throttle valve, and how is theeombustion proceeding ? These points are the second item to be studied.

Above in the section of Study I, the infiuence of eetane vaiue upon

engine performance was discussed in detail by using the experimental

results obtained by operation of an engine of direet injection type.

The infiuenee of cetane value upon the combustion and the perfor-

mance of an engine with pre-eombustion chamber will also be ex-p}aine d.

Experjmental Studies on Some Combustion Problems of Diesel Engines 425

2. Test engine and experimental method.

The engine employed for the experiment was of a single cylinder type

as is shown in Fig. 35, where the dimensions of its combustion chamber

aye also given. The dimensions of the pre-combustion chambey andthe connecting passage are almost the same as are usually found inordinary autoinobile engines. The fuel injeetion pump and the fuelnozzle also have the same dimensions as are used for automobile engines

of the same eylinder volume-the diameters of the plunger and thenozzle are 7mm and 1mm, respectively, and the cone angle of thenozzle is 4 degrees. These dimensions were empioyed in order to ob-tain the practical conditions of an actual engine.

'

>ISI. ..

Pick-up

ii

l i'

iii

1-;

oFindieater

x.

l

lix

ll""

!

illOi

Pick--up eCittdteafor

D="o.. n-iLiE%=t6･2s f==,S:g:5sM,,,,dida4

g'l-5･iilllli iji't'l/li'ii,:oll,Z, -,l"l,:igi,74℃-9j.itk,,Rb

XIZ=87L2c: ='o.4olE FuelvalvgX,kh:oEtle

Fig. 35. Dimensions of tbe Combustlon Chamber.

The experiments were undertaken at a constant speed of 1400 r.p.m.

for two kinds o£ eonstant load of 10 BHP and 4 BHP by changing thetiming of f,pe] injection. The incticator ,diagrams of boeh combustion

chambers and the pressure difference between the chambers werereeorded by electromagnetic osw'illographs.

A commercial light oil of eetane value 53 was at first used in order

to investigate the general items, and the infiuence of cetane value was

studied by using seconda.ry reference fuels in the next step. Finally,

a number o£ experiments were performed to eompare the effects of

t

426

a throttie

follows:

valve and

Tamotsu KuROIwA

apintlevalve. The range of experiments is a.s

Fuel valve Load Fuel Test numberIndieatordiagrams

Throttle valve

))

))

))

tp

JlPintle vaive

eJ

4

10

))

))

))

))

4

10

Light oil

)J

SRF 50

SRF 70

SRF 40

SRF 30Light oil

)1

(V)

(VI), (XII)

(VII)

(VIII)

(XI)

(IX)

<xv)

(XIII), (XVI)

Fig. 37 a

Fig. 36 a, 36b

Fig. 38 b

Fig. 36 c

Fig. 38e

Fig. 38 d

Fig. 37b

Fig. 36 c, 36d

A series of test number is assigned for convenience in explanation.

3. Experimental results and discussion.

The eurves of fuel consumption rate for different timings of in-jection and ignition were obtained for each experiment, and a eom-

parison of the' per£ormance was undertaken. Further, the appa.rentcombustion progress was computed from the indicator diagrams ofmain combustion chamber, and the origin and the proeess of making

a difference in performance were elaxified. Some o£ the indicatordiagrams of both combustion chambers were analysed to obta･in thecombustion progress in eaeh ehambey, and a detailed combustion pro-

gress was obtained.

Performace eurves

Indicator djagrams and eurves of combustion pro.crress

Curves forexplanation

10 BHPFig. 36 Ligh't oil

10 BHPFig. 38 Throttle valve

4 BHPFig. 37 Light oil

Throttle valve

Pintle valve

SRF 70 ,, 50 ), 40 ,, 35

fThrottle valve

1?intle valve

VIXIIec

XIIIXVI'･-

VIII'rc'

VII

XIIX"･

v÷:-

xvee

".",."-, II

",".".". 2j

i'l

il------i--- 1)..,...","･ 2i

Fig. 36'

Fig. 38'

Fig. 37'

ac indieates the eurves of combustion 'progress in eaeh combustion chamber.

Experimenta!StudiesonSomeCombustion?roblemsofDieselEngines 427

Besides, for the purpose of convenience in eomparing the indicator

diagyams and the eombustion progress, figures of explanation are pre-

sented for the maximum pressure differenee between the two com-bustion ehambers, the period of reverse flow, the maximum eombustion

pressure, ancl the eonstant-volume degree.

The exp]anation o£ the results will be given by following the groupheadings shown below.

(i) Influence o£ the timing of fuel injeetion.

(A) Performance curves (eomparison to an engine of direct lnjeetion

type).

The eurves of fuel consumption rate, exhaust gas temperature andthe angle of ignition lag for various timings of fuel injection are shown

in Figs. 36, 37 and 38. When these results are compared with thoseof an engine of direct injection type (Study I, in Fig. 14) with thesa.me cylinder, revolution, and load, the following diseussions can be

derived:

(a) The ignition poin't (of the main eombustion chambey) for the

minimum fuel consumption rate is considerably retareded in comparison

with that of an engine o£ direet injection type, and it is located atabout 10---40 before the T.D.C. Detailed discussion will be given later,

Since the angle of ignition lag is usually almost the same ox slightly

less than that of an engine of direct injection type, the timing of fuel

injection for the minimum fue} consumption xate is also retarded.

(b) The influence of the timing of fuel injection upon .the fuel

eonsumption rate is geen to be rather larger than in the ease of an

engine of direct injeetion type when the performance curves o£ thetwo types are compared. Therefoye, the possible range of the timingof fuel injection seems to be narrower for an engine with pre-com-bustion chamber, but a detailed study will prove that this is not really

true, In the case of an engine of direct injection type, the point of

ignition for the minimum fuel eonsumption rate is located considerably

ahead of the T,D.C., and the gradient o£ the fuel consumption curve

increases rapidly a£teT around the T.D.C. ' Since the pressure rise becomes steep and the knocking becomessevere when the timing of fuel injection is set for about selected at

around the point of minimum consumption rate, the timing of injectionis usually regulated so that the ignition occurs at axound 20-30 before

428 Tamotsu KuROIwA

ce/fghIOB.H.P.'Fue[CeniumFtlcnRete

vs,deP giq"hgPe/･otoFln/ect/onvS,lgAltienPalnt(MeinCo#,b.Chernbet)

25oebt

[400r.prm.

'

a

2uae

h4.

rp-1

n.t'"fAf'

･dXLo .x '

"kh Yc 7 Np

t.

za

'

2e

oe

,,v3.eo

27oo

ocA-

2v'

l5

-2F-2 l5-10-5O 10r/ v

-t

EnhtugtTemperetvie18o

llgnltionLog

"O-t-ys.Vmin

--"---

Cemb,CSambet

vs.Pre-Comli,ChambeE'"

,bNX. .-n-a[D

so

ue

42e

4oe

80

60

ne'

.li･": n.

@-pt-@wa

Fuels

LightOilrht.ttl

Veeves.P..F50

LigntO"P/ntteVo[ve

io nSitEg::Ne,

s-A--.--di--

's'

-----i ra "---Je-25-20V5dO-5o5OI

Fig. 36. Fuel

and

C.FiNt Angle :n

Consumption

Ignition Lag.

JFetb--

Rate, Exhaust Temperature,

"%,rh-

･3SO 4eH,p.

et

1400r･p･m.

FuelCensumpL/onReLe

vs.Ig:;t;enPeintys,beg!nr,FngPotntoF14/ection(Me/'nComb.Chembet)

'

34ee

6o=

x.(*

t6.Yx

't

'

n

o/

-"pt--.

-.".

/

-25-20-t5'-10-5e5

J3"e

3200

31eo

ec.A･

20

t5

fo

5

o

lgnlt/anLeg

o/-

Exh:"tTemperetutee

et"-p"dnetmvi.Mn/nCemb.Chemtet

"v:,PtemConb.Chamber ip----mot

220

200

lge

-JO-25-20-i5-IO- OStOt5Fig･

Crofik Ang(e ln deq[ees

37. Fuel Consumption

and Ignition Lag.

Rate, Exhaust Tempevature,

Experimental Studies on Some Combustion Problems of Diesel Engines 429

the T.D,C. Therefore, the £uel eonsumption rate increases rapidly, i£the timing of £uel injection is retarded £rom that position even a little.On the other hand, in the case of an engine with pre-combustionchamber, the point of ignition for the minimum fuel consumption rate

is placed at around 10"-q-･40 before the T,D,C., and the pressure rise

around that point is not very steep as will be seen on the indicator

diagrams. Therefore, the timing of fuel injection for 'the sake of

attaining the minimum fuel eonsumption rate is usually used £or thepractical purposes. Sinee the fuel consumption curve is almost fiat

around the point o£ the minimum fuel eonsumption rate, a small de-viation of the timing of injection does net-affect it. In this respect, the

author believes that an engine with pre-combustion ehamber is usuallyconsidered to be rather insensitive to the timing of fuel injection. One can see on the eurves of fuel eonsumption rate against theignition point of an engine with pye-eombustion cha.mber that they are

rather steep in cases of both early and late timings o£ ignition, Arapid inerease of fuel eonsumption rate for an early timing is caused

by a rapid increases of cooling loss and a reverse fiow from the main

combustion chambex to the pre-combustion ehamber. The detail ofthis reversg fiow will later be further explained. Sinee the point of

ignition is rather difficult to be exactly determined on the indicator

diagrams of an engine with pre-eombustion ehamber, the curves offuel consumption rate against various ignition points were obtained by

determining the beginning o£ combustion from the computed resultof combustion prog.ress, The results obtained in this manner gavea steep slope at the side of reta.rded timing of ignieion. In cases of

retarded timing of fuel inject2on, however, there should be a periodwhen a very small amount of fuel burns gradually at first, and a elear

rapid eombustion £ollows a£ter that as will be seen on indieator diagramsor curves of combustion progress. Therefore, it might be reasonableto take the point of ignition as the beginning point of this main rapied

combustion The eurves of fuel consumption rate obtained in thismanner are indicated by broken lines.

(e) The angle of ignition lag £or a pre-eombustion chamber issmaller than for an engine of direct injeetion type, when it is eom-pared with respect to the same timing of fuel injeetion, because the

compression ratio is high. The angle of ignition lag £or main com-bustion chamber is almost same as that of an engine of direct injection

type when the timing of injection is early, but it becomes smaller when

430 Tamotsu KuRolwAthat timing is retarded. There is a differenee of about 20 in crankangle bet ween the ignition points for main and pre-combustion cham-

bers, even if that difference varies slightly with the kinds of fuel.

The significant point of the ignition lag for an engine with pre-

combustion chamber exists in a larger variation of ignltion lag againse

the timing of fuel injection than in the case of an engine of direct

injection type. This is a result of the increases of pressure and tem-

perature in the pre-combustion chamber at around the point of fuelinjection whieh are larger than in the case of an engine with single

combustion eharnber. These inereases o£ pressure and temperatureare caused by the severe infiow of the working medium into the pre-

combustion chamber. In the cases of pintle valves, the angle of ignition lag eonsiderably

decreases eompared to the cases of throttle valves and the variation

of ignition lag against the timing of fuel injeetion a]so decreases ap-

preciably, because the ignition lag is related to the amount of fuelinitially injected.

(B) General discussion of indieator diagrams.

A series ef indicator diagrams obtained by changing the timingof fue]. injection for eaeh experiment is shown in Figs, 36 a, b, e, d,

37 a, b and 38 a, b, c, d. A remarkable variation of the forms of in-

dicator diagram can be seen in these figures as a resu!t of slightchanging of the t･iming of fuel injeetion. In the ease of an engine of

direct injection type, a rapid pressure rise appears even £or a con-siderably retarded timing of fuel injeetion, but in an engine withpre-combustion chamber a slight retardation of fuel injection causes

a rapid decrease of the maximum eombustion pressure and fina]ly itbecomes not to exceed the compre'ssion pressure. Even in this case,there is not a remarkable increase of fuel consumption rate. Thiskind of matter can not be experienced for an engine of direet injeetion

type, and the present author was rather surprised to notice that the

fuel eonsumption rate was not so bad even under the condition o£ sucha strange indicator diagram. Therefore, an investigation of the origin

of this phenomenon was first undertaken.

As the first step, a number of apparent combustion progresses

were eomputed from the indicator diagrams o£ main combustionehamber, and the general aspects of this phenomenon became almostelear. Thexefore, the conclusions obtained by looking over these series

p.pt.r

!t',:{ce'hDm't2'tkMeT :t,..:abD.,:L?edict"H,Z'',e']'L-IFnv

conOt..,,,.i"@tl

LlghLO:1t:

10eHPe,f4W.rhfuttPilmpen!thli

ae-7an1

f"etNenteTArcttttttr.",Vl

I'

･･

tlixs Nsss@SNtX

iL

,.-r>g-gTlr

.-1'

INNk".

xX N!P;-Pv' ::::t

txLNNxsN

HNshts

'

Apptt-"tCemfuIL,tnPregrtF

:-r:DE:I;rtsn:tmb-r±

N- -H -. -.

.L-i

o L ctanLAnatelndegtees

Fig. 36 a.

@its-'NiptkuPh;ntnd

Z-･:-

1fieT"toiDi-gTtot"ipt,.:;,.Pr--tptTbust'e-ct1"b--@･ Ir-

:t

IJWtiJ

Jt

tt

/MtTnCk!-ts.t

H!lt!:t

lt"kem/sLtJJ

:tILLiSmnx/li'Nkwui'''lt

@tL"l

L-tEltt

tLti]t

ilN

v@xxZLtmb'xxttt

::.-lil

fll'X.-

11stx X@NhLxtl

tttXtXN

it

NlsxxY/1

'

--.OSEtl

iliislltu/c.,

ll

SNjJrmt xtV

/i' s

i."t@

ntX---AL''i-'lm-eele'fr,L1,･-･mo-ks":AppettdcomL"suanptogniXSL''pm)x.sNPt-Pv/Pt""ltO'iStE::;tcS::tilni"xttsx..o-..diPTt:trhktx/lv'n'...s".--".x.Nla

-.-.

"Uss,t tt- ]h s/ .r!b---JE {ve '-

..e2e

:rf:s----L-

xx

t21,/-xt-X''.r.1"---.L--TtrtP--------"--u-t-tpt-L--7uKL.--t-------lt-J---.--

,."k..:.-x--J'- ,.,

-M t/'txpt

.l..J.--------:-------7:

,."'Z't,-..HJvy･1.'-J--H.SWtitx#i-"lstmo)

..M'k-pmfunt.tfi

-.---

-r.../t.-.-u.H."･--T.'Tv :;oMl:.tj:::d･-m

,./"aT.,T-mpoFVvo-ngM-e,Jm,/il,:t[il'[:l,ee;:llr"fl.eb.i

-;J...---Mt:AChenb.tHvt

.-J--"L---------'-'-- pt..te-SLit;efieke'o-t

H=H:-Hv- " i

Cienk AngTe ;n deQrees

Fig. 36 b.

i#lldeO;IF!-enetuadietDTiguaoF&.'i,O.:';Cor@"terctsr-CXIMilna"mbe,

1･lI

,Xst.9,:,lo'..st--tP"trvMsthP

ftit"riou/Je.t tt'ilrttl

--@slir

Nttufrrwls;tieeN1

moNlt

s(@Lsst;x"txX;"NNxXvtttiLNIti2ltltH

'snxDN

NeNsJn

nvt tNLav,.PAppst-ntCeptimUe-ho;

g'ls.,.,,.,.,Sli':I,1':;."."r:Oe'.`k"t:t,,,.,Fuelleltttien:sK<R

-1

1lw

CrenkArrglelndegreef

Figs. 36 a-36 d.

Fig. 36 c.

Indicator Diagram and Combustion Progress.

1

laaKtketh'-pteelZiMtlAtumbei,. lii:

F

@Lt-t nekolaOi-}thmolPt.-whFtlo-CSsmSH

tix

s

Elll)7.'de

opmettTatLTgHtO"1'

1e:,a･,'tl-t9,:s:rtd,FultPLtrssstc"h"l'ictt7mmd

4a"i

@mtNHt hs/x.@s

/t/.FXX@ )m

t.t.:'x/

xiiitd-ig,I'

ma

eeca/mv,llN

N:tihx1

,NNxto

SIV

,N-NEJtt'

pt-M"

tNNN.

Nx,]tLtlli

t/NsKttisNN

il.x .ONstl

ewxNll

'envJl

X(A

Nstx

P.rfditltFptltAltttT:'s

trNxH'AFpitenLCembvtllcnPraptt

S."'asg ' tShs- - -i

x-Im""",t-U-L

"":

moo

./'/

N%s:

m---symam

,Z

.Z'

:--op-,.::---.'

../HJ

tLLL-t

----"m-rrr---J--eH-L-

f.H.-"----'----u..Tt;TtnpeFWoa:ngM-H,LM:nAblnCNsrn"t

T.:T.nt?cSWdilegM"iumlhFr-"temSviL:ptCLidut

Ht:dnbutt:enProsm"fiM-Emtwt

if --sc--s-t...----."".-i- eeo

Crank Angle tn deg,ees

Fig. 36 d.

(s)di 4diavO,ljrihcF

kEpa.A-r

torrmeTcidl /nxLehtO[d,`

-Wetf4eOteht ,Ll@ p-,ntsP,oprtaofPN-tpmfu:lpttu"

:tt

i-

f"PlaptrSe±[Ai".Tny-ef-ti-tse l.be.,o :ttt

LL,

h@N"i1Nh

"

/LlH

lzJ1

Lzx

N@ ti"li x tT

-"id

Nx xlN

xNt1 x@s .ts sxOx.. -N

s@ t" N `ENXIJ

:t2X'tuHXX

L1lt

Xl'

tN

sXljlt

IN

s,tNNIIII

IN'N

NsNtttt Nlxslt

xXXIrx}<,

IXt

'x'

NxtxxXltlx<s

sNsz1

g:AFptv-+Cpmku"p.v"

p.-R.hapt.DJfmx.NL.tr"-e4Atulim

･":Jn

t"

ttsl/Ns vt)'vis tl--SXs

mu

SN

L s

e)

Hi.Cv-S---hasnTT: ri-pelV-inotadim Ms,sasinlitr

"- Ls': tylti:-Me,naLmS"t T

H.Cpmbete:enPmge-riicTt; T"-".olV-,egM"pm,Sn Pro-tdehlatlopC-pmbev

,../.---.x--X.

"-H xt"N u

scoe

tod/i'f"N:'"'+""-{.-.:L=.M--Edwt-nthde.ttp

e-t. I,r.xH=H.SILx.-T--.NT.-=.-

Y- mo.I'sSL---J--t-r-"-------H--------+-TTT wn -"- t4t;----J-----L"Tt

za

ff --"71 tu -----tL-u ! 1 nytt4""eJ---. -J---"--HUU-Lr"-'-

PtTl

Cte- Anote in degtees

Fig. 37 a.

.@ltSktndicatttOr-ptrmofmu1detritpmelthieCkMS"p..eemSv,tiencrt"S-t

za-tlc:"･'-

D

i

If@ssisM:NNt

Xrll

x..y.X/xt""

iL'

vij XN@ssilki-Xt1xXl

ec )t-

sNSXX@N,Ktl

NAII'

lb

itiN XlNitPas- ''"s 7

N;

st

,XNLvetNsl1'N -xNkN

NNLNN.xs

x!1Sg:Appu,rtC"mtasLlptPT"ptrtL

NN L.Pt-Pv:bPei

,w.?e 1" 10 IO

CrankAogieindegrees

Figs. 37a-37d.

Fig. 37 b.

Indieator Diagram and Combustion Progress.

!

i

ExperimentalStudiesonSomeCombustionProblemsofDiese}Kngines 431

of indicator diagrams and apparent combustion progresses will be de-

scribed at first without going into the detailed Q.xplanation about each

of them,

(a) Pressure differenee between the two combustion ehambers.

-

So'i

9L 6o

'f

g/.' 40

t"

xt

tt"ol,

s

ioo

i

..-"t x

×

'O;vedi:"iect;lanErts"ne

ftrshtOR

leo

"t/.ts

N:---k'

@pmr'E-'-D--

fOBHP..ThratEleValva

.s

.A--

i"

"ghton

/tisx,,

tw@-a--or

70BHP

lt -

b.

tsre

di.li"ptl･l;gliil

op5

tl

?slA

G

'C.A,@A/h

ALts

rR"(04

LighLO;1

?,,eSr,

"60at

'- 5o

-A--!iO

am

(gb

su

ts.N

lrXsm"

@A P,..I

pm

l5m'a"o

qNos'L'

io･-=,B

tpmNX

cotnp,a,t:s.goat,)l5

to

ANNN 1

@)CN,i

c

X,K,ll

N

.Efiko-ni

xBas<5o

'vx

NcA

ec

D EEs"`c/-l,--r

'n"rationDFtReve"seFInptli

,th-te2frFLe-ttr-atie(Pf'Pt)mop.;Lqtrcr:-e-'-:

ttih'"nfHetect"i"sttinX

deee N 'R,

N

o

R-M"7t.

eo"r==g4.'

-

co25t.ii5t e O-ISO5

in

BeginningPeintortniection;ndegrees

Fig. 36'. CurvesforExplanation.

kes･:.

¥.:t

clL

g

ee

:g af

rk

t{ss

tfi

.e

(gz)

4BH?ThretLleVeVve

LsofeO;l

<Ei9,95-4BHRPinlleL;ghtOil

Valve

nohfeofin,n･bel(olisun,bi.

(7fkc;t)/

hao

e

3es

.

'Ra"geo}ta"t.F-elCon)-mth'lk

(e.,aj:tr,t)

foo

,gs

ooe"

X8

aue

o.e--

tgA

5tx,q

pmurrf

.oo

090

as

ft''

760-o

e e

ePlm-n'

ig64af,

}b).@e (ttMbPFns.4n"･1

Rbwt.

1

se

e1

m/

'

Duxat,'onoS

Reversenont

Fartoelnt'L'ete.ff.ftsAMli,'e,

e

lt-n}nL

e.

p gecR-A}tux,

.

e o

e! eLe#tt"aet e

!

2S20l5 teso252ois te,seBeglvnlftg Polnt ol tniectlen ;n tiegreeg

Fig. 37X. Curves for Explanation.

432 Tamotsu KURolwA The pressure difference between main and preL･combustion chambers

during the combustion period can not become- very large when theordinary dimensions of connecting passage and the fuel injection system

arQ employed, The maximum pressure differences are shown in Fig's. 36', 37' and38', diving the period of eombustion into two parts, initial and later.

It sometimes happens that the pressure difference o£ about 6-10 at,appears just for a short time at the initial part of combustion when

the timing of injection is willfully advanced. However, the maximumpressure difference that appears between the two ehambers is almostthe same or less than the one that occurs durlng the compression stroke,

when the timing of fuel injection is selected in the range usually used.

Therefore, when a usual timing is used, the l<inetic energy of the infiow

from the pre-combustion chamber into the main ehamber is not verylayge, and the distribution of the fuel would not be very good in a

flat type of main combustion chamber, This faet was already provedby Schnickieh(i) after his analysis of the gases tal<en from various parts

o£ the main combustion chamber. The thermal efficieney and outputof an engine have been improved in Japan since several years ago bychanging the shape of main chamber into a t,vpe constx.ueted around

the connecting passages. The fundamental reason of these improve-ments exists in the supplement of small kinetic energy by the change

of combustion chamber.

(b) Combustion veloeity.

Looking at the diagrams of combustion progress obtained from the

indicator diagrams o£ the main 'eombustion chamber, one can see thatthere is almost no difference in the eombustion velocity of the main

part of the combustion proeess even in the cases of strange indicator

diagrams. There is also no differenee in the after-burning. Accor-dingly, there was not an extreme inerease in fuel consumption rate,

even if the timing o£ fuel injection were retarded so that the mainpart of combustion would begin as late as 150-200 after the T,D.C.Of course, a decrease of constant-volume degree by the shift of theperiod of combustion to the side after the T.D.C. as a whole and a

litt}e increase o£ the £uel consumption rate would aecompany. Still,sueh a deerease of constant-volume degree were comparatively smallas will be seen on the curves of v,ih in Figs. 36', 37' and 38'. These

(1) "Die Verbrennung im Vorkammermotor" A.T.Z., Jg 34, Ht 251411940.

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 433

l<inds of phenomena can not be seen in an engine of direct injection

type. These results may be brought about originated by the infiowof very unstable activated fuel into the main chamber, beeause the

£uel is thermally cracked by the high temperature in pre-combustionchamber. The above results are noteworthy features of an engine with pre-

combustion chamber from the standpoint of the eombustion proeess,and they also account partially for the insensitiveness of this type of

engine to the kind of fuels. Sinee the injection of the unburnt 'ffuel

into the rnain combustion chamber in an engine with pre-eombustionchamber is similar to the air injection oE fuel in an air injection Diesel

engine, the purposes have been forwarded only to the point o£ obtaining

a good distribution o£ fuel. However, a design of eombustion chambermust be perfected hereafter which will make use of the eharacteristics

described above.

(c) Point of minimum fuel consumption and reverse-flow phenomena.

It has already been clarified that the ignition point £or theminimum fuel eonsumption rate exists at a point far later than in the

case of an engine of direct injeetion type, and that it is loeated at

about 10-40 before the T.D.C. However, that the fuel consumptionrate would becorne minimum under this condition ean not be expeeted

from the standpoint of constant-volume degree of combustion. For the

purpose of investigating the origins o£ this peeuliair result, the timingof fqel injection was advaneed. When the timing of fuel injection wasforced to advanee, a severe noise naturally aceompanied, but the most

troublesome points encountered were the high temperature of thecombustion gas that flows out from the pre-combustion chamber and

repeated failures ln respect o£ valve leal<age eause･ d by a thermaldeformation at the exhaust valve part, where the gas of high tempera- 'ture hits directly. Another diff}culty in the eontinuation of the ex-

periment was that the same £uel consumption rate could not be obtainedso easily as before when the timing was once advanced in this manner.

This all means that the engine was overheated to a considerable extent.

There£ore, the main part of the reason that a decrease of thermaleMciency originated against the intention of improveing the eonstant-

volume degree seems to be a large amount of cooling loss. This is an

acceptable matter in an engine with pre-combustion chamber whichis accompanied by a violent fiow of g'as,

434 Tamotsu KuRolwA A reverse flow into the pre-combustion chamber caused by thehigh pressure produeed in the main combustion chamber during theperiod of eombustion is also a matter to be considered. The reversefiow occurs only at the time when the timing of injection is advanced

and it becomes severer with the degree of advaneement of the timing

of £uel injection. The reverse fiow does not occur at all when thetiming of fuel injection is late. The point of the minirnum fuel con-

sumption rate is loeated at around the border where the reverse fiow

begins to oceur as will be seen from Figs. 36', 37' and 38'. The lossof the reverse fiow is eomposed from the thi'ottling loss and cooling

loss that oeeur when the combus'tion gas Teady to work in the main

combustionchamberflowsbaekintothepre-combustionchamber. Whilethe reverse fiow is oceuring the outfiow of fuel from the pre-combustion

chamber stops, and the objeet o£ inereasing the constant-volume degreemust necessarily be saerifieed.

Looking at the diagrams of combustion progress, one ean see thatjust after the reverse flow it becomes usually slow, and that the end

of eorr}bustion is also retarded in some diagrams, The temperature of

exhaust gas a]so pxesents evidence for this faet by showing a tendency

to increase when the timing of fuel injection is extremely advaneed.

(d) Constant-volume degree of eombustion.

The constant-volume degree of eombustion was computed upon thebasis of the following equation by using the diagrams of the apparent

combustion progxess that were derived from the indieator diagramsof main combustion chamber, The results are shown in Figs, 36', 37'and 38t.

Congtant-volume degree of combustion:

rp,aih =: 2i S elelH. rpgia'ela

where

H = total amount of eombustion heat a == crank angle v,i.== rate of lowering of thermal eMcieney at an arbitrary

position of combustion,

== 1- rlelarmi/1- rkl-i

r. == compression ratio at an arbitrary position of combustion

ExperimentalStudiesonSomeCombustionProblemsofDjeselEngines 435

vth == theoretical thermal eMciency

== O.580 (for T= 16.25, R=1.8)-ll :=:1.311.

For reference, the values o£ eonstanVvolume degree obtained inthe section of Study I £or an engine of direct injection type under the

conditions o£ the same output and fuel are tTansferred to these figures. Comparing these two curves of constant-volume degree, one cansee that the gradient of the curve is not steeper for an engine with

pre-combustion ehamber than in the case of an engine of direet injeetion

type, when the timing of fuel injection is retarded £rom the positionpractically and usually used, This is also one of the reasons that an

engine with pre-combustion chamber is insensitive for the timing offuel injection.

The comparison between these two types o£ engines at the positionof fuel injeetion practically used also shows that the constant-vo}ume .

degree of an engine with pxe-combustion chamber is considerably lowerthan that of an engine of direct injection type. The thermal efficiency

oi an engine with pre-combustion chamber is worse than that of anengine of direct injeetion type because of the losses of throttling and

cooling as well a.s the lowering of the constant-volume degree o£ com-

bustion, However, the lowering o£ constant-volume degree should notbe understood a.s an unfavorable result, since it was caused by re-

tarding the beginning point of combustion but not by the increase o£after-burning. Looking at the series of the diagrams of combustionprogress, one ean see that a rapid pressure rise appears at the initial

stage of combustion and the cooling loss increases, when the timingof fuel injection is early, and that a worthless Iowering of eonstant-

volume degree oecurs as there is a part of very slow combustion atthe initial stage of combustion, when the timing of injection is retarded.

It is also obvious that the combustion proeeeds at a uniform velocity

from the beginning to the enct in the neighborhood of the pQint for

the minimum fuel consumption rate. This kind of combustion maybe deemed as preferable for an internal combustion engine. The

problem o£ advancing the period of combustion withont failing tomaintain this sort of good,progress of combustion is still remains tobe settled.

(ii) Effeet of cetane value of fuel.

The experimental results for various cetane values of fuel areshown in Figs. 38, 38', and 38a, b, c, d. Fro'm these results, the fol-

'

436 Tamotsu KuRolwA

'

tmp(e;=t'".h,F.7oee'iOewPolMOOtrm

/t'h@rndSeelertwof''XlM-iqpa"r''t'ndxt.e#tolifa.moF,'lt' ie--

./±:i,t.

F".t"mlivdllopcumblt,t- ilI

.ct+

.efL' .,:x.k:'

FuelPL'nveosth

dt=7.mmt'Fut4NotittMrettlet

Sinm,-'

r.-Jftsiit

-xof

/@--NNsNs1IOXNNI

iXStN@"xtxt

]' N.O!snc'1xxl p=.

ll-

''

y.

T/--t-ulsN

s' Jt

m) INtstNtxi"oet

lt

eA"LSLINtNtNt

m,YtNINttNxtN]ttsth

J'

"tuxH:NsN

PIFIelSs

-t"AtptTtneComEuttlafiPtogni

rktrJet.DllF-tt-{-belww

asCtssmb-"

-hLMH

-'-Lli-.slro - st- v x lssis-s--'

tt

"ssstamMselF-Itth]rLe"t

h-- -

tu'ro IlTIxttca'e'r' e7oHlt".bl･

ntt

sw ./･

/H.ff.Z-tt.t-..----

if

,LJ.H.---J

l.-''-'

ff-

tt

"-J

-k'e'

/-J-Jr--";-----P--!---4-J

"-±-tJv;T.:Tva?,eiWdir"o･NdiSLm

-'S'mr4Mn;riCha}.f'rv:Ttmp.oFVol-hnM"izamtn

.P"-tnboltlaneslaveee'

riH.:CwhTLkuP-tht"la

h whCN,abvl

sua.ie "im n H'="t+HvCrank Anela la eegreet

Fig. 38 a.

ffji"' triJTcxtorD;tst;mclV;L,-,

tl rnv J@@ MslnC-tmL,r -

-.tnditetabiig#.=oT /i!.i .tt '

S,R,F.50f ,-hxN(B) Pt--coqbust:e"Chtmbtr "･ --

t'

'

i-t

'

'fOfihPeH400rfmtt ･/ t - m.".t

;:7f-'elfameea:cht @ A

-+ )::::dt-7ny.･t .

SIO ' utfi,etnettleMotttt1

I..t4't;x tx@ tli o tt

.s-

: ltNlNN Nx@ .sxO =v.is -f , , '

tL?HV,:.Y};AVJ 1HN'

N

xN'nvNxtt

Nslss! NslKNli

N/

-sssN

xN

t" -Jl t NNrfX.

"'sltttLrX

rmts;i!Nxlt

x

x

N":Xa-n:NN

Appt"thtCentSuit:eth

Pftst:"D[FF-ttnt-

'ktl"h eeLEChzaLks

P.epts

l･

'''xtym'Ktt xs

Ns=

xPitidiblS

ts

N/･

/ ', tx

F"tllnl,ctteAk

's-fi. i- vs ' ･:.". . x

"-ioo la v

1 1l

erandKAneletndggrees

Fig. 38 b.

Iowing diseussions can be derived as to the effeets of eetane value.

(a) Indieator diagrams and the pressure difference between the two combustion chambers.

The lines of pressure rise caused by combustion in both combustion

chambers approach very near in the case when a fue} of high eetanevalue is used, and the two lines are somewhat apart when the eetane

value of fuel is Iow. There£ore, the pressure difference between thetwo chambers is comparatively small in the former case, and it is com-

paratively large for the Iatter.

In the case of a high cetane value, the initial rate of pyessure rise

in the pre-combustion ehamber is slight, since the ignition lag is sma}l.

If the pressure rise in pre-combustion chamber is gradual, the pressure

Experjmental Studies on Sorne Combustion Problems of Diesel Engines 437

1

va["tatefnytme ua !f@te rN-ka"rp"r:ttt

r.eMthtcrlptltfi-

Pt--aelu"htcum-L-t"pa

dy-deJ@ t･-y.@ Ftttt

'

.x l-S.R.F,4ou ''

'I1ee,- t@ t-

@e ".::.

':f"afiHP"ld!odo-

fdik'vvSlth

x11 11i" N)v1

l'k, r'tt

d,h,@ r"-m./ ,,t i

;.

TX::

xlku4t11z'

xzlLrm-xtti

ErpsjA

tl Lxx@xLj

.' x.･--.Ots'1LLHtJ

maL1lttt tt-

L

tt 't-

1'.te=Ss'ss

nt

--h-.-tHtAp"in-lee-Sinules-qut

..p.-athatu"tt..-LdJ--e=thCla!l"

'-rd ' tLle -. i ' i .ro,Fi.ILe"l:on' ' tt

M.''

Ftt't

J ' ' ' r

--..w N /rev

'td te v-"CtatN-Pehdegtees rt

Fig. 38 c.

rn-laletPitgtiqoFl

tP,..,MtineSdthS-4,LXkiaw. ' t

-.

,".

-

:'"i

.:-"':f:: 1-

pa -.f(

,, {st@XlIE11lexl

fi,.ziH.ILs{Li4

"'w-xt'

e/el

sx'sltNIJ

s/i"Nx@stl1!NLpLl

iw

t'xllx@iENNt'NtEiMXd.h@.tNx-:hNltlNl't":

an

nv'"iHe--t.1t" 'sN

ssltt.jN":AppttthLCon.bpttlcnPthpttiL.za1xt'..Pt'Pv-rltss"i-DirFttnnce

PttTdibitwd-nea'thenmS"nLIP)olF--ttnittk:enX

tx

'

' "r lxis/ftt

t" .-m

'lt

' .t-s----HS-

,.O

'' X.1M.t.t.tt

Ht''

ss.Lt-･----

"IV.-Jl

utu

)s)'-r.

xfiSsxIl''.L..±IX--ttttUT-t

et1"'`.Jts.,/..

bM,,/

.17"''xxptttl'---fSi--H.t--hf'

v'NH.-h-.thTt:Tt.thp.eSVetLTnlM.dlunt

lndei,"Ch-mbtr

N;TT,T-MEolWett-gM,drupttn'SLpii-{emL-ilTa-CHemett

-x--!Hg-+--N----t

x---

L.-:L--tV6"MEUStt.''--s[.fi

emo./'HI:moo

.1fi'H,:q･tdn-

Z'"""""'-.gxtL.L---"lvr-L----t-.-".to

re--Combe±L;tnProv"stnn±:nCStm-tComUtLl"ntw"r:Fn-{"mbo-tie"ChtmStt

H="""vk-t.--N=-

.,v-we14 Wt;tOtO'Jt･Jpa' ---+r.".-----:t CTenk Nngle Tn degfees

Fig. 38 d.

differenee between the two chambers must be small. Furthermore,if the eetane value of a fuel is high, its ignition property and com-

bustibility may be good even at the time when it flows into the main

combustion chamber and the combustion will proceed continuouslly.Consequently, the indicator diagrams for these two chambers naturally

have a tendency to approach eaeh other, and the pressure difference

between the two chambers becomes small. On the contra.ry, in the case of a Iow cetane value fuel, the pre-

ssure rise in the pre-combustion ehamber is quite rapid and the pre-ssure difference beeomes eomparatively large.

The initial rate of pressure rise in the main combustion chamber

becomes almost the s?.me as that in pre-eombustion chamber, when

438 Tamotsu KuRolwA

trjptIPIh

ThrettTeVol"e

26ce

250

10B.H,p.

et

ltLoot.prm.

FuelConsumptionRcte

vs,atginn[ngPoinLeFhitctio"

'

vs,rgn;tlehPelnt

palnCcrnb,Chew,bo) tbe-

""s"k

zb"k

Z''o

o-2goo

-

s

uo' --etttt

xxVeb sgg

so

--

-..-"-x"

-25-20-t5-iO-5e'510l5tgeitTonLeg AV eC

Exheu!tTempeEt±uee

t-t

5oa

23eo

22oo

felo

15

ID

5o

" o nyg-t-

esD

eso

..in-e' '

, 7-t,-'a-@

--vs,M:inCtmtsiCh"mbor

vs,PtemCemtsCHembet"t-/-

F"els

SRF70SRF50SRF40.

SRF351

44e

420

-oo

3SG

.

'

-30-25-20-5` -･-IO-SO5IO.,5 Ctevx' Ao"t･e :ny degJEon

Fig. 38. Fuel Consumption Rate, Exhaus't

Temperature, and Ignieion Lag.

4- .G

'f'

･g･

"g

g $ E･

gg

B･

ees,innlnsPein:cFtnkcUo#i"doptees

Fig. 38'. Curves for Explanation.

the connecting pa-ssage between the two chambers is comparativelyIarge as is the usual case. The latter part of pressure rise in main

combustion chamber diffeys by the pyogress of combustion, and the

maximum pressure sometimes exeeeds the maximum value of thepressure in pre-eombustion cha.mber in ce.rtain easqs of fuel and in-jection timing.

tu.L-.Gietti"jetfithTvpe

7. sRF?Oi@,ee reQ

'

?t.xiiSnt4j

!o.o

ops-

'. t... 1-

t

']lscSlo[[iiH)p-.-

'x toeHplhraLTIfclee

'," K'rt,v,ss)tOfiHP

Thrott:eVaVtil

SRiSO

-''Thro"leVINR

in'fj5

･toe

ase

ta

ts

SRf701

"'

. ts･

.

ijm

aos

mo

QSs

oso

nj

as

as

e,

'1 5 '

.Rangeorpt,-n.Fqet

tss-n?t-n-

.

o e

o

e

.AOOS

c

mo

oPt"U,-ttt

.rt-hftAt

-i"-,-rl'ft.lat,Ptase-afF

'a@ A.. 'C4.

fs-pt-i

ose

Za

aO

Ehni

is so So 15

4

'i'

m 4e le

af

eet n

-o

al

lo

eo

a

t 5 sE.e

-di I

P"rah'oneSRevefieFidbl

ts

ona'' s

E

EFarh.-n)sueCStrrtnyuny.Pn.ttsJ

o[ of

tA-A)...

f

:

[(2v FGS,)tu4

7cfi.Pi)ri,t

t s s.

.

t''Lf",:;,::1'.S:-e`,･

. ,' '

.

'

% se?s?osstos o?orf 2S20 l5e

ExperimentalStudiesonSomeCombustion?roblemsofDieselEngines 439

(b) Reverse fiow.

The change of the duration of reverse fiow in correspondenee to

various timings of fuel injectiQn is shown in Fig. 38L A reverse fiow

does not oceur in the case of a late timing of fuel injection, In the

case of an advanced timing of injeetion, the duration of reverse flow

rapidly increases for a £uei o£ high cetane value; i'ts inerease is notso large for a fuel of low eetane value. When a fuel of SRF 35 isused, the reverse fiow did not occur even if the timing of injeetionwas considerably advanced. As will be sgen on indicator diagrams,in the ca,se of high eetane value, the pressure rise in' pre-eombustion

chamber and the pressure difference between.the two chambers arecomparatively small and the pressure resulting from combustion inmain chamber easily exceeds the pressure in pye-eombustion chamber,

For a fuel o£ low cetane value, on the other hand, the pressure inmain chambey can hardly exceed the pressure in pre-combustion cham-ber, since the pressure rise in the latter is very large and the com-

bustibilityoftheunburntfuelintheformerisnotverygood. ,

(c) Apparent combustion progress,

The comparison of the diagrarns of combustion progress for £uelsof various eetane vaiues shows that the combustion velocity of the main

part o£ combustion, after the initial rapid one, is large and thae thecombustion finishes early, when the cetane value of fuel is low. These

results may seem to be against the ignition property and the com-bustility. However, the combustion velocity was promoted because of

the Iarge pressure difference between the chambers, which resultedin an in.crease of fuel infiow into main chamber and of the possibility

of distributing the fuel into air. Also, the duration of reverse fiow

governs the combustion velocity of the main part of combustion whenthe timing of fuel injeetion is ear]y. In the case of a fuel of SRF 35,

for example, the combustion velocity at the main combustion is appre-

ciably large and the eombustion finishes early, sinee the pressure

differen¢e between the two chambers is large and since there is noreverse fiow. However, in the case of SRF 70, the combustion is slow

and the constant-volume degree is also small.

(d) Performance curves.

The comparison of performance curves is shown in Fig. 38.

The main reasons for the higher £uel consumption rate in the use

440 Tamotsu KuRoiwAo£ a fuel of high cetane value exist in an insuMeient distribution offuel into air due to the small pressure difference between the twochambers, a low combustion veloeity, and a low constant-volume degree,

In the case of low cetane value fuels, the pressure difference be-

tween the chambers is large, and the distribution of the fuel is good.

Also, there is not a large reverse fiow in this ease. Consequently the

constant-volume degree of combustion is high, but the flow of ga.s be-

comes severe, and the cooling loss increases. Thus a tendeney ofinereasing the fuel consumption rate xesults. Especially when thetiming of fuel injeetion is early, the combustion beeomes very rapid,

and a vibration of gas in the combustion chambers appears (especially

in the main chamber) and the cooling loss extremely increases. There-fore, the fuel consumption rate increases very rapidly.

A fuel of about 5e cetane value is the case between these extreme

ca-ses, and it gives the best fuel eonsumption rate. The reasons may

be that the size and the form of the pre-combustion chamber and thesize of the connecting passage employed in practical engines weredetermined by testing with a commercial light oil of about 50 cetanevalue.

In the case when a fuel of high cetane value is used,.it .is..neees-

sary to make the connecting passage small and the pressure differenee

between both chambers rnust be large. The distribution of fuel inmain combustion chamber must a]so be good in this case. When a fuelof low eetane value is used, on the other hand, the conneeting passage

must be large and the pressure difference between the two chambersmust be small so that an extreme increase of cooling loss would beprevented.

(e) Comparison of the point of minimum fuel consumption rate.

As wa.s described above in the section of "Infiuence of the timing

of fuel injection", a late timing of fuel injection Iowers the constant-

volume degree. If the timing of injection is advanced in order toincrease the eonstant-volume degree, an extremely large cooling loss

and a reverse fiow oecur, and the purpose of improving the constant-

volume degree can not be effectively attained. Therefore, the timing

of injection for the minimum rate o£ fuel eonsumption is not loeatedat the position of the maximum constant-volume degree, but it is plaeed

near the position where a reverse fiow begins to start. Furthermore,

the pressure rise resulting from combuseion is not so severe for this

Experimental Studies on Some Combustion Problems of Diesel Engines 441

timing of injection, so it can be used foic a practical operation. This

best timing of injection is almost the same for all cetane value, and

the same timing of injectlon can be used even if that value of fuelis changed.

(iii) Throttle valve and pintle valve.

The rate of injection by the fuel injection valve was as shown in

Fig. 39 at the fuel pump revolution of 700 r.p.m. for the amount ofinjection equivalent to 10 HP output, when the valve was tested by

injecting a £uel into open air. The rneasurement was performed byreceiving the injected fuel in a number of thin ceHs arranged around

a dise running with a high speed.

,E

vv

I:-

Jg

Lti

ehrm

r.p.m. oF Fuel Pump;

Arrvount eF laiection;

Type oF Fuel Pump;

,

Length oF Fvel Pipe;

Fuel Velve;

l!Ixl40o

Equ;velent to FuIl Loe'

Bosc}b dp=7mm

750mmthrottle velye 1rnm, e=!20 bt.

pintle velve lmm, e=l20 at-

Nxx '

V-PintleValve

NXN Ax

,ThrottlVelvex

/

'

N '

Crank Angle ;fi, degrees

Fi.d. 39. Measurement of Fuel Injection Rate. (Injeeted into Open Air)

The comparisons of performances of an engine with throttle valve

or pintle valve are shown in Figs, 36, 36', 37 and 37'. From theseexperimental results, the following conclusions can be derived,

(a) Initial stage of combustion (rate of pressure rise and maximum

pressure on indicator diagams).

The effects of a throttle valve evidently appear at the initial stage

of combustion in a pre-combustion chamber; both the rate of pressure

442 ･Tamotsu KuROIwArise and the maximum pressure are low. Howevey, since the periodof fuel injection is usuaily longer (350---400 cran}< angle for the full

load) in the case of an engine with pre-eombustion chamber comparedto an engine of direet injection type, the pressure rise at the initial

stage o£ combustion is originally not so large. In Fig. 40, the points of ignition in pre-combustion chamber with

a throttle valve in Experiment (XII) and those with a pintle valve in

gg l･-

/6

£.

O 10 20 30 40 Crank Angle in degrees

Fig. 40. Ignition I}'oint noted on the Fuel Injection Curve

intheCasesofExperiment(XII),(XVI). ' 'Experiment (XVI) are indieated on the curves of fuel injeetion rate,

One can see, in this figure, that the amount of fuel injected before

the ignition is about 1!5-113 of the total fuel to be injected, which

is not very large. It may be worth-while to note that the amount of

fuel to be injeeted before the ignition is almost the same in botheases at point C where the fuel consumption rate is minimum. Thisunreasonable result was obtained perhaps because the test was carried

out by injecting the fuel into open air. In the case of a praetical

injection into a cylinder, the effect of throttling o£ a throttle valve

Since the sectional a.rea of the conneeting passage between thetwo chambers is compara'tively ]arge and the pressure rjse in pre-eombustion chamber promptly appears also in the main chamber, thepressure rise in the latter is rapid when a pintle valve is used.However, the initial rate of combustion in the main ehamber does notdiffer in cases of both a throttle valve and a pintle valve as will be

seen in the diagrams of combustion progress in each chamber.

(b) Latterstageofeombustion. When a thrbttle valve is used, the 'combustion after reaching the

FgfPsaFE

A/i/NAY

x-Tr.mn

,c

cDt xN-LS A-Nx

xl

x

.t Nx

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 443

point of maximum pressure, which oceupies more than 213 of the total

fuel, gives a ]arger pressure difference between the two chambers, a

larger rate of combustion, an earlier end o£ combustion, a highereonstant-vo]ume degree, and a better thermal eMciency. (ref. theapparent combustion pro.crress in Figs. 36 a, b, e, d, 37 a, b, and Figs, 36',

37,).

'' (c) Comparison at the point of the minimum rate o£ fuel consumption.

The point of the miniinum rate of fuel consumption represgnts agood measure for the comparison between the two eases, sinee thepressure xise at this point is no't so xapid and the timing of injection

for the minimum fuel consumption rate can be practically employed. The ignition points for the minimum fuel consumption rate at full

load are about 40 and 10-20 before the T.D.C. when a throttle valveand a pintle va}ve a.re used, respeetively, As the pressure rise is rapid

and the eooling loss is large when a pintle valve is used, the point of

ignition must be retarded. In the case o£ a light load, the coolingloss does not ehange very much even if the amount of fuel is de-creased. Therefore, the point of ignition for the minimum rate oE fuel

consumption was retarded in comparison to that of full load, and the

ignitioR points for a ]ight load o£ 4 BIIP were found at about 10 before

the T,D.C. for both types of £uel valves.

A pintle valve gives a worse £uel eonsumption rate, whieh comes£rom a large eooling loss and a lower eonstant-volume degree ofcombustion.

Fig. 41 is given £or the purpose o£ eompariAg the £uel injeetionswith the two valves by arranging the diagram of injecting rate of

'

.6 vU ･2- -- fi= tr o g

-2o -io 7:p,C 10 20ec,A. CTank Angle ln degTees

Fig. 41･ ?eriod of Fuel Injeetion and Ignition Point of Min. Fuel Consuption in Experiment (XIO, (XVI).

AAkNN

tN--L.J N(xvl)-e

CXII)-

Nxx

x

444 Tamotsu KuRolwAthe two types of valves so that the beginning points of fuel injeetion

and ignition would coineide with those for minimum fuel eonsumptionrate. One can see, in this figure, that the end of fuel injection of

a pintle valve is later than the one of a throttle valve, As a result,

the constant-volume degree and the thermal etlieiency may havelowered, when a pintle valve was used.

As will be seen from Indicator Diagrams (V)-D in Fig. 37 a and(XV)-C in Fig, 37 b, the point of eut-off of fuel injeetion for a pintle

valve is also later than that for a throttle valve, when the output is

4 BHP. Furthermore, the difference of fuel consumption rates for these

two cases beeomes larger when the output is small, beeause the per-

eentage o£ cooling loss beeomes large.

(iv) Combustions progress in eaeh eombustion chamber,

The combustion progress in eaeh combustion chamber was eomputedfor each three respresentative indieator diagrams with regard to the

following experiments.

Fig. 36 b (XII)

Fig. 36d (XVI)

Fig. 38 a (VIII)

Fig. 38d (IX)

Fig. 37a (V>

10 BHP,

))

')t

'))

'4 BHP,

throttle valve,

pintle vaive,

throttle valve,

)t ' )) '

commereial light oil

11SgeF 70

SRF 35commercial light oil

It is rather diMcult to reaeh detailed eonclusions out of this limited

number of experiments. Still, the following points at least wereclarified:

(A) Combustion in pre-combustion chambey.

(a) The volume o£ the pre-combustion chambe.r oecupies 40% ofthe total clearance volume, and the amount of ai£ contained is suflicientto burn about 70% of the fuel foy full load (1.8xO,4=O.72 with R=1.8),

However, only about 20-30% of the fuel for the full load is burnt in

the pre-eombustion chamber. The absolute arnount of fuel to be burntin that chamber was almost constant withoutregarding to the magnitude

of load. However, a detailed eomparison of combustion progress will

show that the amount of £uel burnt seems to have a tendency toinerease when the output is small because of a large ignition lag.(cf. Indicator Diagram (V)in Fig. 37a),

ExperimentalStudiesonSomeCombust'ionProblemsofDieselEngines 445

(b) The eombustion in the pre-eombustion chamber is governedmainly by the ignition lag.

In the case of an early timing of fuel injection, a rapid combustion

at first occurs beeause of the large ignition lag, and the fuel injeeted

thereafter into the pre-combustion chamber fiows into the main chamber

with the lowering-of pressure in the main chamber without undergoinga significant eombustion.

In the case oE a retarded timing of fuel injection, the initial com-

buseion is less rapid beeause of a sma}1 ignition lag, and the amount

of the total eombustion in the pxe-eombustion chamber usual!y be-

comes small by a gradual progress of combustion.

When the timing of fuel injection is properly selected, the eom-bustion rate is also proper and the total amount of combustion in pre-

combustion chamber is usually large. In the case of an extremely advanced timing of injection, however,

a severe xeveyse flow usually oceurs at the same time and a decrease

of total amount of combustion in pre-eombustion chamber is generallynoticed. The reverse fiow seems to have a trend to restrict the com-

bustion in pre-eombustion ehamber.

(B) Combustion in main eombustion ehamber.

(a) The beginning point of combustion in main chamber appea,rsabout 20 in crank angle after the ignition point in pre-combustionchamber. There should not be an ignition lag as large as 20, because

the working medium with unburnt fuel flows into the main ehamberas a fiame so that the eombustion would proeeed continuously. Thisdiserepancy should perhaps be attributed to the position of the pick-up

of indicator. The pick-up arranged for the main chamber was placedat the side opposite to the conneeting passage. If the sound velocity

of 750mls is assumed for the propagation of pressure wave in the gas

in cylinder, the duration of traversing the cylinder should be 1.220in crank angle at the engine speed of 1400 r.p.m.

(b) There aye two clearly different stages: initial (period uneil

the maximum pressure is attained) and later (period after the maximum

pressure) in the diagrams of apparent eombustion progress computedfrom the indicator diagrams of main combustion chamber. When the

timing of fuel injeetion is early, the initial stage o£ combustion isquite rapid, contrariwise it is very slow when the timing is retarded.

In the case of properly selected timing of fuel injection, there is no

446 Tamotsu KuROIwAdifference in the Jrate o£ combustion in these two stages and the com-bustion progresses uniformly at a proper rate.

An apparent eombustion progress includes the combustion in pre-

combustion ehamber indirectly as well as the combustion in mainchamber. The actual combustion progresses which proceed in the ma.in

chamber are shown on the diagrams of combustion progress in eaehchamber. It can also be seen, in these diagrams, that the rate ofeombustion is almost the same in these two stages of combugtion and

that the combustion is proceeding at almost the same rate £rQm thebeginning to the end o£ eombustion. Also, this eombustion progress isalmost independent of the timing of fuel injection and the cetane value.

(When the timing of fuel injection is ]ate, there is a period of very

slow combustion at the initial stage of the combustion progress in main

chamber. As will be judged £rom the loeation of the ignition pointindieated on the diagram of fuel injection rate already shown in Fig.

40, this period oE gradual combustion must result, because the ignition

occurs mid-way of the throttling of the throttle valve and the rate

of injeetion is quite small £or a little while. A£ter the period ofthrottling. a similar combustion o£ high velocity also appears), The fact that the combustion rate in main chamber is al.ways

almost uniform without regard to the timing of £uel injeetion is animportant feature of an engine with pre-combustion chamber.･In the

case of an engine o£ direct injection type, the combustion becomesvery slow and the after-burning beeomes severe when the £uel in-jeetion is retarded, because the combustion oceurs at a eomparatively

late stage of expansion stroke where the temperature and pressureare low, However, the fact that the combustion rate is almost thesame even for a retarded timing of injection in an engine with pre-

combustion may mean that the fuel enters into the main combustionchamber at such an activated state that a certain lowering of pressure

and temperature can be neglected. It can also be understood that the

fuel gathered at the narrow top of pre-combustion chamber is thermally

eraeked by a high temperature under the condition o£ insuffieient air,and it becomes to be in an unstable and aetivated state.

The distribution of fuel into air is also deeply infiuences the com-

bustion rate, The good distribution of fuel in main combustion bythe large mixing ability of gas entering into the main chamber canalso be one of the reasons for the preferable eombustion describedabove. A]so, the combustion progxess in main chamber must be related

tt

ExperimentalStudiesonSomeCombustion?roblemsofDieselEngines 447

to the amount of fuel fiowing into the chamber.

The velocity of fue] inflow into the main combustion chamber is

primarily governed by the velocity of injection £rom the £uel valve,

even in the case o£ an engine with pre-eombustion ehamber. There-fore, the combustion rate had to be almost the same when the sameinjeceing apparatus was used. However, the amount of fuel is alsogoverned to some extent by the injecting velocity of gas from thepre-combustion chamber. At the same time, the ability of mixing

with air is entirely dependent upon the injeeting veloeity o£ gas. There-fore, the eombustion in main combustion chambey must be related tothe pressure difference between both chaynbers, The pressure differ-

ence between the two ehambers in the latter stage o£ combustion,where the main part of eombustion oceurred, was almost the same,but it became a little la,rger when a fuel of low eetane value wasused. As a result, the rate of combustion was slightly increased bythis effeet of a little larger pressure difference in this case,

4. Conclusions.

As answers to the problems proposed in the preface to this chapter,

the following conclustions can be derived from the above discussions:

(i) In the ease o£ an engine with pre-eombustion chamber, theignition point for the minimum fuel eonsumption xate is considerably

late compared to that of an engine of direet injection type, It is

nearby at the T.D.C. The constant-voluine degree o£ eonibustiod isbad for this timing of ignition. Even if the ignition point is further

advancedi however, an extremely large cooling loss resulting from arapid pressure rise, a severe reverse fiow, and an increasing iCuel con-

sumption rate ean not be avoided. (ii) An engine with pxe-eombustion ehamber is insensitive to the

timing of fuel injeetion, because the timing for the minimum fuelconsumption rate is praetical]y used as 'the pressure rise by combustion

for this timing is not rapid.

The combustion velocity in main combustion chamber is almostconstant, because the fuel flows into the main chamber in an activated

state, cracked in the pre-combustion chamber and the unburnt fuel is

well distributed by the gas injected £rom the pre-combuseion chamberso that the eombustion would easily be advanced even if the timing d-, -of m3eetion is retarded.

Consequently, the fuel consumption rate should not change very .

448 Tamotsu KuRolwAmuch for any timing of fuel injeetion and an engine with pre-combustion chamber is also eomparative]y insensitive to the kinds offuel,

(iii) The pressure difference between the two chambers duringthe period of eombustion becomes large when the timing of fuel in-jeetion is quite advanced, so far as the sectional area of the con-neeting passage is large enough as is usually employed. However, the

pressure difference is unexpectedly small when the timing ofinjection

is se}eeted at the point £or the minimum fuel consumption. rate or ata point iater than that, and it is almost the sgme or smaller than the

maximum value o£ the pressuire difference during the eompressionstroke. Consequently, in the case of an engine with a fiat type main

eombustion chamber, the distribution of fuel in air is not perfeet and

the magnitude of pressure dfference exerts some infiuenee on theeombustion rate in the main chamber. (iv) Even though an engine with pre-combustion chamber is saidto be comparatively insensitive to the kinds of fuel, some differencein the performances was noticed in a fiat type of main combustionehamber as the cetane value of fuel was changed. Concerning the effeets of cetane value, the fuel consumption rate

was best for a fuel of 50tw55 eetane value and it became worse inboth cases of higher and lower cetane values. This fact seems to result

because a group of combustion deviees sueh as the sectional area of

connecting passage, the shape and the volume of pre-combustion

ehamber and the £uel injection system, etc. were determined by theuse o£ usual commercial light oil for many years. In the case o£ higheetane value fuels, the pressure difference between the chambersduring the combustion period is small and the combustion velocity in

main ehamber beeomes a little slower because of an imperfect distri-

bution of £uel. When the cetane value is low, on the other hand, thepressure differenee becomes large and the eombustion velocity be-eomes high, but the eooling loss increases because of a severe fiow of

gas at the initial stage of eombustion and the fuel consumption rate.

Inereases.

(v) With regard to the diffexence of using a pintle valve ora throttle valve the rate of pressure rise and the maximum pressurewere both low in the pre-combustion chamber as well as in the mainehamber and the knocking was small as was expected, when a throttle

vaive was used. This resuited because the pressure in both chambers

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 449

easily balances and the pressure difference between the chambers is

small, when the seetional area o£ eonnecing passage has a usual di-mension of O.4-O.45% of the piston area, Sinee the degree of pressure rise at the initial stage of combustion

is c}osely related to the thermal efllcieney of an engine with pre-

eombustion chamber o£ large cooling loss, the use of a pintle valvewhich is accompanied by a rapid pressure rise is required in oxder toretard the timing of fuel injeetion and the end of fuel injection must

be late so that the whole period of eombustion will be retarded. There-

fore, the use of a pintle valve also gave the worse result in fueleonsurnption rate.

In view of these considerations, a thrott}e valve must be yeeom-

mended for ,the purpose of giving the ignition in the middle way ofthrottling of a throttle valve so that t･he rate of pressure rise at the

initial stage of eombustion will be redueed, for the sake of controlling

the knoeking and the fuel eonsumption rate. Fuythermore, one ean also summarize the fol!owing eharaeteristics

of combustion in a Diesel engine with pre-combustion chamber fromthe above discussion:

(a) Since the £uel fiows into the main ehamber after beingactivated in the pre-combustion chamber, the eombustion in mainchamber does not depend upon any slight dfference of pressure and

temperature in the eombustion chamber. - (b) The injection of combustion gas from t･he pre-combustionchamber is very effeetive for the distribution of unburnt fuel in the

main chamber, and it is also help£ul to promate earlier completion ofcombustion,

The charaeteristies of combustion in (a) and (b) are the mainreasons of the insensitiveness of an engine with pre-combustion chamber

in respeet to the timing o£ fuel injection and kinds of £uel. (c) Since the pereentage of eooling loss is eomparatively large in

an engine with pre-eombustion chambey, it is better to make the re-

duction of the rate of pressure rise at the initial stage of combustion

foJr the purpose of decreasing the £uel consumption rate, too. (d) As the amount of fuel to be burnt ae the initial stage of com-

bustion is small in order to restrict the rate of pressure rise and the

maximum pressure, the most part of combustion oceurs in the later

stage .of combustion a£ter reaching the maximum pressure. Even ifthe combustion progress in this later stage is genera}ly good owing

450 Tamotsu KuRolwAto the characteristics explained in (a) and (b) without regard to thetiming of fuel injection and the kinds of fuel, even a small change of

pxessure difference between the two chambers also infiuenees the com-bustion velocity, because the usual sectional area of connecting passage

gives originally a sma]I pressure difference and the distribution offuel into air is also imperfect.

The charaeteristics for combustion described above may permit to

derive the following suggestions £or the design of eombustion ehambers: (a) So far as one of the characteristics of combustion in an engine

with pre-combustion chamber exists in the £acts that the fuel isactivated by passing through the pre-combus'tion ehamber and thecombustion in main chamber occurs quite rapidly, the Iarge throttlingloss and cooling loss which are the weak points of an engine vgTith pre-

combustion chamber may be ab]e to be redueed by enlarging thesectional area of the coonnecting passage, The reduction of pressure

difference between the two chambers by enlarging the seetional area

o£ eonnecting passage can be faeed by constructing a eompact maineombustion chamber around the passage.

(b) When the cetane value of £uel is largely changed, attentionmust be paid in designing the combustion chambers. When the fuel

is o£ a low cetane value, the sectionai area of the eonnecting passagemust be enlarged, so that a large cooling loss ean be avoided as far

as possible by reducing the initial injecting velocity of combustion gas

from the pre-eombustion chamber. In the case of a fuel of high cetane value, the main combustion

chamber must be constructed compactly around the connection pa･ssage,

because the pressure difference between the. ehambers becomes smail

and the distribution of fuel in the main chamber becomes naturallyworse. The deviees o.f shortening the period of fuel injection andenlarging the rate of injection will also be a Tecommendable method

for mastering of the Problem of increasing the pressure difference

between the combustion ehambers. ･

ExperimentalStudiesonSomeCombustionProblemsofDieselEngines 451

Closures

A treatment o£ combustion problems in an actual Diesel enginewas undertaken in the present paper based upon the analysis of indi-

cator diagrams. Such analysis ineludes a tedious procedures of com-putation but stand.s on a simple principle. Some actual problems on

combustion and performance whieh are unable to be solved by a funda-

mentai study were elarified. The computation of combustion progress

obtainable £rom indicator d.iagrams can be performed in comparativelyshort time when a well prepared calculation plan is estab]ished. Thiskind of procedure may be one of the most effective devices in investi-

gation the combustion problems and judging the performances of anactual.engine in the manufacturing factories. The most important thing

in applying this method is to have as reliable indicator diagrams as

possible. The possibility of obtaining valuable information is wholly

dependent on the accuraey of indicator diagrams. The author beforeconeluding this paper would like to stress the vital necessity of the

improvements of the indicator for the further development of thestudies in this field.