Page 2 of 3 FirstFirst 123 LastLast
Results 11 to 20 of 22

Thread: Engine Code P2096 (fuel trim code on Mirage with custom header)

  1. #11
    Still Plays With Cars Loren's Avatar
    Join Date
    May 2018
    Location
    Florida
    Country
    United States
    Posts
    1,225
    Thanks
    324
    Thanked 936 Times in 539 Posts
    Y'all could go back to my review and initial installation of the SC header.

    Yes, it eliminates the pre-cat.

    Yes, it causes this particular code.

    Yes, you can make it happen less frequently with a defouler, but it still happens.

    I quit worrying about it. My CEL is on most of the time, and I occasionally check the codes to see if there's anything new.

    There is a second cat after the one in the (stock) header. The first one "lights off" quicker because it's closer to the combustion chambers, thus reducing start-up emissions. The second one still does most of the work after the engine is warm.

    This particular code doesn't cause any other problems. Doesn't put the car into limp mode or anything. It's just informative, and telling you what you already know. The pre-cat isn't performing well... because it's not there.

    Also, look at the very top of the stock header. That hole is where the stock pre-cat O2 sensor is. The location on the SC header is pretty much the same. Being super close to the combustion chamber means it reacts quicker to the actual running conditions of the engine. That's the one that is used for closed-loop fuel trim.


    Simplify and add lightness.

  2. The Following User Says Thank You to Loren For This Useful Post:

    Top_Fuel (06-29-2020)

  3. #12
    Senior Member Gnestovia's Avatar
    Join Date
    Jul 2018
    Location
    West Virginia
    Country
    United States
    Posts
    114
    Thanks
    4
    Thanked 46 Times in 23 Posts
    This is just an experiment just to see if this works or not. And remember, everything I do is reversible if things go south. I save every little parts bolts and all just incase I have to put the OEM stuff back on even though I dont want to. I made a pigtail extension for the linear air fuel sensor so it can reach. The second sensor will have long enough wires to make it work.

  4. The Following User Says Thank You to Gnestovia For This Useful Post:

    Loren (06-29-2020)

  5. #13
    Senior Member Gnestovia's Avatar
    Join Date
    Jul 2018
    Location
    West Virginia
    Country
    United States
    Posts
    114
    Thanks
    4
    Thanked 46 Times in 23 Posts
    Hang on a second, put a defouler on the first sensor?

  6. #14
    Senior Member Top_Fuel's Avatar
    Join Date
    Feb 2016
    Location
    Ohio
    Country
    United States
    Posts
    3,699
    Thanks
    2,582
    Thanked 2,537 Times in 1,471 Posts
    Quote Originally Posted by Gnestovia View Post
    Hang on a second, put a defouler on the first sensor?
    Start reading here.

        __________________________________________

        click to view fuel log View my fuel log 2015 Mirage ES 1.2 manual: 52.2 mpg (US) ... 22.2 km/L ... 4.5 L/100 km ... 62.6 mpg (Imp)


  7. #15
    Still Plays With Cars Loren's Avatar
    Join Date
    May 2018
    Location
    Florida
    Country
    United States
    Posts
    1,225
    Thanks
    324
    Thanked 936 Times in 539 Posts
    I wouldn't put a defouler on the first one, no. I'd expect that would change how the engine runs, and not for the better. (it would cause that primary sensor to respond slower and with less accuracy)

    The post-catalyst lean indication is either a bad sensor, or related to the missing cat. Missing cat won't affect readings BEFORE the cat.
    Simplify and add lightness.

  8. #16
    Senior Member
    Join Date
    May 2014
    Location
    Country is Europe, state is Germany
    Country
    Germany
    Posts
    1,713
    Thanks
    234
    Thanked 1,158 Times in 670 Posts
    Quote Originally Posted by Loren View Post
    ... The location on the SC header is pretty much the same. Being super close to the combustion chamber means it reacts quicker to the actual running conditions of the engine. That's the one that is used for closed-loop fuel trim.
    Just for understanding:

    The linear air-fuel ratio sensor (oxygen sensor) works practically in real-time, and in the OEM setup it is located where it can distinguish between the exhaust puffs of each cylinder. This is needed for the fuel-air trim of the individual cylinders.
    Consequently, if the linear oxygen sensor would be located where it can only sense the exhaust gas from one particular cylinder, the precise air-fuel trim would only adapt to the injector of that one single cylinder, and not adapt individually to the two others. That creates the situation of the overall trim being off, and that in turn causes the code.
    Last edited by foama; 06-30-2020 at 07:21 AM.

  9. #17
    Senior Member Mitz's Avatar
    Join Date
    Jun 2015
    Location
    Cavite
    Country
    Philippines
    Posts
    510
    Thanks
    31
    Thanked 268 Times in 178 Posts
    Isn't the linear air-fuel ratio sensor located at the exhaust manifold while the oxygen sensor is located after the 1st catalytic converter?

    Name:  AFR ratio sensor.jpg
Views: 451
Size:  93.9 KB

    Name:  Oxygen sensor (rear).jpg
Views: 476
Size:  74.8 KB

  10. #18
    Senior Member
    Join Date
    May 2014
    Location
    Country is Europe, state is Germany
    Country
    Germany
    Posts
    1,713
    Thanks
    234
    Thanked 1,158 Times in 670 Posts
    Quote Originally Posted by Mitz View Post
    Isn't the linear air-fuel ratio sensor located at the exhaust manifold while the oxygen sensor is located after the 1st catalytic converter?

    Yes, exactly!

    The way Mitsubishi placed it allows it to sequentially read the air-fuel-ratio of each cylinder. Thats the whole point!

  11. #19
    Member
    Join Date
    Mar 2015
    Location
    Austin, TX
    Country
    United States
    Posts
    62
    Thanks
    2
    Thanked 13 Times in 11 Posts
    Quote Originally Posted by foama View Post
    Just for understanding:

    The linear air-fuel ratio sensor (oxygen sensor) works practically in real-time, and in the OEM setup it is located where it can distinguish between the exhaust puffs of each cylinder. This is needed for the fuel-air trim of the individual cylinders.
    Consequently, if the linear oxygen sensor would be located where it can only sense the exhaust gas from one particular cylinder, the precise air-fuel trim would only adapt to the injector of that one single cylinder, and not adapt individually to the two others. That creates the situation of the overall trim being off, and that in turn causes the code.
    I'm finding this a very interesting discussion. Foam is asserting that the location of the linear (or wideband) O2 sensor is such that it can use just one sensor to monitor the mixture in all 3 cylinders. I would agree that the location in the exhaust manifold where gas from all 3 cylinders passes by would allow this to be done, but I had not realized that O2 sensors were now fast enough to make this practical.

    To make the arithmetic easier, we could assume a low speed cruise rpm of 3600rpm which gives us 60 crankshaft revolutions/second, which is 30 camshaft revs/sec, and 90 exhaust pulses/sec. This would imply about 11mS of time between exhaust pulses (and about half of this at max rpm).

    I went looking for the response time of linear O2 sensors, and so far have not found any ( a little disappointing). I did read the data sheet for the Bosch LSU4.9, and didn't see frequency response there. I have seen aftermarket tuner information on this sensor suggesting a response time of 10mS, but I don't really consider this as reliable. I would expect that if the ECU and linear sensor were working in this way, then it would need a response time of around 1mS.

    Now for all I know, foama may work professionally in engine calibration and is exactly correct ion this point, so let's just say I'm surprised.

    If the ECM were monitoring the AFR in all 3 cylinders, would the ECU not maintain a fuel trim table for all cylinders? Although I suppose it might well not be able to report out of trim codes for more than 1 "bank" via OBD2.

    Can you post more information, foama?

        __________________________________________

        click to view fuel log View my fuel log 2014 Mirage ES 1.2 manual: 46.0 mpg (US) ... 19.5 km/L ... 5.1 L/100 km ... 55.2 mpg (Imp)


  12. #20
    Senior Member
    Join Date
    May 2014
    Location
    Country is Europe, state is Germany
    Country
    Germany
    Posts
    1,713
    Thanks
    234
    Thanked 1,158 Times in 670 Posts
    Quote Originally Posted by mpaton View Post
    I'm finding this a very interesting discussion....
    Can you post more information, foama?
    To start with, the FSM is taylored to the needs of servicing, not so much for explaining things, although it does provide some info. Mitsubishi is a huge group of organisations that makes just about everything from integrated circuits, airconditioning, jet airplanes, satellites, and everything in between, you name it. The automobile section is tiny in comparison. The R&D in Mitsubishi electronics has huge resources, and the extremely high level of very sophisticated electronics in these cars is one of their best aces.
    Because of being foremost in engine electronics, most details are confidential, and it is hard to get more than general information, they rightfully want to keep their secrets.
    What we definitely know are details taken from the FSM. There is also a russian forum mainly about ETACS that sheds some light on various details. Its like maths. If you now what the equation is about, what results are expected and see most figures, you swiftly get an idea of whats missing.
    We know that the ECM/ECU controlling the engine works with much larger EEPROMS than with other makes. In the ECM of cars made in year 2000 we had maybe a total of a half a dozen points stored in memory, and everything in between was result of sensors for correcting. These worked with lambda sensors and a huge time-lag, in addition to their very limited accuracy. If engine operating conditions were not constant, and that is normally the case, for example you were accellerating or slowing down, computed conditions were miles off from actual, and emissions were bad and economy also.

    In the ECM of these cars we have 8bit technology for the number of stored tables, and it seems there is also 8bit numbers of information sets (256!) stored within each table. We also have electronics capable of recognizing and switching between tables in realtime in milliseconds. This is light years from what was normallity in 2000, and is much more sophisticated than many other makers today. It makes these cars easily pass emissions and yet be economical at the same time. I have tried finding detailed info, but nill success. Here two pics for understanding:


    Name:  sensors.jpg
Views: 348
Size:  51.6 KB



    Name:  MFI .jpg
Views: 346
Size:  82.6 KB


    Last edited by foama; 07-02-2020 at 08:13 AM.

  13. The Following User Says Thank You to foama For This Useful Post:

    inuvik (07-02-2020)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •