Help me with misfire diagnosis

We have a '03 Sentra Spec-V that we've had for about 500 miles. Its been through 1.5 rallyx events. The second event we limped it home with a bad MAF, and we've been fighting and off and on misfire. Now the misfire is here to stay.

The misfire is cylinder 2.

We have switched plugs, ignitor, coil between 2 and 3 to verify they are all working. Also checked spark via jumping to ground on cylinder #2. Spark looks good.

We recently replaced 1 injector due to a bad clip on cylinder 2 - it wouldn't stay connected. First we used a junkard injector, but replaced it with a rebuilt injector as well. Misfire is present with both.

Finally did a compression check - cold.

1 - 125
2 - 145
3 - 140
4 - 140

These numbers are all well below the spec of 181, but perhaps because cold? Motor has unknown miles as it isn't the original motor from the car.

When we bought the car, the headgasket had "been replaced recently". Judging by the rest of the mechanic work on the car, anything the previous owner worked on is highly suspect. Still, if its a bad HG, I want to know before tearing into it.

Changed the oil tonight. Its black as tar after only ~400 miles. Kinda watery, but I don't think there is any coolant in there. No coolant has been going missing either. I poured it in a clear container so if anything separates out of it I will see it tomorrow.

What should my next step be?

How do I verify the injector on #2 is firing? (including the wiring up to the injector)

FWIW, it does not have the precat and the loctite butterfly screw fix was done a while ago (and all the screws are present)

My buddy's '08 fusion had a coil driver go bad inside his PCM. It's a known issue with the Duratec V6 in certain applications. We did all the parts swaps in diagnosing it, and the DTC always stayed on the same cylinder. Could be similar issue on your car? FWIW, in our case the DTC was for coil output, not misfire, so maybe Nevermind.

Did you confirm that the coil, or any coil, on that cyl is actually firing?

I'm thinking AngryCorvair could be on the something. When I worked the Nissan forklift dealer we had a few internal coil drivers fail in thier ecms on the COP/EFI engines. According to the tech support people the hardware of the ecm was identical to automotive units for some non US markets.

HappyAndy wrote: Did you confirm that the coil, or any coil, on that cyl is actually firing?

Yes. See 3rd paragraph, which may not have been clear.

I started the engine. I left the #2 coil connected to the harness but set in on top of the cam cover. Using a long wire grounded to chassis, I held a raw end in front of the coil. Got a very good spark, jumping several inches.

Is it possible that it still isn't igniting under pressure? Any better way to check that?

Cold compression readings are usually lower than hot. Make sure the throttle was wide open, too.

If the misfire doesn't follow the coil, ignitor and plug, that seems to rule it out. Which might be bad, because then you're looking at a bigger problem.

Can you pressurize the coolant system to see if it is leaking?

Not much help on the misfire, but what rallycross events are you doing?

Tight valve lash/burnt valve?

How did you confirm cylinder 2 is misfiring?

Get a noid light or make one and plug it in to cylinder 2 injector plug. crank or run the engine, the light will pulse if the injector is receiving a signal.

I'd do a leak down test, Cylinder 1 looks like it's weak, time to find out why. Anything less than 120 doesn't work to well if at all and your getting pretty close to 120.

Does the oil smell like gas?

Ah, I did not read correctly that #2 had good spark. But consider me dumb and answer me this: did both #2 and #3 have spark during all iterations of the parts swap? If so, then it sounds like injector.

Brett_Murphy wrote: Can you pressurize the coolant system to see if it is leaking? Not much help on the misfire, but what rallycross events are you doing?

I'll see if I can get a pressure tester.

We did the last 2 events with THSCC. Wilson, then VIR. Like 2 runs into the VIR event we lost the maf and had to limp the car home.

daeman wrote: How did you confirm cylinder 2 is misfiring? Does the oil smell like gas?

Disconnected the coil plug when the car was running on 3. Disconnecting #2 never makes a difference in how the car runs. All of the other 3 will knock it down to 2 cylinders.

I think the oil does smell like gas.

AngryCorvair wrote: Ah, I did not read correctly that #2 had good spark. But consider me dumb and answer me this: did both #2 and #3 have spark during all iterations of the parts swap? If so, then it sounds like injector.

Yeah, both had spark during all iterations of the swap.

I'll try the light idea to verify the #2 injector is being told to fire by the ECU.

Brett_Murphy wrote: Make sure the throttle was wide open, too.

We didn't do this. Doh. We will re-check and get new numbers.

The oil smelling of gas combined with the fact that you've got a dead cylinder is a pretty safe bet that the injector on #2 is firing, but still worth a check all the same.

Have you changed the plugs? If so, any difference? If not, swap the spark plug out of 2 to another cylinder. Also, how do the plugs for each cylinder look? I'm betting number 2 is either really clean or really black.

Its starting to sound like a possibly clogged, leaky or otherwise damaged injector on #2. If it's not atomising fuel correctly you won't get a clean burn or if bad enough you'll get no burn. That then means you end up with a bunch of gas finding its way past the rings and into your oil.

I'll be looking out for you at the THSCC events. I didn't get to do many last year.

daeman wrote: Have you changed the plugs? If so, any difference? If not, swap the spark plug out of 2 to another cylinder. Also, how do the plugs for each cylinder look? I'm betting number 2 is either really clean or really black.

Yes, plugs have been swapped (with no change). #3 looked fine. #2 was black.

Under what conditions does it misfire? At all times ('dead' miss)?

All the time.

Sounds more and more like something is wrong with the number 2 injector.

Its a pain in the ass, but try swapping that injector to another cylinder and see if the problem follows.

Improper combustion or running rich blackens a plug. A leaky injector will give the symptom of being rich, it also means that fuel hasn't atomiser properly Which will result in an incomplete burn.

Youve got spark, and the compression seems to close to the other cyls to think there is a mechanical problem.

Its time check injector operation. As stated above a noid light is the place to start. You plug it in to the injector connector, and with the engine running the light should flash. I dont have wiring diagram in front of me, but every system i have worked on had a 12v+ as a constant with the key on and the pcm pulsed the ground. Dont forget the connector condition. Make sure the female terminals are clean and tight. Also you said that you swapped a couple injectors on that cylinder. Check the resistance of all four injectors. They should be the same. If its open its bad, if its way different than the rest its either bad or the wrong part. Another method would be to pull the fuel rail and injectors and crank it and see if you have fuel spraying out of the injector, but that can be real pain depending on how its put together.

If it helps, here is a link on how to run 3 different compression checks. In my own case, it was the "idle" compression check that diagnosed a persistent misfire on one of my cars. The "static" readings were all normal. Diagnosis was a valve seat problem causing partial loss of compression on the misfiring cylinder.

Compression test link

I would verify that the injector ground circuit is not shorted to have a constant ground. It should have a pulsed (PWM) ground while the engine is spinning. If it has a constant ground it would hold the injector open and flood that cylinder until the injector winding got hot enough to short out and either blow the injector driver or melt to an open-circuit state.

Vigo wrote: I would verify that the injector ground circuit is not shorted to have a constant ground. It should have a pulsed (PWM) ground while the engine is spinning. If it has a constant ground it would hold the injector open and flood that cylinder until the injector winding got hot enough to short out and either blow the injector driver or melt to an open-circuit state.

So just check continuity on both harness pins to ground with the car off?

Well, we're only concerned with what it does while running so i would check while running. This is exactly what 'noid lights' are for, and Harbor Freight sells a set pretty cheap. You can use a test light but if it has an incandescent bulb it may not flicker quickly enough to keep up with the injector pulse, which could make a pulsed ground look like a solid ground. So use a regular test light only with a critical eye towards that.

daeman wrote: Get a noid light or make one a

How can I make one of these?

I have a test light, but never used a noid light.

In reply to ProDarwin:

Is the test light led?, If so, you can probably get away with using it.

Otherwise from memory an led and a 750 ohm resistor will get the job done.

Might be worthwhile looking at what it costs to buy one though, I bought a set of 3 for under $20 a few years back

Well... its fixed.

My teammates were working on the car. Did compression @ WOT and got an even 150 across the board. Was still cold(ish) though.

They found a break in the injector harness. Apparently a broken wire right in the injector clip. Luckily we had a spare harness around when we were trying to fix a different injector problem last year. Apparently its running well on all 4 cylinders now.

Yet another problem on this car that is pretty much the result of shady mechanic work.

Now to sort out the remaining 2 issues prior to our first event.