| Top
> EJ20 & EJ25 Guide >
TOMEI Technical
Report on EJ25 |
TOMEI
POWERED EJ25 R&D REPORT vol.1 |
Production
R&D Engineer Noriyoshi Komatsu
Marketing & Advertisements Naoki Yamamoto |
 |
 |
| |
|
Preface |
|
|
Tomei
Powered has begun research and development operations for
the Impreza EJ20 engine beginning in 2003 and has been continuing
until this day. In April of 2007, the full detailed specifications
were determined for the EJ series complete engine. The ultimate
objective of our research and development for these engine
components is the assemblage of the complete engine.
Many hours are dedicated to engineering these products which
make up the vital components of the complete engine kit.
Features such as longevity and endurance are the major benefits
brought to the end user of the automobile tuning industry.
In order to answer to the demands of the3 Subaru enthusiasts
all over the world, we have decided to research the EJ25
engine, and proceed with the development of parts for this
application. We have completed the first part of this R&D
process, which is included in this report.
■Research and Development Process of Tomei Powered
|
(repeat
cycle until performance meets demand) |
|
|
CONTENTS |
|
|
|
 |
|
|
CHAPTER 1 Analyzing the Stock Engine |
 |
Disregarding
the existence of the TGV and the turbocharger size, we have
determined the components surrounding the long block of
the EJ25 to be similar with the JDM GDB applied A. However
the long block is known to be dissimilar in many essential
ways and is expressed below in a chart.
|
|
EJ25
|
EJ20 |
| BORE
x STROKE |
Φ99.5×79.0mm |
Φ92.0×75.0mm |
| COMPRESSION
RATIO |
8.2 |
8.0 |
CONNECTING
ROD
(center to center length)
|
130.5 |
130.5 |
IN
CAMSHAFT
|
240°－ 9.6mm |
240°－ 8.9mm |
| EX
CAMSHAFT |
240°－ 9.8mm |
240°－ 9.5mm |
AVCS
ANGLE
|
20° |
40° |
|
The
most clear and striking aspect that was revealed after this
test was the power drop after 5,500 RPM, displayed on the
graph. We are planning to prioritize solving this issue
as well as increasing overall performance for this project.
|
| |
|
|
|
CHAPTER 2 Prioritize R&D Subjects |
STOCK
EJ255 ENGINE |
Considering
the fact that the horizontally opposed EJ engine requires
to be completely pulled out of the chassis in order to swap
the internals; we have deeply taken in to account several
factors when performing the R&D. These factors include
tuning difficulties, cost, and effectiveness.
Since
Tomei does not produce air suction filters and muffler systems,
we have resulted in using products from a rival manufacturer.
Although vehicles equipped with the EJ25 are all over the
world, and each account has their separate octane rating,
we had no choice but to perform the test with the Japanese
fuel octane standard. |
 |
|
|
 |
|
CHAPTER 3 Product R&D - Exhaust manifold 1 - |
●Prototype
Product Details
Careful thought and consideration have gone into the decision
for the prototype unit. |
| |
①Researching
equal and non equal length
Pros and cons of equal and non-equal length. |
| |
|
Pros
|
Cons |
| Equal |
Each cylinder
condition in uniformity.
(More precise settings can be achieved)
|
Pipes leading
to junction are longer in length
(decreased turbo response) |
| Non Equal |
Pipes lading
to turbo can be kept short as possible.
(increased turbo response)
|
Disrupts cylinder
uniformity |
|
| |
The
ECU setting can compensate for each of the engine’s
cylinder characteristic differences, improving efficiency
to a certain point. The stock ECU is incapable of
this function.
Considering the next modification one may perform
after changing the exhaust manifold is a turbo upgrade,
we have decided to go with the unequal length design;
similar to our GC8 manifold. The reason for this decision
is that we have made turbo response as the priority
above all factors.
|
|
|
|
|
|
CHAPTER 3 Product R&D -Exhaust manifold 2 - |
| |
②Various
piping layout |
| |
The
GDB equal length manifold resulted in a long design
due to its complicated layout. To compensate for
this structure, the pipe diameters were decreased
to 38mm. However, the exhaust pressures were too
high at the upper RPM range, and the gas flow suffered
significantly. This restriction of flow can be noticed
more clearly especially if a larger turbocharger
has been installed. Also, the 2.5 liter engine has
more exhaust pressure than the GDB 2.0 liter unit.
The main pipe was made to be 42.7mm and the support
piping at 60mm due to the limited space.
The center piping of the 2.0 liter non-equal exhaust
manifold was bottlenecked to 55mm, which further
engine bench testing will prove if the “bottle
neck” was necessary.
(The
“bottleneck” portion increases exhaust flow velocity
right before the turbocharger, aimed to increase
turbo response.)
|
|
 |
 |
| |
|
|
|
|
CHAPTER 3 Product R&D -Exhaust manifold 3 - |
| -
Results & Observations |
|
Tomei’s
bottle neck design had no effect on the output, with
power increasing all the way up to 5,600 RPM. Since
the exhaust pressure was made to be similar to the
stock specification, the turbo response should not
be hindered in any way, but improved. On top of this
improvement, more power is expected to be available.
A decline in power can be seen at anything above 5,600
RPM due to the overall manifold shape.
|
|
It
is prospected that the turbine is almost at full capacity
since the stock manifold exhibited a trend of higher
exhaust pressure and similar power output as our prototype
unit. In regards to the “bottle neck” area of the
support piping, no notable differences were seen in
terms of power and exhaust pressure. However, we had
decided to disregard (take off) the bottle neck area
to spare room for further tuning advancements. Therefore
we will not have any bottle neck designs in our future
tested products.
|
| |
|
| |
|
Main
Piping |
Pipe
Merge Section
|
Support
Piping |
Weight |
Stock |
Φ42.7
Sufficient |
Φ45
Sufficient |
Φ45
|
12.4
kg |
TOMEI |
Φ42.7 |
Φ60 |
Φ60 |
8.9
kg |
|
|
|
|
|
| |
 |
 |
|
|
 |
|
CHAPTER 3 Product R&D -Bolt-On Turbocharger 1- |
| |
The
third step in the STAGE 1 pertains to bolt-on turbochargers.
The selection of the turbocharger assembly depended
on the following 2 points listed below. |
|
| |
A) Space Restrictions |
 |
| |
The
two problems when placing a turbocharger in place
of the stock area is 1.) suction pipe near the surge
tank, 2.) turbo compressor housing is very close to
the block, limiting the size of the housing to 50.5
or 52.6.
|
| |
B)
Over 400 Horsepower + Characteristics Similar to Stock
Turbo |
| |
The
final decision of keeping the vehicle in the 400 horsepower
range was made because we wanted to keep the vehicle
at a streetable level. The 52.6 compressor housing
was used since it is very difficult to produce 400
horsepower out of a 50.5 housing. The exhaust housing
was changed to M77 to fight decreasing response. The
final outcome was the M7760 turbo assembly.
|
| |
| |
COMPRESSOR
WHEEL |
COMPRESSOR
HOUSING
|
TURBINE
WHEEL |
IN
DIA
mm |
OUT
DIA
mm
|
TRIM |
INT.
SIDE
|
EXH.
SIDE
|
OUT
DIA
mm
|
EX
DIA
mm |
TRIM |
| VF39(STD) |
46.5 |
60 |
60 |
48 |
40.8 |
48 |
53 |
82 |
| ARMS-M7760 |
52.6 |
68 |
60 |
53.5 |
42 |
49.1 |
56 |
77 |
|
|
|
|
|
| CHAPTER
3 Product R&D -Bolt-On Turbocharger 2- |
| ●
The Bolt-on turbocharger was evaluated with these criterias
in mind. |
| |
■
If the boost can keep up at the upper RPM range
■ If the primary exhaust pressure (engine to turbo)
is similar to stock turbo
■ If the compressor outlet temp does not keep rising
depending on impellor revolution.
■ If stock boost (1.3lg/cm2) has capacity to produce
400hp |
| ●
Measuring Points |
| |
■
Engine Power (taking into account surrounding aspects
that affect power)
■ Boost Pressure
■ Primary Exhaust Pressure
■ Compressor Outlet Temperature |
| |
~
RESULTS & OBSERVATIONS~ |
| |
The
ARMS-M7760 turbocharger was chosen because its size
and dimension allows it to be swapped out in place of
the stock turbo placement. However, we have begun to
realize that reaching the 400 horsepower mark with this
turbocharger may be a bit difficult. According to the
graph, with the boost pressure at 1.3kg/cm2, the EVC
duty is already at 70%. There is still room to increase
boost pressure but this will result in increasing the
primary exhaust pressure, which leads to high compressor
outlet temperature. Therefore we cannot increase the
boost pressure any higher. |
|
|
|
|
| CHAPTER
3 Product R&D -Bolt-On Turbocharger 3- |
●
Characteristics of the ARMS M-7760 |
|
■
Primary exhaust pressure is very similar to that of
the stock rating at the lower RPM range. |
|
|
*
Power sensation characteristic is not too slow or
sudden, but very natural.
* Ease of use and smooth drivability experienced.
|
|
■
Response similar to stock turbocharger |
|
|
*
Increasing turbocharger size usually hinders throttle
response, however we have avoided this issue without
sacrificing this aspect. This was made possible by
carefully selecting the most ideal combination of
components. |
|
■
25 Horsepower maximum power improvement |
|
|
*
Out of all the turbochargers we have researched and
considered, this unit provided an improvement of 25
horsepower or more.
* This number was confirmed without sacrificing drivability
or response. |
|
■
Boost maintained at the upper RPM range |
|
|
*
The new turbo setup maintains boost pressure even
above the 5,500 RPM mark, which greatly affects the
prospected tuning course of this engine. |
|
■
Completely bolt-on |
|
|
*
This turbocharger will include all necessary accessories
such as gaskets and hardware for a straight bolt-on
capability. The entire kit is ready to be installed
straight out of the box.
|
| |
|
|
|
| *
The entire testing procedures have been completed, labeling
this turbo as the Tomei ARMS M7760. This turbocharger is now
in the preparatory stages of production and sales. |
|
|
|
| CHAPTER
3 Product R&D -Bolt-On Turbo Graph & Photos
|
 |
 |
 |
Turbine
Test Results |
|
 |
1.3
kg/cm2 Boost setting |
 |
|
|
 |
|
CHAPTER 3 Product R&D
- Poncam 1 - |
| ●
AVCS Structure ＆Research |
 |
| |
Upon
comparison of the AVCS functionality range between the
EJ255 and EJ207 engines, we have found that they are
different. The component was then disassembled and analyzed
/ compared, to find out that the hardware was completely
identical. This brought us to realize that everything
was controlled by the ECU. The maximum AVCS advancement
angle for the EJ207 RA Spec-C unit is 45 degrees, with
over 2.0mm of piston to valve clearance (p-v clearance).
|
| ●
Selection of camshaft specifications for testing |
 |
Since
the AVCS function is present, A great amount of duration
is not necessary since the AVCS function is existent,
so 2 types of conditions were brought clear. |
|
|
~
A setup with the most lift |
|
|
~
A setup with the most duration
|
| |
|
|
| |
|
|
IN
|
EX |
DURATION
|
LIFT |
P-V
CLEARANCE @ 45° |
DURATION |
LIFT |
P-V
CLEARANCE |
STOCK |
240 |
9.6 |
2.96(ATDC
10°) |
240 |
9.8 |
7.85(BTDC
0°) |
TESTED |
252 |
10.4 |
2.23(ATDC
5°) |
252 |
10.4 |
|
260 |
9.8 |
2.22(ATDC
5°) |
264 |
10.0 |
77 |
|
| |
|
The
retardation and center angles are set at stock (IN-125°
EX - 115°) |
|
|
|
|
| CHAPTER
3 Product R&D - Poncam 2 - |
| ●
Measuring Points |
| |
・Engine
Power (taking into account surrounding aspects that
affect power)
・Boost pressure
・Ideal AVCS angle advancement |
| |
*
It has been confirmed that the turbocharger cannot maintain
boost at the higher RPM levels, therefore the value
of 1.3kg/cm2 was decided. |
| ●
RESULTS & OBSERVATIONS |
| 4
types of tests were performed including stock settings |
| |
■ 252-252
|
Shared
same qualities as stock at the lower RPM range. As 5,200RPM
as the boundary, some power improvements were exhibited.
|
| |
■
260-264
|
The
graph looks similar to the 252-252 spec, but is offset
at the higher RPM area.
|
| |
■
252-264 |
Shares
same characteristics as the 252-252 in the lower range,
and 260-264 at the higher RPM range. |
|
All
tested versions exhibited a peak of 6,000 RPM boundary
limit, which is believed to be caused by the lack of
intake air at that RPM range. The duration cannot be
increased without the modification of the intake components
(surgetank or intake plenum, intake port/polish, etc).
If we were to change the camshafts,
the next candidates will be 252-252, 252-264 however
considering the balance of the intake and exhaust sides,
252-252 might be more desirable. |
|
The
camshaft specifications can be made up by changing the
AVCS duration, but we have provided 2.0mm margin of
safety for inappropriate settings and also in case any
of the teeth on the timing belt accidentally skip. Increasing
the lift may result in interference with the camshaft
journal caps. |
|
|
|
| CHAPTER
3 Product R&D - Poncam Data & Photo - |
 |
Camshaft
Test Line Graph |
| |
 |
 |
 |
| |
|
|
 |
|
CHAPTER
3 Product R&D - EJ207 Head Modification 1 - |
| ●
Head Structure Inspection |
| |
Due
to the air intake restriction we had experienced earlier
during the camshaft testing, we have decided to inspect
the EJ207 and EJ255 head assemblies. As seen in the
photo, the EJ207 head was equipped with larger ports.
The similarities of the IN/EX valves enabled us to compare
the intake ports of these two heads by swapping each
unit. |
| |
* |
The
EJ207 required alterations to the combustion chamber
due to different cylinder bore diameter. |
| |
* |
Since
the same camshafts cannot be used for both applications,
the EJ207 head was fitted with Poncam (IN250-9.6 EX256-9.8).
|
|
Intake
Port Comparison |
| |
|
EJ207
Combustion Chamber Modification |
| |
We
have prioritized the flow bench test results because
of the differences in the camshaft specifications, and
engine bench tests were performed as well. This test
was performed as backup to confirm the special engine
changes (high RPM range power output) and the power
figures were used as a reference. |
|
|
|
| CHAPTER
3 Product R&D - EJ207 Head Modification 2 - |
| ●
Results & Observations |
| |
According
to the flow bench results, the differences between the
EJ255 and EJ207 heads were further increased as the
cam lift became bigger. The EJ207 head proved to have
10% better flow than the EJ255 with the use of the prototype
camshaft (10.4mm lift). According to these findings,
the differences in terms of performance becomes more
clear as different factors such as intake port size,
higher lift camshafts, and turbo sizes come into play.
|
| |
The
engine bench test reflects directly with the flow bench
results, with no proof of any power drop at the higher
RPM range; even though both units were equipped with
similar camshaft profiles. The stock engine provides
great engine response but lacks power at the higher
RPM range. To surpass 400 horsepower, head modification
is necessary. There two options for this; to modify
the ports on the EJ25 head, or divert to using the EJ207
head. |
|
Flow
bench test results
|
| ■
Problems |
| |
*
EJ25 Head Modification |
| |
Jig
templates and computer programs are required for the
Numerical Control Machine (NC Machine) to make each
port even flow. This will all require a good deal
of time and expense. |
| |
*
EJ255 Head Modification |
| |
Since
the ports do not to be modified, the combustion chamber
diameter need to bne increased. There is a sufficient
amount of hand labor involved but to increase efficiency
and accuracy, the Numerical Control Machine is required
to be used. Also, since the pick up sensor for on
the camshafts are different, the stock ECU cannot
be utilized with this head. |
| |
|
|
|
|
|
| CHAPTER
3 Product R&D - EJ20 Head Modification Data & Photo
- |
 |
 |
Comparisons
Between Two Head Assemblies |
|
|
|
|
|
CHAPTER
4 Product R&D - Summary - |
| |
|
|
|
|
|
CHAPTER
5 - Regarding Other EJ255 Components 1 - |
|
|
CONNECTING RODS |
TIMING BELT |
 |
*
H-Beam
* Rod & cap made from different material
* Oil ports
* ARP bolts
* Knockpins to prevent dislocation
More.. |
 |
*
Longevity against heat & water
* Prevents belt slippage prone to EJ engines
More.. |
|
|
|
HEAD
GASKET |
BELT GUIDES (COMING SOON) |
 |
*
Super grommet construction improves surface pressure
* Exclusive rib design
(Improved seal around water seal)
* 2 thickness available for different compression
More.. |
 |
Since
the boxer engines are horizontally opposed, there are
incidents of camshafts not lifting. This product firmly
secures the belt vibration to prevent unwanted troubles.
|
|
|
|
VALVE SPRINGS |
BAFFLE STIFFENER |
 |
*
Elliptical design adopted for high lift cams
* Improved valve movement
More.. |
 |
*
Circulates oil and prevents build up
* Strengthens crankcase
More.. |
|
|
|
VALVE SPRING SEAT |
INDIVIDUAL THROTTLE BODIES |
 |
*
Easily adjust valve spring seat load
* Prevents surging
|
 |
*
Increased capacity surge tank
* 48mm throttle bodies for high response
* A must for the big EJ fanatics
|
|
|
|
|
|
CHAPTER
6 Proceeding Vehicle R&D Projects - Arise - |
| |
 |
| ENGINE |
EJ207 |
| PISTON |
TOMEI
Forged 92.5mm |
CONNECTING
RODS
|
TOMEI
Forged H-Beam |
| CRANKSHAFT |
TOMEI
79mm Stroke |
| CAMSHAFT |
TOMEI
IN 250-9.60mm, EX 256-9.80mm |
| DISPLACEMENT |
2123cc |
| HEAD
GASKET |
TOMEI
1.2mm 93.5 Bore |
| COMPRESSION |
8.0 |
| TURBO |
TOMEI
ARMS B8446 |
| INJECTORS |
TOMEI
850cc |
| EX.
MANIFOLD |
TOMEI
EXPREME |
| CATALYTIC |
TOMEI
EXPREME |
|
|
Used
as the R&D vehicle for the EJ20 engine components since
2003, the Tomei / Arise Impreza has been competing in various
time attack events. This engine was built as research for
the Tomei complete engine emphasizing on reliability as
well as great performance. Becoming the winner of the 2007
Hyper Car Meeting 10 lap race (AP Battle Open Class), it
proved to be an all around player.
|
 |
 |
|
|
|
|
|
CHAPTER
6 Proceeding Vehicle R&D Projects STPOWERED |
| |
 |
ST
Powered is a Tomei Powered authorized dealer catering
to Singapore / Indonesia / Malaysia / Thai. Their
democar, equipped with a EJ25 along with various components
that make up this research and developme t report.
The stroker kit, which is still under development,
was fitted into this chassis as the test vehicle.
ST-Powered’s continuing support in providing us with
valuable data and feedback makes them one of our prominent
dealers. They have recently taken 2nd place in a renowned
time attack event, and their success can be best celebrated
with the Tomei stroker kit.
|
PROTOTYPE
PISTONS Φ99.75
PROTOTYPE CRANK 81.0mm STROKE
PROTOTYPE CON ROD 128.8mm
TURBO TOMEI ARMS B8446 |
|
|
EJ26
In Development |
|
|
|
|
|
IN
CONCLUSION |
| |
|
The
final objective in Tomei Powered’s researching and
development of engine products is the complete engine.
Only the necessary components are developed to reach
each project’s target objective. |
|
For
this EJ25 project, countless trials and errors have
been repeated, but we have created the best engine
that a customer can use with the utmost satisfaction.
|
|
Special
thanks goes out to Mr. Michael Ferrara of Dsport Magazine,
which he has generously agreed to support us by exposing
our hard work into his publication. This will be the
stepping stone to informing the entire world about
Tomei Powered, and our passions.
|
|
Some
aspecs are difficult to state by text, so if there
are any inquiries concerning this report please feel
free to contact us anytime. |
|
Lastly,
we would like to thank everyone for the opportunity,
and especially for the future opportunities to come.
|
|
|
|
