My 05 evo 8 started stumbling occasionally in the last couple weeks and throwing the random misfire code with the check engine light. I’ve got 6,000 miles on the car and this hadn’t happened before. I searched the forum and read through most of the random misfire related responses without conclusion and started looking at what I’d done different in the last couple weeks.

I’ve been adding rows and columns to my exede maps and scaling every cell for low boost at low mass air flow and increasing boost, ignition retard, less fuel, with increasing mass air flow. The result of these complicated maps is very snappy throttle tip-in combined with heavy 3500 rpm torque at full throttle, and big top end. Unfortunately it’s too much for the factory ECU / piggyback EXEDE combination. Even though the difference between cells was quite small, the fact they were all different required more brainpower than the factory ECU had. I’ve since gone completely the other direction and have only 3 or 4 rows of cells per map (5,10,100% MAF) and the top couple rows with the same values to establish the shape of the curves. Not only is the random misfire problem gone but the car runs smoother all the time. This and the research I’ve done about the myriad of conditions that can cause random misfire leads me to believe the stock ECU needs more ram or better clock speed or increased cooling and a reflash from the factory isn’t going to solve the problem. I really think this is a hardware issue.

 

It has been suggested it is the CAS that is the hardware problem...

 

pm'd ya

 

So are you suggesting:

1. That the P0300 issue has NOT been fixed on the 05s? and
2. Xede can solve the P0300 issue

Or

1. Your misfire issue was only due to the map in the Xede and was fixed after you made the changes you described above

Did you try disconnecting the Xede to see if the P0300 went away?

Also what is CAS and how is this a ECU hardware related issue? Can you put it in layman terms for peeps like me who don't understand the techy jargon.

 

1. Yes
2. Not necessarily, but it might eventually

CAS = crank angle sensor

My 03 throws P0300 pretty regularly. There are changes in the new v3 firmware that change the way the CAS is replicated, so that may alleviate the problem. However, even bone stock cars throw this code, and it could be down to the fact that the CAS sensor itself only has a 4 tooth gear to get its position reference from (which is VERY rough accuracy IMHO).

Hammerr : do you know what your AF's are?

 

Will replacing the CAS fix the issue?

 

There is 2 ways to get a P0300 error. One is the drivetrain vibration and is not XEDE driven. The other seems to be a few cars that run the XEDE. So if you still get a p0300 without the XEDE installed then you have the drivetrain vibration problem.

 

Nope because currently there is nothing to replace it with....

 

I believe that mine is harmonic induced (drivetrain, catless, etc....).

 

So, drivetrain vibration is causing a misread on the CAS, causing a misfire creating the stumbling I feel?

 

No the CAS problem is XEDE related. The drivetrain vibration will cause a harmonic frequency that will trigger the knock sensor and create a misfire. That is why the guys who suffer with the P0300 end up getting new transmissions.

 

The CAS is a hall sensor (magnetic). I don't see how it could possibly be affected by a drivetrain harmonic. If the CAS fails, the ECU will try to guess the crank position to keep the engine running. Has anyone actually replaced the CAS? It would be interesting if that fixed the problem.

The knock sensor is not a timed device, and there is no signal it could produce that would cause a misfire.

It is my opinion that the misfire may be due to a fault in the ECU itself. While misfiring, it will continue to misfire until the code is cleared with the scanner. This makes no sense because the presence of the code in memory (should) have no bearing on actual the operation of the engine. Since clearing of the code from memory suddenly clears the misfire, it seems the fault lies somewhere in that mechanism.

 

I believe when the knock sensor "hears" a hard knock it will cause the ECU to cut the spark and the fuel to the effected cylinder.

 

No, typically when the knock sensor triggers a signal, the ECU will retard spark advance globally, by a fraction of a degree per crank rotation until the signal ceases. The ECU does not cut fuel/spark, and has no means of determining which cylinder is experiencing the knock.

 

If the problem is with the ECU then why does Mitsu replace the transmissions on effected cars? Seems like it would be a lot easier to swap out ECU's. It is basically proven that the misfire is due to harmonic vibration, but what triggers the misfire? Especially since the misfire can be cylinder specific. Something is telling the car to cut fuel and spark to a particular cylinder. Ted, states the ECU does not cut fuel or spark ao what does?