I've had my 2007 V8 Vantage Roadster (4.3L) for six months. Now that it's cooler out, the thermostat rarely goes above 1/4 of the gauge so I knew the thermostat was bad. Replaced it and confirmed the rubber seal was escaping, jamming it open. Now it warms up nicely and stays steady at 50% of the gauge. After driving it a day, the CEL came on with P0171 and P0174 codes (lean LTFT in both banks.) I have a VelocityAP tune and airbox delete but otherwise stock with 14,500 miles now. Cleared the codes and watched STFT and LTFT using a bluetooth OBD2 reader and iPhone app. This showed the LTFT start out at 0% and then rose to +3%, with STFT bouncing around -1% to +6% depending on load. After 20 minutes of highway driving, LTFT on bank 2 rose to 5%. 10 minutes later, at much slower speeds in town, it went over 12% and then triggered the CEL with just P0174 this time. Cleared it and back out on the open road, it calmed down to +6%.

Since STFT never got up to +10%, I'm surprised the LTFT went above 12%. Anyone know what this indicates? I have no idea how long the thermostat has been bad, so potentially it's run rich during open-loop for many miles and damaged cats or O2 sensors. Car always runs great.

I'm starting with the easy stuff first. Checked all my hoses that had to be moved during the thermostat change (nothing to report.) Cleaned the throttle body. Next is changing the MAFs. But I'm guessing it's none of these things...

 

did the CEL come back on?

 

Although this is for DB9, some good troubleshooting procedures for fault codes.
https://www.dropbox.com/s/5oi7g4kvv2...-2004.pdf?dl=0

My guess would be a vacuum line or connection is loose letting air in, but that's just a guess.

matt

 

These codes could be due to your Airbox Delete. When you took off your intake to perform the thermostat swap, you probably put them back on at a different angle which put the MAF in a different position causing the lean codes. Check out this thread: https://www.6speedonline.com/forums/...74-15-v8v.html

Basically, the solution is to put some air straighteners in your intake so that your MAF can get a more accurate reading.

 

Today the codes came back. I can see the STFT running -3 to +3 to mostly 0% at steady state, so the STFT is normal. It's just the LTFT that is up, and it's now starting at 10 and occasionally flickering to 7 on the highway. Once I slow down, though, it jumps up to 14+. I'll try to shoot carb cleaner at the hoses to see if the engine ingests it (air leak.) And then go back to the stock ECU tune. Easier stuff. The air straightener thing...it seems kind of snake oil to me. The filters haven't changed shape. They may flow 2% better? And again the STFT looks normal. I would think the ECU uses the MAF for that real-time so wouldn't that also be off? Just looking for a bigger data set - more examples. Getting to the MAF sensors and air filters on these cars is a royal PITA. The inner fender liner should have been designed as two pieces.

 

Counter to their name, MAFs can't actually read the mass airflow coming into the engine, they rely on knowing the diameter of the tubing and then read the velocity of the air inside the pipe and calculate the mass of incoming air based on that. In the typical setup, a lot of the turbulence in the air is ironed out by the longer tubing prior to the MAFs, but you've gone and removed them with the airbox delete. If you now have a non-uniform velocity profile coming into your intake, the MAF may be reading a higher velocity than the average and reporting to the ECU, which makes it think that more air is coming in than actually is. Your O2 sensors are then analyzing the exhaust and reporting back that it's lean, throwing a code. An air straightener will attempt to clean up that intake velocity profile slightly to make your MAF readings more accurate.

 

When I got my AM it was occasonally throwing the too rich \ to lean bank codes, in addition to complaining about the evap pump. It also had a very weird occasional stall under very hard braking. After getting it back home, changing the evap pump and doing some routine maintenance I discovered that the air intake to manifold body was actually loose and slightly ajar. After I cleaned the throttle body and plate and reconnected it the stall completely went away, but, now it likes to throw the too rich \ too lean codes a lot more frequently, and occasionally the idle gets really rough temporarily. After digging around in the AM manual last night I found a blurb about how the fuel trim tables actually have a base and then a learning adjustment table ( or something like that, it was late) .
In my case I suspect that the learning adjustment table completely adapted itself to the leaking intake condtion and now that I have fixed that the fuel trim tables are completely out of wack.
What the AM manual states ( its long and I cannot copy it directly )
"As fuel system components age or otherwise change .. The adaptive fuel strategy learns deviations while running closed loop. These deviations are stored as long term trim corrections in an RPM / Load table OR as a air mass function. As components age ... trim levels reach a limit and are no longer allowed to compensate with a calibratable deviation.... and CEL...

Now basically the way I am interpreting this, in relation to my situation is. The engine was run and learned to interpret the air intake leak as normal and built a correction into the system. Now that I have fixed it the table mappings are completely out of wack and it cannot add enough correction to get the fuel trim in line.
So my attempted solution to this is going to be to disconnect the battery ( at the battery not the disconnect switch ) to clear the Keep Alive Memory and then do some driving along with the misfire detection learning procedure to build new tables.

Will this work? Im not sure, but if it doesnt I will still get to clean the battery terminals!

 

From my readings last night, ( see above ) the MAF sensors actually take in to account a variety of situations including strong cross winds while driving ( manual specifically discusses this situation ) , so I would expect the system to adapt for non - uniform velocities between sensors.

 

From the symptoms, I would look more for a vacuum leak. The vacuum leak will pull more air into the engine than the MAF is measuring, so the correct amount of fuel (for the MAF) is actually lean for the actual air mass going through the engine.
You note the trim goes way up when coming to a lower throttle setting after a highway run - since the throttle body is a restriction at idle, the engine will tend to pull more air in if there is a leak somewhere. I would very carefully check for damaged gaskets, broken or disconnected vacuum lines. Old school way is to spray brake clean at any joints or gaskets while running, and listen for an RPM increase. Or, a smoke test might help find it.

Matt

 

I think you misunderstood, you're correct in that they can adapt to non-uniform flows between the two branches, and if this happens to a certain extent it would throw a P010a or P010f. But a non-uniform velocity profile within the pipe, there's just no way for the MAF to know if that exists or not.

 

I would do a smoke test

 

I have to admit that I went thru this too (same codes, changed thermostat). Made the smoke machine out of a jar, cardboard and charcoal. Smoke billowed out of one of the hoses I missed. Can't believe I'm so blind..

 

So far:
Propane all around the engine looking for an air leak. Nothing found.
New MAF sensors - no change
Cleaned Throttle body/replaced gasket - no change
Back to original ECU tune - no change
Disconnected battery (cleared fuel trims) twice - no change
New secondary O2 sensors - no change

I'll try the smoke test for finding an air leak, then re-install the stock airboxes. Last ditch effort (and most expensive) is the cats. Interesting to see the primary O2 sensors were replaced a few years ago at only 9k miles.

 

Had the same issue when I was doing all the maintenance. I introduced vac leak through the throttle body, even though I was sure I had done it right. A smoke test will ID quickly.

Use OBD read first to:
Get readings off Short and Long Fuel Trim under different load. (-5 to 5%)
Get readings off MAF (~7g/s)
Get readings off Fuel Pressure (~380kPA or 58 PSI)

This will isolate or eliminate possible areas. For the smoke test, I used a cheap 1.00 cigar and a fuel transfer pump from harbor freights and introduce the smoke into the intake manifold by removing the line from the PCV valve connector. You dont need to be a smoker to use this method, just light the cigar and insert into the pump and start pumping.

 

it would be helpful to know if the entire black rubber seal on the thermostat was recovered when it was replaced. IOW you might have a blockage somewhere in the engine from a piece of that thermostat rubber seal.

...just a wild guess from a mechanically ignorant owner...