gpzkat

I have read many descriptions of cold start/idle problems on this site and elsewhere, but none seem to match the issue I'm having. My issue is not just for a minute or so like others seem to be. I describe symptoms in gory detail below. I wonder if it is something that has plagued anyone else's G37, but the closest I saw was on a Bimmer forum and it was a dead-end, info-wise, as is so often the case.

My car is a 2009 G37 Sedan (Journey with S package) with 79K miles. Original plugs/wires/injectors. It has had any work done on the engine, just oil changes and air filters.

Problem: When the outside temperature is chilly - under 60F, it runs like a plug wire is disconnected for about 15 minutes. Power is down, throttle response is laggy, and I have to give it more throttle than normal to get going. At 4K RPM and above, there is a loud droning noise in the cabin and a lot of vibration from the drivetrain, felt through the pedal and steering wheel mainly. It sounds and runs like a lawnmower.

Even with the coolant temp gauge at normal range, i.e warmed up, it still runs badly until it's driven for a full 15 minutes. Once it smooths out it runs like new the rest of the day. During morning commute the day is warming up so the ambient air temp is where I'm focusing. The next morning, if cold, it happens again. If it's a warm morning though, it's smooth and quiet through the rev range.

I'm not sure when the problem started as my wife was driving it daily (many short trips), and she didn't notice it, but boy, I sure do.

I had an independent shop's master technician (so they said) who spent 15 years at Infiniti/Nissan dealer look at it. He could not figure it out and went down some false leads, IMO, looking at the symptoms, not the cause. He said it could be an imbalanced driveshaft. He noticed the vibrations, but that diagnosis makes no sense to me, why would a driveshaft problem affect the performance of the motor and not happen with a 10-degree difference in temperature outside?

I cleaned MAF and Throttle bodies - they were pretty grungy. After this the CEL came on, I had codes pulled and it said 'high idle'. The light went out after about an hour of driving and it hasn't come back on. Prior to that, the codes don't indicate any misfires, and the check engine light had never come on. There's no change in the behavior. I also bled the cooling system, thinking the coolant temperature sensor might be involved - but again this doesn't jibe with the symptoms related to the outside air temperature that I have subsequently been narrowing down on.

For anyone who read this far and has an idea, I welcome the thoughts.

Rooster Tail

Only cold not wet? Seems like there is some valve / actuator /sensor that is sticking in cooler weather, but is not very close to the engine, and it takes 15 minutes to warm up enough to operate. Would you describe that as accurate?

Rough idle / running could be vacuum leak, bad electronics. Or even the VVT not behaving correctly. Though the latter would almost definitely throw a code.

My initial guess was something in the evap system. That's a complete WAG.

Good luck, let us know what you find.

crookncastle

just because your temp gauge shows normal range does not mean your engine is warmed up fully and the pcm running at closed loop. since you stated you drove it for 15 minutes that should be enough to have the t-stat open and vehicle now running in closed loop parameters. as for the the cold temp causing the symptoms do you have a scan tool to check the air mass with? its possible one of the intake air temp may be faulty in the air mass and providing inaccurate feedback to the pcm creating driveability issues that you may be experiencing. highly unlikely for it to be a vacuum leak since that would be a constant problem.

SonicVQ

Your issue is a little unusual and will require special tools to troubleshoot. Here are my thoughts:

Have a look at the mode $06 misfire counters to see if one cylinder is higher than the others. That might give you a cylinder to focus on.
If you are looking for a good app to gather that data, OBD Fusion (with a cheap OBD scanner) does a good job interpreting the mode $06 data. Torque Pro does not.

Also, using OBD Fusion to look at your fuel trims. Add the short term to the long term to get the total fuel trim. If the total fuel trim if > 15%, there might be a leak in the intake or PCV system.

Pull out all the plugs and have a look at their condition.
In some cases, the spark plug tube can get filled with engine oil and that may cause that cylinder to misfire.

My next step would be to use a scope and look at the ignition coil current.
If you scope the common ignition coil ground, you will be able to see all 6 coils in operation and easily determine if a coil is not operating properly.
(have a look at this: )

I would also check fuel pressure during a cold start. It should be around 51 PSI at idle.
This does require a "fuel tap" since there is no easy way to "tap" into a return-less fuel system.

If you don't have the time or required tools, you could always ask a shop for these tests.
The good news is that the problem is always there when cold, so it shouldn't be too difficult to determine the cause.

Good luck!

gpzkat

Thanks to all your replies, I'll consolidate my reply.
Rooster Tail - just cold is enough, it doesn't need to be a rainy day. I don't think I've had more than one of those in the month or so I've been noticing it (in LA area).
crookncastle - I don't have any of these special tools, I was hoping the tech had what's needed. They said they didn't find any 'old misfires' so I assume they had a way to check.
SonicVQ - I don't even have a socket wrench that'll fit down the hole at this point, so I'll start by getting one and pull plugs. Not sure which reader they used to check misfires, can ask if I decide to try them again or someone else, but I've been burned 2x with no result so far.

Surfnazi

Have you changed trans and diff fluid yet

gpzkat

I had both changed along with coolant and brake fluid flush about a month ago the first time I brought it in for diagnosis. No difference. It's noticeable even at idle in park when revving it, so it seems to me to be isolated to the motor. I can just about guarantee a carbureted vehicle with old fashioned points/coil ignition would never have this problem, as I've grown up and old on them and have never heard of such a specific situational problem. So I'm thinking it is related to the ignition/computer/fuel injection or some combination of those.