Injectors firing when motor is sitting, start problem
 

Hello maxima .org!

Long time no post, hope this works. I’ve searched a dozen times and have not found anyone with a solution to this specific problem.

I have been fighting a start problem for the better part of a year. It’s intermittent at best, and seems to be the worst when the engine is cold. It cranks and cranks but fails to catch. If I’ve cranked for a while it blows a cloud of gasoline vapor out the tailpipe when it does finally start.

The only reason I have been able to live with this so long is that I live in a pretty hilly area and it does not skip a beat if you bump start it. (funny looks from neighbors sometime, though)

I’ve played around with grounding the starter and transmission case, as well checked all the injectors and coils. I get 2 codes, one for the knock sensor and another for the EGR (not too shabby for 350k on the clock). I was messing around with it the other day and cranked it over with the hood up. I ran the starter for about 30 seconds with no start, then let off to let it cool with the ignition still on. Then I hear a little clicking under the hood. The injectors were cycling even though the engine was not turning!

So I dug around a little and find that the POS crankshaft position sensor governs the timing of the injectors. So I unplug the sensor and the injectors stop. OK cool, dead sensor, right? Wrong. It bench tests fine and the same thing happens with a new sensor. Any thoughts on another possible cause?

Thanks!

The clicking your hearing is probably the IACV. It's normal for it to make noise even when the enigne isn't running. The CPS does indeed control fuel injection timing, but if the engine isn't spinning, then there is no way an impulse would come from that injector. When they go bad, they don't detect the cam position and signals stop coming.

Have you ever replaced your ignition switch? It's a known problem item with 4th gens and causes hard or no start conditions.

^ Excelent post!

I'd say it's likley to be the ignition switch too. It could be the KS, but that doesn't sound right for what it is doing. Have you cleaned the IACV? Might not hurt to do so (especially with 350k on the clock! Whew doggY). :) This "vapor cloud" you mention sounds like possibly a burnt valve, car use any oil?

The sound is the injectors. I confirmed this by the screwdriver stethoscope (sp?). This continues until either the ignition is turned off or you unplug the POS CPS. You can hear the IAC independent of the injectors. The car purrs like a kitten once you get it started, and bump starts with no trouble, so I figure that maybe the sensor fails or work intermittently under cranking voltage, then works fine after the engine starts and all the electrics smooth out. Does that make any sense?

The ignition switch is pretty new, maybe a year or two old. The contacts on the start position burnt out, and 1 in every 5 attempts would power the starter. How do you clean the IAC valve? It uses a little oil, maybe a cup per oil change, but I think I can chalk that up to a leaky front (? Pass side) main seal.

if it's leaking it and not burning it ... it's more than likely an oil pan gasket. but upper or lower? hope lower! lol instructions for cleaning the IACV are here ::: http://forums.maxima.org/showthread....=iacv+cleaning <---- great thread with good pictures! and as noted in that thread, it can't hurt to do it.

as for the injectors, maybe you have a questionable ECU that likes to think on it's own. i don't think Nissan ever intended to put AI into our cars, but they may have on yours. lol i'm really not sure what would cause the injectors to continue to fire otherwise (besides a hot signal from the ECU to the CPS all the time) but even then ... i dunno

I hadn't even considered the oil pan gasket. You're not kidding about the upper gasket. The lower gasket has got to be a heck of a lot easier than the main seal!

Thanks for the link on IACV cleaning. If I get some time I'll do that. Probly do the TB while I'm at it. I'd hate to have to put a new ECU in, especially without knowing whether it would do anything. I may run by my Nissan shop (for anyone in Nashville they're called INSAN and are really good), tell them what I'm seeing, and see if they can shoot the problem. Or throw the JWT ECU in there:cool:.

Any other thoughts?

Ignition switch in the on or start position provides 12 VDC that allows the injectors to fire. Is the ignition switch in either if these positions when your injectors are firing? Injector firing signal is controlled by ECM. If your injectors are firing while the ignition switch is in a position other then on or start I would suspect a faulty ignition switch.

Sorry just read more of your posts sounds like your ignition switch is fine.

Sounds like the Engine Control Module is firing those injectors. It's probably due to to some faulty input to ECM? I would not expect the injectors to fire simply because the ignition key is in the on position.

FSM States:
BASIC MULTIPORT FUEL INJECTIONSYSTEM
The amount of fuel injected from the fuel injector is
determined by the ECM. The ECM controls the
length of time the valve remains open (injection
pulse duration). The amount of fuel injected is a
program value in the ECM memory. The program
value is preset by engine operating conditions.
These conditions are determined by input signals
(for engine speed and intake air) from both the camshaft
position sensor and the mass air flow sensor.

Check mass air flow sensor next?

Throttle position sensor fault? Just trying to make a somewhat educated guess.

And what ... raise your post count? You posted 6 times on this thread as the last person to post. :confused:

I still think it has to do with your ECU mate.

Also ... a bad MAF would act differently (and through a code) ;)

Wow, this is a good one...............

I would suspect a bad ecu also, but several circuits need to be checked first.

The question I have for you, is do you know if the injectors are firing in batch mode or sequential? Only easy way I know to determine that is pull the manifold and invert the fuel rail and watch them spray, allthough you could try watching 3 noid lights plugged into cylinders 2, 4, 6. However I think it's safe to assume it's firing them in batch mode because---

Ooooh, ooooh, oooh, I just realized something.............

The injectors are fired in batch mode upon initial cranking, their firing is determined by the reference crank sensor (and maybe also the pos, I don't know for sure, all of this information is speculative for me since the fsm doesn't go into detail on any of this, so I'm making some (very) educated guesses here), than once the ecu can determine which cycle the engine is in via a reading from the cam position sensor, it will than use the position crank sensor on the bellhousing to fire each injector individually at the right time.

Now, funny that you have a very badly leaking front main seal, right next to the reference sensor........ Hummm......... Might want to do a couple things: Clean the teeth on the crank pulley, remove the sensor, clean out the electrical connection, and go from there. Hopefully that (as well as maybe replacing the reference sensor) makes it go away. If not I would check the wiring between the ref sensor and the ecu very closely before finally deciding it's the ecu, you would also want to make sure that ecu pin# ?? is getting 12v when the key is in the start position, it's the start signal to the ecu and I can only assume it's used to detemine a few things listed above. Maybe it's getting 12v even when the ignition is not in the start position...

AH-HA!

My haynes manual said the POS CPS controls the timing of the injectors. While that quote does not directly contradict haynes (the cam sensor is used to sett duration, but not necessarily timing [kinda like VTEC vs. variable cam timing]), it gives me another thing to check.

Thanks

Dude I just printed your post and will give it a try tonight or tommorrow. I may need to buy everyone a beer after all this is done--THANKS!

Keep the ideas coming, I'll let ya'll know if anything works!

Reprint it because I just added some stuff. F the haynes, fsm FTW VQ35de.com/esm

A quick explanation of it is basically at first the ecu won't know if the crank is at TDC for the #1 cylinder for what cycle, it could be TDC on the compression or exhaust stroke, so it fires the injectors in batch until the CPS gives a few readings saying "hey, this is tdc on compression stroke, now you can use the POS sensor to fire each injector at the right time" And the more I think about it, it probably fires them in batch and doesn't look for the cam sensor right away anytime the ecu sees 12v at the start pin, might want to check the start pin sooner than I thought.

Sure, trying to get more posts than you have, oh this is counts as a post also. :-)

I would suspect a sensor or something external to the ECU. The ECU basically is a solid state device. It is probably the most reliable part of that whole engine management system. It's usually input sensors or output devices that cause problems they are the ones exposed to the elements such as vibration, dirt, chemicals and water. The wiring to the sensors/devices is also much more exposed.

Hey JtzMax now I'm really catching up on you........... I really don't care about my post count I just like to enter information as I have time and as I think of it.

^^^ Just razzin ya a bit mate, all in good fun. :)

You may be on to something with it being a sensor, but one way to tell is to map out the ECU, right? Well ... I think anyway. If it is a sensor, I'd think it would give a code, but not all of them do. I think there is a way to use the ECU to read / clear codes, but I don't recall how. Pmohr or Dgeesman might though! Need one or both of them to post. I know I saw it in the FSM, but it was a while back and I just ran accrossed it, didn't actually use it.

So, if it is a sensor, lets figure out which ones run what operations. CPS (crank) = injector timing, KS = timing retardation / advancement for spark, CPS (cam) = also affect timing and injector firing ... Here is a really good site, not sure if what I just mentioned is right on target, but this site is pretty comrehensive. http://www.aa1car.com/library/1999/cm69910.htm

Come to think of it, has anyone mentioned MAF yet? Have you tried to clean the MAF ever? With what if so ... and how long ago?

^^^ Understood, I figured as much.

As stated from FSM in my post#10 the camshaft position sensor and the mass air flow sensor appear to be two very important inputs related to fuel injection. Cleaning MAF is probably a good idea.

Hi anyone knows the solution for this problem? I have EXACTLY the same issue,and I'm afraid of taking the car to inspection as it doesnt start correctly.

Hmmm..... 2 codes and a hard start. I think ya might have a problem w/ the main harness. I had a similar problem at my job and thats what ended up being the problem. Nissan has a TSB on the harness. It`s been knows to cause all kinds of problems .. have someone try to start ya car at the same time .. wiggle the main harness and see if that helps. A32 MAXIMA`S were known for that problem. But I suspect u might have a short somewhere

Had a similar issue with my maxima. After start it would studder a little bit then start running fine. In my case an injector had failed open and was pooring raw fuel into the cylinder.

Please see:

Gasses = compressible
liquid = non compressible

Not good for an air engine to have liquid in the cylinder. You can diagnoe the injectors to see if they are working properly easily enough by checking the impedance of each one I believe.. Perhaps someone else on here has the details on that process or can prove I am full of it :)

I still haven't figurd this out. It acts a lot better in the summer months, but its starting to cool down so I was checking if this thread had any activity.

Several folks suggested that you check the injectors for impedance. I did so and they all check out fine. I still wouldn't rule them out as a possible cause, since I imagine there could be something mechanically causing one to fail open.

GAMERA30: Could you post the name and date of the TSB on the wiring harness? Or better yet, post a link? I'll prolly PM you since you posted this a while back. That would be great if this turned out to be a recall of some kind and the dealer would take care of it for free-fifty-free!

Let's keep this oldie going. (I'm referring to the tread as well as the Maxima!)

Did you take my advice yet about checking the start pin on the ecu and make sure that it doesn't see any voltage when it shouldn't be?

I thought you were gonna post up how to find the start pin on the ECU. Or at least I think that's what I was thinking. I've slept since March....

I wouldn't know where to start unplugging the ecu without frying it or something (or maybe backprobe??), much less finding the start pin. Is there a thread I could read going through this? The only thing I've done with the ecu is clear codes with the screw. While I never turn down a good excuse for taking the max apart, I need someone to tell me that it will still work after I mess with that stuff.

Keep 'em coming!!!

You can find the start pin from fsm in EC section,page 108-->.
Just some things found on my car :
Start signal fine
No spark when turning the key,but if I take a push start it catches right up.
I tried to check voltage from ECU pin#1(ignition signal) when cranking the engine and it showed something like 0.16V maybe,dont remember exactly. I think it is a little low...?
So,what are the possible causes for this? I was thinking about replacing both CPS (POS) and (REF), and rewiring them to ECU. This is because when I tested the CPS(POS),it showed battery voltage when touched with a screwdriver. Could this no spark thing be caused by immobilizer? Does it cut off spark or fuel or start signal? or what?

PROGRESS!!!! :)
 

Hey guys n gals!

Here's a little update. It left me stranded on new years eve. Early in the morning, about to take off to work (prolly in the lower teens) and I crank it an no start. I keep trying until the battery is dead, then my neighbor jumps me and even with the gas on the floor in the "clear flood" position, nothing happened.

F%$^, F%^& ME!!!!!!!!!!!!!!!

So I'm like fine, this POS doesn't owe me anything. Called up enterprize, they came out and got me and hooked me up with a Mazda 6 for about $16 a day (I'm always renting cars for work and they're usually 30-40 daily even with corp discounts--that's a bargain!). And stop at the store on my way in and get one of those portable jump starters, and charge it while at work.

After work I head home and start messing around. First line of business is to get it to start (to get it out of the street and into my garage) without smoking yet another starter. After hooking it up to the rental for about an hour, it would crank, but not fast enough to clear the flood and start. So I got the jump-starter out and hooked it up too. Still no beans.

Then I try hooking the negative terminal of the jump starter to the transmission housing. It cranks over noticeably faster, then finally catches and comes to life.

SOOOO... If it works better with the ground hooked to the transmission than with the ground on the neg battery terminal, that leaves one thing: SH#$%y ground, right? So off to autozone I go. I picked up a 36" 4 gauge solenoid-to-starter cable, that comes with pre-fitted copper connectors that are just too small to fit the starter bolt through. I hooked one end to the neg battery terminal, and the other to one of the differential housing bolts. I made sure to loc-tite the crap out of the diff bolt, just because it would be a real pain if it fell out.

Now it starts a BUNCH better. Still not perfect, but has not given me any trouble, even this morning when it was about 15 degrees.

So here's the theory behind all this. The spot where the ground strap ties into the or the engine/transmission is rusty or loose. It doesn't cause any trouble except when you crank it over, because then you're trying to push what, 300 or 400 amps through a lousy connection. So the voltage "drop" causes the engine case, etc to approach battery voltage (the ground gets a little positive), and all hell breaks loose for the ignition, FI, etc. In my case I think it was two things: 1) something having to do with the injector timeing got screwed up (maybe it did'nt know to fire the injectors in a batch), or 2) the reduced primary voltage at the coils caused a weak spark which could not light off the mixture unless you just lucked out (which would eventually happen after several attempts).

So that's kinda what antti was saying in the previous post. I was getting a spark when cranking, but not nearly the arc damn near a foot and snapping like a whip action that you got when it was running.

So the moral of the story is--nissan grounds may not work too well after a gazillion miles and 12-13 years of abuse. A $6 part from autozone is all it took, at least in my case.

Please, let's hear some comments!

Yea, the grounds tend to go. For the engine you've got the main ground where the negative cable comes straight from the battery and the two ground cables on the LIM.

For the trans, you've got one tiny ~20 gauge wire going from the frame rail to the case, plus the ground it gets where the bellhousing mates to the block.

The tiny ground cable to the trans is useless, it tends to corrode up anyway as it's just a spade connector that's open to the weather (removed mine a long time ago, just gets in the way). The block-bellhousing is also common to completely suck, especially immediately after the trans is dropped for whatever reason.

I was a little nervous about the bellhousing-block being a good connection. Ideally I would have hooked the ground into one of the bolts holding the starter in palce, since these are freakin massive and go right into the engine block (not to mention right on the starter). But the hole on the lead was to small to put the starter bolt through, and my lead wouldn't have reached that far anyway. I really figured 36" would have reached anywhere I wanted to. Not to worry, I have like 10' of 2 guage wire in my garage that may finally be put to use.

that's really not uncommon. Did you sand the mating surface where the bellhousing meets the transmission? Oxidation builds up there and does not conduct the ground to the engine.