I'm putting this out there because I know some of y'all are/have been professional mechanics and maybe someone can give me an idea. First off I'm an ASE certified Technician(all A series tests except Automatic Transmissions) I use a Snap-On Modis scanner that was updated about 6 days ago. And a long list of other fine tools to do electrical diagnostics. Which is and has been my primary job for more than a decade. But we all know you can get a problem that you get so frustrated with you can overlook something and asking someone else is the best way to go. So here it is. 2003 Expedition vin W 4.6 2v with 86189 miles. It is setting a p2198 HEGO B2S1 stuck rich and no other codes. There are no TSBs that I can find on this and the computer has the proper(as far as I can tell) programming.There are no aftermarket devices installed. Freeze frame says it occurs after warm-up, light load, low vehicle speed. Every time I test it the O2 sensor in question switches just fine. Customer takes it and 3 days later the codes resets. There are no misfires and the plugs, boots, and #6 and #7 coils are new(replaced for misfires about 500 miles ago). My first test revealed that the front O2 sensors were getting slow. The customer ok'd replacement. This had a definite improvement in idle quality while in drive but the code came back. It had a noise at start-up. Tore into it and found a loose right side timing chain and worn guide. I replaced both chains, all 4 guides, both tensioners, both cam gears, and the crank gear. I happened to have 4 new injectors for another job(also an '03 vin w motor), so I installed them on bank 2, just to see. That didn't do it either. I have checked the lifters and cam lobes, all good. Compression is good. Fuel pressure and volume are spot on.The harness is good all the way back to the computer. The o2 sensors respond to air and propane so the computer is seeing the sensors properly. O2 heater circuits are operating properly. It runs strong for a 4.6 throughout the entire RPM range. I get a BARO hertz reading of 155(152-153 is normal around here) but, it achieves this with other MAF sensors. The LTF trims (both 1 and 2) are normal(-2 to 2) with various loads and engine speeds . Short term trims seem to bounce around a little more than normal but don't get out side the -10 to 10 range except for idle in gear. They drop to -13 to -14 and hold steady. What am I overlooking? Throw ideas/opinion at me please. I'll try anything. Well.....My boss says I can't put it on the train tracks and let train fix it, so, almost anything.
Well you seem to have all the bases covered. Everything that popped into my head you mentioned in your post. I'm a Chrysler dealer tech and do drivability all day so I feel for ya. I'll think about this some more when I get to work but its one of those problem vehicles we get stuck with now and then. Hope your getting paid for your time.
Did you do a pin pull test on all the female connectors in the circuit(s) of question? Just take a replacement male connector of the correct type, insert and remove it into the female pins of the connectors, make sure they have significant 'drag'. It tests for a possible intermittent open. Replace any that have weak or no drag. That has bit me before on stuff like this. EDIT: I forgot, also check the EVAP system. Make sure you are not getting fuel vapor when you are not supposed to, creating a richer than expected situation.
The first thing I would do is clear KAM to reset the fuel tables and see what it does when starting fresh. If your total fuel correction is more than +/- 15% then you are rich(-) or lean (+). P2198 is for bank 1, front sensor(upstream), which on a Ford is the passenger side of the engine. (sorry if that is obvious, just making sure we're on the same page) That code is generated when the system is correcting for a rich condition that last for a predetermined amount of time(usually more that 5-10sec). It sounds to me like your O2 signal voltage is staying high, possibly intermittently, long enough for the code to generate. I have had wiring issues with some of those on the RH back of the engine where the loom comes around the head and over the valve cover. In that case, the wire was broken inside the harness and was intermittently setting a P2195(just like yours except lean). The only other thing I can remember seeing that caused something like that was a harness that had not been properly put back into its stays on the engine and had fallen down onto the manifold, shorting the O2 heater circuit wire to the signal wire, producing an intermittent 7 volts on the signal wire. I have also seen the heaters on them short to signal as well but you have new ones installed, so other than the wiring issues or a PCM failure, or fuel contamination(some additives are conductive), I think you're going to have to catch this thing in the act. If it's indicating that rich for that length of time, you should be able to recreate it. I would look for the normal switching of the O2's to either stop or bias high and you want to make sure it is just one bank of the engine that the concern actually exists on. If both banks are trending rich then the problem is engine wide and not limited to one bank even if the code is just for one. The other side could be just below the threshold but still be richer than normal. Don't overlook the EVAP system and PCV system for rich concerns. Fuel pressure needs to be verified as well(48-70PSI). One other thing to look at is the fuel damper/regulator on the driver's side fuel rail. Unplug the vacuum line while it is running and see if fuel is leaking past the diaphragm. If you have access to ALLDATA, the entire pinpoint test should help you isolate it, but it does jump back and forth between diagnosis for lean and rich conditions. Good luck, I know the frustration all too well.
EVAP, EGR, and fuel regulator are all good. The connector drag test is a really good idea and gives me the thought that maybe an intermittent connection causes the engine to correct rich because of a false lean and when it comes back the O2 says way too rich for long enough to set the code but not long enough for the freeze frame data to reflect the problem. I checked the harness all the way back to the computer. Looked physically fine and ohm tested each wire for cross feed even with wiggling the harness. The Modis scanner pointed out that the harness is known for problems behind the right valve cover and I looked at that area quite closely. Fuel contamination is sometimes a concern around here but the problem has happened over several tanks of gas. Fuel pressure is a constant 51 PSI under any load condition I can test with the gauge connected. It also passes the fast and slow leak down tests. The problem does seem to be engine wide. With it warm, in gear, at idle both banks show a short term trim(after KAM reset) of around -13%. But the lack of P0172 and P0175 codes and only a p2198 and not a p2197 is baffling. It has to be such a short period of time that it happens it is almost impossible to catch. I think I need to build a break-out style connector for both o2 sensors so I can record the readings as I drive with the Modis lab scope. Thanks for ideas. Gives me something to look at tomorrow.
Finally caught the problem on the Vantage Pro. The heater circuit for both S1 sensors was cutting out at random but it was so fast it never set any other codes. I traced the problem all the way back to the computer driver for the heaters. ECM and reprogramming repaired it. Thanks for the very helpful ideas.
Good stuff. So much easier when the problem actually occurs for us. Every day I have to explain to customers if it doesnt act up for me and no faults.... well you guys no the story.
