This is a brief CR FAQ regarding the EJ series engines used on various Subarus sold into the North American market. I will start with engines that I have personal experience with, and may be expanded to other EJ series motors as information becomes available.

Q: What are the differences between the 2.2L and 2.5L engines?
A: This is definitely a frequently-asked question. One of the best places to start with answers is the NASIOC FAQ's.

Here is a summary of commonly-debated numbers in the NASIOC forums:

Q: Didn't Subaru redesign the engine over the years and change compression ratios?
A: First, see the answers to the previous question. Most current information available is from numbers published by Subaru, or from heresay. It is the latter that has prompted me to take measurements to confirm the numbers.

Q: What are the piston and cylinder volumes? Has anyone cc'd these engines?
A: The NASIOC archives (look for the "search" link) are usually a good place to find this type of information, but I have not been successful from searching the archives. There are plenty of questions, but not many answers. I'm certain that some tuning shops have this information, but I have not seen their data.

This table summarizes some measurements and calculations for stock configurations:

This table summarizes some measurements and calculations for several custom configurations:

¹ From published data.
² From actual measurement. Measurement errors are estimated to be ±0.25cc using a 10cc capacity graduated medical syringe, Pennzoil 10W40 motor oil, and visually sighting the meniscus. Ambient temperature was about 10°C.
³ From Andrew (mcgyver) at NASIOC.
¤ From http://www.sdsefi.com. Notable pages are http://www.sdsefi.com/air17.html and http://www.sdsefi.com/air21.html.
§ From Larry (ImprezaRS dot com) at NASIOC.
ß From Cobb Tuning at NASIOC.
* Estimated from published CR data.

Note that both EJ22 and EJ25 pistons are flush with the deck at TDC, therefore I have assumed that quench volume is determined solely by head gasket thickness. I have done some "what-if" scenarios to see how much of an effect a thicker or thinner head gasket has on the compression ratio to an otherwise stock engine.

I'm not pretending that this is performed using rigorous experimental procedures, but accuracy should be enough to provide ballpark numbers within two significant figures.

What do we learn here?

• My 1997 EJ22 measurements and calculations are very close to the official number of 9.7:1 CR, and are certainly well within measurement errors.
• The offical word from Subaru on the MY00's compression ratio is 10.0:1. This is somewhat less than measured, so maybe the official number is an "effective compression ratio" as a result of intake valve overlap?
• Changing head gasket thickness has a significant effect on compression ratio.
• Yellow entry above is a popular hybrid using an EJ22T block with DOHC heads from a MY98 EJ25 2.5RS or MY97-99 2.5GT or Legacy Outback. Note that the use of a thin SOHC head gasket provides a more favourable compression ratio than the original thick DOHC head gasket.
• Cyan entry above is another popular hybrid using an EJ22T block with SOHC heads from a newer EJ25. This has been reported to be a highly reliable combination due to a low compression ratio and good-flowing heads.

Last updated: 3 May 2003.