patman

Hi guys,

I have been lurking here a while, but I am running out of ideas, so I am hoping someone can either give me some more or clarify how the misfire detection works.

Background is I bought an RX8 a month ago. I knew some stuff was likely to be wrong with it as I got it cheap, but after fixing all the minor things, I did a compression test and found out it had a broken side seal.

Pulled the engine and rebuilt it. Housings were not great but OK, and now with about 250 miles on it, it is building compression nicely. The problem is ever since I reinstalled the engine I have had a pretty much continuous flashing CEL and 'Cylinder 1 Misfire' code. Having never driven the car healthy, I don't know if I am having any loss of power, but as far as I can tell, the car runs fine. It isn't quite fully broken in yet and still idles a tiny bit rough and smokes a little on startup, but I wouldn't have thought this was enough to cause a misfire, at least not 100% of the time. If I clear the code, it comes back on in ~30 seconds, and I can't notice anything abnormal during that time.

So far I have:
Checked compression at 100 and 250 miles (still building but looks good)
Changed coils, wires, plugs
Changed air filter
Checked fuel pressure (60psi)
Changed ESS
Changed MAF sensor
Checked timing on all four leads
Cleared codes and NVRAM various times
Checked for vacuum leaks
Tried grounding coils with a jumper wire
Tried unplugging knock sensor
Looked in cat and it is empty, so PO must have gutted it

None of this stuff made any significant difference.
My question is, is there something I am missing here? Is the misfire detection really so sensitive that it would be alerting constantly just from the engine not being fully broken in? I expect this engine will take another 1k miles or so to fully break in, and will probably never make 'like new' compression, which I am OK with, but I didn't think the ECU was going to whine about it the whole time.
My understanding is that the misfire detection is just based on a torsional variation calculation from the ESS, is this correct, or are there other sensors/signals that can cause this code?

I have been an rx7 guy and have been building and modding rotaries for years, but this is my first renesis. Hopefully someone here knows some stuff I haven't figured out yet.

Anyway thanks for the help,
Pat

RIWWP

iTrader: (2)

You got all the normal items.

Is the misfire consistently P301? Or do you ever get P0300 or P302?

If it IS restricted to just P301, then it's going to be something specific to the front rotor, which isn't common with most of the misfire causes.

Do you have any freeze frames of the misfires? When is it occuring? How did you validate that you didn't have a vacuum leak?

There are still a number of items possible, everything from a flaw in the engine rebuild to an incorrectly hooked up fuel injector (they are easy to get wrong when re-installing the engine).

patman

It is always p301. Sometimes also p300. When the engine was just built it took about 50 miles to make decent compression and it threw p302 then, but I have not noticed it since.

It is possible that i screwed something up in the build, but I have done this a lot of times and have not messed one up yet. I did have a little trouble getting the apex seal corner pieces to stay in place. I do not have a mazda compression tester, but I am getting nice even pulses of 95-100psi at 190-200rpm on both rotors (warm, have not tested it cold) It runs the same warm or cold. Same tester usually reads about that level on a healthy rx7 engine, and read 80-80-80 and 80-80-30 on engine prior to rebuild.
Have also not had any issues starting, although it does crank a little longer than my rx7, but with exception of one isolated time where it took a few extra cranks (right after I installed new plugs and wires, which was a little weird) it seems consistent every time.

I thought of looking at fuel trims to try to see if it is an injector, but I don't know how much fuel trim at what speed/load would indicate a problem. Maybe someone could give me some insight on that one? I did run some injector cleaner prior to the rebuild, and have been premixing since I got the car. This is supposed to be my DD and I have no plans to mod it, so I'm not crazy about messing with it, but I could stick my wideband from the rx7 on it if there is any reason not to believe the OBDII AFR readings.

I do not have FF data, it never works when I try to get any on my phone, but I will see if I can figure that out. Checked for vac leaks using the carb cleaner method. I do have a setup that I use for boost leak testing, maybe I will put 15psi to it just to be sure. I will also check wire colors on the injector plugs if I can get my boroscope down in there to see them.

EDIT: I also did try swapping coils between rotors today with no effect, so I didn't get a bad coil.

RIWWP

iTrader: (2)

You can use the rear O2 to test the front O2, plus the OBD2 mode test. The rear O2 will be dithering around stichiometric when the front is dithering around 14.7.

Do you have any freeze frames of the misfires? When is it occuring? How did you validate that you didn't have a vacuum leak?

patman

Sorry, must have been still editing when you replied. The CEL flashes ~95% of the time when I drive the car. If I clear it it comes back in ~30s, no matter what I do. It sometimes will idle for a while, but as soon as I rev it or drive at all it comes back on.

RIWWP

iTrader: (2)

Yeah, a proper vacuum leak test would be better.

That consistent of a misfire suggests something rather permanently out of sorts. Have you triple checked that the plug to coil wire order is correct? A swapped plug wire could produce this effect.

patman

Yep, I am 110% sure they are in the right place, and I checked wire colors to all the coils as well.

dannobre

iTrader: (3)

Injector wiring?

patman

Front rotor injectors all have the correct wires. I can't get to the rear ones without taking the manifold off.

Started the car this morning to pull it into the garage, and threw p300 and 301.

Freeze frame:
Engine Load 27.8%
Coolant Temp: 0C
Fuel Trim Short term/long term: 0%
RPM:2104
Timing 11.5*
IAT 1C
MAF 12.91 g/s
Manifold Throttle pos: 15.29%
Barometer: 14.65psi
Cat 183.5C
Voltage 14.84
Load 23.53%
commanded lambda: .93
relative throttle position 3.92%
abs throttle pos B 32.16
accel pedal pos D 31.76
accel pedal pos E 20.78%

None of that seems particularly useful to me, but maybe it will throw a flag to one of you guys. I will check for vac leaks and recheck compression after lunch.

RIWWP

iTrader: (2)

Yeah, I'm not seeing a problem there either. Looks about right for just post-start.

You might need some assistance from one of the guys that has done engine rebuilds and multiple engine replacements. I'm moving this thread to Troubleshooting for some more exposure for you.

patman

OK, did another comp test, this time I used a pressure transducer and my oscope to make 100% sure I wasn't missing a pulse on my manual one. I didnt bother to calibrate the sensor but results are about the same as before, and minimal variation between faces.

I am going to do a vacuum test now, but at this point I am going to just go drive another 500 miles and see if it gets any better.

patman

Drove another 150 miles this evening. Still flashing CEL all the time. I did notice that at WOT sometimes it would go out.

I also did some poking around OBDII data. Looks like my fuel trims are staying right around 0, and I didn't see anything else weird. I was seeing 35-40 degrees of advance at ~4500RPM, maybe 20% throttle, seems a little high to me. I am assuming that the ECU thinks everything is healthy or it wouldn't advance it that much. I am running 89 octane with ~.5oz/gal premix. Not sure if anyone can confirm this is normal timing that I should see?
One thing I did notice is that at a 75mph cruise, my cat temperature was running 905C (1661F). I assume that is measured at the o2 sensor? Seems pretty hot to me for o2 sensor temperature in an empty cat. I also have always thought this car seemed quieter than I expected.

I am wondering if the cat may have failed and clogged up the mufflers or the resonator (rather than the PO gutting it as I thought)? Can anyone confirm if this is a reasonable assumption based on what I am seeing? I am going to see if I can find a way to do a lap around the neighborhood sans exhaust tomorrow without catching anything on fire and see if it still throws the misfire code.

RIWWP

iTrader: (2)

That cat temp is rather high, but that is also something we can expect with constant misfires. Since misfires basically means unconsumed air and fuel in the housing, and they are burnt in the exhaust stream instead.

Clogs tend to only be when the 1st cat element disintegrates and packs into the 2nd. Some other possibilities, but once the cat elements are gone, I don't know of a condition in which they could restrict the exhaust enough to cause misfires at all, much less at idle. Any restriction that is bad enough to cause misfires will be a high RPM, since you are trying to pack so much more exhaust through the piping.



My gut tells me that there is a problem with the rebuild, but it's just a guess based on not being able to come up with another reason that hasn't already been ruled out. I'm hoping someone else has a different opinion or idea.

1.3_LittersOfFurry

iTrader: (4)

MY only experience with a constant flashing cel was after replacing the front main seal. 20 Brake stomp was the only thing that fixed it.

patman

Have done the 20 brake stomp about 50 times. I do apparently still have the older ECU firmware though because my oil pressure gage does not sweep.

EDIT: this brings up a question: is it possible that there is a problem with the NVRAM on my ECU and it is eaither not clearing or corrupt or something and that is why it is throwing the misfire code? I don't know anyone else with an rx8 to be able to borrow their ECU or anything.

ShellDude

iTrader: (12)

1660 isn't all that high for the cat under load… you know more about building than me so I really don't have much to offer. if you're at 500 miles and you're STFT and LTFT are still at zero then the ECU isn't going crazy trying add/subtract fuel.

Have you tried just dropping the mid pipe completely and running out thru the headers? It'll be loud but you won't have to worry about diagnosing a CAT issue if the problem doesn't go away.

I think Dan had a good point about validating injector wiring too…. might as well add ignition wiring and ground while you're at it.

patman

Pulled the engine this weekend, everything on the front rotor looks fine. Still have no idea what is wrong with this stupid thing. Guess I will check all the injectors while I have it apart. If that isnt it, only thing I have left is the ECU. Anyone got a spare one I can borrow for testing?

EDIT: injectors are all fine.

j9fd3s

since the ECU is coded to the car and keys, you can't swap ECUs without reprogramming.

patman

that sucks.
I am trying to find a better rotorhousing before I assemble this thing, as the front housing was a little questionable and at this point I want to rule everything out, but otherwise I am pretty much at the electrical tape over the CEL and forget about it stage.

hoosier1104

iTrader: (2)

I would call a dealership and give them your VIN. They can check to see if there is a firmware upgrade for the ECU.