Project Scion FR-S Part 4- Exhaust and Tuning Frustrations

By Mike Kojima


When we last left off, we had just installed some bolt ons onto our Scion FR-S's motor and picked up some quick power.  Since we had gotten power gains so easily we were confident that the car would respond well to more mods and were ready to test and install additional parts.  To make for more accurate testing, we typically rebaseline a car everything we install new parts.  We strapped the FR-S to Technosquare's Superflow dyno and were both surprised and dismayed to see that our car had somehow dialed out nearly all the power we had previously gained.

We double checked our readings and the dyno's calibration and we are certain that the car did indeed lose power from our last test.

Our stock power levels were 145hp and 119 lb-ft of torque and after a few weeks of driving our power levels had returned to 147 whp and 122 lb-ft of torque for a big WTF.  We didn't (and still don't) know why this happened, perhaps the increased VE of our intake and exhaust caused the super high compression 12.5:1 FA20 to have some detonation (which we could not hear) to which the ECU reacted negatively or perhaps there are some torque limiting algorithms in the ECU itself which so many late model cars suffer from.  Determined to keep adding power, we continued with our testing.

Want more Project FR-S? MotoIQ Project Scion FR-S

For the next step in uncorking the FA20 motor we acquired these front pipes from Berk Technology.  We got a catted downpipe and a race only uncatted pipe to test.  The Berk pipes are made from smoothly mandrel bent 2.5" diameter 304 stainless steel and TIG welded.  The catted version features a metallic cell high flow cat.
It's a little hard to see in the picture but the Berk part is much larger in diameter with smoother bends.  The stock part gets crushed down in a few places as well.  The stock part has a resonator in the back.  We wonder why some of this stuff was not left off.  If they were then you probably wouldn't need the cabin noise amplifier thing that we removed in our last installment.
First up is the catted downpipe.  It installs easily with just two bolts in the front.
Two bolts in the rear and the mount in the center.  Since our car was new, we were able to reuse the stock gaskets.
The Berk piece fit well and really looked like it would produce some worthwhile gains.


Page 1 of 5 Next Page
Bookmark and Share
Dusty Duster
Dusty Dusterlink
Thursday, January 17, 2013 2:46 AM
I'm wondering if a standalone ECU is going to be the only solution to get consistent results on California's 91-octane swill.
Thursday, January 17, 2013 5:00 AM
Why not try water injection?
Mike Kojima
Mike Kojimalink
Thursday, January 17, 2013 10:00 AM
Wrecked, I have a lot of experience experimenting with water injection on high compression NA motors. It doesn't work too well because water is too effective and NA cylinder pressure is not enough to make up for it. You always end up with a power loss and a loss of fuel economy.

I was going to do a masters thesis on super high compression with water injection to reduce knock and NOX but my initial experiments proved so bad I didn't pursue it.

E85 works exceptionally well when injected directly into the cylinder but E85 is hard to get in our area. Visconti has gotten excellent results with it.
Thursday, January 17, 2013 10:31 AM
Yea, I noticed that although there was a huge push for E85 a few years back and California is normally an enviromental whore I rarely ever see even a single pump with E85.

Might be easier to convert it to just run on biofuel...

(that was a joke btw. Please don't do that to this project)

Thursday, January 17, 2013 10:47 AM
Are alot of the dash and body electronics similar to older subarus? Also, might you be able to pull the zero mile motor and sell it for enough to swap in a turbo ej25 perhaps
Mike Kojima
Mike Kojimalink
Thursday, January 17, 2013 10:52 AM
Toyota fixed a lot of the dumb Subaru stuff so this car is actually a pleasure to work on.

Mike Kojima
Mike Kojimalink
Thursday, January 17, 2013 11:06 AM
Mr T. Please note that there is nothing intrinsically wrong with the Nameless header, it is our low octane interacting with the ECU that is wrecking havoc on our numbers. In parts of the country with 93 octane, it has produced over 10 hp.

The headers is also a prototype and Nameless s currently doing development on it. The production part will make more power and we will test and report on it.
Mike D
Mike Dlink
Thursday, January 17, 2013 11:16 AM
"Unlike a Subaru Impreza the FR-S is a pleasure to work on."

"Toyota fixed a lot of the dumb Subaru stuff so this car is actually a pleasure to work on."

can you please explain these statements? looking for some evidence of these statements.
Thursday, January 17, 2013 11:25 AM
"can you please explain these statements? looking for some evidence of these statements."

have you ever done something "simple" like change spark plugs on a 2008+ WRX?

Thursday, January 17, 2013 11:27 AM
Paradox of the day: Is it still a downpipe if it never goes up in the first place?
Mike Kojima
Mike Kojimalink
Thursday, January 17, 2013 11:29 AM
Work on a STI, then work on an FR-S and you will know what I mean.
Mike D
Mike Dlink
Thursday, January 17, 2013 12:16 PM
"Work on a STI, then work on an FR-S and you will know what I mean." not really a valid response as i have not had the chance to work on a FRS and probably wont any time soon. all i wanted to know was the things that are easier, a quick list would have sufficed. especially since i come to this website for "more signal, less noise"

"have you ever done something "simple" like change spark plugs on a 2008+ WRX?" from your comment id say it is more of a pain than doing them on a GC8, which was easier to do by dropping the motor rather than have to deal with the minimal head to frame rail clearance. but is the FR-S easier? it looks as if they may be easy to do, but based on the limited clearance from the pictures provided here, and the cross section picture over at http://www.speedhunters.com/2012/05/getting-to-grips-with-the-toyota-86/ i dont see the FR-S being that much easier to do spark plugs on. could be wrong on that tho as i havent worked on a FR-S.
Thursday, January 17, 2013 12:27 PM
What about running octane booster?

So many cars these days are calling for 95RON from the factory (at least in AUS) that you'd hope they start increasing what you can get over there
Thursday, January 17, 2013 1:56 PM
would california's winter blend gasoline have anything to do with it?
Mike Kojima
Mike Kojimalink
Thursday, January 17, 2013 2:44 PM
Mike D, the list is pretty long, I would say that every single bit of the mess under and on top of the STI intake manifold is now not an issue on the FR-S. The wire harness and plumbing is now really straight forward and not in screwed up layers that are all stuck together like the EJ25. The plugs are still sorta lame but look not to be as bad but I haven't tried to take the FR-S plugs out yet.

The STI is sorta like, good thing the engine is easy to take out because it is lame to work on in the car. It's to the point where I am thinking of making a mil-spec engine harness just to lay the wires out in a way to make the car easier to work on, that is if I keep it.

There will always be guys that dispute how hard it is to work on a STI but I asked the guys at Cobb who work on them every day if they have a hard time to see if I was smoking crack and they tell me how much they hate working on them.

They also told me my car is really cleanly laid out and done well. They told me they spend a lot of time cleaning up other peoples and shops messes on Subarus.
Mike Kojima
Mike Kojimalink
Thursday, January 17, 2013 2:49 PM
Chris, what I don't like about our smog legal octane boosters that are easy to get is that they leave thick orange deposits on everything. Plus octane booster would be sort of a bandaid on a problem that I would rather get figured out.

Rich, that is a good thought that didn't occur to me. We were testing when California did an emergency shift to winter blend gas to reduce fuel prices which were nearing $5 a gallon at the time. It was still pretty hot so the volatility of the winter blend in relatively warm weather was probably making things worse.
Mike D
Mike Dlink
Thursday, January 17, 2013 3:28 PM
Thanks Mike! That is exactly what I was curious about. The STI, is probably more of a rats nest than the GC8 I owned. I don't think I've read anywhere the differences such as you've mentioned. I know about subaru design flaws, one would hope that at the time of design there was a perfectly logical reasoning behind it. But there really aren't many places you can get a more indepth and I guess "techy" stand point from a review. I enjoy the coverage on the mods and the car. Maybe in a future article you could cover such differences that toyota has brought to the table.
Thursday, January 17, 2013 4:31 PM
The octane booster in a bottle doesn't really do much. Maybe 1 point.

Meth/E85 injection should help significantly vs. just plain water injection. Some paper from MIT back in the day where they did a study on it.
Thursday, January 17, 2013 4:40 PM
Would mixing a few gallons of 100 octane in be a relatively painless way to fix our gas? I know some 76 stations have it.

Thursday, January 17, 2013 5:05 PM
Try looking into the oil temps. Perrin found about a 10hp range from cold to hot.
Justin Banner
Justin Bannerlink
Thursday, January 17, 2013 5:48 PM
SubiNate, I believe you can only buy race gas at a race track now in California. I haven't noticed if 76 does indeed have it, so I'll check the next time I need gas.
Dan DeRosia
Dan DeRosialink
Thursday, January 17, 2013 6:02 PM
I think I've worked on so much old terrible stuff that the Subaru wiring stuff seems like a joy in comparison. 1st gen RX-7s have so much weird unnecessary stuff underhood (or maybe I've just never worked on a street one) and good god my brother-in-law's Celica Alltrac. Compared to that, I can actually figure out what the Subaru wiring is there for, which is such an amazing novelty.

Also, man, this has to be irritating. This and the Noveske article just make me glad I don't live in California (weather forecast going to below zero this weekend here makes me wish I did though)
Thursday, January 17, 2013 6:10 PM
Is this the typical Toyota CAN-BUS ECU crap? I'm wondering if it is? On the newer Toyotas, you can "tune" anything you want, but after a few days, the stock ECU fights it...and eventually wins.

It was like this on my IS350..and other newer Toyotas I've had experience with.
Mike Kojima
Mike Kojimalink
Thursday, January 17, 2013 7:45 PM
Haha, first gen RX-7. Mine was easy to work on when I threw that stuff away! I got lucky and sold it before we had difficult smog certification. I am sure that the next owner probably had to junk the car as it would have cost a lot to get it back to stock. Jeff's FD was pretty scary before the LS. My other favorite car not to work on is the twin turbo z although the only really tricky part is working on the turbos in the car. The main difference between the EJ25 and the FC20 is the Toytaru has a nice clean and logical layout of wiring, vacuum lines and plumbing under the manifold and the wire harness isn't all stretched out and tied into stuff just to piss you off. Changing injectors is easy.
Mike Kojima
Mike Kojimalink
Thursday, January 17, 2013 7:47 PM
Khiem, the MIT research was for direct cylinder injection of E85, not injection into the intake tract. My post grad experimenting had it as a big flop.

RRocket, our IS-F doesn't seem to suffer from "now you see it, now you don't" power.
Thursday, January 17, 2013 8:55 PM
@Justin Banner

The station I remember having race gas was a station at the I-5 and 269 interchange east of Bakersfield. Maybe they have it because of their proximity to the socal race tracks.
Thursday, January 17, 2013 9:00 PM
I never felt it was too hard to work on 02+ WRX/STi, never removed the engine though. Then again, my first performance car I worked on was a WRX and I bought two STi after that so...

I will say, removing the STi intake manifold and replacing the TGVs is a major pain. Not only are there some difficult bolts and nuts to line up or put on, but squishing all that crap under the intake manifold back in place plain sucks. The other stuff like spark plugs, you just gotta know what to move out of the way first (like washer bottle).

It is interesting to hear that the BRZ/FRS is easier to work on, maybe some of the Toyota layout know how will make it to some other Subaru models.

I think there is a new E85 station in Huntington Beach if you had not seen it yet.
16990 Beach Blvd
Huntington Beach, CA 92647
Mike Kojima
Mike Kojimalink
Thursday, January 17, 2013 9:03 PM
How do you like the wire harness bolted to the bottom of the intake manifold? grrrr.
Thursday, January 17, 2013 10:33 PM
@Marc "have you ever done something "simple" like change spark plugs on a 2008+ WRX?"

Ha! If you think changing spark plugs on an Impreza is hard, have you tried changing FRS/BRZ spark plugs?
The manual has 16 pages dedicated about how to change the plugs. The procedure involves raising the engine up in the air, etc. Good luck.

I have two turbo Subaru cars (including the '08+ WRX you mention). But for a RWD roadster IMHO a MX5/Miata is a much better car. Better suspension/handling and much easier to work on and make it fast.
Thursday, January 17, 2013 10:39 PM
@Justin Banner
" Thursday, January 17, 2013 8:48 PM
SubiNate, I believe you can only buy race gas at a race track now in California. I haven't noticed if 76 does indeed have it, so I'll check the next time I need gas."

Not true. The choice of selling 100 octane is up to the gas station. I know stations in LA which do have 100 octane pumps, the closest being a 76 one.
Friday, January 18, 2013 5:09 AM
"Ha! If you think changing spark plugs on an Impreza is hard, have you tried changing FRS/BRZ spark plugs?
The manual has 16 pages dedicated about how to change the plugs. The procedure involves raising the engine up in the air, etc. Good luck."

christ, that's uncalled for. i know on the wrx i ended up usinga thick sheet of rubber around a pry bar to move the engine over about 3mm so i could get the coilpacks off instead of disassembling half the top end.

they could have simply made the clerance 3-4mm more on each side without too much trouble or used shorter plugs/valve covers. there's a bunch of ways that would have avoided that issue but since it was designed in a computer by people that never get their hands dirty, this is what you end up with.
Dan DeRosia
Dan DeRosialink
Friday, January 18, 2013 5:45 AM
I will say that after seeing Crawford's pictures from an FA20 teardown, I'm really kind of wondering if it'll bolt to the older transmissions. FA20 with a turbo in a GD chassis car seems like it would be nice. I mean, given an ECU that can make it run properly. And yeah, the 1st gen RX-7 is so much easier to work on with a new wiring setup. I should actually do a better job on the next race one though; it's not exactly what one would call mil-spec wiring.
Mike Kojima
Mike Kojimalink
Friday, January 18, 2013 6:59 AM
I changed plugs in a GD without shifting the engine but I had to use a plug socket with a hex on it and a lot of time. It probably would have been quicker to lift the engine.
Friday, January 18, 2013 7:21 AM
RRocket: "Is this the typical Toyota CAN-BUS ECU crap? I'm wondering if it is? On the newer Toyotas, you can "tune" anything you want, but after a few days, the stock ECU fights it...and eventually wins."

it's funny you say that. i use the 2gr-fe (3.5L transverse) in an MR2 and with headers and an intake the motor makes about 70HP more over the choked stock configuration. it's been that way for over 30k miles now and the ecu hasn't taken the power back.

Friday, January 18, 2013 7:46 AM
Plugs in a subaru arn't that terrible if you have the right extensions and a swivel socket. The only suabru I havn't done them in is a gc but I can't imagine its much worse than an GD or GR. Once you do them once or twice there's nothing overly complicated about it and all the coil packs come right out if the rotate them to the proper orientation(one or two must be turned 180) before trying to slide them out.

Do you think its going to be necessary to replace the pistons with lower compression ones in order to add reliable forced induction to a frs/brz?
Friday, January 18, 2013 8:21 AM

If you're comparing the FR-S to the mx5 I have to disagree, despite the suspension design differences the FR-S runs at approximately the same level stock that a spec mx5 runs. A modded BRZ spanked a track prepped S2000 time at Streets of Willow Springs with the same driver.

It's a very capable car. If it's doing this well with only a few months of development I'm excited to see where things go in a year or two once it's quirks have been sorted.
Friday, January 18, 2013 8:58 AM
With the mods so far (airfilter, exhaust, header, ECU, ... 91+ octane), what would be a conservative estimate if you slap all the parts together ?

Looking forward to the next part of Project FR-S!
Mike Kojima
Mike Kojimalink
Friday, January 18, 2013 9:46 AM
Amp0412, please come over next time I have to work on my Subie!

Me80, people in parts of the country with better gas are getting close to 200 whp which would be about 190 on our dyno.
Saturday, January 19, 2013 1:33 PM
while hOpes & Dreams are found in the mechanicals & tune..
the Sexx & Magic has always been in the *Fuel*

i asked Moto @ club4ag a lil about the FRS/BRZ ECU and how it functions,
as i view it as a PRIME candidate to illustrate the principals of fuel blending & associated inherent chemical combustion characteristics..
especially with the high degree of monitoring & additional control over the 4stroke process the system offers
He relayed to me that the ChiefENG said the stock ECU (D4S) was capable of constantly monitoring the DI, combustion rate, & temps
and would determine the spark ignition point closer to TDC accordingly..
(less theoretical ignition advance)
Chief gave the "Ok" to anything from 87pump to E10-E15 and possibly even higher alcohol blend ratio's (lowering of actual stoich)
Moto said he actually tried running the [87]pump (E10 i assume) thru his FRS..
and it ran fine, tho a bit down on power??
Which i personally feel is a Reflection more so of the chemical makeup of the Cali[87]
as opposed to Cali[91] which probably has more Aromatic(BTX)content to rate higher on that A-K-I scale = (R+M/2)
*Aromatics = Benzene-Toulene-Xylene(s)
Which are the commonly used High-Energy bearing components of most blended hydrocarbon(read:gasoline) based fuels..
So Energy Density per gallon is most likely greater in the "high octane" grades of gasoline..
Thus the major contributor to the increased power on dyno charts..
And not so much the "octane rating" in itself in this case..
BTW: in my experience Unocal76 sells a 100octane with a very low energy density.
Try to Realize that "octane ratings" tell you absolutely nothing about critical combustion characteristics like flamespeed, combustion rate, & BTU's per gallon..
Selling a "rating" can be a very profitable thing!!

if Enthusiasts want a better combustion to make power..
it helps to Start by thinking about the equation, Fuel --> Forwards
as opposed to thinking... parts & tune <--- backward
enthusiasts have become so enamored by the Thought of economy based fuels doing high performance things for them..
that nowadays they Demand a silk purse be made from a sows ear..

once again i reiterate my simple solution for a more Performance minded fuel blend for the California people..
Cali [91]pump E10 + VP's M1(100%methanol)
@ [14:1] = E9M6
15% total alcohol = less energy density than pump E10 of any grade
BUT enough flamespeed to demolish the thermal efficiency of any engine optimized to burn gasoline
Lucky the D4S can constantly monitor and adjust on its own to burn many Different fuels efficiently!!

on a side note..
and i'm willing to bet Toyota's map for AirCon on setting 2 or 3 makes a supercharger like power band when u hit 10% meth in your fuel
i know it does in my Scion Xa & V8 Tundra
Monday, January 21, 2013 1:33 AM

If you have not seen the GearWrench swivel magnetic spark plug sockets, they are pretty awesome and come in a Subie size. The link below is not magnetic though.


Coincidentally, a friend just bought a JDM long runner intake manifold and we got to compare it a bit with the USDM one. It appears a bit better thought out. On the USDM there is a pretty large wire loom that goes over the turbo inlet, while the JDM one routes these wires beneath. Also, on one side the JDM one only has "half" an injector shield.

But yeah, too much crap bolted and zip tied to that manifold no matter what country it's from, lol.
Monday, January 21, 2013 2:27 AM
Hi Everybody,
I am following Motoiq for quite a while now to have seen a few articles where pump-gas became quite an issue.
I don`t know the whole story about US pump-gas and why there is mostly only low-octane fuel available in the States.
For me as a german where 95-octane fuel is standart it`s just not understandable why they still make you fill up your car with "water" and make you run on it especially with more and more high-compression engines on the market now.

I bet I`m not the first one to ask but why is it like that?!
It must have some reason since we here (at least in the southern part of Germany) even get 102-octane gas at some stations.

I just had to get that off my mind!

Monday, January 21, 2013 2:39 AM

the fuel here is actually quite nice. everywhere except California which has excessive pollution laws requiring adatives that aren't conducive to high performance engines.

here in the midwest it's easy to get 96 octane ((RON+MON)/2) you have to keep that in mind, many countries rate in RON only. a fuel will usually have a higher RON number than it's MON number so you can't compare both indexes to each other directly.

and yes, just like Germany you can easily get 100+ octane fuel at many gas stations but certainly not all of them. it just comes down to sales. if the market demands it it is available. I've seen all the way to 118 available at a gas pump but those are usually near larger lakes since some boats have some seriously large motors and they eat through enough fuel to be worth stocking.

Monday, January 21, 2013 2:44 AM
here's a good read about RON and MON: http://www.refiningonline.com/engelhardkb/crep/tcr4_29.htm

as you can see, i'd rather have 96octane RON+MON/2 than 102octane RON. it'll usually have a higher resistance to detonation.
Monday, January 21, 2013 5:00 AM

Thanks a lot for the answer and now I can see things a little clearer since I was also wondering why some cars though making
the same power and torque on paper require higher octane "euro"-fuel compared to US-spec models who seem to run fine on US fuel like i.e. the R35-GTR.

Monday, January 21, 2013 5:17 AM
Just a short comparison 100 oct. RON equals around 88 oct. MON.
Since Euro-Fuel is rated in RON it`s not even that good as I thought.

Monday, January 21, 2013 5:22 AM
it really depends on the exact fuel mixture. but that seems to be a commonly accepted conversion ratio.

making the 91 octane available in california closer to what you'd see sold as the 95 at your pumps.

Monday, January 21, 2013 9:12 PM
@Vmax, the R35(especially 2012) has boost issues in California every now and then with the stock ECU and tune.

The rules on race gas aren't too complicated. If you pay highway tax on it, aka 100 octane "street" race gas-76 etc, you can pump it into anything.

If its race gas with no highway tax, you are not allowed to pump it into a street car. The laws got a bit more strict in California a couple years ago. A pump/seller has to keep records of where their fuel is going.

Don't confuse our RON+MON/2(PON) fuel ratings, with only a RON number. Most of the world only uses a RON number, the US uses the PON.
Tuesday, January 22, 2013 2:21 AM

Thanks for your input! I heard about the new laws on race-gas which I think is weird since its better quality than pump gas and burns better so I think if people want to spend the money on it why not as long as its unleaded.


p.s.: Sorry for getting off-topic here!
Tuesday, February 19, 2013 1:48 AM
A.K.I. = anti knock index
is the Correct term is for (RON+MON)/2

idk what that "PON" is..
Tuesday, February 19, 2013 8:22 AM
Its a pretty common term. Its the one I hear referenced. It is what we get out of the pump in the US hence the "pump octane number". RON+MON/2 is clearly marked under all pumps here.

"(R+M)/2 (Pump Octane Number) as the name implies, is determined by adding together a fuel's RON and MON ratings and dividing by two. This is the common rating found on pumps in North America."


"Anti-Knock Index (AKI)
In most countries, including Australia and all of those in Europe, the "headline" octane rating shown on the pump is the RON, but in Canada, the United States and some other countries, like Brazil, the headline number is the average of the RON and the MON, called the Anti-Knock Index (AKI, and often written on pumps as (R+M)/2). It may also sometimes be called the Pump Octane Number (PON)."


So its like saying hood or bonnet. Just because one place uses one term, try and be smart enough to know there is also another term used for the same item.
Saturday, March 09, 2013 8:53 PM
Hey Mike, do you guys have a video of how the Nameless + Greddy Evo3 exhaust sound like? I know that the Nameless is still in R&D and I already have an EVO 3 and want to hear how the combo will sound like.
Post Comment Login or register to post a comment.

MotoIQ Proudly Presents Our Partners:

© 2018 MotoIQ.com