1. Attachments are working again! Check out this thread for more details and to report any other bugs.

O2 sensor #1Upstream Sensor?

Discussion in 'Gen 2 Prius Technical Discussion' started by tkc100, Jul 12, 2024 at 10:31 PM.

  1. tkc100

    tkc100 Junior Member

    Joined:
    Jun 4, 2016
    45
    10
    0
    Location:
    Arizona
    Vehicle:
    2007 Prius
    Model:
    Two
    First, before I get into anything else let me express my sincere gratitude for the support this community has offered me.
    If you have seen or followed any of my other posts, you know that I am stumbling my way through an education.
    I have owned 4 Priuses but this is the first one that I have attempted to learn its inner workings. My years of working as a mechanic/technician was before OBD was a thing. So, I am learning (enjoying) about a computer on four wheels and all of the information the ECU is capable of imparting.
    I learned some time ago that it is best to only address one subject at a time and ideally only one question.
    What follows is only one subject but as you can clearly see I am struggling to understand the upstream O2 sensor.

    There are four separate SAE PIDs supported by my Prius referring to what I am assuming is the sensor upstream from the Catalytic Converter.
    1. Fuel/Air commanded equivalence ratio
    2. O2 Sensor lambda (Bank 1, Sensor 1)
    3. Short term fuel % trim (Bank 1)
    4. O2 Sensor voltage wide range (Bank 1, Sensor 1)
    5. Catalyst temperature (Bank 1, Sensor 1)
    a. I am guessing that this is a secondary function of the upstream O2 sensor inferred by either the variation in voltage or resistance.
    Questions:
    1. Are all 5 of these PIDs referring to the same sensor?
    2. Fuel/Air commanded equivalence ratio, and O2 Sensor lambda (Bank 1, Sensor 1) are these the same thing?
    a. If so why 2 PIDs
    i. Lambda 1 is a stoichiometry
    ii. X = < 1 rich X = >1 lean
    3. Short term fuel % trim (Bank 1)
    a. + or – 10% normal operating range. The closer to 0 the better.
    4. O2 Sensor voltage wide range (Bank 1, Sensor 1)
    a. >3.0 V injection volume (rich) ?
    b. <3.35 V injection volume (lean) ?
    c. I am not at all sure of these figures. It seems like a pretty narrow operating range for a “wide range” sensor requiring no inline resistance and a very sensitive ECU
    5. Catalyst temperature (Bank 1, Sensor 1)
    a. ≤ 750ºF
    b. What would be considered excessive?
    c. What should the temperature differential be between the upstream and downstream sensor readings for a healthy properly functioning catalyst converter?
    Thanks again, Looking forward to reading you response
     

    Attached Files:

  2. mr_guy_mann

    mr_guy_mann Senior Member

    Joined:
    Sep 3, 2020
    3,454
    1,501
    0
    Location:
    NJ-USA
    Vehicle:
    2006 Prius
    Model:
    Four
    So, let's first look at a normal "narrow band" oxygen (lambda) sensor. It's a zirconium dioxide sensing element (the classic design is a thimble shape) that's coated on both aides with platinum.

    The inside of the thimble is exposed to atmospheric air. The outside is exposed to exhaust gases. Add heat and hydrocarbons to the platinum, and you get a catalytic reaction. When that happens with 2 layers of conductors separated by ZiO2, you have a galvanic cell. The output of that cell is dependent on the difference of O2 between the ambient reference air and the exhaust stream.

    You need heat to have a reaction so they add a resistive heater.

    Now the term "Lambda" refers to stoichiometry or perfect Air-Fuel Ratio, where all of the air reacts with all of the fuel. For 100% straight gasoline, that's 14.7 lbs of air for 1 pound of fuel. That is a Lambda of 1.0, a "lean" ratio is higher than 1.0, while "rich" is less than 1.0.

    The actual AFR by weight will be different for different fuels, but Lambda is always 1.0 when you have a stoichiometric mixture (E85 is about 9.8:1, as ethanol has much less heat energy per volume than gasoline)

    At 14.7:1, an O2 sensor outputs around 450mV. At 14.6:1, an O2 is near 0V ( less than 100mV), and at 14.8:1 it's near 1.0V (over 900mV). The ECM reads the voltage and uses Short Term Fuel Trim to drive the O2 reading back and forth across 450mV.

    That's the limit of this type of sensor - it's only good AT "more than Lambda of 1", or "less than Lambda of 1".

    The upstream sensor on a Prius is a "4-wire" Air-Fuel Ratio sensor. 2 are for the heater, and 2 are for the sensing element. Instead of reading voltage, they "flip" the sensor design around and use current. The ECM maintains a 300mV differential across the sensor. Depending on the O2 concentration in the exhaust gases, the sensor will help or hinder current flow.

    At Lambda of 1.0 it's 0mA, below 1 (rich) the ECM reads negative mA, and above 1 it reads positive. At normal steady throttle, the AF sensor output is typically 0 +/- 0.1mA (100microamps). It will go just over to +2mA on fuel cut.

    So the ECM interprets the AF sensor amperage reading and displays a "upstream O2 sensor Lambda".

    The AF sensor CAN read about 12:1 to 18:1. That lets the engineers program the ECM for something other than "either side of stoichiometry". Maybe they want Lambda of 0.95 under WOT, or 1.05 at light cruise. (5 wire AFR sensors can read 10:1 to 20:1)

    That's the "commanded" Lambda or AFR (or whatever). It's the target. The upstream S1 A/F sensor lets the ECM see if the result hits that target. STFT is the means to make adjustments to stay on target.

    I do not know exactly where they get catalyst temperatures from. It could be interpellated from S1 and S2 heater current readings (increased temperature causes increased resistance = lower current), or it could be algorithms based on RPM and load. There's a bunch of variables that affect what the temperature difference is across the catalyst. If I knew the particulars of how it worked on Toyotas, I might bother to use the data - or not

    Now for whatever reason, Toyota engineers decided to put the AF sensor reading as a "made up" voltage instead of current. 3.3V is Lambda of 1.0, lower V is rich, and higher is lean. Typical in-use range is 1.8V to 3.8V, and 5.0V for fuel cut.

    If you go to generic OBD2 data, you can get AF sensor current values.

    Posted via the PriusChat mobile app.
     
    #2 mr_guy_mann, Jul 13, 2024 at 11:50 PM
    Last edited: Jul 14, 2024 at 12:02 AM
  3. tkc100

    tkc100 Junior Member

    Joined:
    Jun 4, 2016
    45
    10
    0
    Location:
    Arizona
    Vehicle:
    2007 Prius
    Model:
    Two
    mr_guy_mann

    Thank you so much for your well thought out and detailed response. You are helping me with my education, and it is starting to make sense.

    Please correct any of the following that I may have misunderstood.
    (Bank 1, Sensor 1) Upstream before the CAT wide band
    Lambda 1.0 = 0mA > .01 = Rich < 1.0 = Lean
    3.3V = 1.0 > 3.3 V = Rich <3.3 V = Lean.
    All four of these PIDs are referring to the same sensor. Saying the same thing, just in different ways.
    1.Fuel/Air commanded equivalence ratio
    2. O2 Sensor lambda (Bank 1, Sensor 1)
    3. Short term fuel % trim (Bank 1)
    4. O2 Sensor voltage wide range (Bank 1, Sensor 1)

    (Bank 1, Sensor 2) Downstream after the CAT narrow band
    Lambda < 1.0 = Lean > 1.0 = Rich

    However, I do still have one question. I thought the upstream bank 1 sensor 1 was responsible for STFT and sensor 2 was responsible for LTFT.
    Thanks for clarification.
    Where does the LTFT come from?

    P.S. Today I learned about the CAN bus and I now understand how a faulty instrument cluster circuit board could wreak so much havoc in seemingly unrelated functions.
     
  4. mr_guy_mann

    mr_guy_mann Senior Member

    Joined:
    Sep 3, 2020
    3,454
    1,501
    0
    Location:
    NJ-USA
    Vehicle:
    2006 Prius
    Model:
    Four
    Close. 1 is what the ECM wants (is programmed to achieve). 2 and 4 are the AF Sensor reporting what is happening (the result of whatever fuel mixture that just went through the engine). 3 (STFT) is a direct reaction to AFS in order for 2 to "match" 1.

    So STFT is there to keep AFS on the programmed target. LTFT is a "learned" compensation that is there to keep STFT near 0%. Say STFT goes to +10% "for awhile". LTFT will step in (go toward +10%), and STFT will go back toward 0. (ST + LT gives you total FT).

    LTFT has different values for different RPM and load values, and is stored in ECM memory (as long as it has 12V battery power). So LT @ idle is different from LT @ high idle, vs 2500 rpm cruise vs 4000rpm accel, etc.

    AFS voltage - above 3.3V is "lean", below 3.3V is "rich".

    The B1S2 (downstream) O2 sensor is primarily there to monitor catalyst function. The ECM is programmed to adjust fuel in order to "feed" the cat with a slightly oscillating (rich to lean to rich) mixture. The cat needs CO, HC, AND O2 in order to work.

    The cat has cerium in it to "store" the O2, which gets used up to convert the CO and HC to H20 and CO2. So there's "not much" O2 left - the downstream sensor will read (somewhat steady) "about" 700mV when the cat is healthy and lit off. (Lambda doesn't really apply to the downstream reading).

    Caveat - some manufacturers DO use the downstream sensor to adjust fuel control (sometimes, and it is poorly documented). Toyota is one of the earliest to do this, they just do not on the Gen2 Prius.

    Posted via the PriusChat mobile app.
     
    #4 mr_guy_mann, Jul 15, 2024 at 8:37 PM
    Last edited: Jul 15, 2024 at 11:05 PM