Interesting problem - P0170, O2 sensor, water?

[blmqzjc]

We have a 2001 1.8T with the AUG engine. About a week ago, wife was driving and got a cell - P0170 (fuel trim malfunction). Did some research around here on that (so I'd be educated when talking to someone about it), and made an appt with the dealer (was getting an 80K svc anyway).

Dealer ruled out the vacuum lines, check valves, and the MAF, so the O2 sensor was next ro check. Sounds like the same protocol most on PW would recommend. In the meantime, many more codes were being thrown (random misfires, etc), car was running like crap, no power on acceleration, etc. Dealer said all these new codes were probably a "snowball" effect off the original P0170.

Dealer removed the rear 02 sensor (behind cat), and said it was definitely toast, so they replaced it. When car is first started, it idles and runs great. Only when warmed up does it start spitting and sputtering. This continued even with the new 02 sensor in place.

Dealer removed the "brand new plugs" and put some kind of scope into the engine. Said there was evidence of "rust" on the pistons (said it looked like mushrooms - weird). However, there is no evidence of water/coolant getting in (except for the rust), head gasket apparently is ok, coolant level is fine, no white smoke out the tail, etc. In addition the last two Blackstone UOA's show no antifreeze, no water, insolubles at .2, and fuel % < .5, all very good. So where did the rust come from? Could it be a contaminated tank of gas? Last tank fillup was Sunoco, about 50 miles before all this started up.

Dealer pulled the fuel filter and said it looked pretty bad, so they are replacing that. They are also doing a fuel injection service/cleaning to see if that will help, since they are concerned about contamination in the fuel rails, injectors, etc. If not, they are suggesting dropping the fuel tank, and thoroughly cleaning it out. I guess this is all assuming that I just got a lousy tank of gas, and if so I guess the measures taken will remedy that.

I'm already into this for over a grand (incl. the 80k svc), and could be looking at a lot more. I have confidence in the dealer, and they are really trying to keep the costs down, which I appreciate.

My question to you guys - any of you ever seen anything like this, or have any ideas? I was sure when it first started that it was probably a vac line, or a dirty/bad MAF, or an O2 sensor at worst. Well the rear O2 was bad (as described above), but that appears to be effect and not the cause. Dealer said the bad O2 was "white" - not sure what that means. I fear the new O2 that was just installed will crap out in a few hundred miles if the root cause is not uncovered. But car is really not driveable safely right now anyway.

All ideas appreciated - TIA.

[Sharky]

It all sounds like horse shit to me. To see what the deal is, I would want to see how fast the rear O2 portion of the readiness code sets. Also the front.

Gut says you need a MAF and that when they bore scoped your cylinders they didn't know what the fuck they were looking at anyway.

[blmqzjc]

It all sounds like horse shit to me. To see what the deal is, I would want to see how fast the rear O2 portion of the readiness code sets. Also the front.

Gut says you need a MAF and that when they bore scoped your cylinders they didn't know what the fuck they were looking at anyway.

Thanks for the response, Sharky. Educate me for a moment. Dealer said the MAF was OK, but let's assume they were wrong. Is it possible or likely that a bad MAF would crap out the rear O2 sensor? Dealer said the O2 was definitely toast. Also, one of the most interesting things in this whole scenario is that the car runs fine until warmed up, then starts acting up. If it's stone cold, it can take 4 or 5 mins till the problem occurs. If engine is relatively warm, it can start spitting and sputtering in less than 30 seconds. Would a bad MAF act like this (not causing a prob till warmed up), or would it be noticeably bad as soon as you started up the engine?

Thanks again.

[Sharky]

What happens is the MAF being bad will cause the P0170 or ones like it. People stupidly assume since the O2 sensors measure mixture, it is a O2 sensor problem. O2 sensors seeing something wrong is pointing at the effect, not the cause.

What is even stupider is they replaced the REAR O2. All this sensor does is measure how well the cat is working, and while it is true the cat probably isn't working too well since the car is running like ass, the O2 has nothing to do with this at all. The sensors are always white like that because they work at 600 gillion degrees.

The car starts and runs fine cold because the O2 sensor input to the ECM doesn't figure into much at this point. This is called open loop operation. When things warm up, the feedback from the O2 becomes more important. When the O2 sees weirdness, it tells the ECM, and the ECM makes adjustments to try to make what hits the O2 more appropriate. The ECM is not able to get the desired effect at the O2, and it sets fuel trim codes. And ALL of what the ECM is looking for, and what it expects to see at the O2 sensors, is based on what the MAF tells it. If the MAF is lying, even wildly so, the ECM never suspects this at all. This is why you never hear of MAF codes, just the aftermath on the other end of the engine, there at the O2 sensor.

The MAF on your car is highly suspect. It sure as hell isn't the REAR O2, good God...

[kenblasko]

Try this, disconnect the MAF sensor so that the ECM is forced to use defaut values. If the car runs better with the MAF disconnected when hot, it's probably a bad MAF. I know this works for the V6 but not sure how the 1.8t will perform with it disconnected. If you have a tip, tip shifting may be harsher with the MAF disconnected.

The only other thing to do is to find someone with a VAG-COM and monitor certain blocks of data with the engine running to see if anything looks unusual when the car is acting up, although you would think the dealer should have already done this.