2 Engine Banks have nearly fixed opposite fuel trims

Model: Mazda MPV (Europe/General market)
Engine: GY 2.5L V6
Generation: 1999 MY
Transmission gear: Automatic

The vehicle is NOT OBDII compliant.

There are no DTCs in memory.

Complaints: Fuel efficiency, cruise control, mild overheating at idle which goes away when engine is revved

History: head gaskets done two months ago apparently due to over heating

Other diagnosis: The heater elements in both O2 sensors seem to be defective as it takes up to 10 minutes after a cold start for them to start switching. Revving the engine makes them start switching faster. ECU is supplying voltage on the heater circuit and ground is good. Front O2 also seems a bit lazy sometimes.

So the thing that's freaking me out here is this:



At idle, LTFT1 stays at about 0%, and LTFT2 at about -10%, with a little variation. However once rpms are raised they go to +14.82% and -16.41% respectively and stay at exactly those numbers, regardless of how high I raised the RPMs. The short terms are all over the place.

The positive fuel trim at high rpms on bank 1 made me suspect a dirty MAF so I pulled it and it was extremely dirty, but after cleaning it there was no change. I also suspect maybe a bad fuel pressure regulator on bank 2, but these fixed opposing long term fuel trim numbers are confusing the heck out of me and I kinda want to figure that out before pulling things apart looking for a fuel system issue.

Could the injector harness be wrongly fitted swapping banks 1 & 2 ?

Reversed O2 connectors?

Disconnect what you think is Bank 1 Sensor 1 and make sure that PID goes dead to verify.

Andy.MacFadyen wrote: Could the injector harness be wrongly fitted swapping banks 1 & 2 ?

What makes you think that? Have you seen something like this before?

I'm wondering to myself how I could even check something like that....

MerleRezo56 wrote: Reversed O2 connectors?

Disconnect what you think is Bank 1 Sensor 1 and make sure that PID goes dead to verify.

I'm pretty sure that's not the case, just because before the bank 2 O2 starts switching the bank 2 STFT is 0. You can see here:



However, I'll take a look at that video cause it looks interesting.

Thanks!

Paul mentions in the video that he's also seen this problem caused by timing issues. Do you know if he has a video on that cause?

OK, I've got some more definitive results. Took a VE measurement and got 83.64%. Calculated load PID was 71.24%.

Max Torque for this engine is at 5000 rpm, and I got the above numbers at 4962 rpm, 88.99 g/s, 38*C IAT, 92 kPa at 754 meters above sea level.

Intake vacuum is also weak at 15 inHg and it only gets up to 19 inHg on a snap.

The question is, is this enough evidence to tell the guy who did the heads to pull off a bunch of stuff to check the timing?

With the engine warm you should try testing the response of the fuel trims and downstream O2 sensor to pulling the output of each of upstream O2 sensors rich and lean in turn. If you see major change in fuel trims on the opposite bank it points to cross wiring issue.

SailorBob wrote: OK, I've got some more definitive results. Took a VE measurement and got 83.64%. Calculated load PID was 71.24%.

Max Torque for this engine is at 5000 rpm, and I got the above numbers at 4962 rpm, 88.99 g/s, 38*C IAT, 92 kPa at 754 meters above sea level.

Intake vacuum is also weak at 15 inHg and it only gets up to 19 inHg on a snap.

The question is, is this enough evidence to tell the guy who did the heads to pull off a bunch of stuff to check the timing?

83% VE isn't horrible, but 71% Load is a bit low. :huh:

A plugged cat could do the same thing. If it's the same 2.5L as in the US market, they're manifold converters that fail all the time. If the engine was burning coolant prior to the head gasket repair, then that certainly wouldn't help.

I'd suggest a relative compression test next, if you haven't done that already. That'd stand a decent chance of showing a timing problem. From there, you could do some in-cylinder testing for a plugged cat or timing issue. Since bank one is showing lean, I'd suspect a restriction on bank two.

Two main reasons for opposite fuel tims:

1 - On V-engines with dedicated catalitic converters and one air metering device for both banks an exhaust restriction like a melted cat will cause this.

2 - jumped timing on one bank will cause this as well.

With regards to point 2 I believe Paul might have referred to CAM/VAVE timing and not to ignition timing.

OK, so things get more interesting. The guy had the water pump replaced while on vacation in a city 2 hours drive away, and in the process the guy played around with some of the vacuum hoses apparently.

I found a solenoid missing a vacuum hose, and I couldn't find a dangling one, so I just put some electrical tape over the nipple in the mean time and took some measurements. Vacuum at idle is still about 14.9 inHg, and at around 2000-2500 rpms goes up to about 17.5 inHg.

So now the fuel trims situation has changed significantly:



At idle, bank 1 FT is around 0%, and bank 2 ( the front bank ) is about -10%. Once it gets to about 2500 rpm LTFT1 stays around 0%, with STFT1 is all over the place. LTFT2 goes to -16.41 and never moves from that number regardless of rpm. STFT2 goes as low as -20%.

What's even more interesting is that at WOT bank 1 goes full lean, while bank 2 goes full rich, which can be seen in the above graph.

What I'm wondering is if there are separate fuel pressure regulators for each bank? I'm thinking maybe the vacuum hose for the bank 2 fuel pressure regulator isn't attached, but I think it's all under the intake manifold and I didn't want to start pulling it apart.

It also strange that the fuel trims no longer go in opposite directions.

There's some other strange stuff going on too. The engine is idling at around 850 rpm, while the ECU is requesting 750. The ECU is keeping the IAC at only 15%, and the spark advance around -2.5 degrees.

Another thing that stood out was that the bank 2 lambda is oscillating twice as fast as the bank 1 lambda:



Lot's of weird stuff going on.

The solenoid with the missing vacuum hose was what Mazda calls the PRC, Pressure Regulator Control solenoid. Told the guy about, don't know if he's done anything about it. But that explains the negative fuel trims on the front bank...