At the end of my rope. P300 code has turned into possible engine replacement. Please help.
 

Sorry this is so long but it's been an odyssey with, what has been otherwise, one of the best cars I have ever owned and I'd really like to keep owning it. So here goes...

2005 Mazdaspeed Miata, 60K miles no previous mechanical issues outside of oil changes, brake pads and general maintenance.

About a year ago car starting feeling sluggish. Rough idle, underpowered when accelerating until it would all of a sudden 'wake up' and power would return. No check engine light. It was a gradual type of change so I didn't really put together something was wrong at first.

6 months ago on long trip it basically lost all power. Practically undrivable. Still no check engine light but hooking computer to computer to car showed intermittent P300 code (random multiple misfire). Replaced plugs, coilpacks and wires and it ran better. Not great, but better. Still registered p300 internally though, just not enough to trigger CEL.

At that point I figured I would wait for whatever it was that was causing the occasional misfire to actually fail so my mechanic wouldn't have to spend a lot of time hunting down possible reasons for the misfire and just start throwing parts at the problem.

Sure enough, about a month ago I got a solid CEL. Took it first to my local Miata specialist whose worked on the car before. They changed crank position sensor, tried new plugs coils and wires, all with no change. Still P300. They threw in the towel and said go to the dealership as they have more robust diagnostic equipment. Their belief was that possibly flashing computer could solve the problem. Something that they didn't have equipment to do.

At the dealership, Diagnosis from their techs and backed up by Mazda techline was that it's an underperforming fuel pump and or failing fuel pressure regulator. They replace both and car starts running great - but still P300 code.

At this point I am getting impatient and the dealership is baffled but, to their credit, they keep searching on their dime (if they fix the problem, I will be back on here to sing their praises). They find what looks to be damage to crankshaft pulley assembly. Missing teeth and signs that something either hit it, or some other part around it came loose and caused the damage. Keep in mind, at the time they found this, they had had the car almost 2 weeks, so I am guessing it wasn't super obvious, or just in a really hard to see place. This is the first tangible evidence of something actually wrong! Confidence is high that this is actually the problem. They replace the pulley assembly, put it all back together and .... still code p300. At this point they also try another crank position sensor.... no difference.

This is where it gets really bad. The engine runs great, no signs of oil or coolant leaks, no weird stuff coming out of tailpipe but now they are theorizing that the crankshaft could be damaged and causing crank position sensor to be giving incorrect position readings. In other words, the car isn't misfiring at all, it's just getting bad readings because of bad crankshaft. They say this but have no way to confirm it. That means, basically putting a new engine in the car on a hunch.

My counterargument to this is that if there was damage to crankshaft, enough to make it screw up crank position sensor, shouldn't there be some external sign like oil leaking or an unusual vibration??? Shouldn't the car not run well?

There's been a bunch of other stuff checked. Basically the whole checklist for sources of P300 have been checked and double checked (oxygen sensors, catalytic convertor, cam position sensor, etc). Outside of that, compression tests came back fine, egr valve tested good. Only other thing wrong on the car is that the canister on gas tank was not draining correctly but there is no evaporation code so it's thought that this isn't contributing to anything, other than maybe a slightly rougher idle or harder startup.

Any help/advice/solutions would really be appreciated as I am looking at not only having to get another car (not worth it to put new engine in I am betting) but on top of that having not much of a trade in to put against it.

EDIT: Flexplate was incorrect name for part. It's actually the pulley assembly attached to the crankshaft.

UPDATE 1: Some things I left out of original post:

Premium fuel used, at least 89 octane, mostly 93. Recently been going to gas station that uses 0% ethanol.
Fuel filter was checked and was fine.
Car was checked for vacuum leaks (smoke test). No leaks found.


UPDATE 2 (4/22/2016):

Well, I've gotten my car back. Still have the error code but the car runs great. It's night and day. I went in to the Dealership (Stevenson Mazda) expecting a giant bill for all the time and parts but they were extremely fair, so much so that I went back and gave the folks I dealt with there some gift cards to show my appreciation. The plan right now is to bring the car back when there is a regional tech in town to take a look at it. In the meantime, I have my car. Parts replaced so far:


-Plugs/Coils/Wires
-Fuel Pump & Fuel Pressure Regulator
-Crankshaft Belt and gear assembly
-Camshaft Position Sensor
-Crank Position Sensor

When you say crank flex plate, are you referring to the flywheel on the back of the engine, or the timing trigger on the front of the enigne?

Did you actually replace the cam position sensor at any point? They're notorious for failing more frequently than anything else on your list.

I had P0300 not long after getting my 04 MSM with similar mileage to yours. Cleaned the bog solenoid (didn't see that one on your list), then changed plugs, plug wires, crank position sensor, and cam position sensor sequentially. Fairly certain the cam position sensor was the ultimate fix.

Crank damage to affect the CPS...when you look at the crank damper while it is running, is the pulley turn straight or it oscillate on the side or moving closer/further to the bloc ? When moving it manually with a bar, do the teeths distance to the CPS change when you turn it? If not, the crank is probably not the issue. You can compare piston height with a precise reading device, it can show a bent rod but i would be surprise if compression was reading good without any strange noise or metal in the oil...

Is the engine is a 2.0l like the base Mazda 3 with a keyless timing gear ? Do the valve timing has been checked because if it slip, the refference between cam and crank will be off. Or if the chain get too much loose it can make issues by disturbing the refference between cam and crank...

I had a similar issue with a different problem a long time ago with a klde v6 engine. It was feeling correct because it was my first v6 but during a dyno day, the car showed only 103whp. We tried to check the timing advance but the numbers was all around the place with a timing lamp. The damper was loose !....I replaced the pulley with a new key and dynoed it again to 136whp!!!

My bet is they are referring to timing trigger on front. If flywheel was mangled, I'd think it would have been noticable on startup. Plus, that would be way harder to be damaged I would think. It's on my list of questions for when I call them today to find out specifically which part they are talking about.

UPDATE: Service writer was actually referring to pulley assembly attached to camshaft which makes a lot more sense.


Cam position sensor has not been replaced. I've brought it up several times and the response is that it's been tested and it works. I though, keep coming back to it as well. Will bring it up again today.

UPDATE: Cam sensor replacement may happen today.

I plan to ask them if they've actually checked timing with timing lamp to see if the random misfire is actually happening or it's an incorrect sensor reading. Also to to see if timing is actually correct. Will report back when I know.

UPDATE: They checked timing using time lamp and it seems fine. Unfortunately the misfires are happening most while car is being driven and rpm's above 2-3k so makes checking timing with lamp impossible.

No idea about timing belt but I don't think it's due to be changed until 100K miles? That makes me think timing belt isn't the issue, unless something really unusual happened to it, or I got a defective one.

UPDATE: I brought this up with service writer and yes it is due but they don't think it's contributing to problem because timing looks correct. See updates above in this post.

UPDATES:
Things I forgot to mention in original post:

Premium fuel used, at least 89 octane, mostly 93. Recently been going to gas station that uses 0% ethonal.

Fuel filter was checked and was fine.

car was checked for vacuum leaks (smoke test). no leaks found.

since your car make stange behavior, i wouldn't discard anything related because it isn't supposed to break. The timing and VVT on new cars isn't suppose to fall the whole life of the engine. (Sale argument) but on the MS3/6speed we have to do it because the chain hit the valve cover after about 100-150kmiles
But to my knowledge is the engine a 1.8 or 2.0l ?
1.8 = reliable belt that can still slip teeth or get loose
2l = chain like the speed.

The MSM has the same BP 1.8L as the rest of the NB Miatas (More or less. Some minor differences with VVT on some model years). Timing belt replacement intervals are every 60k miles.

It's a 1.8 liter but only 60k miles.

UPDATE:
I talked to the service writer and they tested timing with timing lamp so they don't think belt is an issue. Unfortunately, at this point, the misfire is most prominent under load and above 2-3k rpm. In other words it pops up when driving the car which makes using timing light under those conditions impossible.

lol BPT or LF ?

Anyway i would also check wiring of the CPS. If it rubb and strip to occasionnaly ground to the bloc, it can do that kind of behavior.