adrial

I had a problem with my factory wiring harness back in 2006, where the power to the AFM was being cut out only when the car was warmed up. It was so bad that it would just turn into an undriveable car with a bouncing idle, with power being cut to the AFM every ~5 seconds. After going thru all the usual suspects, I finally troubleshooted it by hooking up a multimeter to the AFM output, then to the AFM power wire. Cured it by running new wiring to the AFM direct from the DME.

I would suggest hooking up one of your datalogger's analog inputs to the MAF output, then input (or even better if you can do both at the same time). At the very least, you'll cross one more thing off the list.

refresh951

Jon,

At about 3 lbs of boost (with car warmed up) in open loop what is your AFR?

Van

Gotcha on the colors. Would intake air temp (intercooler effectiveness) cause a difference in combustion?

fast951

Analog #1 is MAF. Where on the graph do you see it having a steady decline?

Based on the logs, I noticed the MAP sensor is declining. Hopefully you are seeing something we missed.

Apparently, with engine up to temp (no heat soak), the car is fine. When the engine is heat soaked, the problem surfaces... It's time to add/eliminate more variables..

Bill

Mine has been doing this exact issue for the last year. I have been unable to find the issue and it's driving me crazy.

Techno Duck

After further testing, i can 100% say the problem is heat related. Basically as long as the car does not sit at idle and heat soak it will be fine. When the car is fully warmed up, as soon as the car sits at idle for about a minute or more, it will start happening. If i shut the car off, i can come back 5 minutes later and it will start happening also. I think it is highly unlikely the issue is DME or KLR related, but will still be putting in known good units hopefully thursday.

Here is what has been done, checked..

-Intake track pressure tested to 17psi, all leaks fixed..everything hose and line under the intake has been replaced, throttle body was rebuilt recently
-TPS is new.. less than 500 miles on it. I have swapped in my old one which was working also
-O2 sensor.. installed aprox. same time as TPS. When run disconnected the lean spot between 3-3200 gets leaner, so i think this says it is working
-ISV, just replaced it with a new one today. Reset idle speed aswell
-MAF and air filter have been cleaned
-MAF 12v switched and ground were checked
-FPR is an adjustable by Aeromotive. I have ran with the vac. line disconnected and the car still goes lean between 3-3200 rpm. I think this rules it out.. also i believe this was happening prior to the rebuild when i installed it but i cant say for certain
-DME temp sensor, installed maybe 2 years ago, well under 5k miles on it
-Fuel filter replaced in '09 and has 6k miles on it about

Some notes about the lean spot..
-Lean spot does not seem to correct it self. If i try and hold constant engine speed around 3-3200rpm in the lean spot, the car will continue to run lean as long as i keep it in that spot.

-Another point is i cannot get it to go lean when bringing the RPM's up in neutral.

-The AFR under boost is normal (what i expect to see) as soon as i pass the 'lean spot'.

I have no clue what else it could be that is sensitive to heat..

-Alternator; original with 126k miles. Voltage regulator replaced i think 3 years ago.. probably about 20k miles on it since. I ran the car with every electrical item on.. head lights, fog lights, driving lights, rear defroster, blower on 4 the problem was not any worse or better. I do not notice a drop on the voltmeter when the problem happens either. Still, maybe the alternator does not like 3-3200rpm?

-Fuel injectors; aprox. 20k miles on them, installed the same time as the MAF about 3 years ago now. Maybe a heat related issue?

-Fuel injector harness; everything 'looks' ok and there are no other issues with the car other than the one lean spot. I have the LR replacement harness, just have not had time to install it.. maybe tomorrow.

-Fuel pump; also original with 126k miles. I do not think this is the issue because the AFR under boost is normal, drops down as expected with the current tune on the car.

Anyone have any other ideas? Any comments about the components listed above?

This is a very irritating problem. The car runs absolutely awesome when the lean spot is not happening.. i almost cracked a smile when i was on the highway earlier after just starting the car. Even when the lean spot is happening, the car runs good other than the lean spot.

Thom

Probably not what you want to hear, but considering all investigations you have done it sounds like a fault in the wiring harness.

Thom

The IAT sensor is in the MAF and the CLT sensor is the DME Temp sensor, which can be reasonably ruled out according to what Jon said.

Were both crank sensors checked?

Techno Duck

Both crank sensors are relatively new, cant remember exactly when but pretty sure i replaced them last year. I used the Bosch BMW part # ones with the slightly longer harness. I gapped them using the tool Sharky used to produce, never had any problems with starting or the car cutting out.. but maybe when the car gets hot something funky is happening? I have the LR crank harness i never put in also.. (won this at 944 Fest last year!). Seems unlikely though but i am willing to take any suggestions right now.

A problem with the engine harness is looking more and more likely at this point. Wish it were still possible to get a new engine harness.

I did some driving last night with a buddy of mine. He suggested i brake boost and see if the lean spot still happens between 3-3200rpm. If it did, this would suggest it really is a RPM dependent problem. I tested this a few times on the highway and the test was inconclusive, it was difficult to keep the car in the RPM range with a ~200rpm window going off the tach / wideband. I am going to try this again tonight with my laptop hooked up though so i can do it based on load column instead. Another test i thought of while doing this was simply accelerating from say 2500 to 3500rpm while brake boosting. This will shift the load columns which would tell me for sure if it is RPM or load dependent. I was going to test this last night but passed a highway patrol while brake boosting at about 80mph.. the car practically did a nose dive because not only was the braking done with my left foot which is 'numb' to brake pressure but the track pads i have on were sufficiently heated up from all the brake boosting.. .

Any thoughts on the fuel injectors being the culprit? Has anyone ever heard of a fuel injector having a heat related issue at a specific duty cycle?

How about a bad MAF sensor? I know they get dirty but do they ever give a bad reading at a specific load? Seems unlikely with what the datalogs show, but i am totally out of ideas.

So close to having this car running good, must resist temptations to sell it or put in an LS1.

fast951

Based on the logs, and comparing the MAF voltage betwen when it's running fine and when it goes lean, there is no indication that the MAF is causing it. It was the first thing I examined, you can see the signal on your charts earlier in this thread. The only thing I noticed was the MAP is reading lower, indicating more vacuum??

Adding load to the engine (left foot braking) will help eliminate a few variables... Just make sure you cool the brakes in between runs. And don't ask me why

Tom M'Guinn

There used to be a notorious 3k hesitation in the n/a cars, with varying theories about what caused it. Have you confirmed that your bell housing ground points and all your wiring to the MAF (power and ground) are squeaky clean?

Techno Duck

Tom, i actually just cleaned the bell housing ground today. Did not test drive yet but will be in a few minutes. The ground was cleaned last summer when i was having issues with it running uber rich (turned out it was just a dirty MAF). It still looked perfect but i cleaned everything with a wirebrush anyhow.

The MAF wiring was ruled out, i ran the ground wire directly to the battery with no change. Also i moved the power wire from the test port to the cycling valve connector.

Tom M'Guinn

Since it's rpm related, have you checked the voltage on the battery at 3k? Maybe the alternator is faultering at that rpm range, taking the dme a second to adjust? Just grasping at straws maybe, but you may want to put a multimeter on the battery and see if anything interesting happens at 3k?

blown 944

I chased a miss for a few months that turned out to be relatively new injectors. Happened overnight out of the blue. New injecors fixed this problem. It only appeared under low load conditions

Techno Duck

Alternator was on my list of things but just ruled it out today. Hooked up my multimeter to the battery and wedged it inbetween the windshield and dashboard. Nothing abnormal, stays around 14v the entire time even with the lean spot.

I am going to order a new set of flow matched 72lb injectors in a little while. I am totally out of ideas, the last thing left i can think to replace is the injector harness which i will try and do tomorrow if i have time. Going to order the injectors now in the interest of getting them before i head back offshore for another week. It will suck having to wonder for the next 4 weeks if it really is the injectors. Atleast i can rule them out, even if they are not the cause atleast i end up with a spare set or i can just sell them to someone looking.

Spare DME and KLR's should be here today or tomorrow.. will swap those out also but i think the problem lies somewhere under the hood due to the way the heat soak is such a big factor.

Really hoping its just the injectors.. i am so damn close to getting this car running 100%, this stupid lean spot is killing me.