ramiles

Since updating the motor on my L98 Vette several months ago have had a frustrating problem getting good O2 sensor data and closed loop. Need your recommendations. Detailed chronology is attached in second post below. The short version is the O2 wiring checked and several different sensors were tired without being able to get normal mV and crosscount response.
I next ran the troubleshooting procedure on SES Code 13 – O2 Sensor. FSM has you ground the O2 sensor and scan the mV and then has a decision point of 200 mV. Above 200 mV the FSM says indicates “faulty ECM connection or ECM” – for less than 200mV FSM says indicates bad sensor. Mine was 253mV. So I checked continuity on the sensor and ground leads from the ECM – they looked good - and then ordered a rebuilt ECM from Ecklers.
Want to make sure I am properly trouble shooting and not missing something else – are there any $6E closed loop parameters, wiring problems, or other gotchas I am missing? The ECM is 21 years old, was removed from the dash to hook up an emulation cable, and has been resting on the passenger floorboard for a while – It could certainly be the problem, and might explain why no sensor or wiring combination seemed to help.
----------
Sep 2006 - Removed AIR system and installed Stainless Works headers, and installed a Caspers heated O2 sensor, with the heater circuit wired into a relay on the driver side fender. Got a BIN from Alvin for those changes and car ran fine. Wasn't data logging then but assume the O2 sensor worked OK and was going into closed loop, as I didn't get any SES lights.
Nov 2006 - Started data logging was using a $32 mask – that template didn't have a closed loop flag, and I didn't know enough to check it with diagnostic mode on the ALDL – so didn’t get any closed loop data points then.
June 2007 - Updated to new heads, camshaft, lifters, TB, big mouth manifold and runners, and injectors. Reinstalled the same O2 sensor - had some roughness and idle problems and it threw a code 13 and a code 45. Was running a BIN from Alvin made for the motor changes. Started more data logging with a $6E and saw that closed loop was intermittent and generally the fuel trim was rich.
July 2007 - Tried to get Alvin to update the BIN – and he requested Datamaster log files. Sent these and after reviewing them, he said there were problems with the O2 sensor. Researched this forum and 3rd Gen DIY PROM and got smarter on NB non linear response, mV ranges, and crosscounts. It was clear that the O2 sensor was lazy and / or erroneous.
August 2007 – Tried various attempts to fix the O2 sensor. Checked the voltages going to the senor – had 12v and gound to the heater circuit, and 170 mV going to the senor. Removed the units and the heater checked good manually. Replaced the O2 Sensor with another a Denso unit, with no improvements, then replaced the sensor and wiring completely with another new Casper’s unit – none of these changes brought a consistent O2 sensor capability.
August 2007 - Then had the car dyno tuned with a wide band sensor – base fuel trim was adjusted and runs well, and we locked out the closed loop by setting a closed loop temp constraint to 300F. Next Tried a GM 3 wire O2 sensor, 25312179 aka AFS-74, and wired it up – reviewed the mV in the data log and could not get regular response to engine / throttle movements. Starts at 451mV and then after a minute or so goes to 253mV or so and stays there.
August 2007 – Troubleshot according to the SES Code 13 procedures in the FSM. The procedure has you confirm open loop and then ground the O2 sensor terminal scan the O2 mV. If less than 200 mV it directs “faulty oxygen sonser connector or sensor”, if 200mV or above it indicates “Check for open oxygen sensor signal CKT 412 or 413. If CKT 412 or 413 OK, it is faulty ECM connection or ECM” . Mine was reading 253mV. Saw ground continuity at D6 from the ECM (sensor signal ground), and good continuity through D7 (sensor signal) . We had pulled the motor for the engine work so I thought maybe there was some wiring issue – wasn’t sure where this D6 engine ground connects on the engine. I looked in the FSM for the ground and saw that it (tan wire CKT413 ) it joins a wire (brown CKT486) from the ESC module but couldn’t verify these on the engine. Ordered new ECM.

JP86SS

iTrader: (1)

Possibly high resistance from the header to the motor ground.
Gaskets insulating the connection and coated bolts not giving connection.
Just throwing some items out there.
Jp

ramiles

Thanks JP - the bolts aren't coated but I'll check that out. Get a new ECM in a day and if that doesn't fix it then its back to square 1.

RFmaster

One way to get around NBO ground return path issues is to go to a four wire O2 AFS-75. I would also check all my ground circuits connections, and lugs. BTW, with four wire I get around 75 mV when I ground my O2 lead at sensor connector. Since you have checked it with WBO I can not thinking of other reasons for NBO being lean.

//RF

RednGold86Z

iTrader: (1)

If the ECU main grounds are bad, the current tries to go through the O2 sensor ground and raises the potential (= some volts show up there). Also, if the wire for the O2 sensor to the ECU is corroded (= medium to high impedance), the O2 bias voltage (450mV) in the ECU will partly make its way into the signal, even when the wire is grounded outside.

ramiles

These are great ideas and inputs - appreciate your time helping me understand how and why the O2 isn't working.
RednGold86Z, want to understand more about your note. After grounding the sensor, I saw 450mV, (the bias voltage you described) in the scanned data logs before starting - once started it went fairly quickly to 255mV and stayed there.
What is the electrical basis for the FSM proceedure checking for above or below 200mV when grounded to troubleshoot to either the sensor or the ECM and its wiring? When looking at the basic schematic below I guess the scanned O2 reading measures something other than potential between the ECM ground line (D6) and the 450mV bias voltage (D7), otherwise wouldn't it stay the same wether it was grounded or not if the sensor was not in the circuit adding or subtracting voltage?
Thanks

RFmaster

The FSW procedure indirectly checks resistance of C413 and C412. The problem is that single wire NBO relies on exhaust manifold (or headers) to provide low resistance path to engine ground (lug). For the most part it achieves this task, but with age and with modifications resistance may increase. Increase in resistance screws up NBO readings.

C412 (ppl wire) has a DC voltage supplied by ECM (450 mv or their about) O2 sensor output, due to integrator switching, is constantly swinging so in grand scheme of things NBO voltage is an AC signal riding on top of DC bias. However for ECM to measure it correctly the return (ground path) must also provide lowest resistance possible.

The way to solve the aforementioned problem is to provide a ground return wire along with signal wire to the NBO like it is done in later OBD-2 cars. These use a heater to keep NBO at operating temperature as well ground return wire to keep circuit resistance to a minimum. Later four wire NBO, such as AFS-75, can be retrofitted into earlier single wire applications (AFS-21).

//RF

ramiles

Update - O2 Closed Loop RESOLVED
Replaced the ECM with a rebuilt from Eckler's and the O2 crosscounts went into what appears to be a normal rythm. I reset the Closed Loop Temp enable (we had used that constant to lock out closed loop), and car went into Closed loop and stayed there. Yahoo!

Appreciate anyone with Datamaster reviewing the file to confirm O2 Closed Loop operation (and any other suggestions). Also what are good parameters for the three Closed Loop enable timers and temps. Thanks!!
DataMaster Log File Link for Review

RFmaster

Thanks for resolution update - enjoy the ride!!!

//RF