Misfires when warm

[ParsedOut]

Hey all, need some advice with my 87 M20B27...

After about 20 minutes of driving from an overnight sit and during mid-day restarts, I get a pretty bad intermittent misfire. Usually happens under moderate to high load, if I limp it along I can usually avoid it for the most part. I've noticed that before the car warms up the idle stays at 600-650 but when warm it goes up to 900-950. Also when the car is overnight cold, it fires up with 1-2 seconds of cranking but when warm it takes 4-5 seconds.

What I've already done is listed below.

Inline fuel pump (255lph)
Fuel filter
Hose clamped the in-tank rubber connector
Fuel tank and pickup screen are clean
Replaced fuel injectors with fresh rebuilds
Tested FPR (vacuum unplugged goes up to 2.5 bar, idles at around 2 bar)
Tested fuel pressure, 2 bar at idle and goes up to 2.5 bar under load
AFM tested using the 9v voltage drop test (not the resistance test)
CTS tested within spec
Coolant temp switch tested within spec
TPS tested within spec
Unplugged O2 sensor (was getting bad "stair stepping" acceleration along with the same misfire when plugged in, need to troubleshoot that another day)
New plugs (NGK copper)
New wires, cap and rotor
No obvious vacuum leaks
Replaced the main relay and fuel pump relay with known good relays

Based on everything that has been replaced and/or tested, I'm at my wits end. I'd say the only things left are the ignition coil or the DME.

Any other thoughts?

[Golfer_705]

Once the engine is warm and the miss becomes present disconnect you CTS it should put your ecm to cold start and not rely on the O2. If the miss disappears and the car runs replace your O2 sensor

http://www.r3vlimited.com/board/showthread.php?t=286917

My issue. Had several M20b27's in the car before that the same test work for.

[ParsedOut]

Once the engine is warm and the miss becomes present disconnect you CTS it should put your ecm to cold start and not rely on the O2. If the miss disappears and the car runs replace your O2 sensor

http://www.r3vlimited.com/board/showthread.php?t=286917

My issue. Had several M20b27's in the car before that the same test work for.

Thanks for your response. The O2 sensor is already disconnected. The DME should be running off of the stock maps which are a bit rich but shouldn't be the cause of the misfire.

[ParsedOut]

With my fuel pressure gauge hooked up on my way into work today, I noticed when the car is cold the gauge sits at 2.25 bar and goes up to 2.75 bar under load. However, when the misfire happens I can see the needle "fluttering" at 2.75 bar, basically just looks like it's vibrating instead of holding rock solid like it does when it's cold. Would this be a cause or effect of the misfire? Maybe air in the lines somehow?

[bmwman91]

It may be the ignition coil. If the insulation inside is breaking down, when it heats up any cracks in there might expand and allow internal arcing. See about borrowing a known good one if you can.

The fuel pressure reading thing is a little odd. Not sure what to say about that. Is your battery under the hood or in the trunk? Maybe consider checking the condition of the fusible link (applies if battery is in the trunk).

[ParsedOut]

I'm hopeful it's something as easy as a coil... but thinking about it more, wouldn't that cause the tach to drop rpms when it's misfiring? When the stumble happens, the throttle goes dead, the car bucks and sputters but the fuel pressure remains consistent and the tach does not do anything funky, same with the econogauge... If it were a ground or power supply issue, it seems like it would cause the gauges to go wild. Feels kind of like a fuel cutoff rev limiter. If I hold WOT it'll pop and sputter, let off, then give it light throttle it'll kick back in and drive "fine" until it gets more than about 10% throttle.

I know the TPS or AFM seem like the obvious culprits here, but they've been tested fine. I even disconnected the cold start valve (to run the fuel pressure gauge) to eliminate that as a possible failure point. I'd say something under those certain conditions is pulling power from the injectors... DME is seeming more and more likely.

[Golfer_705]

Thanks for your response. The O2 sensor is already disconnected. The DME should be running off of the stock maps which are a bit rich but shouldn't be the cause of the misfire.

Disconnecting the o2 will put the engine into "limp mode" it will run rich and also cause your symptoms like yours.

With the O2 connected. Engine up to operating temp. When the issue occurs pop your hood and disconnect your Blue CTS harness. The engine will stay in "cold start" not "Limp". The o2 signal wil be used but only secondary to the "cold" programing. If the issue goes away replace your O2 Sensor.

Cleaning the wiring is a great idea.

If you have a DVOM/multimeter with the batter disconnected measure the ohm reading between your - battery cable and the chassis ground you suspect. If the reading is high Check the chassis grounding point, if it's bad simply clean and retest and retest. It's a lot faster that taking everything appart to cleaning and getting unsure result.

As for the power side you can do the same as suggested for the grounds, battery disconnect but connected to the batt pos cable.

You could preform a voltage drop test on you positive side, it's quicker than taking ohm reading but more accurate.

Try unplugging you CTS you may be surprised.

[ParsedOut]

Your info regarding the O2 sensor is incorrect. Unplugging simply causes the car to run off the stock DME maps instead of using the realtime data to trim the fueling. Correct it'll run rich, but not nearly enough to cause this issue.

As for the other tests, I appreciate the response and will double check my grounds.

[Golfer_705]

Your info regarding the O2 sensor is incorrect. Unplugging simply causes the car to run off the stock DME maps instead of using the realtime data to trim the fueling. Correct it'll run rich, but not nearly enough to cause this issue.

As for the other tests, I appreciate the response and will double check my grounds.

Sorry if the terminology is wrong, just know it worked to diagnose several M20b27&25's O2S&HO2S in the past after going through all the tests you have. It may be worth a try. Disconnecting the O2 helped but didn't resolve my miss's. With the engine warm disconnecting the CTS did. I replaced my O2 in all cases and the issue resolved. It may be worth a try.

[ParsedOut]

Sorry if the terminology is wrong, just know it worked to diagnose several M20b27&25's O2S&HO2S in the past after going through all the tests you have. It may be worth a try. Disconnecting the O2 helped but didn't resolve my miss's. With the engine warm disconnecting the CTS did. I replaced my O2 in all cases and the issue resolved. It may be worth a try.

Thanks, no luck on that either :)

Tried a new ignition coil, still no good. I think I'm done trying to track this down, M50 swap here I come...

[NJ2.5rser]

Look into the head at all? Sounds like a floating valve or a bad rocker arm. Try taking the valve cover off and taking a look at everything. Try a comp test and and leak down while the motor is warm. If it's something in the motor that should give you a good idea. It's not the most likely cause but I've seen rocker arms go bad and floating valves before in the m20s


Sent from my iPhone using Tapatalk

[ParsedOut]

Pulled the eta motor and swapped in an M50. That's one way to fix the problem! Honestly I still believe it was sensor related, CTS, TPS or AFM. Just didn't want to spend the money if the motor was coming out sooner or later. Turned out to be sooner than later. :)