General OBD2 Diagnostics

Discussion in 'General Discussion' started by AndyB1976, Wednesday 7th May, 2014.

  1. AndyB1976 Valued Contributor ★ ★ ★ ★ ☆

    United Kingdom Andy Aberfoyle
    799
    541
    5
    I've just started toying around with the 'full version' Torque app on Android and was considering starting a thread on DIY Diagnostics. This is unlikely to be good as a 'real' software and hardware setup or like the service CJ offers on diagnostics, stickied at the top of this forum.

    This idea if you can call it that was born out of the issues I had going back to the hesitation/misfire thread a couple of months ago and even with all the helpful advice given, the troubleshooting was all a bit spanners n hammers so to speak. Maybe a digital diagnostics could have pointed to the issue sooner or least collaborated with the manual diagnostics.

    As the majority of us aren't mechanics then maybe collectively we can look at information obtained and the mechanics among us can offer advice on the information we post. Obviously I am not suggesting we limit this thread to just Android/Torque so any systems you use can be freely posted.

    This can also include basic skills to obtain this information. In a nutshell the Torque (app) logs this information as a CSV file, open it in a spreadsheet using strictly the COMMA as the separator. The spreadsheet used and chart creation was done in the free Open Office suite.


    TorquePIDtest1b_zps1b18b082.



    So this was the first date capture and log which was just a short run of around 1200 seconds (20mins). Around halfway ( 600secs) you can see my foot down on the throttle then immediately off, I hope to dig into the Fuel Trim information as this may have shown the issue during my hesitation/misfire issue (fuel line leakage). I can't remember what I done around 280secs , maybe turned the engine off but left the key on as coolant temp dropped and RPM died completely.

    TPS seems to always sit around 10%, don't know if this is how it should be. Throttle body is clean and don't have any idle issues.

    I don't have any known issues with the car, this could be viewed as a baseline of sorts!
     

    Attached Files:

    Last edited: Wednesday 7th May, 2014
    Loading...
    DeviateDefiant and Ichiban like this.
  2. SpeedyGee Administrator Staff Team

    England Speedy Birmingham
    14,999
    5,595
    4
    Good stuff Andy.

    You could do with lambda readings on the graph also as well as MAP reading. With the accelerator fully or almost fully pressed, you'll notice that the short term fuel trim drops to zero, this is because the fuel system has gone into open loop mode. So you could do with putting Fuel system status on graph as well, although, this could be surmised from the throttle position and STF value.

    The 280 second mark does look like you switched off the engine.

    TPS at 10% would be when the car is sat idle, this is fine.

    Also you would really need to know at what points along the graph the hesitation occurred.
     
    Loading...
    AndyB1976 likes this.
  3. AndyB1976 Valued Contributor ★ ★ ★ ★ ☆

    United Kingdom Andy Aberfoyle
    799
    541
    5
    Thanks for the input Speedy.

    I had the 02 Voltage but never thought about the MAP - I'll add that PID on the next test.

    TorquePIDtest2voltage.

    I tried to add Fuel Status but couldn't see the PID, although I can view it realtime and watch it switch between Open and Closed loop. I was under the impression once Closed Loop was attained it remained but you can see it switch between under acceleration/deceleration in some circumstances.

    I know the f18b2 is supposed to cut fuel supply if the engine is decelerating and the RPM remains above 1100 so I was hoping to see if Fuel Status mode would reflect this or some other parameter.

    EDIT

    Just to clarify regarding the hesitation, that was a past problem and the car is now running fine. I am just looking at data from the car at the moment and how it relates or reacts under normal driving albeit extreme throttle and gear choice to see how the engine management performs. If any of us have engine issues in the future then we can maybe use this thread as a data dump or a discussion point.
     
    Last edited: Thursday 8th May, 2014
    Loading...
  4. SpeedyGee Administrator Staff Team

    England Speedy Birmingham
    14,999
    5,595
    4
    PID should be Mode 1 PID 03

    Open/closed loop refers to whether the feedback system is utilising the O2 sensor/s or not.

    Open loop will occur during engine warm up.
    Open loop will occur due to fuel cut during deceleration.
    Open loop will occur if there is a fault that stops feedback system being utilised (pending or current faults).
    Open loop will occur due to high engine load.
    Closed will occur if none of the above situations have arisen.
     
    Loading...
    AndyB1976 likes this.
  5. ArcticFire-Account Closed Banned Getting Started

    Scotland Graham Scotland
    3,521
    1,051
    7
    I agree with the principle of your thread, sensors are there for a reason so naturally you should be able to tell a fair bit about how your car is running from reading the data.

    A test you can try with regard to fuel trims, although I don't think it's applicable to you as your LTFT looks almost perfect (hard to see from the graph due to the scale differences):

    1. Set to record RPM, LTFT and MAF rate and also show these dials.
    2. Then get the car rolling very slowly and change into second gear
    3. Accelerate up to redline only in second gear but not WOT, you only want about 25% pressure on the throttle
    4. You need to reach 100g/s airflow though during this process so it may take a few runs to get right without going into Open Loop.

    Then check out your fuel trims. Long term fuel trims should settle within +/- 5%.

    • Positive trims mean that it's adding fuel so can indicate a vacuum leak at the manifold / gasket as your engine is receiving additional (unmetered) air and therefore adding fuel to compensate.
    • Negative trims mean that it's taking fuel away so can indicate a leak somewhere after the MAF sensor as the engine is receiving less than the originally metered air (by the MAF sensor) and therefore has to pull fuel away accordingly.

    I don't know if disconnecting the battery resets the ECU on the Honda but if you LTFTs don't settle then you can always reset the ECU and then do about 50 miles mixed driving (but not WOT) to let them settle again. Sometimes more miles is needed, the more the better usually before going WOT.


    Also, check your Commanded vs Actual AFR. :Smile:
     
    Loading...
    SpeedyGee and AndyB1976 like this.
  6. AndyB1976 Valued Contributor ★ ★ ★ ★ ☆

    United Kingdom Andy Aberfoyle
    799
    541
    5
    Loading...
  7. ArcticFire-Account Closed Banned Getting Started

    Scotland Graham Scotland
    3,521
    1,051
    7
    LTFT looks damn near perfect :Niceone:

    How about a Commanded vs Actual AFR?
     
    Loading...
  8. AndyB1976 Valued Contributor ★ ★ ★ ★ ☆

    United Kingdom Andy Aberfoyle
    799
    541
    5
    I'll need to take another run and log those PID's :Wink:
     
    Loading...
  9. SpeedyGee Administrator Staff Team

    England Speedy Birmingham
    14,999
    5,595
    4
    In that last graph, what were the situations under which the STFT went very negative ?
     
    Loading...
  10. AndyB1976 Valued Contributor ★ ★ ★ ★ ☆

    United Kingdom Andy Aberfoyle
    799
    541
    5
    I can't remember exactly what I was doing, although I was deliberately going WOT, then closed, and also selecting gears that were too high - I was basically trying to force to engine to react to a somewhat abnormal driving style. I logged this info over 1 week ago so its a bit vague of what I was doing and the terrain I was on.

    Looking at the first graph it appears my throttle is closed at around this time.
     
    Loading...
  11. AndyB1976 Valued Contributor ★ ★ ★ ★ ☆

    United Kingdom Andy Aberfoyle
    799
    541
    5
    Took another run today with the MAP, Fuel Status added. Started off with the engine a little warm.

    I have the O2 data too, I presume Sensor 1 is the one driving Short Term?

    TorquePIDtest4MAPstatus.
     
    Loading...
  12. SpeedyGee Administrator Staff Team

    England Speedy Birmingham
    14,999
    5,595
    4
    Yes that's right.
    All looks quite normal to me.
     
    Loading...
  13. AndyB1976 Valued Contributor ★ ★ ★ ★ ☆

    United Kingdom Andy Aberfoyle
    799
    541
    5
    Nice one Speedy!

    So those negative short term fuel trims appear to be occurring primarily during closed loop so does that imply that during open loop the system runs on a tad on the rich side, then the 02 sensors see this (during closed loop) and pull the short terms back down?
     
    Loading...
  14. SpeedyGee Administrator Staff Team

    England Speedy Birmingham
    14,999
    5,595
    4
    In any feedback system you will see this back and forth swinging as the feedback system tries to compensate for whatever it is there to compensate for.

    In this case we are talking about Fuel Trims and we see that for the most part the STFT (short term fuel trim) swings between +10% and -10%, which is we want to see.

    There's a couple of instance where it goes beyond -10%. Let's take the instance around 228 seconds, in this case STFT went down to about -18% (negative fuel trims mean ECU wants to back off a rich condition) and we see that the throttle position was steady but intake pressure increased (so load on engine increased), so the ECU must have have ramped up the fuel trims to accommodate the increase in load. The feedback system decided that the increase was too much and started to back off the fuel trims by applying negative STFT. That is perfectly normal and the feedback system is just doing it's job. The same happens again around 312 seconds.

    So what's happening between 332 and 362 seconds ? STFT during that period is bobbing back and forth around the -10% range, you've got steady throttle, very low engine load, steady RPM. Well if we remember that the ECU has a number of different fuel maps for different engine loads, different ignition timing values and in our Honda engines different VTC angles. So what's happened in this case is that the ECU has chosen a fuel map that is tad rich in nature, so again the feedback system kicks in as it should and applies some negative STFT to keep the stoichiometric (A/F ratio) value around
    14.7

    The LTFT (long term fuel trim) is like a running average stat of the STFT, in the more modern cars there's actually more than one LTFT but we won't confuse matters by talking about that.

    In this case Andy, we can see that the LTFT is close to zero, so over a period of time we know that we aren't constantly actually applying a lot of positive or a lot of negative fuel trim. This is what we want to see. Hence everything looks fine.
     
    Loading...
    AndyB1976 likes this.
  15. AndyB1976 Valued Contributor ★ ★ ★ ★ ☆

    United Kingdom Andy Aberfoyle
    799
    541
    5
    Cheers for the detailed reply Speedy.:Hey:

    Its quite interesting (to me lol) what sort of information can be gleaned from the engine and being a novice at this stuff I was under the impression that Fuel Status was simply open loop until warmed up then closed loop thereafter.

    However as you said in post 4 you can see that open loop mode will be met in other conditions



    Then whilst trying to create the PID for logging and understanding the data fields in Torque I found this wiki entry, which although you will already know explains (to me and others) why the graph for the Fuel Status isn't a binary on off (open/closed) situation and why my graph infact goes to value 4 or would have went higher if there was a fault.

    ValueDescription
    1Open loop due to insufficient engine temperature
    2Closed loop, using oxygen sensor feedback to determine fuel mix
    4Open loop due to engine load OR fuel cut due to deceleration
    8Open loop due to system failure
    16Closed loop, using at least one oxygen sensor but there is a fault in the feedback system
     
    Last edited: Saturday 10th May, 2014
    Loading...
    SpeedyGee likes this.
  16. ArcticFire-Account Closed Banned Getting Started

    Scotland Graham Scotland
    3,521
    1,051
    7
    Aye don't worry about STFT, it's the LTFT which are important. Have you checked your commanded vs actual AFR yet?
     
    Loading...
  17. AndyB1976 Valued Contributor ★ ★ ★ ★ ☆

    United Kingdom Andy Aberfoyle
    799
    541
    5
    Not yet @ArcticFire , thats next to look at lol.

    There are quite a few things that can be plucked from the engine eh, all that going on oblivious to the driver. I guess these and maybe EGR would be useful too?
     
    Loading...
  18. ArcticFire-Account Closed Banned Getting Started

    Scotland Graham Scotland
    3,521
    1,051
    7
    To be honest the commanded vs actual is more of a tuning thing but still no harm checking it out. You want them to be as similar as possible.
     
    Loading...
  19. AndyB1976 Valued Contributor ★ ★ ★ ★ ☆

    United Kingdom Andy Aberfoyle
    799
    541
    5
    No harm at all, and also useful to capture information as a baseline when the engine is running without issue.
     
    Loading...
  20. Nels Moderator Staff Team

    @AndyB1976 , @SpeedyGee and @ArcticFire, I am fascinated by the info that can be obtained.
    However, :Whoosh:it's like another language to a novice like me. :Unknown:
    How did you learn about this field and is there a resource available that I can read/watch to learn more?

    I would not dream of advising anyone about this data, that's for you guys, but I really do want to learn more to help my understanding.
    Cheers,
    JD
     
    Loading...