So, I've been battling a Code44 left side lean which I can not figure out. Initially I thought it was a bad o2 sensor connector, followed a bad plug wire (which there was) and also did a heated o2 sensor conversion in the process.

This all started after putting a new set of headers and o2 sensors on the car, but I suspect the car has been running lean for some time. Finally got datamaster running and did a quick log the other night to find the passenger side is running lean too. I see that IAC position is at 0 during idle which usually indicates a vacuum leak, however I cant find one anywhere.

Hoping somebody can take a look at the log and help me out


https://drive.google.com/file/d/0B4E...ew?usp=sharing

 

Well, I would like to help you but I am away from a PC right now. I would like to know what the BLM is? You should connect the O2 sensor signal wire to a graphing meter and with the engine running, inject propane into the intake either through a vacuum line or the snorkel. If you get a large voltage spike on the graphing meter, the o2 sensor is processing oxygen change. If during normal operation the BLM is clocked at 160 and the O2 sensor is reporting a voltage that is above .450 volts, there is a problem with the way the ECM is interpreting the O2 voltage.

Also, please could you tell me your MAP voltage and MAP converted vacuum pressure at idle?

If your car has a MAF sensor, I'd like to know the voltage or gm/sec of that as well.

Another thing you can watch for too is injector base pulse width. If you have a data log of when the car was in good running condition you can compare the idle values of the injector BPW. The injector BPW should increase as the BLMs increase if there is a problem with the o2 sensor.

 

I reviewed your data log on my computer with Data Master. It looks like it's just a bunch of gibberish and inaccurate sensor data. Could be the way my program is set up, could be the way you recorded it, could be a problem on your end or a problem with your ECM, but I could not read any accurate sensor data at all.

RPMs were around 5500 the whole time, MAF was in the 130-180 range the whole time, fuel trims were below the minimum limit, coolant temp was -40* the whole time, car was in open loop the whole time, etc. Looks like a corrupt log to me.

 

You need to use DataMaster DA version to view the datalog. The datalog is viewing correctly using version DA

 

First thing is try closing your throttle blades by adjusting the Idle set screw on Throttle body. Lets see if your IAC counts go up. Your Log does indicate higher BLM on the Left side , but there is O2 correction being applied by ECU to try and adjust. There is no spark retard.

 

Thanks. I got the log to show up. What I noticed is that the BPW on the left bank is significantly higher than the right side. I watched the BLMs as you drove off, didn't look like the patterns of a vacuum leak to me. I'd say either a misfire on the left bank or a clogged injector or something on the left bank at this point. I would pull the plugs to see if the left side is running the way it should be as if the computer is doing it's job properly or if the computer is getting a false reading somewhere and dumping fuel into the cylinders.

The BLMs on the right side were also pretty high as well...

 

Thanks for your input, alot of good thoughts flying around.

I've always understood that the idle set screw is not something to ever be adjusted?

I may be able to get the plugs pulled tonight if there's time and report back.

A few other things i've noticed:
- there is a backfire during shifts (not sure if it's due to the new exhaust or not)
- When cruising coming to a light, i'll push the clutch in and it will stay rev'd around the 1000-1200rpm mark. It comes down to the 750rpm area only when I'm at a DEAD stop. I would think rpm's should drop to 750 whenever the clutch is depressed.

Could the IAC be gummed up or not fully seating causing extra air to be introduced?

 

Some exhaust systems will backfire a little bit. It can also be caused by some exhaust leaks. Typically if that is the case they wont be very loud. A backfire that sounds moderately loud or like a gunshot can be an indication of retarded ignition timing, rich air/fuel mixture, or a misfire.

Judging by what you just said, it would give me more reason to believe the spark plugs would need to be read.

Also, my car sometimes comes to a high idle when I stop and the idle RPMs drop down over a few seconds. I don't think that should be the cause for concern right now. Quite frankly it's not very much of a problem unless it is excessively noticeable. But I still think the original issue you mentioned needs to be addressed first.

 

Plugs were all tight along with the wires on the drivers side. They all look good to me....

 

You need to adjust the Idle set screw to close the throttle blades , This is so the IAC counts come up to at least 40 - 50 so that it gives the IAC some room for control. I have had personal experiences where the backfire is caused by fouled plugs. This may also be causing the high BLM as unburned fuel is read as a lean by the O2 sensor. Great idea to start by pulling/ reading the plugs.

 

Is it my eyes or do the plug gaps look massively large. Please verify that the gaps are 0.045

 

So it looks as if there is a real problem here where the o2 sensor is reporting accurately.

I would try to do a vacuum leak smoke test using a black and mild cigar, 3 foot 3/8" vacuum hose, sealed metal container, and air compressor.

Punch two holes into the metal container. One that's big enough to squeeze the vacuum line into, and another one to fit the compressor nozzle into.

Find a port on the intake which is large enough to fit the 3/8" vacuum line over. Light the black and mild cigar and slip it into the vacuum hose. Then slide the vacuum hose with the cigar into the hole in the metal container.

Remove the front of the snorkel from the air box or somewhere where you can fit a rubber glove over it to seal off the intake. Use a zip tie or two to secure the glove.

Apply compressed air into the metal container and clamp off the vacuum hose with something after the intake is pressurized. Watch for smoke to appear.

To me it appears to be unmetered and unevenly distributed air as the cause of the problem. Something to cause a significant fuel trim change like this should be pretty noticeable, not indicative of a small vacuum leak but rather medium sized.

If that checks out okay, I'm going to give you the procedure to test the MAP sensor.

 

I agree. They do look out of spec. Adjust them to .045" first and then data log again and see if the BLM moves significantly closer to 128.

But I would think if the gap was excessive that there would be fouled plugs.

 

Thanks Frank. I'm a bit nervous to start messing with the idle set screw because everything i've read says it's sealed to discourage adjustment. Isn't closing the throttle blade by adjusting this screw just masking a problem somewhere else? (i.e vacuum leak)

The plugs are actually gapped at 0.050, I thought that was the factory gap. It may look large in the picture because it's hard to see the needle tip coming off the ground strap. It's a dual tip DENSO design. Maybe it wouldn't hurt to knock the gap's down to 0.045 before I datalog again.

Creative way to smoke the system I'm thinking this may be the next step. Though, i'd probably have a shop smoke the intake track since it can cost more than 1 hour of labor.

 

It honestly works a lot better than any expensive smoke machine I have ever seen.

 

Well...went out for a cruise last night and the service ride control light came on. Got home, pulled the codes and here's what I got:

module 1:
- H41 (ecm serial data circuit loss of communication)
- H74 (LED dimming output open or short to ground)

Module 4:
- H44 (lean)

Module 9:
- H72 (serial data malfunction)

So it looks like I have more problems than originally thought. I'm going to check the ecm ground above the oil filter and check its tight.

 

ECM ground was nice and tight. Since they are history codes think I'm going to cross my fingers and hope the serial codes dint come back

 

Well it's been a little while since i've started this thread, but alot of work has happened.

Step 1
- Cleaned engine bay grounds and the grounds underneath, above the oil filter and bell housing.
- In the process found a damaged fusible link to the alternator which I also replaced.
- Cleaned ECM connections

Data logged the car and still found it to be running lean with IAC counts 0 at idle.

Step 2
Thought it may be the aftermarket PROM in the car, so I took it out and replaced it with the stock PROM. Took a datalog which confirmed it was running the same and the PROM was not the cause.

With AC on IAC counts jumped up which indicated to me it was functioning properly. The only thing left in my mind was a vacuum leak.

Step 3
So I broke down, took it to a local shop who smoked the intake and found a leak in the rubber hose between the EVAP purge solenoid valve and the throttle body.

The line was replaced and it now idles great and the engine speed settles nicely coming to a stop. BLM's also came down but are still not at 128, but close.

They also indicated there was a small leak at the throttle blade bushings, but nothing serious and didnt recommend replacement.

I'm just glad to be back up and running, just in time for dream cruise

Still have history codes for the serial data malfunction (H72 module 9). Even though they are history codes and i've tried to clear them they remain and the "Service Ride Control" light remains on....I'm hoping there are no issues with the ECM...

Thanks for everyone's help!

 

I want to update again for future readers.

Found both exhaust manifold gaskets to be leaking on the bottom side of the flange. Ordered a set of percys aluminum gaskets and they are leak free now Popping of the exhaust during shifts is gone.

Also, cleared all the history codes and they have not come back. I'm thinking during the data log process it triggered serial data codes?? Either way it's running great!

 

Thanks for the update. I wish more would do it