Building an FA20 Race Engine for Michele Abbate: Part 1- Planning the Internals

by Mike Kojima


We have known Michele Abbate ever since she was a college student doing track days with her trusty Mazdaspeed Protege and have followed her career as friends as she has progressed up the ranks from recreational driver to a Scion Sponsored driver in SCCA's STU class. 

In GT1, Michele has to run against some heavy competition, mainly V8 powered domestic machines in her Edelbrock Supercharged Toyota 86. Although she has been doing pretty well in the class, if she has any hope in doing okay at the annual Runoffs, her engine is going to need a big injection of power. 

Here is where we are coming in. Up to now, Michele has been running her car with a bone stock FA20 engine at a very low boost street blower drive ratio. Although this is decent power, she needs a lot more to be competitive in one of SCCA's fastest classes for production-based cars. 

Our goal is to build her an engine strong enough to take more supercharged boost while keeping great reliability. We will also be increasing the engine's VE to make more power at higher RPM as well. It's a tall order, but we are gonna try to pull this off without losing any bottom end power- just making the engine's powerband broader. How are we going to do it; check it out below!


Road racing is very hard on parts compared to street use or drag racing. The engine is flogged, not just for a few seconds, but for a half hour straight for sprint races 3 to 4 times a day at the track!  If the car is used for enduros, the clogging can last for hours.

Since Michele's FA is going to be subject to forced induction stress for long periods of time in road racing use, we decided to source what we feel is one of the best connecting rods- period. We obtained a set of Pro-H beams from Carrillo. The benefit of an H-Beam is higher resistance to lateral torsional buckling, which is the primary mode of failure for most I-beam structures under compression.

The H-Beam is for extreme use and is stronger for its weight because its resistance to buckling loads is greatest in the direction in which the rod is stressed the most. The H Beam is a more expensive design because the forging is now a multi-step process.  


A Carrillo rod starts as a near net shape blank of forged proprietary steel alloy consisting of chrome, nickel, moly and vanadium alloying elements which meets all VAR (vacuum arc re-melt ASTM standards). Vacuum remelting reduces impurities and pockets in the base metal to reduce internal defects, which acts as stress risers and points of potential failure.

Carrillo rods are heat treated using a maraging process, resulting in an ultimate tensile strength of 195,000 psi with a yield point of 182,000 psi. Maraging means that the alloy used is a maraging steel alloy. These alloys are known for strength without the brittleness associated with many strong and hard steel alloys. A Maraging alloy is precipitation heat treated to convert most of the austenite (a less strong form of steel) to martensite (a stronger form that the heat treating distributes through the steel's crystal structure). 

All Carrillo connecting rods are CNC-machined from an oversized blank forging resulting in the proper grainflow and are surface machined to ensure the removal of any potential inclusions or surface conversions such as scale generated by the forging process. These inclusions can cause stress risers that reduce cyclic fatigue strength.

A final multi-stage shotpeen operation completes the rod. Shotpeening greatly improves fatigue strength by forming a tough fine-grained surface skin on the part. Carrillo Rods are fully machined, Magnaflux inspected and balanced as a set to +/- 1 gram before being shipped to the customer.


The Carillo rods for the FA engines have a unique offset cap to allow the bolts to be tightened in a boxer engine where you can only have 90-degree access to the rods, due to the engine configuration. The rods' caps' mating surfaces have a unique sawed surface to reduce the shear load on the rod bolts that are experienced due to the offset angle of the cap. The main caps have an outer "cheeked" contour that gives strength to the rod cap under side loads as well.

The small end of the rod has oil reservoir countersink on the top of the rod's small end. This helps the small end of the rod get better lubrication. The rod's small end bushing is aluminum silicone bronze with a very smooth surface finish of 4RA in addition to lapped sides and a radius on the outer shoulder to be easier on the wrist pin.

With any rod, the bolts are the most highly stressed and critical part. Carrillo Pro-H rods come with CARR fasteners. The CARR fasteners are made from multi-phase TM material, giving them a higher Ultimate Tensile Strength (UTS).

Carrillo rod bolts are a proprietary design with asymmetrical threads that have more engagement on the thrust side for better strength and more accurate torque readings. The base bolts are made of super strong H11 tool steel. H11 is a steel alloy with high amounts of chromium that exhibits a high degree of tensile strength and toughness. The H11 bolts have an impressive ultimate tensile strength of 220,000 PSI.


Page 1 of 5 Next Page
Bookmark and Share
Monday, November 27, 2017 5:53 AM
What is the FA like in terms of block strength and stiffness compared to EJ? I have read an awful lot about proper torquing of the EJ and its distorsion from fastener load (both block and head fasteners). Within the "home-tuner" crowd, there does not seem to be a definitive and conclusive answer (which is not really rare or surprising). Some swear by oem fasteners and oem torque values, saying that aftermarket fasteners torqued to aftermarket-recommended torque level distort the block's critical surfaces and dimensions too much, causing or at least aiding to a future failure (be it a blown headgasket or a crankshaft bearing trouble). Most real professionals seem to be past this issue, but keeping their secrets of course, so the enthusiast level people still argue about this and it is a point of uncertainty in many an engine build.

I have a personal experience with one stock EJ20 blown headgasket (after some overheating problems so it is hard to blame the engine design) and two instances of a spun conrod bearing on two different nearly stock EJ20 engines. Not having all the details about said engines and equipment necessary, I refused to conclude a cause of failure of these failures, although I would very much like to know.
Monday, November 27, 2017 6:16 AM
I have rebuilt several (>5) stock under-16000-mile FA20-s.
Failures include but not limited to blown head gaskets, spun rod bearings and melted pistons. Cars included both manuals and automatics.
It's worth mentioning that those cars were exclusively on-track rental cars. Track sessions not more than 5 laps (<10mins).
On blueprinting the only thing common between all those blocks was the trueness of the crank bore.
Monday, November 27, 2017 6:30 AM
Interesting. Were those naturally aspirated or boosted variants?
Dan DeRosia
Dan DeRosialink
Monday, November 27, 2017 6:52 AM
I'm sorry to be that guy, but I think you mean STU? I mean no disrespect but at the level of competition in GT-1... someone brought out a Huracan Super Trofeo and barely was under the 115% limit to be allowed to run the race.

That said, holy cow, I didn't realize Piper had US distributors now. Also, didn't reaize the Carrillo rods had serrated parting lines on the big end. Going to be looking forwards to seeing what you guys can do with it.
Mike Kojima
Mike Kojimalink
Monday, November 27, 2017 7:46 AM
Nope GT1, she was running in that class for sponsorship reasons and points for her other ride which was not ready at the time.
Mike Kojima
Mike Kojimalink
Monday, November 27, 2017 7:50 AM
twist, I have some misgivings about these engines robustness (like all Subarus) and am worried about oiling. We will be running a deeper pan, baffles and an accusump. I am very curious to know about your spun rod bearing experiences.
Dan DeRosia
Dan DeRosialink
Monday, November 27, 2017 9:30 AM
GT-1 in anything but a Trans Am tube car is going to be a really ambitious undertaking. Props for shooting for the moon, though.
Monday, November 27, 2017 10:06 AM
RedX, completely stock cars. Even the tires are the same eco-friendly Prius tires. No slicks or semi-slicks.

Mike, all bearing failures were on the #3 rod.
On disassembly all rod bolts loosened with the same torque, so i deduce that on a NA FA the rod bolt is not an issue.
The sump on the other hand is silly small and shallow, especially when compared to an EJ one. Sadly they are not interchangeable.

IIRC the narrower and wider main bearing bores had different diameters (by about 0.002mm)
Mike Kojima
Mike Kojimalink
Monday, November 27, 2017 4:18 PM
twist do you feel that the sump is a major contributor for spinning the bearings?
Monday, November 27, 2017 9:49 PM
Hmmm..... buddy of mine spun a bearing in his FR-S a few weeks ago... Just driving in the canyons. On Michelin super sport tires, but stock suspension.
Tuesday, November 28, 2017 5:34 AM
Mike, I believe that a deep baffled sump is the most cost-effective solution to the FA20 oil starvation issues. Next inline is an accusump. After that dry sump.

In my view, there are actually three contributors to the starvation issue.

1. The sump is too shallow and flat and without any effectice internal baffles. Keep in mind that the oil level is somewhere in the middle of the upper aluminium oil pan sub-assembly. This becomes a problem in long hard corners during which oil is sloshed into one side and can only drain back from one (inside corner) head.

2. IIRC #2 and #3 rod journal oil passageways shared the same main bearing oil gallery, while #1 and #4 had a dedicated main bearing oil gallery.

3. The VVT is a massive oil pressure drop, once activated.

Sorry for my English.

Mike Kojima
Mike Kojimalink
Tuesday, November 28, 2017 8:41 AM
twist, is there a sump and baffle that you have had good luck with?
Tuesday, November 28, 2017 10:15 AM
At the time there were no off-the-shelf deep sumps for the FA20 in Europe. Due to finacial constraints we had to make do with 8mm water jet cut spacers for the sump and a bit thicker for the pickup. The factory exhaust manifold and heat shields needed persuasion to suit. Also oil level was always kept 300ml over maximum.
That seemed to be sufficient for that application.

If I´d have to build one for proffessional level racing today, I would choose a deeper sheetmetal sump and add an oil temperature sensor in the sump.

That may require a custom exhaust manifold, because most headers on the market are made for factory style sump.
Mike Kojima
Mike Kojimalink
Tuesday, November 28, 2017 8:44 PM
Wednesday, November 29, 2017 6:27 PM
can you still wpc treat the king bearins with that coating that they have? or would it kinda negate the coating?
Mike Kojima
Mike Kojimalink
Wednesday, November 29, 2017 10:52 PM
it actually strips the coating
Thursday, November 30, 2017 2:48 PM
when's the next article on this build coming out? I'm curious what you guys are gonna WPC...
Mike Kojima
Mike Kojimalink
Thursday, November 30, 2017 5:50 PM
Lots of stuff!
Post Comment Login or register to post a comment.

MotoIQ Proudly Presents Our Partners:

© 2018 MotoIQ.com