posted on September 09, 2013 10:31
Project FR-S: Getting Torque Everywhere with an Innovate Motorsports Supercharger
Our FR-S is a difficult car to deal with in the engine department. The ECU with a lot of command authority has been driving us nuts. We have put several thousand miles on our FR-S over the last few months and it seemed like the car was once again dialing out a lot of our power over time.
If you can remember from previous installments we had done a accumulation of several hundred dyno pulls trying to get repeatable dyno results, removing and re-installing parts many times. Being curious we decided to dyno our car again. To our dismay the car had returned to nearly stock power levels again.
We were thinking that the erratic power readings might have been due to the tune we had done on the car that perhaps wasn't very suitable to our 91 octane gas. To test this thinking we flashed the ECU back to stock and still experienced wildly erratic power on the dyno.
What we ended up doing with the help of Innovate Motorsports was to reflash our ECU with a stock program with a less aggressive ignition timing multiplier constant. This gave us consistent readings but we were disappointed to find that our cumulative mods only made 5-6 more whp than stock. At least the runs were stable.
Our theory is that the FA20 engine with its high 12.5:1 compression is not really happy with California's crappy 91 octane pee water. Possibly a less aggressive tune might actually make more power. It works for cars like the EVO and the STI!
Being confounded with our efforts to gain power, we decided to take a big step by going forced induction. Innovate Motorsports has just come out with a twin screw positive displacement supercharger kit for the FR-S and BR-Z in addition to a tune that is stable with California fuel.
We have liked the torque everywhere, wide powerband, bigger engine feel that a postive displacment blower gave on some of our other projects and our FR-S surely needs more power everywhere!
The Innovate Motorsports supercharger kit is very complete and comes with the supercharger, manifold and every single belt, bolt, clamp and hose needed for a complete install. The kit uses a Sprintex S5-210 Supercharger which uses a positive displacement screw type compressor. Screw compressors have internal compression, unlike conventional roots blowers which makes them inherently much more efficient.
The blower manifold has every boss and bolt hole cast into it for a very clean OEM like install. There is nothing loose or unsecured or out of place due to the kit. This makes the installation a snap because typically half of your time gets wasted when you are trying to reroute hose and make brackets and stuff on a lot of kits. Not this one!
The rear of the kit is the same, great fit and finish with attention to detail.
The manifold has all of the fittings for the OEM vacuum lines in the proper places.
Monday, September 09, 2013 11:19 PM
It got rid of that big torque drop at 4k rpms too. That's major bonus.
Tuesday, September 10, 2013 4:07 AM
Looks like a good kit? How about water injection instead of an intercooler?
Is the FRS ECU strategy different to Subarus of past? In other Subarus increasing the advance multiplier to 1 wouldn't reduce the ability of the ECU to pull timing. It would just increase the total timing right after an ECU reflash.
Tuesday, September 10, 2013 5:47 AM
what made you guys choose a supercharger over a turbo kit? is response a higher priority than power? not impressed with the price/performance/reliability of current turbo kits?
Tuesday, September 10, 2013 6:08 AM
Geeze, what a pain to tune. Subaru knows that this model is going to be often modified and should have run an ECU that adapts towards standard modifications, not against them.
Also, I think it is fair to call BS on the factory 200 hp number. If FI can't even get the WHP number back to crank hp then something is wrong.
Tuesday, September 10, 2013 7:23 AM
I'm sure many others are having the same tuning issues you are and there must be an ECU in the works rather than just a flash. This is just a small jump in power compared to cars of the past, and I'm wondering what the tuning ceiling will be. Have you guys tried running a tank of "imported" 93 octane or other fuel through the car? CA gas is notoriously bad for tuning. Maybe I'm just always looking for a workaround rather than solving the real problem :/
Tuesday, September 10, 2013 7:34 AM
Maybe Subaru did what they thought was best to protect themselves as far as warranty claims. The BRZ is so popular, every backyard tuner might take a crack at it. This isn't a low strung/low compression motor to start with, so maybe it just doesn't have a lot of latitude for inept tuners messing around.
If i was Subaru, i could see making the factory ecu do conservative correction, while assuming something like KPro will allow people who really want to tune to go to work (assuming something like that can exist for this platform). At least with a KPro mod, Subaru would know if the factory ecu has been altered/tuned. It's inevitable that the BRZ/FRS will gain good ecu tuning tools given their popularity, no?
Tuesday, September 10, 2013 7:45 AM
I think you are right about the warranty issue, I just think that it is silly for the car manufacturer to try to protect the tuner. You modify certain things, it voids the warranty. I never understood why that is such a difficult concept for some tuners.
Also, you are correct that a full stand alone is not installed (yet?) and maybe these guys just need more time to flash certain parts of the ECU to fix the adaptive process.
I thought I remembered a project WRX from SCC where the ECU had actually added hp on the stock vehicle after significant hard driving and they couldn't reproduce the results after resetting the ECU.
Tuesday, September 10, 2013 8:02 AM
I was thinking more like Subaru protecting themselves from the tuner than protecting the tuner. February tunes don't necessarily work in August unless you plan for it. It's pretty easy to imagine a scenario where a tune works in colder weather, but summer heat brings pinging and there are engine problems. The car owner is then able to bring the car in for warranty service. i know this is a potential problem with any car, but i think there is more latitude in most engines. This is an inexpensive car (relatively speaking) with an audience that will want to modify things but without the know-how. When an engine is pushed as this one is from the factory, are there timing changes that have to be made to run I/H/E? I have no idea, but a ton of people probably don't even know to ask that question. When I said backyard tuner, i was using the word tuner in a joking way - i meant someone with the ability to fit a CAI only because someone said it would make the car faster (and with no other understanding or consideration of what else might be involved).
I'm not sure if the S2000 received the same attention as the FRS/BRZ, but it's the only other similar car i can think of that's so highly tuned from the factory. Cars i can think of from my school days in the 1990s like the EG/EK civic, an NA rx7/300zx, a 240sx, sentra se-r, etc., had a relatively mild tune from the factory and it was safe to add I/H/E without much in the way of tuning considerations.
Tuesday, September 10, 2013 8:11 AM
12:1 compression and 91 octane fuel. Direct injection or not the engine isn't happy. You have to tune with our fuel to know just how bad it is. If feels worse than 91 octane. When we went from 92 to 91 it was a huge difference.
Tuesday, September 10, 2013 8:19 AM
If the Ecutek timing and fuel control strategies work well with the supercharged car, wouldn't they be able to be applied to the NA car as well? I know that the Ign. Mul. Constant was changed on the NA car but what else was modified, besides fuel and timing target tables, on the Ecutek flash that got the consistency? If the 91 Cali fuel is that bad I would be worried about long term issues with knock/other issues as fuel quality, and driving conditions vary as the ECU won't be technically as quick at stopping those issues.
Tuesday, September 10, 2013 9:24 AM
The fuel and spark map when NA were left alone. We just changed the intake advance multiplier. Since it is 1 it will not increase the amount of advance, simply reduce how aggressively the ECU can remove timing. It can still remove it though.
Tuesday, September 10, 2013 9:58 AM
Some of the things I plan on trying are a water to air intercooler that Innovate has in the works right now. We will get one of the first production pieces. Modify the intake tube so we can use the less restrictive Greddy airbox and filter. Water injection and a smaller pulley. Custom tuning. Oh yeah an oil cooler because some of the power fade could be oil temp related based on experiences others are having.
Tuesday, September 10, 2013 10:17 AM
Just out of curiosity, what station are you getting your fuel at? I ask because I know Chevron's fuel is higher quality and I was curious to see if you have tried that. Maybe you just need to visit you local Autoslut and get some octane booster to see what kind of results it will pull out, although having to go and buy some every time you fill up doesn't seem like the most "luxurious" idea.
Tuesday, September 10, 2013 10:27 AM
An intercooler and adding straight water injection to combat detonation is what I was thinking as well.
Tuesday, September 10, 2013 10:48 AM
We use only Chevron in MotoIQ's FI project cars.
One thing I hate about octane booster is the weird orange residue they leave over everything. I have always felt that this couldn't be good for the O2 sensors even if it says O2 sensor safe.
Tuesday, September 10, 2013 11:08 AM
This thing is BEGGING to have water/methanol sprayed into the intake. I think the power gains you guys saw are only the tip of the iceberg. That Sprintex 210 blower should be capable of *a lot* more power.
Tuesday, September 10, 2013 11:36 AM
I understand the use of the advance multiplier but was curious if this was modified on the Ecutek map for the supercharger as well as any other constants or factors used by the ecu for timing and fuel correction. If they weren't it would seem you would be having the same problem with the supercharger kit down the road where you would be seeing power lose after some driving time.
Tuesday, September 10, 2013 11:40 AM
@ lastminute, Protodad, etc: the FR-S is so hard to tune compared to other Subaru-powered vehicles because it's not running Subaru engine mangement. If you remember in my glorious FR-S review (http://www.motoiq.com/MagazineArticles/ID/2505/Tested-Scion-FR-S.aspx), the FR-S is running Toyota's D-4S engine management, which is just as recalcitrant to modifications as other Toyota products, probably even more. Anyone who's thrown breathing mods at a newer Toyota without the ominus dominus from a TRD reflash knows this pain.
@ SupraHKS: Even with Chevron and other high quality fuels, it's still high quality shit. The Arizona/California/Nevada "91" (ACN91) octane fuel is so horrible that major tuners release unique maps for it. The formulation itself, mandated by our best buddies at CARB, does something to the fuel that makes it perform worse than it's octane rating suggests.
Also, octane boosters aren't incredibly effective since octane works by averages. So that $8 12oz can of "NOS 100 OCTANE!" against 13 gallons of 91 pisstane results in something like 91.1... :)
Tuesday, September 10, 2013 12:42 PM
We currently have the .7 multiplier with the supercharger. We are probably going to change it to 1.0
Tuesday, September 10, 2013 2:57 PM
ah, that makes sense. malice or ignorance? does toyota purposely make ecus that are difficult to work with, or is it just that they haven't made any efforts to make tuning easier? i guess i haven't read much about this because there hasn't been a toyota i've wanted to own or tune for quite some time.
is there a standalone that keeps street legality (or fakes it) with respect to obdII compatibility? my understanding is that kpro is good for this in the honda world
ps., can't believe i forgot something from that glorous review. guess my mind was so blown, i couldn't retain the details :)
Tuesday, September 10, 2013 5:59 PM
Short answer is no.
Wednesday, September 11, 2013 7:07 AM
Supercharging Subaru engines has gotten a lot more common in the past few years, and I'm glad to see it! It makes my '00 RS/C look like less of a freakshow! :D
Not being well versed on modern ECU emissions and such, are there any piggybacks that could serve to tune the FR-S/BRZ, or would that just be adding problems?
Wednesday, September 11, 2013 8:32 AM
Piggybacks don't really work all that well. They fight the stock ECU.
Wednesday, September 11, 2013 8:52 AM
Wow...I was shocked to learn this kit only costs 3700.
Wednesday, September 11, 2013 10:39 PM
Element tuning has a standalone out http://elementtuning.com/store/#!/~/product/category=439245&id=1509847
Thursday, September 12, 2013 10:00 AM
Sigh...This kit gets produced/imported from Sprintex in around a year, and USDM GE8 Fit guys are still waiting on ours 6 years later.
That's what we get for buying into grandma car chassis though.
Thursday, September 12, 2013 10:05 AM
Oh, and definitely keep us posted on the gains this car gets from smaller pulleys!
Thursday, September 12, 2013 10:05 PM
Some Guys in Aus are using a combination of ecutek, haltech ps1000 and a flex fuel sensor to get the required flexabilty on e85 till some one develops a complete stand alone soloutiuon.
Sunday, September 15, 2013 9:23 PM
As someone who works in Toyota, I think I have a theory as to why your having problems with the repeatability of the power. Sadly TMC (Japan) did all the engine cal for the FR-S so I cannot be 100% positive.
So I think it runs the newest version of our (well, Denso's) ECU which has a really complicated knock control system. Basically to sum it up, Toyota takes the stance that the computer should adapt to the engine so that there is always a little bit of knock (a little knock isn't harmful and lets you get a little more power). It is an active system, so it adds timing until it knocks. This is unlike most systems I have seen, they are usually passive (so pull timing when there is knock). Toyota's system then looks at the waveform generated by the knock sensor, the intensity of the knock sensor output (Vmax), and the time gate it happened in crank angles.
This system is also brandy new, before the knock system just looked at the intensity of the sensors output, not waveform matching. It also has tons of variables that need to be calibrated, so that may be a factor too.
So the point is that I think the poor CA gas may be working the knock system a little more than TMC thought it would, so if I was a betting man and you were able to log ign timing during those pulls i bet you would see it vary quite a bit. Furthermore, the system does work over longer terms to learn the engine and knock sensor so that may explain your power drop off over time (or the current tuner software isn't fully overriding the stock system, it is really complicated).
Just an idea!
Monday, September 16, 2013 5:53 PM
Don't worry, the GF will still have plenty of support. Unlike my GC...
Tuesday, September 17, 2013 3:05 PM
@hydrochloric, I think we're talking about different things. I'm referring to the 09+ (GE) Honda Fit, for which Australians have a Sprintex PD SC kit, that doesn't work on USDM Fits.
Some people have been able to get custom turbo setups working, but not without major issues, major expense, or both, so I and others have been waiting on Sprintex to deliver for a while...
Also, I wrote 6 years when I meant 4. Must have been thinking in base 12...
Wednesday, April 09, 2014 5:07 AM
For whatever it's worth, this thing's WHP/L is right in line with more expensive but similar engines (NA, no cam profile switching). Toyobaru should have hooked this up with Toyota's VVL-i system and 500ccs more displacement. A torquey 200whp in a 2800lb body is pretty much perfect and in line p/w wise with "faster" but significantly lardier hardware
Wednesday, April 09, 2014 7:21 AM
IIRC, Toyota was concerned about not getting the revs they wanted (north of 7000rpm) with 2500ccs.