I'm about to give up on my car.

[Kroze]

1999 Maxima w/128k miles

The car runs on 5 cylinders when it's cold (as if I leave it parked overnight and started up the next morning) but once it has warmed up, it runs and act perfectly normal.

Check engine code: P1320 & P0306
P1320 = Ignition coil failure OR crankshaft position sensor
P0306 = Cylinder #6 misfiring.

Parts I've replaced in an attempt to fix it:

- Replaced Idle Air Control Valve (twice)
- Replaced the crankshaft position sensor (REF). Near the oil filter.
- Cleaned Throttle Body
- Replaced all 6 spark plugs and coil packs.
- Moved the spark plug & coil pack on cylinder #6 to a different cylinder.
- Checked for air leaks/loose hose.

None of those attempt work. Again, the car runs perfectly fine once it has warmed up.

Any idea what's wrong? Could a clogged fuel injector be the cause? What's a quick and easy way to test it?

 

[96Firebird]

Sounds like you've checked for spark, you can pull the plug and carefully check to see if it's sparking just to make sure though. Now, the next step is fuel or compression. You can do a compression test on that cylinder, the piston rings may be warped and will only seal when the engine is warmed up? Long shot, but who knows.

I'm leaning more towards the fuel side, especially if cylinder 6 is on the end of a fuel rail. Could be some buildup in the rail, or could be a clogged injector.

 

[Pacfanweb]

How quickly does the miss go away?

 

[LTC8K6]

Move the fuel injector?

 

[Jimzz]

Have you done a cold and hot compression test on cyl 6?

 

[Kroze]

Have you done a cold and hot compression test on cyl 6?

No but thanks for reminding me. I do have a compression test tool but not sure how you go about getting a fuel injected cars that uses coil packs (the instruction said to disconnect the distributor).

So do I remove all 6 spark plugs and coils and just crank the engine with the compression tool on one of the cylinder?

 

[natto fire]

I would do the above suggested about visually ensuring you have spark and then go to fuel, compression last.

If I had to guess after passing visual spark test, their might be something wrong with the injector that is causing poor atomization which higher coolant/fuel temps are able to help overcome?

Does the cold-engine roughness just suddenly go away, or slowly? What was the condition of plug #6 when you pulled it to change to the new plugs? Even though you replaced coils, there could be a problem with the wiring harness to the #6 coil. Although a harness problem is not likely to be temperature based so consistently, it is worth it to visually confirm spark.

 

[halik]

Have you done a cold and hot compression test on cyl 6?

This. Sounds like a mechanical problem. I'm betting low compression or bad valve seat (leakdown would be a good idea also).

 

[halik]

I would do the above suggested about visually ensuring you have spark and then go to fuel, compression last.

If I had to guess after passing visual spark test, their might be something wrong with the injector that is causing poor atomization which higher coolant/fuel temps are able to help overcome?

Does the cold-engine roughness just suddenly go away, or slowly? What was the condition of plug #6 when you pulled it to change to the new plugs? Even though you replaced coils, there could be a problem with the wiring harness to the #6 coil. Although a harness problem is not likely to be temperature based so consistently, it is worth it to visually confirm spark.

Why would fuel delivery or spark be a function of the engine temperature? That's a complete long shot. Compression should be the first thing to check.

 

[Kroze]

I would do the above suggested about visually ensuring you have spark and then go to fuel, compression last.

If I had to guess after passing visual spark test, their might be something wrong with the injector that is causing poor atomization which higher coolant/fuel temps are able to help overcome?

Does the cold-engine roughness just suddenly go away, or slowly? What was the condition of plug #6 when you pulled it to change to the new plugs? Even though you replaced coils, there could be a problem with the wiring harness to the #6 coil. Although a harness problem is not likely to be temperature based so consistently, it is worth it to visually confirm spark.

Just got done doing the compression test on a hot engine.
Cylinder #6 = 180 psi
Cylinder #4 = 170 psi
Cylinder #2 = 170 psi

Cylinder #1 = 180 psi
Cylinder #3 = 185 psi
Cylinder #5 = 195 psi ???


Should I now wait for the engine to get cold and do the compression test again?

Spark plugs looks fine to me, what do you guys think?

 

[Jimzz]

Yea do it on the cold engine as well since your issue is only when cold.

The hot numbers look good but if it runs well when hot then it just confirms everything is ok when hot.


Yea don't see anything on the plugs but if they are new it takes time for them to get gummed up again in cases like this.

 

[dtgoodwin]

Could be a leaking intake manifold gasket. Normally that would set a bank 1 or bank 2 lean code, but not always. In GM vehicles I see exactly what you describe. Misfire when cold, but when warm, they can deal with a minor leak and run without misfiring. I'd bet you would find a small leak for the runner on that cyllinder. You could spray some carb cleaner in small amounts near where it enters enters the head when it's running and see if it stumbles. If so, then you've found your culprit. https://www.youtube.com/watch?v=tt2pprGMYaE&feature=g-u-u may help.

Good luck.

 

[jolancer]

easyer to test the injectors on a 6cyl then compression. unless the shrader valve is in a very difficault spot. If the cylinder is not completly dead when cold but intermitten misfire, i'd also suggest testing injectors first. sorry i dont have time atm to go into great detail. but...

first easyest - disconnect the vacume line from the fuel pressure regulator when Idle, cold when your having the issue, that will raise fuel pressure a bit so if the injector isn't dead but struggling, it may start working at that point (at idle).

the coils in the injectors can be tested with mutimeter/ohms just like ur coil paks, this is a less usefull test, just the electrical.

hooking up a fuelpressure gauge to the fuel rail is what will tell you the most, needle flucuation, pressure, leak down... theres prollly tones of tutorials on how to use one online.

Oh and may seem like a dumb question but i have to ask cause i know people who have put 93 octane into there cars thinking there getting some sort of benefit or maybe even helping to eleviate such issues, but ironically its he oposite.... if your running 93 in a regular engine, filling your tank with 87 maybe all it takes to fix intermitten misfire.

 

[Kroze]

easyer to test the injectors on a 6cyl then compression. unless the shrader valve is in a very difficault spot. If the cylinder is not completly dead when cold but intermitten misfire, i'd also suggest testing injectors first. sorry i dont have time atm to go into great detail. but...

first easyest - disconnect the vacume line from the fuel pressure regulator when Idle, cold when your having the issue, that will raise fuel pressure a bit so if the injector isn't dead but struggling, it may start working at that point (at idle).

the coils in the injectors can be tested with mutimeter/ohms just like ur coil paks, this is a less usefull test, just the electrical.

hooking up a fuelpressure gauge to the fuel rail is what will tell you the most, needle flucuation, pressure, leak down... theres prollly tones of tutorials on how to use one online.

Oh and may seem like a dumb question but i have to ask cause i know people who have put 93 octane into there cars thinking there getting some sort of benefit or maybe even helping to eleviate such issues, but ironically its he oposite.... if your running 93 in a regular engine, filling your tank with 87 maybe all it takes to fix intermitten misfire.

So I just tested out the compression on cylinder #6 when it's cold and it's similar as when it's hot...


Anyway, how do you test out the injector? I measured it with an ohm meter and it registered 11.8 ohms. The car is running perfect as of right now. The car seems to only does it (run poorly on 5 cylinders) when I leave it park overnight. If I drive it for a little bit, the car will then run perfect for the rest of the day even after it cools down.

 

[Kroze]

Could be a leaking intake manifold gasket. Normally that would set a bank 1 or bank 2 lean code, but not always. In GM vehicles I see exactly what you describe. Misfire when cold, but when warm, they can deal with a minor leak and run without misfiring. I'd bet you would find a small leak for the runner on that cyllinder. You could spray some carb cleaner in small amounts near where it enters enters the head when it's running and see if it stumbles. If so, then you've found your culprit. https://www.youtube.com/watch?v=tt2pprGMYaE&feature=g-u-u may help.

Good luck.

I replaced the intake manifold gasket recently and it doesn't seem to leak from there. I sprayed water on it when the car is running but it doesn't change the idle one bit so I don't think it's leaking.

 

[halik]

So I just tested out the compression on cylinder #6 when it's cold and it's similar as when it's hot...


Anyway, how do you test out the injector? I measured it with an ohm meter and it registered 11.8 ohms. The car is running perfect as of right now. The car seems to only does it (run poorly on 5 cylinders) when I leave it park overnight. If I drive it for a little bit, the car will then run perfect for the rest of the day even after it cools down.

Swap the injectors around and see if the problem moves with it. Leakdown test would be a good idea also. Does it keep on misfiring when it's cold and you open up the throttle a bit?

 

[phucheneh]

Plugs look fine, coils look okay but I can't exactly inspect them for cracks through my computer monitor. If you had a constant ignition misfire, the plug for that cyl would be wet. A lack of fuel doesn't give such an indicator. But basically-

Step 1: Stop throwing parts at it.

Step 2: Find out the specific criteria for the ECM setting those codes.

P0306 is easy. There is a lack of power output on cylinder six. This could be a half-dead cylinder all the time. Could be an intermittent dead miss. The thing to know is that it sets this code based on the acceleration/deceleration of the crankshaft (through CKP) as the engine rotates through the firing order. Basically, it's seeing that cylinder six's combustion is not at powerful as the other five.

It's not a code for a coil. Or an injector. Ect. It could literally be anything that makes that cylinder not function as effectively as it should.

However, your other code (manufacturer-specific as denoted by P1xxx format) is for an ignition problem. I do not know the exact criteria- but you should try and find out before you start replacing more stuff.

The two codes are probably from the same issue, but it's not guaranteed.

...did you already have a thread on this where I gave instructions on how to check the coil wiring for #6 and got ignored? :/

 

[phucheneh]

...yeah, I think you did. 'cause P1320 is for an ignition coil primary winding.

If swapping coils between five and six doesn't move the misfire, then it's a wiring problem. With one of two wires. Period.

Well, or a bad coil driver in the ECM. But that's unlikely.

 

[phucheneh]

The previous thread is important info that saves everyone time:
http://forums.anandtech.com/showthread.php?t=2278035&highlight=

Switch coils as recommended above.

If the misfire does NOT move, you have a wiring issue. Double-check power with the key on...test light works better than DMM for this, because it verifies the wire can carry a load. >12v on a DMM with no load on the circuit can mean one tiny strand of copper is touching- no more is required to get a 'normal' voltage (or resistance) reading. You'll of course be using a known good ground.

If that passes, it pretty much has to be the negative from the ECM. If you want, you can check the wire the same way- unplug the ECM connector and stick a small probe (don't damage the terminals) in the appropriate pin on the ECM connector. Connect that to ground with a jumper wire. Do the test light check again. If it doesn't light, or is dim, there's a bad spot in that wire. (if it works fine, the driver in the ECM is likely bad)

If it's the wire, rather than rip open the engine harness, just cut the wire at the ECM connector (leave as much good wire coming out of it as possible) and overlay a new wire to the appropriate coil.

Again, you can do voltage and continuity checks, but with the nature of your problem, I doubt you have a completely severed wire. So doing those two checks with a test light are the way to go, and you WILL find the problem.

Again, first test- you're using the B+ wire at the coil connector for power, and a known good (e.g. battery terminal) ground to see if it lights a test light.

For the second test, it's harder because constant ground is never supplied to that wire. To actually check it as the ECM is firing the coil, you'd need to backprobe the coil connector and look at the signal with a scope. My way circumvents this by manually applying ground at the ECM end of the wire (with the ECM disconnected). Then you can do the check in the same way- use coil signal wire as ground, use battery terminal for your positive. If it can't light a test light, you know that one wire is not capable of carrying the current needed to energize the primary winding in the coil (which then induces the high voltage in the secondary winding, which fires the plug).

 

[JCH13]

If you replaced the crank position sensor it may be possible that it is slightly out of adjustment, if it can be adjusted. Either way I would double-check that you set the sensor gap correctly.

 

[jolancer]

Sorry phucheneh - but ohm meter = test light ...shoot me later

honestly though judging by Kroze responses - i think he'd have much better luck, and would be cheaper to have a 'real' mechanic find the prob under 30min, then it would have been to buy a 6pak a coils and iridium plugs :/

 

[phucheneh]

Sorry phucheneh - but ohm meter = test light ...shoot me later

honestly though judging by Kroze responses - i think he'd have much better luck, and would be cheaper to have a 'real' mechanic find the prob under 30min, then it would have been to buy a 6pak a coils and iridium plugs :/

A DMM set to ohms measures such by putting out a TINY amount of current. To be really basic- it kicks out a few electrons and sees how long it takes them to get back.

A circuit that reads 0 ohms (or close to it) is not necessarily a circuit that can support an actual load. Like a primary coil winding. Or a light bulb. When a more significant amount of current is drawn, the weak spot in the wiring will magically go from zero ohms to...well, a lot of ohms.

Think of it this way: you have a pipe that's half a foot in diameter. You pump some water into it with a drinking straw...hey, look, all the water came right out the other side.

....doesn't mean the pipe isn't clogged. A new pipe and a 90% clogged pipe are both going to flow your tiny water supply just fine.

 

[JCH13]

A real (not cheap) multi-/ohm-meter won't have that issue. It will measure contact resistances and other circuit resistance issues just fine.

 

[phucheneh]

well, geez, damn piece of shit Flukes...

edit: If you really want to play this game, give me ten bucks for shipping and I'll mail you a few '0 ohm' battery cables.

 

[JCH13]

well, geez, damn piece of shit Flukes...

Flukes are usually good, but I've never run into the sort of magical "0 ohm turning into non-0 ohm" situation you're describing.

I use them fairly frequently at work measuring various battery components and test circuits looking for 1-20mV losses and 0.01-0.1ish ohm resistances. Never let me down thus-far.