Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

~.A~ TheEngineering
Resource For
Advancing Mobility 400 COMMONWEALTH DRIVE WARRENDALE, PA 15096

850523

Silicon Nitride Swirl lower=Chamber for

High Power Turbocharged Diesel Engines
Sumio Kamiya,
Mikio Murachi,
Hiroshi Kawamoto,
Satoshi Kato,
Seiho Kawakami
and Yasumitsu Suzuki
Toyota Motor Corp.

International Congress
& Exposition
Detroit, Michigan
February 25 - March 1, 1985
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

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

Papers published prior to 1978 may also be copied at a per paper lee of $250
under the above stated conditions.

SAE routinely stocks printed papers for a period of three years following date of
publicalion. Direct your orders to SAE Ordp.r Department.

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

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

ISSN 0148·7191
Copyright © 1985 Society of Automotive Engineers, Inc.

This papl'r is suhjert to revision. Statements and opinions ad-
vanl'l!d in papers or disrussioll arc the al:thor's and arc his
responsibility, not SAE's; however, the paper has been edited
by SAl for uniform styling and formaL Discussion will be
printed with the paper if it is published in SAL Transartions,
lor permission to publish this paper in full or in part. contad
the SAL I'uhlkalions Division.
Persons wishing to submit papers to be considered for pre-
sentation or publication through SAl: should send the manu-
snipt or a .100 word abstrat'l of a proposed Illanusrript 10:
$enctary. I' nginecring Al'livit}" Board, SAL.
Printed in U.S.A.
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014
Silicon Nitride Swirl lower=Chamber for
High Power Turbocharged Diesel Engines
ABSTRACT
This paper describes application of
sintered silicon nitride to the swirl lower-
chamber in order to improve performance of
turbocharged diesel engines. Various stress
analyses by finite element method and stress
measurements have been applied to determine the
design specifications for the component, which
compromise brittleness of ceramic materials.
~laterial development ,.;ras conducted to evaluate
strength, fracture toughness, and thermal
properties for the sintered bodies. Ceramic
injection molding has been employed to
fabricate components with large quantities in
the present '!'vork. Quality a.ssurance for the
components can be made by reliability evaluation
methods as well as non-destructive and stress
loading inspections. It is found that the
engine performance with ceramic component has
been increased in the power out put of 9PS
as compared to that of conventional engines.
SILICON NITRIDE SINTERED ~~TERIAL has potential
for high performance applications for gas
turbine engines or automotive engines because of
its high temperature capabilities, Hear
resistance and resistance to chemical attack.
It is expected to make engine performance
improve by application of sintered silicon
nitride to the combustion systems for diesel
engines, since an increase in temperature for
engine operation will be possible. New
techniques must be developed to succeed in
application of ceramic component to automotive
engineerings.
In this paper, several techniques will be
discussed which are essential for development of
ceramic swirl lower-chamber. including design
methods. material evaluations, fabrication
process, inspection methods and reliability
evaluation methods.
0148-7191/85/0225-0523$02.50
Copyright 1985 Society of Automotive Engineers, Inc.
850523
Sumio Kamiya,
Mikio Murachi,
Hiroshi Kawamoto,
Satoshi Kato,
Seiho Kawakami
and Yasumitsu Suzuki
Toyota Motor Corp.
ENGINE PERFORNANCE OBJECTIVE
The engine employed in this work is 2L-THE
electronic controlled turbocharged diesel
engine carried on "TOYOTA eROHN Il • A cross
section of this engine is shown in Fig.l.
Haximum pOtver out put can be increased by
increasing in the temperature of combustion
chamber as shoHn in Fig.2. The temperature was
measured by inserting a thermocouple into top
end of the glow plug. The inside surface
temperature of swirl lower-chamber can be
estimated to be ~lOOoe lower than that of
combustion chamber. The maximum operation
temperature is limited to ~930oe in the case of
ordinary metal chamber. since it may be cracked
due to strength degradation above this
temperatures.
In this work, the objective of the maximum
temperature of combustion chamber is determined
to be 970 0 e by use of ceramic swirl chamber,
where engine performance in power will be
expected to be increased from 96 to l05PS of
the order of top class as shown in Fig.3.
PROBLENS IN DEVELOPNENT
Sintered silicon nitride is high
performance structural material. but this
material has disadvantages for its brittleness
and large strength variations. This is
attributed to the fact that the strength of
ceramic materials depends on surface flaHs or
internal defects. Figure 4 ShOHS typical
example for failed component after engine
durability testing in early stage of this
development. Therefore, it is necessary to
approach from vieH points of both material
evaluations and design methods so as to employ
ceramic component in engine applications.
Design methods must be considered to
compromise brittleness of ceramic materials.
Fabrication process is required to improve
strength. to reduce strength variations, and to
minimize defects. Moreover, it is essential to
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

2 850523

Fig. 4 - Failed ceramic swirl

lONer-chamber

Fig. 1 - Cross section of engine employed in ensure component reliability by several

the present \vork inspections and reliability evaluations that are
suitable for ceramic materials.
The development for these techniques Hill be
discussed belm.."

110 CONPONENT DESIGN HETHODS

The shape of ceramic chamber is similar to

ordinary one of Ricardo Comet NK-V type (Fig.S).
This shape has advantages for good performance
100 Ceramic and noise characteristics. External surface is
<1,----;,(. application straight type to reduce stress concentration.
Service stresses applied to the s\virl chamber can
be generated by mainly thermal stress and
90 combustion stress. Other stresses are associated
Hith fixing forces from cylinder head gasket or
the forces generated by thermal deformation of
900 950 1000 cylinder head. The model for these stresses is
Temperature of combustion illustrated in Fig.6.
chamber Ce)

-Fig. 2 Conceptual illustration for maximum

poi-ler out put
Metal ring

\
220 2L-THE
110 2L-THE
~ 210 0
E 100 1 z
-200
in 0
2L-TE g. 190 0
f 2l-TE
~ 90 .s 180
E
~
c 8
§ 170
.~ 80 .~ 160 C
8 Ceramic chamber Metal chamber
:ll 70
°A ~ 150 A

" 1.5 2.0

Engine d~placement (Q)
2.5 3.0
" 1.5 2.0
Engine displacement (Q)
2.5 3.0 Fig. 5 - Comparison of the shape of sHirl
Imver-chamber

Fig. 3 - The objective and comparison of engine

performance
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

3
850523

400

500

o'- --''''o;'£.' _ ~
~~ITi ~§
300 4001 600 500 400 300
Time 500

(fT ; Thermal stress (Unit, 'C)

O"c ; Combustion stress
O"F ; Fixing stress Various stresses Fig. 8 - Temperature distribution calculated by
FEN
Fig. 6 - The model for service stresses

STRESS ANALYSIS BY FINITE ELENENT NETHOD-

Figure 7 shows three dimensional solid model for
FEt-! analysis in ceramic sHirl chamber.
Temperature distribution was analyzed by FEM,
Hhich Has based on the results for temperature
measurement in the component under engine
operation (Fig.S). The temperature of inside
surface of the chamber becomes 10\oJer than that
of combustion chamber because of the existence
of boundary layer in periphery of the component.
HOiVever, it can be recognized that there exists
relatively large temperature gradient towards
external surface of the sHirl chamber. Figure 9
ShOHS distribution of thermal stress calculated
by above results. Large thermal stresses can be (Unit, kg/mm')
generated in the upper and lmver out side as well
as throat region in the component. Naximum Fig. 9 - Thermal stress distribution calculated
stress was estimated to be~290~Wa as tensile by FEN
stress.

The present swirl chamber is designed to

Fig. 7 - Three dimensional solid model for FEN
The service stresses Here also measured
using strain gages under engine operation. The
result is shown in Fig.lO. It is apparent that
the stresses are increased according to an
increase in engine speed, and the swirl chamber
is subject to dynamic stresses. The ultimate
stresses were estimated by the combination of
above measured stresses and FEN analysis, as it
is difficult to measure the ultimate stress
distribution due to temperature limit for strain
gages.
install the outer metal ring of shrink-fit to
reduce stresses applied to upper and lower
regions. Figure 11 shows FEM stress analysis to
determine the shrink-fit clearance as well as the
thickness for upper region of the chamber.
Relatively large shrink-fit clearance and small
thickness are favorable to reduce applied
stresses. Optimum values for them were
determined by above FE~1 analysis follmved by
engine durability confirmation.
STRESS ANALYSIS FOR FIXING FORCE -
Figure 12 shows deformation pattern for a
cylinder head by temperature gradient, 'vhere the
head is deformed elliptically by compressive
stress. This is consistent ,,,rith the result
obtained by strain measurement under engine
operation. The clearance (A) in Fig.13 is
necessary to reduce stress generated by such a
cylinder head deformation.
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

4 850523

Other stress must be considered which is due

to the force from cylinder head gasket. The
clearance (B) between bottom surface of the
Sirain gages chamber and lOt-ler end of the cylinder head is
particularly important (Fig.13). Figure 14
Cenler of shaHs the correlation bettveen this clearance and
cylinder stresses applied to the external surface of the
chamber. It can be seen that the stresses tend
13'
~
" 200

I
E 100
300

/fl
.____fl
-' - to decrease according to an increase in this
clearance.
It is necessary to minimize the stresses due
to these fixing forces as for initial stress •

STRESS ANALYSIS FOR A FITTING HOLE IN THE

CHAt-mER - The fitting pin prevents from rotating
~ / the sHirl chamber as \olell as making it loose
into the cylinder head. Figure 15 shows FEN
analysis for stress distribution around the
'& 6°t---t---t---t---ilr fitting hole in bottom surface of the chamber.
e15 40 !- C"yc...li_n_d_ri_c_alrd_i_rec.c_l_io..,n,-,<-_ _-t-
The result indicates that stress concentration
is assumed to be not serious problem, because
higher stress can be generated besides the
~ .,.,fJJ fi t ting hole.
~ .---6~
~
l!
~l __ o
0 1---+--..,,..-,'-:-.,,--.,.1---""<'1- Engine front
to- Axial direction

--20\---+--+---t---+-
t ,,
& ,
e 20 !---.,r"""'=:-..Jill-,..,-,"":;'-."...-'-'1'-
.'!
I -.1.-8-'
\
,,
~ / CYTdriCal d1ireclion
\
,I
,,
~ 0 I,
.§ ---'0T--9-~ __ ,
- Axial direction 0 , Before deIOrma~!on
gj -20 f---+-.:.;;.::.:;:.:-=.::::::.::r~--1- , " In. '
~ I I I I , \ Afler delormalion :
,
,
,,
\ Valve seal }
o \ lower hole /
u "'-,..,..- ,,;,/ I
o 2000 4000 I
I

Engine speed (rpm)

Fig.IO - Pleasured service stresses applied to Engine rear

s\oJirl lower-chamber

Fig.12 - Deformation pattern of cylinder head

40
DRegion I
30 r-
r- E2J Region
2

In
b-
-~r 20
"'E §
el-
10
r
llj-l~
~ ~ -20
ii= (Ouler diameter
'" u
.; 0 -30 of chamber)
en A B C D
I Large Small
II
- -
Small Large
- (Inner diameter
of metal ring)

Fig.II - FEM stress analysis for determination

for thickness and shrink-fit clearance Fig.13 - Assembling for s\virl Imver-chamber
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

850523 5

The properties of sintered material can

be affected by the amount of oxide additives.
.....
Cylindrical 0
direction ,'- Ir
ok
~~
. ,..::::-.
Strain
gages
Flexural strength has slightly increased from
715 to 745NPa according to decrease in additives
[rom 5-5 to 4-4H/O. Small <lmount of additives
r·~ is favorable to improve high temperature
o I Axial direction, properties as glass phase can be decreased, but
Clearance-
relative density \.Jas 1m.] in the case of less
!~O
yc additives than 4-4w/o in the present work.
Hence, it Has concluded that the oxide additives
01 Has sufficient to 4-4w/o.
I1 t
Allowed mnge

Fig.14 - Clearance and applied stresses in

external surface of swirl lower-chamber
99 e
---.
::!2. 90
0 63b=718MPa
..........
70 m =18.7
>-
-
:<::: 40
:0
C\'l

Without fitting hole

.0
...
0
20

10
,
e
e

Cl. e

4 • Additives
e 5%Y203
2
e 5% MgAQ,204
400 500 600 700 800 900 1000
Stress (MPa)
With fitting hole
Fig.I6 - Flexural strength at room temperature
Fig.I5 - FEN stress analysis in bottom surface
of sHirl lower-chamber

FABRICATION PROCESS

:-lATERIAL DEVELOP1'IENT - Silicon nitride

pmvder has average particle size of O.S.-.Q.9fI ID,
and a-Si3N.\ content is more than 90%. Yttrium
oxide and spinel were used as for sintering
additives. Silicon nitride were mixed with
oxide additives in ball mill under ethanol.
Testing bars were molded in die by pressing
follm.;ed by cold isostatic pressing. The
testing bars Here machined to final size of 3 X 4
X 32mm after sintering for 4hrs at l750°C under
N2 ambient. Flexural strength by three point
bending Here lTieasured by span 30mm and cross-head
speed of O.5mm/min at room temperature. Flexural
strength of sintered material has 7l5Npa Hith
Heibull parameter of 18.7 as shmm in Fig.l6.
Hicrostructure for the sintered body is composed Fig.I7 - Hicro-structure of etching surface of
of fine crystalline particles Hith the order of sintered body
betHeen submicron and a [eH microns. The
rectangular crystallines are also observed in
Fig.17.
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

6 850523

Thermal properties of sintered body are Fracture toughness Has measured by vickers
summarized in Table 1. Thermal conductivity indentation method developed by A.G. Evans et
can be increased according to decrease in ale (1) and G.R. Anstis et al. (2). Figure 19
additives, Hhich is presumably due to thermal shaHs dependence of fracture toughness on
scattering effect of glass phase. Increase in temperature in the range up to BOOoe in air. The
thermal conductivity is advantageous for defect size Has estimated by plotting (KIC/(J3b)2
reduction of thermal stress applied to the swirl versus defect size C as shotvn in Fig.20. Defect
chamber. size Has nearly 40 II m at the most, and the size
Has consistent Hith that of observed pore at
fracture surface ~lS shm10 in Fig.21.
Table 1 - Thermal properties of sintered body in
It is important to knoH strength
the present Hork
degradation after exposure at high temperatures
since engine components are used under severe
conditions. There Has no significant change
Thermal diHu. Specific heat Thermal
Temp. Additives in flexural strength after annealing of sevFral
coefficient capacity conductivity
ee) ("Yo) (m'/s) (KJ/I(g·k) (W/m·k)
100hrs at 1000°C in air.
These results all suggest that the material
developed in the present Hark has sufficient
5-5 8.164Xl0- o 0.6737 17.75 mechanical properties in application for tIle
20 present diesel engines.
4-4 9.275XlO- o 0.6904 20.60

5-5 3.66Xl0- o 0.933 11.03

~
590 <"'li,""
4-4 3.82Xl0- o 0.988 12.13 6
__
E -
--z
:2: a-~
~

o ~

~
:::.:::"
In Fig.lS. flexural strength at high en
en
temperatures is shovm for the material Hith Q)
r::: -
4-4w/o additives. Strength degradation can be .s::: 4
en

-
observed above :::::;80QoC, which attributed to
softening of glass phase existing in grain ::;,
0
boundaries at high temperatures.
....
Q)

-
LL
::;,
U
~ 2 I I I I

800 T T 0 200 400 600 800

r I\
~

'"
a.
:2: Temperature ('c)

- 600
~

.s::: Fig.19 - Dependence of fracture toughness on

en temperature
r:::

-....
.... 400
Q)

en
Cii
0

1\ ~

E
0.10

::;,
><
200 Specimens 3X4X32mm E
~
Q)
u::: Loading span 30mm
N
---. 0.05
D
M Pore
0 500 1000 ~ Coarse
grain
Temperature ('c) "
'-'
0 20 40 60
Fig.IS - Flexural strength at high temperature
Defect size Cum)
Fig.20 - Correlation betHeen fracture toughness
and defect size
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

850523 7

treatment of as-molded body utilizing thermal

decomposition. Figure 23 shaHS the Height loss
during heat treatment of the sample \>lith heating
rate of lOoe/hr. Rapid thermal decomposition can
be observed above """350°C. The suitable heating
pattern was determined based upon above data.
(a)
Heating rate must be slow enough to prevent
cracking and blistering of the body after binder
removal.
The suitable sintering conditions including
t----i
temperature, hold time, and heating rate are
1000 11m different between testing bars with simple shape
and complex shaped components. Therefore, final
condition was determined by the result on the
component itself.
The strength of the s~oJirl chamber fabricated
by injection molding was measured using testing
bars cut from the sintered body. Strength
distribution along axial direction is shol.·.'11 in
(b)
Fig.24. where strength variations is quite small
and Heibull parameter attains 20.3. This
suggests that injection molded body is quite
homogeneous in the present Hork.
I----t
lO/Im
Fig.21 - (a) fracture surface,
,,
, 0
(b) fracture origin (pore) 4.5
;:,
~ +---Visc. oc Torque--+',
INJECTION MOLDING PROCESS - Ceramic X 4.0
injection molding is suitable for processing -;- ,,
, 0

complex shaped components with large quantities .!2 3.5 ,

(3). This method is composed of [allot'Jing
0
~ ,,
I 000 0

process steps, and total processing control is 3.0 ,

~ o 0
essential to succeed in this method, as each one
influences subsequent step.
ill
2.5 a
O 0'
,, 0

:>"
'" 0

,,,
0 ,0
0 0
( I ) Nixing to obtain homogeneous mixture 2.0 0
0
0
composed of organic binders and ceramic
pOt,Jders 36 38 40 42 44
(n) Injection molding
( 1II) Binder removal Torque (N-m) X10'
( Ii') Sintering
Fig.22 - Torque and viscosity for mixtures
In this Hork, polypropylene based thermo-
plastic binders were used as well as minor
plasticizer and additives. The mixing of binders
",ith ceramic pOtoJders is important to obtain
sufficient flow or viscosity suitable for
100 o·
subsequent molding. The flo\1 characteristics for
the mi;"ture can be evaluated by its torque \"hich
is measured using high-shearing apparatus.
Figure 22 shO\"s flo\1 characteristics for the
e:.
~
80

60
/
•
mixtures prepared by various conditions. The :E /
torque indicates rather 10\.] value and is '"
'Qj /.

.-
40
proportional to the viscosity above melting point ;: <i/.
of binders, Hhen homogeneous mixing is .:lttained.
This data is useful to determine tlle final mixing 20 .------
conditions including ceramic pOHder
characteristics.
.------
Olf"'----'---"---'---'--.......-~-'_J
One of the examples for defects gener.:lted by 150 200 250 300 350 400 450 500
ceramic injection molding is "Held line" \oJhich is
Temperature ('e)
caused by inadequate adhesion at the front of
flow or entrapped air. This type of defect can
be eliminated by consideration of injection Fig.23 - WeigJlt loss during heat treatment of
pressure, injection rate, and hold time. mixture
Binder remov<:ll was conducted by heat
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

8 850523

99

to
99
Grinding direction
90 r- 0
0
~ F')6",'O ,,,, 0'
70 m=20.3 90 , <;:.;,
~
2f2
50-
l
./ 0
t.
~
80
70
60
6",=501MPa
m=12.6 d'
j7'
~ 20- 0
0 Specimens; :c111 50
40
~ 10
~ 5
0 1.2X1.7X22mm
Loading span; 1Smm
.
J:l
0
30
!/ ~o Grin~direction

~
0
20
Cl. 0
!?
~ 21-
a. 1 I-
10 /='
, 0
0/ 6,,=635MPa
m=17.7
450 500 550 600 650 700
1 , 1
Flexural strength (MPa)
500 600 700 800 900 Fig.25 - Flexural strength according to
grinding direction
Flexural strength 0"" (MPa)
0
Fig.24 - Flexural strength in axial direction ~
M
3 Grinding wheel
of sHirl chamber E 0/ SDC170R858
E
HACI-1INING PROCESS - As sintered SHirl ---E
M

E
chamber is ground at the upper and lOHer surface 2
as Hell as the external surface to ensure good 0
:;::
...
o~
tolerance in final size. The grinding of the
external surface of the chamber is important, ro
since the maximum stress is applied to the 0'>
external surface along cylindrical direction as c: 1
discussed before. It is known that the grinding
'5
c:
.;::
direction influences the strength of ceramic
C) 0
materials (4). Figure 25 ShOHS that the strength
of testing bors \oJith tensile stress normal to the
grinding direction can be about 20% 10\.J81' than o w,;-..l.--'---L--.J'---'---....L.J
tllose with tensile stress parallel to the 234 567
grinding direction. It is desirable that the
ceramic chamber is ground "'ith conventional
Stock removal rate
cylindrical grinding, as stress is applied along (mm 3 /mm·sec)
Fig.26 - Stock removal rate and grinding ratio
cylindrical direction. In the present work,
inrced centerless machining was adopted in
consid2ration of productivity, grinding
~
o
E
alloHan;:es, cycle time, the numbers of machine, E 1.5
and the size of a diamond Hheel. The oscillation
method was employed to avoid uneven \<lheel wear
---
Sf
~

during grinding process.

TIle objective of the present development is '"
~
to raise productivity as Hell as to reduce
machining cost. H<1chining conditions must be
-;,
c:
to
determined so as to increase in grinding ratio '6
and to extend dressing interval. because diamond c:
.;::
Hlleel wear directly influences grinding cost for 0>
the ceramic chamber. Hheel Hear is Hell knmm to :Jg
c:
0.5
be affected by stock removal rate (5). l-!oHever,
there is few data on correlation betHeen stock
removal rate and grinding ratio in ceramic infeed
'"c:
0>

center less machining. ~

It can be recognized that rapid decrease in Ol-C--L_--L_-'-_-'------J
grinding ratio occurs as stock removal rate 2 4 6 8
becomes larger than a certain value as shotvn in
Fig.26. Figure 27 indicates that grinding force Stock removal rate
is also increased rapidly above the same stock (mm 3 /mm'sec)
removal rate as in Fig.26. These suggest that
Fig.27 - Stock removal rate and tangential
grinding condition changes at this value of the
grinding force
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014
850523
stock removal rate. In fact, it can be observed
that grain release is much progressed in the case
of occurrence of rapid increase in grinding force.
Hence, stock removal rate has to be lOHcred so
as to prevent rapid increase in grinding force.
Dressing interval is extended Hhen the
grinding is conducted under the condition Hhere
grain release and regeneration is stable. The
frequent dressing must be avoided, since it is
equivalent to ;~rinding \.Jheel wear. The
correlation betHcen the number of grains and
cumulative stock removal is Shm111 in Fig.2B.
This data can be obtained by counting the grains
in trnl1sferred surface of grinding t<theel. One of
the examples for the transferred surf<h~e is shm·JO
in Fig.29. The existence of grinding conditions
Has confirmed t"here the grain release and
regeneration occurred at constant rate from
Fig.2B. It Has possible to make dressing
interval extend by determination of grinding
condition from these results. Pres":nt grinding
conditions make it possible to assure the
component quality anci to reduce machining cost.
The final conditions are summarized in
Table 2.
n Regenerated grains
~ Remaining grains
2
o cr.;:'--''''''-r.c.d.-l''''''--Ul..':--tL2+_--l~
1 2 3 4
Total stock removal mmJ/mrn
Fig.28 - The numbers of grains in diamong Hheel
surface
• RELIABILITY EVALUATION HETHODS
!~
, !
~
11>."
I----f
O.2mm
Fig.29 - Transfern~dsurfaceof Hheel after
machining process
9
Table 2 Hachining condition for cl~ramic SHirl
chamber
GRINDING WHEEL SDCI70858
REGULATING WHEEL A120R3R
GRINDING WHEEL SPEED 1800mlmin
REGULATING WHEEL SPEED 12m/min
OSCILLATION SPEED 840mm/min
INFEED RATE 1.0mmlmin
INSPECTION IIETHODS
It is necessary to screen the ceramic
chambers since sintered silicon nitride has
large strength variations as compared to
metallic materials. Several non-destructive
inspections toJere employed Hhich include X-ray
radiography, fluorescent penetrant and ultrasonic
inspections. Internal flato.rs more than -:: : : 200 fI m
can be detected by X-ray radiography, nnd
fluorescent penetrant inspection is useful for
sharp flm"s existing in all the surface of the
component. Ultrasonic can detect internal
defects Hith the size above -::::::50/1 m, ,md
external defects above ~ 100 I' tn in the present
"ark (6).
Hm,.rever, these non-destructive inspections
can't ensure all the regions in the SHirl chamber
because the failures of th!:. chamber occurred
after durability testing from tIle regions in
".rhich (lbove inspections could detect no defects.
Therefore, the regions applied higller stresses
can be ensured by thermal-stress loading and
mechanical-stress loading inspections. Thermal
stress loading toJas conducted by heat-treatment
for inner surface of tIle chamber using propnnc
burner, ,,,hich generated stresses higher than in
x10~
engine operntion. Nechanical-stress loading t,Tas
performed by application of inner pressure to
the SHirl chamber.
Quality assur".Il1ce for all the chambers :IaS
been made by these inspections in the present
toJork .
..
•
LONG TERN DURABILITY OF THE SHIRL CIIAHBER -
It is essential to assure component reliability,
particularly in the case of ceramic material for
engine parts. For example, it is said t]lat life
prediction for ceramic materials can be made by
follmoJing equation using stress intensity factor,
crnck propagation rate and Heibull probability
theory (7).
r 2{\!'n(1-p)-1}~1 ( 11-2 B
i:jnL=t,ll
li\(jTlY2(1l-2)KI~C~ J T
'"
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

10 850523

The component strength Has measured by

L;life, n;cruck propagation parameter,
m;Weibull parameter, a;applied stress,
Pjprobability, K1c;fracture toughness,
A,B;constant, Y;shape parameter
HOHever, as engine components are subject to
various loading llistory, it is not sufficient
to assure'strength of the component on the
supposition of a single axial stress as in above
equation. Hence, several reliability Cestings
~vereperformed as folloHs.
Load 490MPa
Span 3mm
o
circular shape specimens cut from the component
along cylindrical direction. FEl>! analysis and
strain measurement confirmed that strength of
these specimens could be evaluated Hithin a few
percents of experimental errors as compared to
rectangular shaped specimens (Fig.3D). The
flexural strength for circular specimens is shmm
in Fig.3l according to engine durability term.
The strength for upper region of the component
Has o.lso evaluated using ring shaped specimens
Hhich Here stlh,il'(·t Lo tcnsile strl'SS loaded hy
inner pressure as shmm in Fig. 32. It seems that
ther..:: is no SIOH crack groHth since strength
degradation can't be detected after engine
durabili ty testing as shmm in Fig. 31 and Fig. 32.
The estimation for the strength of the component
itself from that of small specimens can be made
by effective volume method based on Heibull
probability theory (8).

Load

~
290MPa Ring
Span 12.Bmm
.J;:=f=::;;:;Uspecimen

, "--~:
Fig.30 - FE~lstress distribution for circular Elastic
shaped specimen medium -.+-
I
~

Testing method Ring specimen

500

0
'"
c..
3400 0 0
8 0
en 0

-
Cutting regions for en 0
circular specimens Testing specimen
...<I> 0
0
0
en 0 0
300 0
~

co 800 2
a. "iii
e c:
-...
.c:
Cl
c
~
<I>
I-

-CI>
en
l:!
600
8
EO
Cb
0

(p B'
Engine durability term
Fig.12 - Tensile strength in upper region in swirl
:::>
X ~ 8 0 chamber according to engine durability

~
CI> 00 Q term

-...
;;::
en 400 0
c INFLUENCE OF PISTON STA}WING - It is
"0 necessary to evaluate the strength and durability
Co for the component under severe environment, as
:::> engine component must be durable under various
0 conditions. One of the examples is the influence
LL
of piston stamping. The large compressive
Engine durability term loading may be applied to the chamber, if excess
deposit can be generated betHeen bottom surface
Fig.3l - Flexural strength in IOHer out side in of the chamber and top end of the piston head.
SHirl chamber according to engine The compressive strength of the chamber is
durability term summarized in Table 3. High compressive strength
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

850523 11

[or the chamber is apparent independent of

deposit material and compressive position.
Figure 33 SI10HS the result for stamping load by 105 2L-THE
piston head, ",here deposit Has attached
intentionally in bottom surface of the chamber,
and stamping load \';85 estimated by measured Boost pressure
strains under engine operation.
These results ShOH that compressive strength
of the chamber is more than 10 times of excess '5
loading from piston head.
-0.100
: :l
o
Compression ratio

Table 3 - Compressive strength of SHirl chamber

95
2L-TE
1:{.f-l--'---'~--'--.l-...<.--'-
..........<--.L...o..-.J
~l /-Jarct.
I't:
""'.;;"'~. no..
Aluminum Alloy Aluminum
Hv83 Ha44
Graphite
Hv257
900 950 1000
Temperature of combustion
@ 56800 -- >29000 chamber eCl

e 130000 -- >107800

--
Fig.34 - Engine perfonnance in pOHcr out put

(2) -- 33700

©) -- 26000 --
(Umt, N) 110
220 105/4000
211/2400 ~
9614000 100
200
'-"',
E ,/ \ gO
:Z 180 1',
~

/ " "- r;;

'" 80 eo
- //--z-;. -
/
::l
~-MOCkdeposit C" 160 / \
0
t-
70 '"
;:
0
140 0-
/
o 0-
1 60
~ 2000
,---;; 00-
120
/
1
- - 2L-THE 50
".3 1000
I1l
I - - - 2L-TE
I 40
~
260
o 100 200 ~
0-
Height of deposit (pm) "-
240 -'"c
0
Fig.33 - Stamping load by piston head 220 :aE
::l

19012000 200 c0 '"

0
ENGINE PERFOR;~NCE
c;;
::l
180 u.
The adjustment of engine performance Has
made in consideration of compression ratio, boost
pressure, injectIon timing and injection quantity 1000 2000 3000 4000
in order to utilize ceramic capabilities at high
En9ine speed (rpm)
temperatures. PO\.Jer out put can be increased
according to an increase in temperature of the
combustion chamber as shO\.Jn in Fig. 34. In these
steps, tile injection quantity and injection Fig.35 - Comparison of engine performance
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

12 850523

Table 4 - Comparison of major specifications

Engine
Numbers of Cylinder
r II
2L-TE
4
·_-T_!I'-TIIE

I i
Bore x Stroke (mm) ¢92 x 92
II
Displacement (cm 3)
I 2446
II ,

Valve illechanism OIlC, Timing ,

Belt Drive
Type of Combustion Swirl Chamber
Chamber
-
Compression Ratio 20.0 21. 0
! -
Injection Pomp Bosch VE
Distributor-type
Injection Quantity
at Maximum Power 52 57
(mm'/st)
l\laximum Pm"rer 96/4000 105/4000
(PS/rpm), ,JIS gross
r.laximum Torque 191/2400 211/2400
(N.m/rpm) , .JIS gross
- --
Boost Pressure 350 400
(mmllg)
--
Nozzle Opening 11.8 15.7
Pressure (MPa)

timing were adjusted to compression ratio, boost summarized as follows.

pressure, and injection rate in order to obtain
maximum pOtver performance. The upper limit for
these parameters t.;as determined by the
limitation of smoke concentration, exhaust gas
emissions and temperature.
Conventional metallic chamber is employed
in 2L~TE, and ceramic chamber is in 2L-THE
engine. Comparison of major specifications and
engine performance are sho,vn in Table 4 and
Fig.35, respectively. The increase in enigne
performance is apparent in wide range of enigne
speeds and the maximum power out put can be
increased from 96 to 105 PS by application of
silicon nitride swirl chamber. In addition,
t:here was no significant change in engine
efficiency or emissions for the 2L-THE engine
with ceramic chamber as compared to 2L-TE
engine.
SUHHARY
Sintered silicon nitride sHirl lmver-
chamber has been developed in order to improve
performance of turbocharged diesel engines.
The reSUl.t::> obtained in this Hork are
(1) Design specifications for the ceramic chamber
have been determined by FEN analysis and
stress measurements for thermal stress,
combustion stress and assembly fixing forces.
(2) Sintered silicon nitride has average
flexural strength of more than 700 tlPa "lith
Y2 OJ and NgA1 2 04 as oxide additives, and
ceramic injection molding has employed to
fabricate the sHirl chambers Hith large
quantities.
(3) Quality assurance and cost reduction in
machining process can be attained by infeed
centerless method and selection of machining
conditions suitable for ceramic materials.
(4) Reliability of the final products was
assured by reliability evaluation methods
as Hell as non-destructive and stress
loading inspections.
(5) FOHer performance of turbocharged diesel
engine ,vi th ceramic chamber has been
increased in 9 PS (7~1) as compared to that
of conventional engine by adjusting injection
ril te, compression ratio and boost pressure.
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

850523 13

ACKNOHLEDGEHENTS

Authors are grateful to Dr. O. Kamigaito

and cO-Horkers in Toyota Central Research and
Development Laboratory Incorporation for their
valuable discussions and offer of some datum.

REFERENCES

(1) A.G. Evans and E.A. Charles, I1Fracture

toughness determinations by indentation",
J. Am. Cl~l-i.ll:l, Soc., 59(1976)37l.
(2) C.R. Anstis, P. Chantikul, B.R. Lawn and
O.B. i'larshall, IIA critical evaluation of
indentation techniques ror measuring
fracture toughness: I, Direct crack
measurements", J. Am. Ceram. Soc., 64 (1981)
533.
(3) B.C. Hutsuddy, "Injection molding research
paves Hay to ceramic engine parts", Ind.
R&D. July(1983)76.
(4) C.A. Andersson, R.J. Bratton, "Effect of
surface finish on the strength of hot pressed
silicon nitride!l, N. B. S. Special
Pubication 562(1979)463.
(5) T. Nakajima, Y. Uno, and H. Yoshinobu,
"Cylindrical plunge grinding of fine
ceramics!', J. Ja. Soc. Prec. Eng., 50, N04
(1984 )
(6) K. Asai, A. Takeuchi, and N. Deda,
"Computerized ultrasonic inspection system
for ceramic precombustion chambers of
automotive diesel engines", to be presented
in tllis cOllgress. (19El5)
(7) A.G. Evans and S.H. Hiederhorn, "Proof
testing of ceramic materials-an analytical
basis for failure prediction", IntI. Jnl.
Frac., 10(1974)379.
(8) D.G.S. Davies, liThe statistical approach to
engineering design in ceramics", Proc. Brit'
Ceram. Soc., 22(1973)429.
 Downloaded from SAE International by Nicolas Braam, Monday, October 20, 2014

This paper is subject to revision. Statell\ents and opinions ad-
vam'ed in papers or discussion are the author's and arc his
responsibility. not SAl's; however. the paper has been edited
by SAt for uniform styling. and format. Discussion will be
printed with the paper if it is published in SAl Transactions.
For permL~sion to publish this paper in full or in part, contact
the SAl Publications Division.
Persons wishing to submit papers to be considered for pre-
sentiltion or publication through SAE should send the manu-
script or a 300 word abstract of a proposed manuscript to:
Secretary, F ngineering Activity Board. SAL
Printed in U.S.A.