I've had my 997.2 for about 8mo. It's got ~77k miles at this point. There was an issue in October where this happened. I started cold, then the engine got rough right after I started driving. I restarted the car and it resolved itself. When I looked at the PIWIS it was the same error codes as now.

As of today it seems to stay. Looking at PIWIS I am getting P0175 O2 sensor is rich AND misfires on 4,5,6. P0304,5,6.

Is the O2 sensor complaining because the cylinders are misfiring or are the cylinders misfiring because the O2 sensor for bank 2 is off?
Is it possible I just have a bad / failing O2 sensor which is making the whole bank 2 misfire? If I recall modern O2 sensors have a relatively narrow operating temp. I understood that they have a heating element to get it there quickly.

Wondering if bank 2 is not (always) getting to temp as quickly as it should. Also I am in NorCal. It's never cold. Today it was almost 40 out (which is rare). I think it might have been similarly cold the last time this happened. Also it would be helpful to know if it's likely the before cat or after cat sensor?

What are some things that would affect all of bank 2 but nor bank 1?

Any suggestions on how to test the O2 sensor (short of just buying one and swapping it)?
Anyone want to throw out some insight?

Thank you in advance

 

Look for things specific to that bank like a vacuum leak on that side's intake manifold, that camshaft position sensor (my guess), a cracked AOS corrugated hose, etc

If you have a scan tool, see what the camshaft deviations are for both banks are. If the one misfiring shows all zeros, that sensor has failed. The reason it works when first started is that the car ignores the inputs from them for the first minute as it runs in open loop then once it starts using the input from them in closed loop it starts misfiring. Happened to me on my 997.1 and was repeatable - ran fine when first started than aft 30-60 seconds would get rough. Ended up testing this a few times when I had to move the car around in the garage and driveway.

 

Oh this is really good.
This is easy enough to test as I can swap sensors on each bank (I assume they are interchangeable) and I can see if the problem moves to bank 1.

As I read this - the other options are some form of vacuum leak, cracked AOS hose, etc after the MAP. My understanding is that this would more likely produce a lean mixture ( there's more air than expected) rather than rich? Is that why you are leaning towards the solenoid or cam sensor?

Any tricks for testing the cam sensor while still in the car?

N

 

You need a scan tool that can read live valive and want to see what they are for Bank 1 camshaft deviation and bank 2 camshaft deviation. If one of them reads 0 that probably a failed sensor and if that's on the back that's logging misfires that's likely it. You can swap it with the other bank and see if the faults all move over there to confirm it. Will look like this if one is bad.




Next see what your mean fuel trims are for banks 1 and 2. The DME will try to keep them at 1.0000 all the time if one reads lower than that, it means it's sensing a rich condition and pulling fuel. If it reads higher than that it's a lean condition and adding fuel. This is what it should look like.



If you don't see these readings, get a smoke test done to the intake and you'll find the cause of the vacuum leak.

Let me know what you find.

 

Thank you for the pictures - much appreciated.
I am doing this through PIWIS so I think I can find the equivalent values to monitor.

I'll let you know what I find out.

 

The cam shaft deviations are 3.36 on bank 1 and 2.25 on bank 2
Neither of them is 0, so I assume that the cam sensors are working?
It seems odd they are different. But I don't know what the unit °KW is

The fuel trim mean value is 10.45 and -24.86 %
I assume this is the DME trying to pull the fuel back on Bank 2.

The actual lambda values hovers closer to 1.00 on bank 1 and bank 2 is closer to 0.88
FWIW its worse at idle than at 3k.
Eventually the nominal value for bank 2 moves up to 1.10 - I assume thats to adjust for some issue.
---

Given that throttle reduces the problem makes it seem like a vacuum leak in the exhaust causing weird O2 readings.

Never learned if the exhaust manifold, before the cat, runs at a vacuum or not, ie would a leak result in air being pulled in or pushed out.

 

Don't know if this will be helpful or not but I had a similar episode last week except it never resolved itself by turning the car off and back on. 69,500 miles on the car. CEL lit up and car running rough in normal mode but ran just fine in sport and sport plus curiously. Either way the problem was diagnosed as misfire on cylinder # 4 due to a bad coil.

 

That was my first thought but when it happens it always the entire bank (not the whole car), never a single cylinder.
So I am looking at things that might be bank specific.

 

So I swapped the solenoids last night to see I can check if the problem moves from one bank to the other. I also cleaned them out with MAF cleaner to see if maybe one was dirty and sticking.

Anyway I put everything back together and tried to reproduce the error to no avail.
Not had a chance to drive it yet but in the garage its smooth, except at idle where its pretty rough.

However before I cleared the codes I realized that DFE had thrown another code P0153 which is the bank 2 O2 sensor sending weird signals. That would imply that the O2 sensor is acting up.

Tonight I will try to look at the voltages in PIWIS and see if anything unusual graphing the voltage of the loop O2 sensor. I assume the two should stay similar to each other.

 

So I ran it again and now the problem seems to have gone away, even the rough idle

The only thing that really changed is that I swapped the solenoids between bank 1 and 2 and the temp.
We had a cold spell where things got below 40F (well... CA) but that is over.

Wondering if the solenoid is sticking when things get cold enough?
Or maybe the O2 sensor? Maybe the heating element in the O2 sensor is not quite up to the task?

 

And now I have good news.

So I went for a long drive and once the car was up to temp or something it started struggling again.
It was having a ton of misfires at idle. Restarting the engine (like Wndows 95) seemed to fix the issue for a minute or two.
But eventually the idle would struggle or even get consistent misfires while driving. It was actually worrisome getting home.

Once I checked the OBD codes they have moved to cylinder 1-3. So I am pretty confident that at least solenoid is not behaving correctly.
Seems likely that the solenoid is not always correctly adjusting the cam timing. The DME expects a certain volume of air but if the intake is not lengthened, the mixture would be rich, and would maybe cause misfires.

Also the Bank 2 is throwing a code for running lean, which I assume is the DME trim re-adjusting after running rich for so long.

The blocker is that I ordered a solenoid but it won't be available until Monday. Oh well. We'll see once I can replace the solenoid.

 

So this looks like it fixed the issue. I replaced the solenoid, cleared the fault codes and reset the adaptations for mixture on the DME. Everything seems to be running now.

---
If I understand the workshop manual, the oil pressure comes in from the end of the solenoid, pushing it out of the cam shaft cover. When activated it allows that pressure through the solenoid to adjust the cam (I assume this is lift since the other solenoid is closer to the timing chain up front?). So I applied 12v and it engages. I have to push from the end to make it close. I assume that oil pressure would do that (maybe?).

However compared to the new one, the sealing bracket is loose. Loose enough that it would allow the part to move 1-2mm out under oil pressure. That *might* be enough to misalign the pressure holes and throw off the lift (or timing). The though is, I swapped sides, I pushed it all the way in. That fixed the issue. *Then* I get the oil pressure high enough to push the solenoid back out again and it barely worked for the remainder of the ride home.

---

But this is of course the simpler explanation is there is dirt in the solenoid and now it's stuck or something. It's interesting that the first time it happened the car was sitting idle for 6 weeks. It failed in the first few minutes, the a restart fixed the issue.

 

Oh and thank you Petza914 for all your advice...much appreciated.