FIXED! 06 Nissan Altima 2.5L Cold Start Random No Coil/Injector Control

Hi Guys,

This vehicles peaks my interest and is why I like doing what I do. I didn't have much time with it today so I made sure to grab some captures before leaving work.

SYMPTOMS
Initial Cold Start, Normal Crank time, After initial startup, engine surges for a few seconds and then smooths out and runs great with no issues. This surging is actually a complete engine cut as if the crank sensor crapped out for a split second. NO CODES regardless how many times I duplicate the issue.

For the first few seconds of engine run time after initial startup, at random times the PCM doesn't send control to all 4 coils and all 4 injectors. According to the waveforms I attached, the cam sensor and crank sensor have no problems. I grabbed the cam and crank signals at the harness sensor connector and not at the pcm.

I have not tested 5v ref for consistency yet nor have I tested for consistent signals at the PCM

What do you guys think could cause the PCM to not send control? Remember, after 2-5 seconds of random no control, PCM eventually controls everything normal and engine runs great the entire time. This also seems to be a temperature related problem. As the engine heats up or after starting the car about 10+ times, the problem goes away but returns as everything cools off.

The ignition coils look like they have been replaced at some point recently. They are numbered and don't look factory. I need more background information on that.

All Cylinders, 1,2,3,4 show same problem at same exact moment

GREEN: INJECTOR CONTORL
YELLOW: COIL CONTORL
RED: CAM POSITION SENSOR
BLUE: CRANK POSITION SENSOR

Waveform capture of Cylinder 1


All waveforms below are of cylinder 4.
Close up of missing ignition coil control, previous missing fuel injector control not shown and is off the screen to the left

Hmm.. So, if I understand what you have presented in that you're missing coil and injector control yet have good waveforms at the sensors.

This only occurs on start up? So, after a while everything starts to function correctly? If this is the case, I'm curious if there is a solding issue that gets better after the circuit warms up. Just brainstorming.


Assuming this, I would do the following:

See if the scan tool indicates either of these sensors are periodic. Sometimes the PCM will have a PID indicating it is operating with a good crank and cam sensor.

I would backprobe at the PCM to see if there is a wiring issue(s) are between PCM and sensors.

I would backprobe the PCM at the injector and coil controls to see if there is a wiring issue between PCM and actuators.

This is a good one that could eat up time, so I would try to find the macro issue first before going micro. I would try to determine if it is a wiring issue or is PCM having issues inside it. i find it interesting that after things warm up, all is good, meaning it sounds heat/vibration related

I'm so curious as to what you find out.

GeekDIYMechanic wrote: Hmm.. So, if I understand what you have presented in that you're missing coil and injector control yet have good waveforms at the sensors.

This only occurs on start up? So, after a while everything starts to function correctly?

You are understanding correctly. Honestly when I first started looking at the car, I was not even going for the lab scope. I did a preliminary eyeball check in the engine bay and saw a loose intake hose after the MAF and a bad PCV hose. I also noticed the Throttle body built up with crud so I cleaned the TB and relearned the throttle and idle air. After all that, the car started up great each and every time so I thought I fixed it. Next morning just to confirm the fix I started the car again and sure enough the problem was back, thats when I broke out the scope thinking this was something deeper.

If this is the case, I'm curious if there is a soldering issue that gets better after the circuit warms up. Just brainstorming.

I had thought of that too, or at least thought of a wiring/circuit problem on the cam or crank circuits. Though I kept thinking heat would make it worse, but I suppose its possible that heat could expand solder and close up a broken solder joint, possibly not making it worse as there is no current flow. Thats a good point.

Assuming this, I would do the following:

See if the scan tool indicates either of these sensors are periodic. Sometimes the PCM will have a PID indicating it is operating with a good crank and cam sensor.

I would backprobe at the PCM to see if there is a wiring issue(s) are between PCM and sensors.

I would backprobe the PCM at the injector and coil controls to see if there is a wiring issue between PCM and actuators.

This is a good one that could eat up time, so I would try to find the macro issue first before going micro. I would try to determine if it is a wiring issue or is PCM having issues inside it. i find it interesting that after things warm up, all is good, meaning it sounds heat/vibration related

Ya I'll be testing at the PCM from here on out. I"ll check more PIDs to see if I can see anything. If I find the harness is good, then maybe I"ll try a wiggle test at the pcm, if that still doesn't duplicate the problem on command, maybe ill take the pcm out and see if I can open it up, maybe re-flow some solder joints. I need to rule out all other possibilities before I make a call on a bad PCM. I will also make sure I have good powers/grounds, though don't think those would be a problem, in circuitry I guess anything can cause an issue.

I'm so curious as to what you find out.

Me too haha

how about removing the pcm from the car. take pcm inside let car sit outside. then the next morning plug pcm back in start it and see if the issue repeats if not then that would almost rule out the pcm. maybe try checking for a software update

Well issue is, though its worse when stone cold, it still did it after letting the car sit inside over night.. I know I never mentioned that fact. It stays around 55 or so in the shop at night.

Though if I brought the PCM inside and left car outside, if the problem didn't repeat, I would assume then that it does have something to do with the PCM as that would be the only "warmer" component. But the other half of this is, if I unplug it, will I "disturb" the possible connection problem possibly giving me false good/bad news.

graywave wrote: Though if I brought the PCM inside and left car outside, if the problem didn't repeat, I would assume then that it does have something to do with the PCM as that would be the only "warmer" component. But the other half of this is, if I unplug it, will I "disturb" the possible connection problem possibly giving me false good/bad news.

I really like this plan. Even if you're potentially fighting a connection issue, you can still be reasonably confident that you've narrowed it to the PCM connector itself.

Love the zoomed-out scope shots, but have you had a close look at the CMP and CKP? Mostly just curious about irregularities or glitches you might find to be consistent with the missing firing events. I'm suspicious of a sensor issue due to their history of failure.

got mixed up on the typing. thats what i meant words got mixed up

I really like this plan. Even if you're potentially fighting a connection issue, you can still be reasonably confident that you've narrowed it to the PCM connector itself.

I could give it a try if it comes down to it.

Love the zoomed-out scope shots, but have you had a close look at the CMP and CKP? Mostly just curious about irregularities or glitches you might find to be consistent with the missing firing events. I'm suspicious of a sensor issue due to their history of failure.

I have one image in my first post that is a close up of the cam and crank. I have analyzed the cam / crank signals close up and at least at that sensor connector, they don't skip a beat. I honestly thought it was going to be a crank sensor issue.

got mixed up on the typing. thats what i meant words got mixed up

Kinda figured, no worries.

How many miles on the engine? Chain noise, chain tensioner oil pressure at start up?

You should have a data pid for cam/crank in sync...Graph this pid on your Verus, together with cmp and ckp count. That way you don't even have to scope for signal/wire integrity at the pcm and you'll move on quickly.

juergen.scholl wrote: You should have a data pid for cam/crank in sync...Graph this pid on your Verus, together with cmp and ckp count. That way you don't even have to scope for signal/wire integrity at the pcm and you'll move on quickly.

I like that idea. Let the pcm tell you when there's hiccup.

I forget how many miles. No chain noise at all, seems cam and crank stay in sync on the scope the entire time.

I will look at the PIDs some more in the verus and see if I can spot anything. I would assume if it knew something was out of sync or if if it knew it was loosing cam or crank signals I would get a code for it. I know thats not always the case though.

Tomorrow I will be attacking this some more if I have time and I will try everything suggested and let yall know. I appreciate it.

TESTs so far;
1. Cam Sensor Signal @ PCM 5ms Time Base (Close up view) = GOOD - No Irregularities, Square waves all look the same width and height.
2. Crank Sensor Signal @ PCM 5ms Time Base (Close up view)= GOOD - No Irregularities, Square waves all look the same width and height.
3. Knock Sensor Signal @ PCM 500ms Time Base = GOOD - 2.5v During Events
4. Checked 3 PCM Ground Wires = GOOD (Loaded)
5. Checked 3 PCM Batt+ Wires = GOOD (Loaded)
6. Disconnected PCM and plugged back in couple times, problem still persisted.

AT THE MOMENT
Removed PCM from vehicle and placed in warm area waiting for it to warm up through out the day. Vehicle outside in 18 degree weather

TEST TO BE DONE
1. Look at all 4 coil control signals at the same time to confirm all drop at same time. I know they do, I just need reassurance.
2. Same test as above on injectors
3. Look at 2 coils and 2 injectors at a time of same cylinders to see when both drop out.

QUESTION
Is there back up strategy in case a coil fails maybe the PCM will cut the control signal to fuel injector or visaversa? I almost thought about this till I saw a injector cut out before the coil on cylinder 4. the opposite happened on cylinder 1. You can see that in a few of the waveforms I attached in the original post.

So I am leaning heavily on Crank and Cam Relationship. I missed this before and I did a test that seems to verify it as well. The oil is topped of as well as I do believe the tensioner run off oil pressure

Now I didn't decipher the PIDs very well in the scanner so it looked like there was no cam/crank relationship type of PIDs but according to a picture on IATN, there is and I noticed those pids just don't know the values. What I do know is my relationship is off compared to Nissans relationship waveform.

There other rough test I did was unplug the cam sensor and start the engine. Not knowing if it would actually start, it did. Long extended crank but smooth startup with no issue. Plug in the cam sensor and start it up and problem cam back. Now I already know the waveform integrity of the sensors are good so I wasn't confirming a bad sensor. Just the fact that it couldn't correlate the cam/crank sync and probably used some algorithm.

The other question is why does it cut control signals only for a few seconds, only during cold startups. After everything is warm there is no issues starting. Could this still be related to cam and crank relationship?

ALLDATA NISSAN GOOD WAVEFORM


IATN KNOWN GOOD



IATN KNOWN BAD


MY WAVEFORM

Just a small update on information

Car is 93K miles on it

I haven't touch it for a couple days. The INT/V TIM DEG parameter at idle is -24 deg to -36 deg. I read this is supposed to be between -5 to +5 degrees. At 2000 rpm the same parameter is around +16-20 deg.

Could the idle timing be caused by a stretched chain and if the timing is that far off could it cause the computer to stop sending control?
When the engine runs good, the timing doesn't change much. maybe a few degrees difference. The -34 degrees is where it spikes to, it doesn't sit there.

The part of this only happening during cold starts is what is really throwing me off and especially for it to drop control. Since after it warms up timing is roughly the same. - 24 degrees

Got approval. Timing Chain and all the components are being replaced next week. I will keep you all updated.

I did open a case with AllData and they explained if the timing is off enough, the computer gives up and stops sending control. We agreed that it seems this car is on the edge of being a crank no start situation. After sending them all my waveform files and gave them all the information I had, they too are leaning on the timing chain. Taking the cold start situation out of the equation and looking at the timing, its off and needs to be fixed. If it gets off any further, this is when a crank no start situation happens. Plug the cold start back in and it might be a oil priming situation on the hydraulic tensioner or the tensioner is fully extended since the timing does come back from -36 degree to -24 degrees after about a minute or so.

Also with these cars, I learned if it is a crank no start due to timing correlation, you can simply unplug the cam or crank sensor and the car should start after some extended cranking.

Glad to see you're making progress on this one!
I was having a hard time comparing yours to the known good, so I cut this together real quick. Hope you don't mind.

Thanks.

I don't mind at all. Honestly I completely missed the timing portion of this problem in the waveform when I first scoped the engine strictly due to the fact that at a quick glance, the known good and mine are very close. So I at first thought the engine was in time. I also skipped right over the Intake valve timing PID in the scanner. Decided to take another look when it was mentioned by a forum member earlier in this thread.

Hopefully after the timing chain and its components are replaced, I can mark this thread as "fixed".

This one is a good learning experience and one for the books. Everyone chiming in here helped a lot to get me thinking some more too

Graywave,

i am glad seeing you making progress not only in this case...

The art of diagnostics in which u are so interested is a learnable skill. In most of the cases it includes thinking "inside " the box.... In many cases it really comes down to observe, gather data and analyze it.

As a diagnostician you must learn to see or sense whatever data is given to you, no matter if scanner, scope, test light, ears, nose etc.
Once you recognize the data as what it is - raw data - you can analyze and relate it to more data and start to enter the diagnostic part of the game. But before you really have to "SEE WHAT YOU'RE SEEING". Once again, this means to recognize the collected data as what it is and as what it can tell you.

Another very important part, especially with scope and scan diagnostics, is don't let become experience become to hinder you in the diagnostic process: EVERY car is a different car, EVERY problem is a NEW problem. Use your experience, don't let it mislead you.

Love your updates, graywave. Much appreciated.

There other rough test I did was unplug the cam sensor and start the engine. Not knowing if it would actually start, it did. Long extended crank but smooth startup with no issue. Plug in the cam sensor and start it up and problem cam back.

This sold me on a timing issue. What surprises me is how how it runs rough on startup only, when I usually see these engines just flat out not start with a sync problem. I agree with AllData - Perhaps the timing isn't far enough out to really confuse the PCM, just enough to make it hesitate at first?

93K miles and a stretched chain? :huh: Do you have any service history on this engine?

Tyler wrote: 93K miles and a stretched chain? :huh: Do you have any service history on this engine?

All I know is our shop put in new ignition coils about a year ago, and its been getting oil changes. Thats about it.

In fact...starting to think back now....I wonder if this was the same altima that came in for an oil change which was extremely low on oil and had a ton of timing chain slapping noise...once the oil change was done and oil topped off, the timing chain became quite....hmmm I have to investigate that and see if it is. Though I thought that car was a different color, but the time line from that oil change to when this owner said this problem started is roughly the same.

Tyler wrote: This sold me on a timing issue. What surprises me is how how it runs rough on startup only, when I usually see these engines just flat out not start with a sync problem. I agree with AllData - Perhaps the timing isn't far enough out to really confuse the PCM, just enough to make it hesitate at first?

Ya I am not sure at what degree the PCM stops sending control but the fact that at start up the timing spikes to -36 deg and while its running good its around -24 deg, tells me it may not be far enough out to be a complete no start. Though what is interesting is I don't see much movement on the scope but that is probably due to the time base I used.