Scion FR-S dyno by Insideline = 173 whp @ 7000 RPM, 143 lb-ft @ 2800 RPM

[phattyduck]

Seems fine to me... 2.5k-3.5k for normal driving (that rev range will get you on to the freeway just fine) and 4.5k-7k for spirited driving.

And I'll second (third? fourth?) the comment that most/all other Subaru motors have that odd dip just under 4k rpm.

-Charlie

[MmmHamSandwich]

Originally Posted by Allch Chcar

Okay, I was just checking since you didn't mention K20 or the last gen Civic.

It was mentioned several times in the conversation though.

[Deslock]

Originally Posted by MmmHamSandwich

2006-2011 Si. These are the years featuring the K20.

FYI for you guys talking about the K20, according to wikipedia the Si had a K20Z3 and the RSX-S a K20Z1.


RSX-S (a bit hard to read, but the blue line is stock):



2007 Civic Si (link instead of embedded image because of its dimensions):

http://www.8thcivic.com/forums/attac...dyno_chart.jpg

[Dimman]

Originally Posted by serialk11r

It wouldn't be cams, a high strung cam has limited VE at low engine speed because it ends up pushing charge out of the cylinder. Intake tuning + intake cam advance can attempt to fix this but you're going to be limited by the exhaust gases "poisoning" the charge.

Because the cam appears to be a high rpm cam, there are no variable intakes, the torque down low approximately matches the torque peak, and the peak torque is quite low by performance engine standards, I think the most logical conclusion is that they intentionally tuned the intake for low rpm. This gives the car "better driveability" (as if you're actually going to floor it while getting groceries :P), and the kick of torque that people like to feel at low speed.

I haven't really thought or read in detail on how intake harmonics work though, so I may be missing something important.

It is pretty co-dependent on the cams/timing, so both statements are 'right'.

How does pressure differential affect flow, Serial? Something like ~ Squareroot(pressure1/pressure2) is I think for rough calculations of flow increase to pressure increase in fuel injectors?

Now consider that the exhaust acoustic pulse can exceed 6 psi (that I think is the measured figure from the ~1960s era book I have) and there are multiple pulses from different cylinders, and they can travel into the intake tract. So it's feasible with the timing we could end up with the cylinder at ~ 6psi below atmospheric and the intake tract at ~6psi above. 12 psi pressure difference can limit the amount of intake reversion on a late closing intake. Or we can use it to fight airspeed frictional losses at higher rpm for better high rpm filling. Where in the rpm range it occurs is a mix of distances traveled by the waves (intake runner and header runner/exhaust lengths) and the time available for the pulse to reach the cylinder through open valves(cams/overlap). Then there is of course sometimes corresponding opposite pressures.

In the AVCS situation it's possible that some of the timing changes can be used to 'catch' or 'skip' the good and bad pressure waves which could have a stronger effect than just inertia of the intake charge.

^ Shit, that's kind of rambly....


(Patiently waiting for Serialk1llr to learn about acoustics so contributions that will lead to epic headers, exhausts intake manifolds and AVCS timing strategies can be made...)

[MmmHamSandwich]

Originally Posted by Deslock

FYI for you guys talking about the K20, according to wikipedia the Si had a K20Z3 and the RSX-S a K20Z1.

They are essentially the same engine. Slightly different intake manifold, and the Z3 has balance shafts that eat 3 hp.

[serialk11r]

Pressure is "specific force" in a sense, so the difference in pressure dictates a difference in force acting on the fluid.

Knowing that you want to see the positive pressure part of the wave to arrive when the intake valve closes, the intake runner length has to be like 3.5 meters long or something to catch the wave the first time it comes back, but the runner lengths you see on cars are more like 1/20th of that, so I need to go think about this more sometime. I'm guessing that a cylinder can utilize the wave generated by another cylinder or something like that, which makes things quite complicated.

At any rate, the intake "frequency" appears to be best matched to the mid 2000s range.

[Dimman]

^ Like I mentioned before, one cylinder is not that bad to figure out on a spreadsheet. 4 cylinders all interacting though...

[CSOCSO]

Originally Posted by ahausheer

Also what octane was used. Car will have more Hp with 93 than 91 but it can take up to a few tanks for the ECU to fully adjust. I've heard of some cars loosing 10 hp going from 91 to 87 because the engine pulls allot of timing.

premium required. Don't even talk about 91.

[Buggy51]

Well... all in all.... is that a good or bad dyno graph? lol

Comparing these numbers to the ones on Scion/Subaru's website doesn't seem to have a huge drop? Just can't for the life of me figure out how to work the graph lol

[Khyron686]

Non-motorhead here, but I remember reading that the engine changes to Direct Injection at higher revs. Is that dip the switchover?

[Dimman]

Originally Posted by Khyron686

Non-motorhead here, but I remember reading that the engine changes to Direct Injection at higher revs. Is that dip the switchover?

No. The D4-S injection is 'continuously variable' in a sense. At low loads it is a mix of both.

The manifold/cam tuning is likely the case. They could likely tune it out but they would probably lose too much low-end to the point that people complain about daily driving it.

[Bahraini86]

Did they perform the dyno-run in 4th gear? Coz in the MT 86 5th is 1:1, don't think redline in 5th is ~100mph!

[Deslock]

Originally Posted by Deslock

FYI for you guys talking about the K20

K20 and FA20 dyno plot http://www.ft86club.com/forums/showthread.php?p=181301

[Chenslee]

That graph shows output higher than the K20 across the board including in the dip, all the way up to about 7100rpm. Based on my experience with the K20 in my '06 Civic, the FA20 in the FRBRZ-S86 is going to be more riot than I have balls to loot in. Then again, that is from different dynos, different conditions.