My car: a 2004 mini justa.

Symptoms: Every thing is fine at idle. If I rev the engine (sitting in the driveway or going down the road) over 3000 RPM, I have a noticeable loss in power, and the check engine light flashes and stays on. I just set a P0300 code, random misfire. I can then simply shut off the car and restart it and it now idles perfectly. That is until I rev the engine over 3000 RPM again. This is 100% repeatable. Seem like I have done it a couple of hundred times since the problem first started last Friday.

What I have tried:

1.) I replaced the in-tank fuel filter. Actually took a couple of tries because I was placing the large internal O-ring in the wrong location. I have an excellent spray out of the fuel test port (which is a different size than the connector on my fuel pressure gauge...rats). Absolutely no change.

2.) I inspected the coil, wires, and plugs. They were not exactly the best shape so I replace everything. Again, absolutely no change.

3.) OK, so now I'm getting serious. I pulled out the O-scope and started to take readings. The power to the coil and trigger circuits appear regular and square. I checked each of the fuel injectors and found that the trigger on cylinder one was no longer operating the injector when the engine was running rough. All of the other injectors showed no changes in the electrical circuit's operation (regular and square). OK, I am getting close.

4.) I swapped the questionable injector with the one next to it. The problem did NOT follow the injector. The trigger signal from the ECU was not present again on the cylinder #1.

I have completely traced the circuit back to the ECU. No breaks, shorts, or questionable areas on the wire.

Since the circuit works fine after each restart and up until the time of approximately 3000 RPM when it throws an engine code (P0300), does this mean I need a new ECM ??????

Thanks in advance for any "advice" or "information"

 

Not sure if it applies here but I had a similar issue with my 05 GTO
turned out to be the throttle pedal assembly put in a new pedal and it went
away as both cars are drive by wire may be worth a shot

 

It's not likely that you need an ECU, I've only ever replaced on R50 ECU in the last ten years.
I would suggest doing a compression test and leakdown test before going any further

 

I had a similar problem with my 06 R53 last April. I did most of the same things you have done as well as a pressure test on each of the cylinders. Initially I was getting a P0301, finally ended up with a P0300. Had me and my mechanic stumped for about a week. Turned out to be a bad engine temp sensor. I spent a lot of cash replacing plugs, wires and coil pack for a $12.00 sensor.

 

Check your oxygen sensors especially the one closest to the engine before the cat.
I changed one and then the problem came back I did the same as you changed plugs, wires, coil. My mechanic finally disconnected one of the sensors and pluged it back and the problem is gone but the check engine is still on. So my problem is both oxygen sensors went on me. I even went to check out a price for an ecu not cheap btw. There is a place in Vancouver that sells them.
I think it might be your oxygen sensors or coolant temp sensor.
My car went into limp mode and does not recover until the code is removed from the ecu

 

Great info!

I have checked the O2 sensors using my AutoTap program (a very handy tool). The pre-cat O2 is switching like expected and the post is steady as a rock (also as expected). I have unplugged and re-plugged the O2 sensor several times over the course of the last week.

The coolant temp seems to be working fine. Again, using the AutoTap I can see the temperature of the coolant in degrees F. Right in the range I would expect it.

Yesterday I removed the intake manifold and inspected everything real close. There are no cracks in the housing and the vacuum lines all look good. I think if it was a vacuum leak it would effect all of the cylinders unless it is right at interface between the intake manifold and the cylinder head at cylinder #1. I flipped over the silicone gaskets in the intake manifold so the best side was next to the cylinder head.

I also pulled off the the valve cover and inspected the valve springs. Nothing broken or obvious here.

I opened up the ECU. I was able to identify the chip responsible for actually driving the injectors. The driver chip for the injectors (ATM43D-446778) has all four injectors integrated into a single HSOP. Before I purchased a replacement on eBay for $4 and $40 shipping from the far East, I got out the O-scope and took a look at the signals directly on the chip. The input signal for the driver chip was wired directly to the micro-controller. The chip is doing EXACTLY what the micro-controller is commanding it to do. OK, so why does the "brain" think cylinder #1 is mis-firing and shutting down the injector?

My next step is to do a compression and leak down test (after the SPECIAL tool arrives). Please keep those ideas coming...

 

OK, the compression gauge has arrived! These are the results:

Cylinder PSI
1 - 115
2 - 110
3 - 110
4 - 115

I even rechecked #1 and got the same reading. I don't know if these are low enough to indicate a problem or not (not much experience in this department).

I have never seen blue smoke on either acceleration or deceleration. I guess that the next thing to check will be the crank sensor and cam sensor.

Does anyone have the pin-out for the Mini Cooper R50 non-S ECU??

 

Have you swapped the crank sensor yet? Also, have you looked at the crank sensor signal as you rev the engine above 3000 rpm? I have seen a VW that had a damaged (bent) crankshaft sensor impulse wheel that was hard to diagnose.

 

Checking the cam and crank sensors are next on the list. My best access point is directly at the ECU connector. I am looking for a pin out on google right now...not a lot of luck.

 

Any luck on this problem? Having the same problem on my '03 R50 at #4 injector.

 

I just went thru the same thing on my R53. After doing everything possible including things you did, I got a leak down test and it turned out to be bad exhaust valves. I am finally after 2 weeks getting it back tomorrow. Get the test done. It will at least narrow it down some more.