Just when I thought I was making progress...1222!

[BennyM]

Help!

Day 0: Install newerish exhaust system. Start car. Car runs quiet and smooth, but still has annoying whistle/whine (relative to engine RPMs) that it's had for nearly a year. Still suspect minor vacuum leak. Drive car home.

Day 2: Car sits. Reset ECU to clear a lingering 02 sensor code (1221).

Day 3: Drive car to run errands. Once car is near normal operating temp, idle begins to fluctuate. CEL comes on. Idle gets worse and it dies just as I park it. I let car cool down. Run the stomp test: 1222. Suspect that whistling/whining vacuum leak has gotten too big for the engine to handle. Start it. Every thing's fine. I drive it home and as it warms up, it starts hunting for idle again. CEL comes on. The car will die if I don't keep applying gas, but when I apply the gas, it feels and sounds normal. I limp home this way, keeping the RPMs up, rolling stop signs.

Day 4: Take apart everything running into the intake manifold: Throtle bodies (2 because I have ASC), air boot, air filter housing, and the ICV and it's corresponding hoses. I check other vacuum hoses for cracks. The ICV hoses are covered in oil. So, I buy a new connector for the top hose that plugs into the manifold and tighten the clamps on the lower hose. I then tape up the connections with Gorilla tape to temporarily cover any potential leaks. I tape up the air boot as well, which has a long gap on the bottom. Put everything back together. Start the car. The whine is gone, but now the CEL light is on the whole time and it hunts for idle a bit even thought it's halfway warm. If I open the throttle for a moment and then let it settle back to idle, it fluctuates dramatically, almost dying, but after 20-30 seconds it'll even out for the most part and keep running. Idle still gradually deteriorates as the engine warms up. I disconnected the O2 sensor, but that made no difference. I tried spraying hose fittings with carb cleaner. No difference.

Day 5: I've made a list of hoses to order from BMA and I will go out shortly to check the sensors. But I'm concerned that it's not as simple as a hose fitting. And it seems like too big of a coincidence that this happened after I replaced the exhaust system. I mean, i've seen it hunt for idle occasionally, but not to this extreme.

Keeping these symptoms in mind, what should I check next? Could a bad ICV or MAF cause these symptoms? How do I clean the ICV?

I plan to order new intake manifold gaskets along with the hoses too.
http://www.realoem.com/bmw/showparts.do?model=HD53&mospid=47404&btnr=13_0391&hg=13&fg=15

http://www.realoem.com/bmw/showparts.do?model=HD53&mospid=47404&btnr=11_1124&hg=11&fg=40

Thanks in advance for your input.
BennyM

[bubba966]

A quick test for checking the MAF is to unplug it. If it runs better with it unplugged, then your MAF is bad. If no difference then it's not the MAF. Either way plug the MAF back in after testing.

Cleaning the ICV is simple. Take it out. Get some carb/choke cleaner and spray it inside the ICV and shake it. When the valve inside is able to move back and forth with ease while you shake it it's clean. You'll also see the varnish & dirt get cleaned off while you're spraying it with the carb cleaner. Just keep spraying it and shaking it until the valve moves freely and you see no more dirt/vernish build up. It's really quite simple and will make more sense when you've got the ICV out and in your hand.

[BennyM]

Small update.

The engine is running very rich. You can smell it pretty easily.

Pulled the oil dipstick while engine was running. No change.

Then, with engine still running, I unplugged each sensor I had access too and plugged them back in. Each time I unplugged a sensor the idle got worse. This was true for the MAF as well. I couldn't reach the ICV while the car was running, but I did measure the resistance per the Bently manual afterward. Here's my findings.

Spec: My readings:
20 ohms 15 ohms
20 ohms 15 ohms
40 ohms 25 ohms

I tested the meter on known, working resistors I had lying around and it was pretty accurate, so I really think the ICV is providing LESS resistance than it should. What does that mean? I'll take it out and clean it when my new hoses arrive.

Also, could an exhaust leak cause this hunting idle behavior? I just found that one connection between the resonator and the cats was dripping water.

[BennyM]

So, idle is better, but as soon as the temp needle gets to the 11 O'clock position, I get the CEL and if I rev the engine at this point and let it return to idle, it just about dies. After a few seconds of hunting the idle levels out. I have replaced the intake manifold gaskets and every vacuum hose, seal, and o-ring I could find. I checked the resistance of the temp sensors and the TPS, cleaned and tested the ICV, and made sure there was voltage to the MAF.

Has anyone found a solution to an open throttle to idle transition issue like this? Again, it only happens when the engine is hot.

My next step is to test the O2 sensor for voltage pulse. I have run the engine without the 02 sensor plugged in and nothing seems to change. I feel like if the O2 sensor were sending a bad reading, then disabling it would make the engine run better.
I also want to test with a different MAF (though I don't know where to find one).


Please help! I'm going crazy here. I don't want to just replace part after part. There has to be a way of figuring this out.

[632 Regal]

still sounds like an intake leak from all you describe.

[MBXB]

I'd check the temp sensors again. Static resistance may be OK, but it may not kick in at the proper temp sending garbage to the ECU.

[BennyM]

Well, I'm not done checking stuff, but I did find that the O2 sensor heater wasn't getting any power, and I suspect that this would lead to screwy readings. The sensor produced voltage in a range of 0.17v to 0.4v (usually right around 0.19v) I also suspect that this is close enough to the acceptable range that the computer would continue to run in closed loop mode, trusting these readings and dumping fuel into the engine. This is my theory for now at least.

The o2 heater relay was bad. I have another on the way, but I was wondering if anyone sees any harm in temporarily jumpering terminals 87 and 30 (which is what the relay should be doing) so I can test my theory. It seems pretty straightforward to me, but then again, there must be a reason they have a specialized relay doing this job. Anyone think I shouldn't run the car this way?

[BennyM]

Conclusion:
For anyone who is curious, replacing the O2 sensor (and all leaking vacuum hoses) did eventually solve the 1222 problem. I think if the computer detects erroneous (or stuck) information from the O2 sensor, it actually cuts power to the sensor heater and ignores the unit altogether. Idle is now rock-solid at all temperatures.

Instead of checking resistance values and voltage levels, I wish I had ignored the Bently and just focused on getting the old O2 sensor out and a new one in. So, a lesson to everyone else: when you get a 1222, start with a fresh O2 sensor. Then replace all your hoses, and THEN consider more obscure problems.

[Tiger]

So where is this O2 relay located?

[BennyM]

So where is this O2 relay located?

It's in the electronics box in the engine bay, front Passenger (right) side...next to the DME, main relay, and fuel pump relay.