Automotive Forums  Where Car Enthusiasts Go to Discuss Cars Online

My issue seems like the ideal case study exercise for a mechanics classroom textbook. Except I am not a student in the class. As a result, any assistance would be appreciated.

Vehicle
2000 328i with 166k

Problems:
SES light is on
Car will not pass smog
Consistent RANDOM MISFIRES code is returned
Consistent MISFIRE code is returned for specific cyclinders (1,2,6)
Inconsistent Fuel System 'Closed Loop' status returned*
Inconsistent power loss
Inconsistent fuel economy
Inconsistent rough idle (regardless of temperature)

* Additional information from scanner's Live Data supplied below

Work Completed
Spark plugs replaced
Ignition coils replaced
MAF replaced
No vacuum leaks found (not professionally smoke tested)

After items 1-3 were addressed, my problems went from being constant to inconsistent but with vastly improved performance. This means that...

When the car runs 'normal', the following things from the LIVE DATA get my attention:
Idle becomes stable AFTER the secondary air pump goes off.
Fuel System status - ‘closed loop’
STFT hovers near 20% or stays at maximum 27.3%
LTFT NEVER changes from 0.0%
Consistent get RANDOM MISFIRE code along with 3 specific cylinder misfires codes as previously described above
O2 sensor voltage averages 0.4V


BUT when the car is NOT running 'normal', the items above reflect the following:
Idle NEVER becomes stable - car randomly shakes (from misfires)
Fuel System status - Open Loop due to system fault
STFT hovers around 9.38%
LTFT NEVER changes from 0.0%
Consistent get RANDOM MISFIRE code
Inconsistent cyclinder specific misfire codes from ALL cyclinders
O2 sensor voltage averages .01V

Questions
Regardless of how the car is running - IAT (internal air temp) reads a little high. I checked this due to info posted here that a high reading could indicate blockage in the intake manifold. Its 61 degrees today and when the vehicle is at operating temp IAT reads 97 degrees. Is this sufficiently high?

O2 sensor voltage, when showing low voltage is indicative of failure; however, at other times (not during the same drive) they operate within normal voltage parameters. Would a failed or failing sensor sometime work properly and other times not?

When the car is running poorly, the STFT value is hovering around 9.8% (assumed to be an acceptable value), the Fuel System is NOT in a 'closed loop' and the O2 sensors are showing a very low voltage - 0.01v (NOT within acceptable range). Can one conclude that when this occurs the ECM is NOT using the O2 sensor data (correct?) and has defaulted to the manufacturer's predetermined settings? If this is correct - what would cause a car that is using default fuel/air mixture ratios to NOT idle as designed?*

* ADDITIONAL INFO RELATED TO PREVIOUS QUESTION - MAF is showing nominal readings

When that car is running 'normal' the STFT value stays at about 27.3%, the Fuel System is in a 'closed loop' (using the data from the O2 sensors) and the O2 sensors show nominal voltage outputs of about 0.4/0.5v. BUT the high STFT value indicates the car is running LEAN (not enough fuel). With such a high short term fuel trimm value, why hasnt the LTFT kicked in?* Is this even something I should consider when diagnosing the problem?

Concluding thoughts for today
MAF is metering air.
We have ignition.
O2 sensors sometime reflect normal operation and other times appear to be failng/failed.
Why am I getting cyclinder specific misfire codes? I could understand it better if the misfires were for cyclinders on the same bank - but they are not (right?)
But then why am I failing my smog? If I were running rich (which I am NOT) then I see a relationship cause I could be passing unspent fuel through the system.
I need more tools or money to resolve this issue and unfortunately, I have neither right now.  

i almost never get much time under the hood of bmws.  so with the info you have posted..

i would probably dig deep into the smog pump air injection system..

i have had failed diverter valves do similar things where they allow air into the upstream exhaust diluting the exhaust flow across the oxygen sensors causing Havoc with the air fuel ratio.


its not the average voltage of the oxygen sensor...  its the sweeps back and forth usually..  although thats probably a wide band oxygen sensor on that model..

it will take me a little while to research how the system operates on that model.    if you were in the SF bay area.. i have learned of a great shop in that area that handles BMWs  

when i looked for a faster answer over at bmaparts..

they show a air pump shut off valve..   that controls the flow from the smog pump... and shuts it off tight as commanded by the computer...

how the computer controls this cast aluminum shut off valve. is with a small vacuum switching valve that is controlled by the computer..  this switches the vacuum from the intake off and on.. to control the air injection systems flow into the exhaust system...

why is the ability to switch the flow important..

the exhaust flow can create lower than atmospheric pressures at some points of operation.. so it needs to be SHUT off tight..  as commanded while the smog pump slows down and stops..

it looks like its only 2 bolts and a hose clamp to remove the shut off valve.. i don't know what it takes to reach it..

the idea is to test the diaphragm where the vacuum hose hooks up for sealing .. when vacuum is applied.. it should HOLD vacuum..   it should also either open or close off the air flow through the large fittings tightly..  no leaks at all when you try to blow through it with it off the car..

this is where i would start.. so it does not cost any money to do this..  just some sweat...

you can look at the parts images at bmaparts  online..  they are at the bottom of the emission controls menu..  

I love BMA Auto Parts!

Thanks again for your input. You are the first to mention this - which is great. Thanks!

One more piece of information regarding my issues. I previously mentioned my IAT in my original post was only 97 degrees - wrong. After driving the car later that night, the outside air temp was around 50 - the coolant temp read 197 and the IAT was reading 150.

Not sure if that is indicative of something - but I thought correcting my information was worth mentioning.

Ill post back soon with my findings.

A sincere thank you!  

plug in your scan tool.. with a COLD engine before you start it..

see if the temp is correct.. with ambient temps ..

you can unplug it.. and it should go to one end of the scale.. short the 2 wires to it with it unplugged and it should go to the other end of the temp scale..  this proves the wiring and connector are intact..   be sure to clear the codes.. as the computer if it fails.. MAY set a default temp...  and you could be seeing a GHOST reading..

intake air temp sensors are available also..  they are generally a thermister that is exposed to the air flow.. i have had a lot of them get damaged..


one more thing..  are all your hoses tight??