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

P0138 testing procedure

Discussion in 'Gen 2 Prius Care, Maintenance and Troubleshooting' started by Kamaloha, May 3, 2021.

  1. Kamaloha

    Kamaloha Junior Member

    Joined:
    Jan 4, 2008
    9
    4
    0
    Location:
    New Hampshire
    Vehicle:
    2006 Prius
    I have an '06 Prius 166K mi that is showing a single P0138 code. I've replaced the downstream O2 sensor without effect. I got a simple OBD2 scanner that allows me to view the freeze frame and the live data. It shows the downstream sensor voltage as stabilizing at 0.88 v (rich) when the engine is warm, with the O2 1 sensor voltage varying around 3.2-3.5v. The freeze frame shows that at the time the code was thrown the voltage was 3.5v (lean) with the #2 voltage at 0.83v (rich). I suspect the downstream sensor reading is valid as my mileage has dropped so it is likely that it is running rich. All engine components including the catalytic are original with the exception of the new downstream sensor I just installed.

    I've read through all the threads on the forum concerning this problem. The most applicable was the one by Jgurley but there wasn't any ultimate finding listed at the end of his thread.

    Reading the Toyota diagnostic procedure it appears that the next thing to do would be to use the Techstream software to force the engine rich and lean and see what the sensor voltage responses are.

    So here are my questions:

    1) Is this the correct course of action?
    2) What is the best way to obtain and use Techstream? Hard cable vs wifi? If you use it, where did you get yours?
    3) Would it be better at this point to just take it to the dealer?
    4) I have a hunch that the likely course of action will end up replacing the MAF sensor. Can you point me to the replacement procedure for it?

    Charlie
     
  2. SFO

    SFO Senior Member

    Joined:
    Feb 7, 2017
    5,286
    4,225
    0
    Location:
    Northern California
    Vehicle:
    2007 Prius
    Model:
    N/A
    Following the workup step by step is a good course of action. Before the cable arrives, you might first try steps 5, 6 and possibly 7.
    Many 'mini-vci' cables will ship with techstream.
    Techstream can also be downloaded online.
    Pretty sure you can only use a "mini-vci" cable with techstream.
    Bought my mini-vci cable on fleabay for about 15usd.
    They will charge you for the diagnostic, which can be applied to the repair quote. Unlikely if the diagnostic or the repair would be affordable.
    Rarely does an expensive MAF sensor fail, best not to touch it unless it sets a DTC. If keen, you could clean it with MAF cleaner and disconnect the 12v so the fuel mapping is reset as well. Cleaning MAF Sensor | PriusChat

    If you want to throw parts at the problem, then replacing the upstream A/F sensor would be my first choice. Many will buy OEM/Denso.

    For future reference, here is the workup for DTC P0138 : https://share.qclt.com/%E4%B8%B0%E7%94%B0%E6%99%AE%E7%91%9E%E6%96%AF%E5%8E%9F%E5%8E%82%E8%8B%B1%E6%96%87%E6%89%8B%E5%86%8Cpdf%E6%A0%BC%E5%BC%8F/repair%20manual/04pruisr/05/2054m/cip0136.pdf

    On a side note, @PriusCamper tested a cheap aftermarket sensor recently, now I'm curious how that worked out.
     
    #2 SFO, May 3, 2021
    Last edited: May 3, 2021
    Raytheeagle likes this.
  3. mr_guy_mann

    mr_guy_mann Senior Member

    Joined:
    Sep 3, 2020
    3,206
    1,339
    0
    Location:
    NJ-USA
    Vehicle:
    2006 Prius
    Model:
    Four
    Look at short term and long term fuel trim. Ideally both should average near 0%. They can go up (+) to add fuel or down (-) to subtract fuel in response to readings from the Air Fuel and oxygen sensor. If both fuel trim numbers (added together or "total" fuel trim) is high, such as +10% or more then you likely have a faulty A/F sensor. Not uncommon on Toyotas.

    Posted via the PriusChat mobile app.
     
  4. Kamaloha

    Kamaloha Junior Member

    Joined:
    Jan 4, 2008
    9
    4
    0
    Location:
    New Hampshire
    Vehicle:
    2006 Prius
    Sorry for the delayed reply, I work out-of-state for five days at a time.

    I don't like throwing parts at things, I'd rather know ahead of time what is wrong. I thought that was the point of OBDCII codes, but in this case they seem to be leading me astray?

    1) I typoed when I wrote MAF sensor, I meant A/F sensor.

    2) I just went for a test drive to look at the FT parameters. After the engine had warmed up, the LTFT settled in between +2.34% and +3.5%. I never saw it get down to zero or in negative territory. The STFT was all over the map, but my eyeball average was in the zero range. The sum is clearly positive, but not so telling as the +10% you mentioned. Still thinking A/F Sensor?

    3) All the mini-VCI's I see on EBay, Amazon or elsewhere are in the $35 range and come without the Techstream software. Is there a better place to look?

    4) I see one You-Tube video that shows replacing the A/F sensor by removing the wiper deck. It doesn't look too hard after that except for the heat shield; the video looked like a bit of a hack.
     
  5. mr_guy_mann

    mr_guy_mann Senior Member

    Joined:
    Sep 3, 2020
    3,206
    1,339
    0
    Location:
    NJ-USA
    Vehicle:
    2006 Prius
    Model:
    Four
    Yes, the A/F is still at the top of my list. The downstream sensor says the engine is rich (I like to make sure the downstream sensor can go above 900mV on heavy accel or with propane enrichment and goes to near 0 on decel/ ICE off) . Lower mpg says the engine is rich. The A/F apparently does not (if it did then fuel trims would go negative). Without having a very expensive exhaust gas analyzer, you cannot verify correct operation of the A/F sensor. So the next step is to try a "known good part".

    Posted via the PriusChat mobile app.
     
  6. ChapmanF

    ChapmanF Senior Member

    Joined:
    Mar 30, 2008
    23,073
    14,982
    0
    Location:
    Indiana, USA
    Vehicle:
    2010 Prius
    Model:
    IV
    Techstream will work with dongles that support the J2534 standard. Mini VCI is one choice.

    The J2534 dongle that Toyota tests and stands behind is the Mongoose Pro from Drew Technologies, but that one's a few hundred dollars.

    There are some other choices in the range between $80ish and $170ish that reportedly work well, might be more reliable than the very inexpensive Mini VCIs. But you can definitely get the MVCIs for very little money.

    When Techstream is downloaded from Toyota, it will check your subscription before it runs; you can officially turn it on for two days at a time for $65. That's still often lower than what you'd pay a dealer to do it, to say nothing of the hassle of making an appointment and getting the car there. Some of the J2534 dongle sellers might bundle a copy of Techstream where the subscription check has been suppressed, whatever your opinion of that may be.

    If Toyota listened for advice from me, they'd probably have an option where you could activate Techstream for a reasonable one-time payment, linked to your own VIN. What I dislike most about the subscription model isn't so much the price, it's the need to be near internet and go through the whole activation rigmarole every time you've got a car problem when you really want to be troubleshooting your car, not a clunky payment web page.
     
  7. Kamaloha

    Kamaloha Junior Member

    Joined:
    Jan 4, 2008
    9
    4
    0
    Location:
    New Hampshire
    Vehicle:
    2006 Prius
    Thanks to the brain trust for the continued help. I just did the steps 5,6,7 in the diagnostic list, more or less; I couldn't figure out how to release the connectors on the EFI-M relay as the wires are quite short, and I was a little concerned I might not get it back together again. So instead I just checked the voltage and resistances at the downstream O2 sensor. I can see that the heater has the correct resistance and that it is getting power OK, so that effectively tests the EFI-M relay and the downstream heater. That covers the majority of what those tests are doing. I just now found a thread that showed the location of the ECM, thus I could do continuity testing there, but since the live data shows realistic sensor voltages I doubt that there is any harness problem. Mr. Guy Mann's reply sounds reasonable, so I think ordering a new OEM Denso A/F sensor is next on the list. I do think it worthwhile to get a mini-VCI, though, but it is disappointing news about the expensive Techstream subscription.

    Semi-related Q: I also have an airbag light. I think it might be related to when I removed the seats to install seat heaters a few years back; I have seen it intermittently since then. Will techstream help me figure that one out?
     
  8. ChapmanF

    ChapmanF Senior Member

    Joined:
    Mar 30, 2008
    23,073
    14,982
    0
    Location:
    Indiana, USA
    Vehicle:
    2010 Prius
    Model:
    IV
    If you don't have Techstream or a code reader handy, your Gen 2 airbag ECU can blink out trouble codes the same way the brake ECU can. There are a bunch of threads around here for how to count the blink codes. Most of the threads are about codes from the brake system, but the procedure is the same for airbag codes, you're just counting the blinks on a different light.
     
  9. Kamaloha

    Kamaloha Junior Member

    Joined:
    Jan 4, 2008
    9
    4
    0
    Location:
    New Hampshire
    Vehicle:
    2006 Prius
    Hi folks, I'm back from work again. My new mini-VCI was in the mail. After an entertaining morning of getting the Techstream software and drivers installed, the software and adapter seem to be working as expected. I got the engine warmed up and performed the Active A/F test (step 2 in the test algorithm). Here are the results:

    Mixture -12.5 0 +12.5
    B1S1 volt 3.72 3.18-3.26 2.34
    B1S2 volt 0.07 0.8 0.91

    As you can see, lean and rich are what we should expect, and so is the B1S1 normal reading; it is the B1S2 normal reading that is triggering (along with the calculated OSC values, which are out of range high too). However neither of the sensors have obviously failed; they just don't match expected values. Steps 3&4 would not seem to be indicated since I can clearly observe the problem which is triggering P0138 code, so there is no reason to think it will go away by resetting it and going for a drive.

    As a reminder, I've already installed a new Denso B1S2 sensor, performed test 5, and I tested the voltages at its connector which confirm the EFI M relay is supplying power to the heater (test 6). The fact that I get good voltage readings in the Active A/F test would seem to indicate to me that test 7 is not needed. Thus I've run out of official steps in the test algorithm.

    As an aside, the diagnostic algorithm has several branches which lead elsewhere in the manual that I cannot follow, not having the manual. Specifically, I don't know how to put the engine into Test mode (I was able to keep it running quite effectively by turning the A/C on) and I can't follow the branch on the bottom of p05-136 to test for intermittent problems. I rather doubt I need to do this, though, since the B1S2 voltage is not in the least intermittent, it is solidly in the high range.

    I did get a new Denso A/F sensor in the mail too, but as we all know it is a lot of work to swap it out. Do the gurus here still think that is the next step to take?
     
  10. SFO

    SFO Senior Member

    Joined:
    Feb 7, 2017
    5,286
    4,225
    0
    Location:
    Northern California
    Vehicle:
    2007 Prius
    Model:
    N/A
    This would be my first step, and may likely fix the issue.
     
  11. mr_guy_mann

    mr_guy_mann Senior Member

    Joined:
    Sep 3, 2020
    3,206
    1,339
    0
    Location:
    NJ-USA
    Vehicle:
    2006 Prius
    Model:
    Four
    I pulled up some of my test records, in comparison I don't like your B1S1 results. In general I see around 3.2 to 3.4V just sitting there at idle, and when I command the mixture I see 3.8-3.9V lean and maybe 2.0-1.9V rich.
     
  12. Kamaloha

    Kamaloha Junior Member

    Joined:
    Jan 4, 2008
    9
    4
    0
    Location:
    New Hampshire
    Vehicle:
    2006 Prius
    Gentlemen, the problem is fixed. I bit the bullet and replaced the A/F sensor, then went out for a test drive, and both the B1S1 and B1S2 voltages have returned to just where they ought to be. I reset the CEL. Let's hope it stays off!

    WRT to replacing the A/F sensor, it was actually quite easy to do, even considering that I went about it the wrong way. Following advice found here and elsewhere on the web, I removed the wiper arms, wiper mechanism and trim below it. This gave me good visual access to the A/F sensor, but it also became clear that I would not have room to swing a wrench from above. I then jacked the car and found that I could reach the sensor quite easily by sliding my arm up alongside the catalytic converter. There is adequate room to operate a wrench from that direction. I had the correct thin-walled O2S socket; however there was insufficient space left above the socket to connect a breaker bar before hitting the tranny. Luckily the socket also has a 1" hex on top. I used a 1" clawfoot on my breaker bar and was able to turn the socket that way. The old sensor came loose surprisingly easily considering it has been in there for 15 years and 166K miles.

    The most difficult part of the operation was unplugging the sensor cable. It was easy to see the plug, but I had a devil of a time getting the leverage to push that silly release catch and get it unplugged. I probably spent half an hour on it before I got it loose.

    Since I was now a Techstream expert (!!) I then read the airbag code as a bad connection to the driver seat. I unplugged and cleaned that connector and the airbag light is now off as well.

    A big thank you to all of the contributors who helped me on this quest!
     
    SFO likes this.
  13. SFO

    SFO Senior Member

    Joined:
    Feb 7, 2017
    5,286
    4,225
    0
    Location:
    Northern California
    Vehicle:
    2007 Prius
    Model:
    N/A
    Had the same experience, it was almost like toyota forgot to torque it down.
    Unclipping the sensor was fairly easy with a second pair of hands to pull from the top, but still required removing the wipers and tray.
     
  14. Kamaloha

    Kamaloha Junior Member

    Joined:
    Jan 4, 2008
    9
    4
    0
    Location:
    New Hampshire
    Vehicle:
    2006 Prius
    Regarding unclipping the sensor, I'm not sure if I needed to remove the wipers and tray, but I did move the ignition harness out of the way by unclipping the left two coils, unscrewing three screws, and lifting it aside. (I've done this before to replace the spark plugs so I am familiar with the maneuver.) I was then able to get under the harness and reach the plug. Since I did have the wipers and tray out I can't comment on if this made it easier or not.

    I'm now several hundred miles and many trips post-repair and the CEL is still out. Hoorah!
     
    SFO likes this.
  15. TEF

    TEF New Member

    Joined:
    Jan 30, 2024
    1
    0
    0
    Location:
    Minnesota
    Vehicle:
    2004 Prius
    Model:
    Touring
    Hi Kamaloha, did your car have a slight misfire issue across all cylinders when this was happening? I think I am having the same issue, but I want to be a little more certain before I take the plunge on replacing the A/F sensor. I am getting 1-30 cylinder misfires per cylinder per driving cycle, mostly when ICE is cold, and when ICE is idling or start / stopping. If it’s warm usually 0-7 misfires per cylinder. Enough to notice a bit at idle, and it shows up on the non continuous monitors, but does not set a fail on the non continuous monitors or set a CEL.

    For some more context, I have a 2004 Prius with about 260K miles on it. My 02 sensor readings are basically the same and behavior is very similar. I also have the p0138 code, as well as a P0446 EVAP system vent control malfunction. I have cleaned the MAF and throttle body, and tried replacing it the MAF. One failed spark plug and coil was replaced about a year ago, and it worked well after it was replaced. The misfires seem to be random and not concentrated on any particular cylinder(s).

    Any help would be appreciated!