Cam only car getting 10MPG, part throttle tune needs work.

[T34418L31]

Quick car rundown -

1997 BMW Z3 with a 2001 LS1/T56 swap, 64xxx when I bought the motor, assume 65xxx now. Car weighs 2940 on a full tank.

Relevant mods -

Lunati VooDoo cam, 232/238, .598/.601 on a 113
4.11 rear end (with proper adjustments made for speedo)
Aeromotive fuel pressure regulator set to whatever stock is (56ish, IIRC)
1 7/8" longtube headers, 3" collectors with O2 bungs in the collectors, 3" duals merging into a 3.5" "cat-back"
EFILive for tuning
All emissions pieces/associated vac lines removed/cap'd/turned off where applicable

The motor is healthy, and the full-throttle tune is pretty well done (car made 389rwhp and 35xrwtq on a Mustang dyno, maxing out the injectors) but the part-throttle tune is proving to be more of a pain.

The car, when pulling from a stop, tends to sputter a bit, cleans up, and goes, but is getting roughly 10MPG, calculated by gallons used to fill up vs distance covered.

The car had not thrown any CEL's for a while, but last night threw codes for P1133 and P1153, H02S Insufficient Switching Bank 1/2 Sensor 1. (It was throwing a few other seemingly non-relevant codes such as "fuel level sensor high voltage" and "vehicle speed output circuit" - things that are more likely to just be swap issues.)

My understanding of the two O2 sensor codes is that it's not necessarily a problem, unless it's never switching. My logs are showing switch points, but at idle it drops and holds at 0.

My fuel trims were contained enough that I can't imagine it would cause 10MPG conditions (I think at worst they were at -8.6%, which I know is slightly high, but still at the point where the PCM can contain it.)

If someone could take a look at the log and tune and offer any advice, it would be very much appreciated. I understand that it's a cam'd V8, but I refuse to believe I'm getting worse gas mileage than (much heavier) cam'd F-bodys

[joecar]

Hi T34418L31,

Your HO2Sx1 voltages show that the front NBO2 sensors are not switching correctly...

correct switching is 3-5 times per second (i.e. 0.3 second cycles, or shorter) with swings exceeding 300 mV (going below) and 600 mV (going above);

your HO2Sx1 voltages are wrong in switching rate and amplitude swing.

It seems to me that your NBO2's are bad... there might be other things going on too.

Do you have a wideband...?

[Taz]

+1 on Joecar's comments .... replace both NBO2s. I recommend using "quality" replacement O2 sensors. I have tried "cost effective" O2 sensors in the past - rarely was this a worthwhile endeavour.

Just replaced the NBO2s on my shop truck (stock 4.8 L / A4) - improved fuel efficiency about 3 MPG on the highway.


Regards,
Taz

[T34418L31]

Thanks for the replies -

the interesting thing, to me at least, is that the car is still entering closed loop. My (possibly/probably incorrect) understanding of O2 sensors is that, if they fail, the PCM will default to open loop mode. Is this not the case?

I do have a wideband, but it's been on the fritz. Hopefully I'll have it working again today.

Out of curiosity, though, what could grenade both O2's at the same time? It seems strange that both would go at once.

Thanks!

Jesse

[Taz]

Hello Jesse,

O2 sensors simply get old / wear out over time. I assume the precious metal in the sensor is consumed / reduced by prolonged reaction with the exhaust stream oxygen.

The ones I mentioned replacing in the above thread had not set any DTC codes, and were still switching from Rich / Lean. The GM tune allows for a fairly large range of values before setting a DTC - regarding the O2 sensors. No DTC does not necessarily mean the sensors are working well, it means they haven't completely died yet.

In your case, there were some O2 DTCs set - leads me to believe your sensors are beyond their service life (though not completely nonfunctional yet).


Regards,
Taz

[T34418L31]

Some updates -

Through this or that, the speedo is off a bit. My office is 11.7 miles from my house as per Google maps and closely verified by my DD stock Honda. The BMW, however, was showing 8.0 miles, or ~32% difference. This, multiplied by the 10.8MPG the car was getting, works out to 14MPG.

I replaced the O2 sensors, and the sputtering of the car cleared up, but I have not checked for CEL's yet. I also got the wideband working, and it's showing 14.3-15.3 at idle, and similar during cruise. At no point is the car super rich (under accel it'll dip a bit into the 12's, etc, but nothing that seems out of place.)

Since the replacement of the sensors, I drove the car a tank (99 miles, 12.X gallons) and got 8MPG, or 10.5 "corrected." The car runs smooth, pulls good, checks out on the wideband, and I have no idea why it's getting such horrible MPG.

Edit: Something else to add - the car only has front O2's, the rears have been turned off. I checked for CEL's and found 8 (two of which were repeats.)

The driver's side front O2 (P0135), P0300 (random misfire,) two swap-related non-issue codes, and P1133 & P1153 again. The O2 sensors I installed were used but working sensors from the same place I picked up my dropout from. The driver's side O2 harness has been extended, by me, to work with the longtubes. I did this on both of the O2 sensors, not the engine harness. I have no problem buying one or both brand new O2 sensors, but I'd prefer to not spend $150 for it to not be the solution.

[joecar]

Check the misfire counters using the scantool... if you get sufficient misfire counts, then that is what is killing your MPG.

[5.7ute]

If you extended the O2 sensor wires by soldering in a new section of wire this will lead to your problem. The wiring needs to breathe as this is where the sensor gets its reference air sample from.

[T34418L31]

The first sensor I extended I did using solder. The second time around, with the second sensor, I extended just by twisting wires and securing with heat-shrink/tape.

I'm tempted to go pick up two brand new O2 sensors and see if there's any improvement.

[joecar]

+1 what 5.7ute said, this is what the Service Manual says.