Thread: EJ207 Info
View Single Post
Old 07-24-2012, 07:14 PM   #4
Scooby Specialist
Member#: 46135
Join Date: Oct 2003
Chapter/Region: Tri-State
Location: NY
02 WRX
Powered by Sti V9 Spec C



The early GRB EJ207

As the GRB bodystyle started for the JDM, Subaru did not yet have a Spec C model. They quickly started an A-line Sti, equipped with an EJ257 and automatic transmission (different than the USDM EJ255 GRB one) and used an EJ207 converted to throttle by wire, canbus, no intercooler autospray, immobilized, dual AVCS.
All GRB EJ207 have the dark blue injectors, replacing the GDB pinks.
The turbo is a VF49, twinscroll. The intake manifold has functional TGV's. The engine speed limit is 8000 rpm

It is possible, but not confirmed yet, that for these engines, the castings were the thick liner, desirable ones, described below in detail, under "spec C".
The part number for the kit of : block half-castings, castings bolts, castings plugs is:
1008AB210. This is the same part number as for the later Spec C engines, suggesting that while the long block is different, the castings are the same, you'd still get the thick liners.

I believe that these were the Rev A and B
The codes of these engines are

The factory notes for these are two, the first is "for EJ207", the second is "for Sti".

So in a confusing way, the first of these engines, went into a car called "EJ207". My opinion is that this refers to the factory prototype, it could be ANY chassis.

The Sti in Rev A was not yet a Spec C

The Spec C EJ207

Sobaru finally rolls out the Spec C. They were definitely putting together the base for a strong engine for this particular EJ207.
These are dual AVCS, topfeed injectors (they switched to dark blue ones), with functional TGV's, canbus, throttle by wire, immobilized.
These came, at least in the beginning, still with the VF49, then the VF53. Then the VF56 for the R205 and then for the S206.
This engine was homologated for Group N.
The intake ports in the heads are much bigger than anything before, by as much as 2 milimeters. The exhaust ports are larger by 0.7mm, but the inner diameter of the exhaust manifold is unchanged.
The block for the Spec C is a thick casting, reinforced block, I call it a 3/4 closed deck.

The total thickness of the liner and liner support, measured at the headgasket seating area is:
EJ257 7.7 mm
EJ207 GDB 9.6mm
EJ207 GRB 13.2mm
Here is a photo:

I believe that these were the revision B Spec C engines.
the codes for these are:

The factory notes for these two are "for Spec C" and then you have couple option packages.

An important note is that the option packages marked OPC are starting to show up again, with the beginning of the Spec C.
I never really decoded the OPC yet, but within OPC 91-93, in GDB, you had oil radiator, no AC, lightweight battery and maybe small windshield washer tank.
So now OPC91-94 are showing up, hinting to a possible GRB Spec C with no AC and maybe with oil radiator.
A new OPC 98 appears..

How to ID your GRB EJ207

This becomes a very important question, if the early GRB EJ207 can be different..
I have partial information available, from the work of dr20t (Mick).
This partial information is important because it points towards the spec C engine.
As in the photo below, it is marked with B25C. Don't forget that there WERE JDM EJ257 Sti, they can also be marked B25C. However, the EJ20 sqare casting mark, singles thisone out as EJ207.

Mick noted that the extra thickness of the liners (and maybe of sleeves too?) is obtained from a casting similar to an EJ257, but with thicker liners, to account for the reduced displacement (and cylinder diameter) pf the EJ207.
Apparently, the coolant jacket holes in the headgasket, line up with those in an EJ257 headgasket.

The pistons are cast, from all the info available at this time.

These were provided by Mick.

There is still a significant amount of TDC to deck clearance with these. The clearance to deck is the same as a V8-V9, the compression ratio with a hybrid based on heads from V8-9 would not change.
Headgaskets are particular to these.
The part number for the kit of : block half-castings, castings bolts, castings plugs is:

Swapping the GRB EJ207

Swapping into a GDB Body
The GRB needs its ECU to work properly. This time, the discrepancy in range between AVCS and other parameters is larger.
The GRB both in US and JDM can only be CANBUS.
CANBUS is vaguely similar to Ethernet.
So your GBD ECU does not have the "network switch" installed, hardware is missing. It could not control a GRB engine.

Caution: the following is theory only, nobody has tried this yet.

-You could use your GDB JDM EJ207 ECU

-You could use your GDB EJ207 heads and manifold, with all associated sensors.

-You would in this case use the GRB EJ207 headgaskets and bolts, the compression ratio will not change.

-There are 3 sensors that normally would be located on the GRB block and their compatibility has to be looked at:
The knock sensor, water temp sensor and the Crank position sensor. The information available suggests that these should work even from GDB.
In fact, K2Kevin posted in this thread that the 32bit crank sensor is the same as the 16 bit crank sensor and that the only other 32bit conversion was to change the plug style for the temperature sensor, to splice a different style plug into the engine harness.
This has been done in reverse (16 bit sensor substituted by 32 bit sensor), by Kpluiten and K2Kevin, in their projects of converting a 16bit EJ207 to a 32 bit EJ207.

-You would use GDB EJ207 spark plugs and coil packs that you already have.

This combo should in theory be more sturdy than a regular GDB EJ207, due to the thicker liners and flow air the same way a GDB 207 would.

From the latest research and clearance to deck info above, the compression ratio of the hybrid will not change.

Swapping into a GRB USDM body

I am seeing two swaps GRB into GRB with one other possibly done with the owner unwilling to post or share any info. I believe that this third and oldest one is a full and complete swap, including wiring, etc, therefore not that interesting or important.

Once again, here is my theory:

Since CANBUS cand be vaguely compared with Ethernet, each ethernet device must have a switch and a software driver.
When you get a JDM ECU, the first step is going to be to get the immobilizer to allow the ECU to start the car.

For the GRB, the latest information collected from Mick and Dave is that you need as a matching kit: the Cluster, Body control unit, ECU, keys and immobilizer unit.
The GRB would already have the antenna built into the ring.
This may sound like a lot, but many convert their clusters anyway

Once you start the car, the problems you may be facing will have to do with the ECU seeing on the network devices for which it has no drivers.
By example: USDM ABS.

The only thing i happen to know is that German VW implementation of CANBUS uses the VAG software to install drivers each time you replace a sensor.
So maybe there is a driver installation device, like a Subaru laptop and there are drivers that they download from a database.

End of theory, beginning of collected data.

One fo the other two swaps mentioned above, resulted into the engine started but idrive and ABS/VDC non functional. The owner believes that swapping a JDM ABS unit may solve that problem.
I have a different theory:
A board member has tried to implement a quick steering rack, made by Qrack in a GRB Sti, the ECU of that car has then disabled the VDC/ABS, due to unmatching output from the steering angle sensor.
A Spec C ECU would expect to find a 13:1 steering rack, and if it finds a 15:1 standard Sti rack, it will then disable the ABS/VDC probably (this is a theory). The solution would be to get a Qrack.
* Registered users of the site do not see these ads.

Last edited by Vlad; 07-15-2014 at 10:19 PM.
Vlad is offline   Reply With Quote