Sudden rough idle

[jfoj]

Not sure I am going to explain step by step how we gathered the data, this has been covered earlier and in other posts on this and other forums.

As for the graph I posted, this was for Pre-Cat O2 sensor performance on an BMW E46 with a 3.0l I6, non wideband O2 sensors and SAP system.

So for anyone trying to figure out what is going on and what we are doing here what we are going.

After a long time dealing with driveability issues and a lot of work with other members on many forums I finally determined that the best way to check and verify a "standard", non wideband O2 sensor is to graph the sensors on warm up. What we have found is there are MANY, MANY lazy and slow to respond O2 sensors that will never trigger any DTC/trouble codes. What happens is these lazy/slow to respond sensors cause fuel trims to climb, poor fuel economy and at times even other performance and driveability issues. This is partly due to the how the Adaptions adjust and shift once the O2 sensor(s) degrade.

Interesting enough O2 sensors are consumable, they need to be replaced like spark plugs. But that being said many times they can last more than 10 year/100k miles. After a lot of thought and data gathering, what has become clear is O2 sensors rarely trigger DTC's or trouble codes unless the heater circuits fail. So we have resorted to cold start graphing of the O2 sensors and we have been able to consistently and reliably identify degraded sensors and condemn them as needed. In many of these cases the sensors may not be the ultimate source of the original complaint, however, in the process of "Whack A Mole" most owners have decided to change out the O2 sensors that we have flagged as questionable/lazy/slow responders.

Due to the fact we can gather data and graph the sensor performance, owners are more likely to replace the sensors if they are in question rather than "blindly" replacing the sensors and having no idea if they were bad to begin with and there was not solid proof the replacements were any better. Now we have proof of before and after O2 sensor performance.

One thing we have also been seeing is Emission Readiness Monitors not clearing and many of these cases have been directly caused by lazy/slow to respond sensors that also had lower overall Voltage swings. All the Emission Readiness Monitors rely on the O2 sensors for almost all aspects of the Monitors changing to the Ready or Pass state.

If you look at the graph of the new sensors, you will see a deep "V" that starts almost immediately when the engine is started and then sharply cuts off around the 66 second mark. This is the SAP pump running and adding additional air into the exhaust stream, then the sharp transition is when the SAP actually turns off. So I can detect how well the SAP is running and for how long with these graphs.

You can see how the old sensors are very sloppy and slow to respond, they have a very shallow or non existent deep "V" and slow overall switching time.

A fresh and active O2 sensor should have switching transitions about once a second. In this graph the new sensors do switch rather quickly and often, then the switching time does decrease a bit as the graph progresses.

I suspect we still have some issues with this specific car, but the Adaptions were not cleared and we will need to wait a few weeks for the Adaptions to adjust and will revisit the O2 sensor switching time then and see if there are other issues with this specific car.

This is a great way to test for O2 sensor performance. The problem is very few OBDII tools allow this type of data gathering. You will likely need a smart phone App or a computer based program to record or log the OBDII PID's and then graph the data. Most hand held OBDII scan tools to not have the ability to record the OBDII datastream. Luckily most of the software/Apps that support the OBDII recording/logging are less than $10 for the App and usually no more than $20 for the interface.

I consider what we are doing as trending, finger printing or gathering a signature of the O2 sensor performance for the E46. Similar testing can be done on other cars, however, the trend, finger print or signature may look slightly different.

The E53 is pretty close to the E46 and E39 so this should be be a typical graph of the E53 Pre-cat O2 sensors. If someone wants to gather data we can compare the E53 and possibly have a good benchmark or finger print.

[jfoj]

By comparison, below is the graph of the Pre-cat O2 sensors for greggo151.

Note the this may not have been from 100% cold start and this graph has more than idle and for a longer period of time.

But I did not have enough solid data and I do not think the O2 sensors on greggo151 vehicle are the primary problem.

It is clear the fuel management system is not working and the O2 sensors are seeing a pretty Lean condition consistently. But more data would have allowed a better view of what is really going on.

[greggo151]

all right, the car is back, and here is the diagnosis/repair none of which should surprise anyone:

Smoke test

Replace lower intake boot. It was torn on the small accordion hose on the inside facing the fire wall...I just didn't see it.

Replace PCV hose from crankcase to dipstick

Flap on intake manifold actuator broke.

I tried out a new independent here in Bakersfield, Precision Auto. Great service today and the car runs great. My son says it has it's power back.

Thanks everyone for the help...

BTW to clear one thing up, I cleaned the MAF after the rough idle started yesterday.

[jfoj]

greggo151,

The problems you described are all NORMAL or STANDARD features of the BMW I6 after about 10+ years.

I would still recommend gathering the 3 Log files even though the vehicle is running better.

Would be interested to see what the O2 sensor health is and what the Fuel Trims look like. I would bet there may still be some lingering issues.

If you can get the Logs I would be happy to look at them.

[bcredliner]

Great that all is well again! I hate those kinds of problems as well.

[jfoj]

I suspect that grego151 may have a lazy Bank #2 Pre-cat O2 sensor, but would need more data to know for sure.

[jdstrickland]

Originally Posted by greggo151

That reading came after I cleaned the MAF...I so hate chasing Idle/Misfire/Vacuum issues.

The plugs are also fairly new, probably 8k on them. Coils I don't know. But this also seems more like vacuum than ignition.

The vacuum issues that arose from the broken bellows could easily be 100% of the problem.

The MAF reports the quality, temp and density, of the air in the intake. You broke the rubber bellows at a point after the MAF, and this causes the engine to run poorly -- it is running lean.

You've reported that the problem is fixed, I'm just saying that the hose (duct) with the flexible bellows was the problem.

It's not clear to me that you were doing some preventative work and actually created a problem, or if the problem occurred spontaneously due to old and rotted rubber.

[upallnight]

Originally Posted by jdstrickland

The vacuum issues that arose from the broken bellows could easily be 100% of the problem.

The MAF reports the quality, temp and density, of the air in the intake. You broke the rubber bellows at a point after the MAF, and this causes the engine to run poorly -- it is running lean.

You've reported that the problem is fixed, I'm just saying that the hose (duct) with the flexible bellows was the problem.

It's not clear to me that you were doing some preventative work and actually created a problem, or if the problem occurred spontaneously due to old and rotted rubber.

The rubber bellow normally cracks due to the age of the components. I think he was having a lean issue before he posted the scan. The scan he posted is a scan of the vehicle after a prior scan was done and any previous trouble codes were erased. This can be verified by this statement in the scan that he posted.

Originally Posted by greggo151

2002 X5 3.0 145k miles...Car drove fine yesterday and this morning, and this afternoon it just threw codes and had a horrible rough idle. It drove Okay at speed, but it just shudders at idle. I didn't see any obvious air leaks. Here is a report.

Diagnostic Report


Created by OBD Fusion - OCTech, LLC www.obdsoftware.net

Date: 12/14/2015 6:35:46 PM

VIN: Not Available


MIL On
Number of Stored Codes: 7
Readiness Standard: None

This vehicle is not ready for emissions testing.

Reason
MIL On
Stored trouble codes have been detected
Number of incomplete tests exceeds the maximum number allowed
Trouble Code Report


5

Incomplete tests are produced when a scan of the trouble codes are erased. Several drive cycles must be recorded before the test are completed and the incomplete changes to complete. This was put into place to prevent people from clearing trouble codes to their cars prior to taking it in for testing. When the incomplete tests show up on the screen that is an automatic failure.

Since the original poster kept on talking about chasing vacuum leaks that lead me to believe that there were P0171 and P0174 codes on the prior scan. The scan that the poster posted did not list any P0171 or P0174 trouble code. You would have to drive a car 10 - 15 miles before the fuel trim would have informed the DME that the engine is running lean.

The actuator for the DISA valve was listed as also broken. That would have also contributed to the poor performance of the engine.

[upallnight]

Originally Posted by jfoj

By comparison, below is the graph of the Pre-cat O2 sensors for greggo151.

Note the this may not have been from 100% cold start and this graph has more than idle and for a longer period of time.

But I did not have enough solid data and I do not think the O2 sensors on greggo151 vehicle are the primary problem.

It is clear the fuel management system is not working and the O2 sensors are seeing a pretty Lean condition consistently. But more data would have allowed a better view of what is really going on.

Since greggo151 cleared the trouble code, he in essence also reset the long term fuel trim. The values for the LTFT can not be consider a good indication of the health of the fuel system and engine. He had a vacuum leak so that was the cause of the lean condition. He should post a new scan of the engine now that he corrected some of the problems.

[jfoj]

Originally Posted by upallnight

Since greggo151 cleared the trouble code, he in essence also reset the long term fuel trim. The values for the LTFT can not be consider a good indication of the health of the fuel system and engine. He had a vacuum leak so that was the cause of the lean condition. He should post a new scan of the engine now that he corrected some of the problems.

Fuel Trims do not get reset when clearing DTC's. Trouble codes and Freeze Frame data and Emission Readiness Monitors are the only things cleared when DTC's are cleared.

Clearing Adaptation will reset Fuel Trims, but for BMW a BMW specific software, tool or Pro level scan tool with Adaptation Reset capability is required.

Assuming greggo151 checks back in here, he should go ahead and get 3 Logs to see what the O2 sensor health looks like along with Fuel Trims.

It appears his vehicle has had very little maintenance for the STANDARD BMW I6 problems.