Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!





A Mechanic's Life - Tales From Under the Hood

1238239240242244

Comments

  • steverstever Posts: 52,573
    edited January 3
    Sorry Doc, got other pancakes in the air at the moment, including some concern about body work on my van. Maybe you can help with this and I'll try to get back to your dissertation in a day or two.

    Van got smashed in the LR tire, new axle and body work. After taking delivery I noticed that the sliding door doesn't fit. They've adjusted it once and I told them to keep trying. I'm worried about structural stability and safety. There was no obvious damage to the door. When closed the top seams aren't "flush" with the roof and it's obvious when you look, especially when compared to the right side slider.


  • thecardoc3thecardoc3 Posts: 3,870
    Need a sharper pic. Plus get and upload pics of the gap in question and of the pillar at the rear of the door.
  • qbrozenqbrozen Posts: 20,417

    '12 EX35 Journey AWD; '14 Jetta TDI wagon; '98 Volvo S70 base; '14 Town&Country Limited; '09 LR2 HSE. 42-car history and counting!

  • steverstever Posts: 52,573
    edited January 3
    Thanks Doc, took a few snaps over the weekend but the car is at the shop today so can't post more anytime soon.

    First pic is the front top corner and the second is the gap at the top of the B pillar. May not mean much without a reference shot of the other side.



  • thecardoc3thecardoc3 Posts: 3,870
    edited January 3
    qbrozen said:
    Yep that's the circuit. Now explain how you test it without the failure being present 100% of the time as well as not even being designed to be accessible which is nicely demonstrated by the video of the service disconnect. You see the two male pins in the service connector? Consider for the moment that the failure with this particular vehicle could be an issue with one of the female pins for the interlock connection of the service disconnect.

    Now consider that the failure could be anywhere in the circuit depicted in the service information and since it doesn't fail and remain failed for a long enough period of time you cannot use the trouble tree linked.

    Got to love how they write this stuff. Under "Trouble are"

    Wire harness or connector
    Power management control ECU
    Service plug grip
    Inverter with converter assembly
    Frame wire
    Inverter terminal cover


    If all you are capable of doing is looking at these you might guess the right thing to look at first, you might guess the right thing to look at as the last possible choice after you looked at the rest. Of course there is also the possibility that just looking at it won't reveal the cause. BTW to "look at" the Power Management ECU you have to disassemble the dash, to look at the service disconnect socket and pins you have to remove and disassemble the battery pack.....

    Now if you choose to test it the right way, you will start putting your plan together with nothing more than this.


  • guitarzanguitarzan OhioPosts: 733
    edited January 3
    Is the only info that the Prius set code P0A0D? Or is there a detailed post that I missed?

    Two questions:

    What are the requirements for the initialization of Ready mode?
    What specifically does the self correct do? (I bet there is no record of this, but you can tell us.)

    I am unclear about the self correct: Are you able to reproduce the failure and have some troubleshooting time before the self correct kicks in? Or do you get no time at all to analyze the failed state?

    If you get zero time to troubleshoot during the failed state because of how fast the fault corrects, then it seems you must look at the Ready mode requirements and start making some pretty fancy guesses to move forward. But then you say guessing is not allowed. So I feel stuck. Maybe answers to the above will help. Or maybe I will just flunk out.
  • qbrozenqbrozen Posts: 20,417
    If it were my vehicle, I'd at least know the history and know if any of these things were recently touched, which is where I'd start first to make sure it was reseated properly. If it occurred without any previous repairs having been done, then it could be anything and I'd be at a loss. If I got to the point I'd have to remove the dash just to find out, I'd probably throw in the towel. Ha.

    Speaking of which, I've done just that on my Volvo and its nonworking AC. I've checked for leaks everywhere else and the last place is behind the dash, a common failure point. I'm not willing to go through the effort on a 19-yr-old base model Volvo with 2 deer strikes against it, so I'm just going with no AC until I donate the car to charity probably later this year.

    '12 EX35 Journey AWD; '14 Jetta TDI wagon; '98 Volvo S70 base; '14 Town&Country Limited; '09 LR2 HSE. 42-car history and counting!

  • thecardoc3thecardoc3 Posts: 3,870
    guitarzan said:

    Is the only info that the Prius set code P0A0D? Or is there a detailed post that I missed?

    No other information beyond : If the problem occurs when attempting to start the car, you can get to ignition on but not the ready mode. If the problem occurs while the car is being driven, the car will continue to run but not restart if it is turned off, unless the problem self corrects.

    You are actually getting the benefit of the description of the failure being filtered based on both product knowledge as well as having located and repairing the fault. Had it been left to the description from the owner it would have been much more cryptic and amounted to "Sometimes the car wont start, but if I try to shut it off and turn it back on a few times it eventually starts up, but even then sometimes it turns back off all by itself"
    guitarzan said:


    Two questions:

    What are the requirements for the initialization of Ready mode?
    What specifically does the self correct do? (I bet there is no record of this, but you can tell us.)

    To achieve ready mode the system has to energize and test the SMR's (system main relays) and part of doing that it needs to monitor the HV interlock circuit for any faults. If a fault is detected in the interlock circuit, the SMRs will be turned off if the vehicle is not moving, or will be allowed to stay energized if the vehicle is being driven. (The assumption there is that since the car is moving it is a circuit failure and not someone disturbing one of the safety switches or covers)

    If I tell you what the "self correct" is really doing I'll be answering the question as to what was wrong with the car. By saying "self correcting" imagine a functioning circuit failed, but then resumed normal operation without any action from an outside force.
    guitarzan said:


    I am unclear about the self correct: Are you able to reproduce the failure and have some troubleshooting time before the self correct kicks in? Or do you get no time at all to analyze the failed state?

    I reproduced the failure four times for a total duration of about 12.6 seconds. That was enough to prove what the failure was. It did however take a week and close to ten hours of driving the vehicle to get that 12.6 seconds of failure.
    guitarzan said:


    If you get zero time to troubleshoot during the failed state because of how fast the fault corrects, then it seems you must look at the Ready mode requirements and start making some pretty fancy guesses to move forward. But then you say guessing is not allowed. So I feel stuck. Maybe answers to the above will help. Or maybe I will just flunk out.

    You need nothing more than the schematic, and a good game plan as each failure event is observed.

  • thecardoc3thecardoc3 Posts: 3,870
    qbrozen said:

    If it were my vehicle, I'd at least know the history and know if any of these things were recently touched, which is where I'd start first to make sure it was reseated properly. If it occurred without any previous repairs having been done, then it could be anything and I'd be at a loss. If I got to the point I'd have to remove the dash just to find out, I'd probably throw in the towel. Ha.

    If you simply guess that it has anything to do with a recent repair and start there, then you are starting off with a biased assumption and while there could always be a chance that you might be lucky and get it right, the odds are actually against you and you will start off checking things that you don't have any proof are the issue. I have even had people add problems to a vehicle that weren't the primary failure simply because they started guessing and just trying things no matter how they justified their reasoning for having done so. Meanwhile if you forget all of that and simply test just like any other first time failure, you would go right at the problem whether it was caused by a recent repair or not.
  • guitarzanguitarzan OhioPosts: 733
    edited January 3

    If I tell you what the "self correct" is really doing I'll be answering the question as to what was wrong with the car. By saying "self correcting" imagine a functioning circuit failed, but then resumed normal operation without any action from an outside force.

    For clarification, I was wondering what the scope of self correct is and what it is allowed to do. The designers may have created this exact problem then taught the self correct to temporarily reset the system. The self correct can only address a problem that it is programmed for so a big hint to this issue lies within it. Is the self correct functionality fully documented or is it just a black box to you?
  • qbrozenqbrozen Posts: 20,417
    If it were my vehicle, I'd at least know the history and know if any of these things were recently touched, which is where I'd start first to make sure it was reseated properly. If it occurred without any previous repairs having been done, then it could be anything and I'd be at a loss. If I got to the point I'd have to remove the dash just to find out, I'd probably throw in the towel. Ha.
    If you simply guess that it has anything to do with a recent repair and start there, then you are starting off with a biased assumption and while there could always be a chance that you might be lucky and get it right, the odds are actually against you and you will start off checking things that you don't have any proof are the issue. I have even had people add problems to a vehicle that weren't the primary failure simply because they started guessing and just trying things no matter how they justified their reasoning for having done so. Meanwhile if you forget all of that and simply test just like any other first time failure, you would go right at the problem whether it was caused by a recent repair or not.
    I completley disagree with your statement that the odds are against me. If you remove a harness during a repair of another problem and then the car has a new problem that can be explained by the fact that you messed with that harness, then I wholeheartedly contend that the odds are squarely in my favor that your problem lies therein.

    '12 EX35 Journey AWD; '14 Jetta TDI wagon; '98 Volvo S70 base; '14 Town&Country Limited; '09 LR2 HSE. 42-car history and counting!

  • thecardoc3thecardoc3 Posts: 3,870
    qbrozen said:

    I completley disagree with your statement that the odds are against me. If you remove a harness during a repair of another problem and then the car has a new problem that can be explained by the fact that you messed with that harness, then I wholeheartedly contend that the odds are squarely in my favor that your problem lies therein.

    Whatever. One of us does this for a living and teaches advanced routines to other technicians. My advice is get a job as a tech and see how it works out for you and maybe in 20-30 years we will see if you have changed your mind or not...
  • thecardoc3thecardoc3 Posts: 3,870
    guitarzan said:


    For clarification, I was wondering what the scope of self correct is and what it is allowed to do. The designers may have created this exact problem then taught the self correct to temporarily reset the system.

    Your on the wrong track, by self correct it only means that what ever the failure is, no outside influence causes the failure to go away, it just does. This isn't a strategy in software or anything like that, its just a short duration random failure.
    guitarzan said:



    The self correct can only address a problem that it is programmed for so a big hint to this issue lies within it. Is the self correct functionality fully documented or is it just a black box to you?

    I don't subscribe to black box theory approaches. I find out how the circuit works and then test the circuit/system just like the module does.

  • qbrozenqbrozen Posts: 20,417
    edited January 4
    thecardoc3 said: My advice is get a job as a tech and see how it works out for you and maybe in 20-30 years we will see if you have changed your mind or not...  


    Haha. No, thanks. I had my taste of it working my way through school. I much prefer what I do now. But I greatly appreciate all I learned from the pros I got to work with. It has saved me countless dollars over the years since.

    '12 EX35 Journey AWD; '14 Jetta TDI wagon; '98 Volvo S70 base; '14 Town&Country Limited; '09 LR2 HSE. 42-car history and counting!

  • thecardoc3thecardoc3 Posts: 3,870
    Here was a measurement taken during the failure. With this piece of information what do you know, what do you not know, and what would you need to know next?


  • thecardoc3thecardoc3 Posts: 3,870
    qbrozen said:

     


    Haha. No, thanks. I had my taste of it working my way through school. I much prefer what I do now. But I greatly appreciate all I learned from the pros I got to work with. It has saved me countless dollars over the years since.

    That's one of the challenges for techs, since most of what they learn comes on the job and from other techs in the trade. What we were all taught was when something didn't work out we were to go back and look over what we had just done to find the mistake. Experience has since proven that to be a flawed approach and that is because it is a different approach then you would normally use for any other diagnostic. The best way to explain this is:

    Picture you were working on something and something happened and either the repair didn't solve the problem or another issue was then revealed and you end up stuck unable to figure out what is wrong. Now you need another techs help, or maybe that tech simply takes over and goes to work on the problem. I do exactly this for other shops and the first step is to forget everything else that has been done or what someone else thinks the problem is and I concentrate on exactly what the car is doing right now and set out to prove why. In other words I start at the beginning without any assumptions and I go straight at the problem, every single time whether they caused it or not. You should see the looks I get when it takes me less than a half an hour to solve a problem that someone else might have been fighting for a couple days......
  • Mr_ShiftrightMr_Shiftright CaliforniaPosts: 54,664
    Doc, maybe I'll fly you out to California for this one---the "check gauges" warning comes on randomly on my 1998 Dodge Dakota SLT pickup, and the oil pressure gauge falls to zero. Those are the only symptoms.

    My "intuition" tells me I am dealing with a voltage spike or variation out of spec, and not a bad sending unit. I really don't think pulling the sender out and testing it for continuity is going to do me very much good here, but perhaps careful inspection of the plug-on might.

    MODERATOR --Need help with anything? Click on my name!

  • guitarzanguitarzan OhioPosts: 733
    edited January 4
    Doc is "frame wire" the negative side of the circuit? Thus node 36 should be ground. I suppose the inter lock function is to complete the ground, and thus we know from the measurement that the inter lock is working correctly.
  • roadburnerroadburner Posts: 9,373

    Doc, maybe I'll fly you out to California for this one---the "check gauges" warning comes on randomly on my 1998 Dodge Dakota SLT pickup, and the oil pressure gauge falls to zero. Those are the only symptoms.

    My "intuition" tells me I am dealing with a voltage spike or variation out of spec, and not a bad sending unit. I really don't think pulling the sender out and testing it for continuity is going to do me very much good here, but perhaps careful inspection of the plug-on might.

    If it's anything like my TJ, it could be the sockets that the gauge panel plugs into. Jeep had a TSB on the exact problem.

    Mine: 1995 318ti Club Sport / 2014 M235i / 1999 Wrangler / 1996 Speed Triple Challenge Cup Replica Wife's: 2016 i3 REX/2004 X3 2.5i Son's: 2009 328i

  • Mr_ShiftrightMr_Shiftright CaliforniaPosts: 54,664
    Hmm...that's possible. Time for the "wiggle test"! Thanks for that.

    My ultimate solution might be to just hardwire an under-dash oil pressure gauge. I'm not getting into wiring harnesss issues with a nearly 20 year old truck.

    MODERATOR --Need help with anything? Click on my name!

  • thecardoc3thecardoc3 Posts: 3,870
    The frame wire insures that the invertor assembly is completely assembled. This video at 1:35 shows the frame wire connector with the service cover removed and at 6:35 the connector is visible (its moving fast you'll have to pause it) then a little later in the video you can see the actual wire (red) at 7:08.

    "36" is the pin ID for the interlock out, which goes to the HV battery pack, through the service disconnect and then grounds at the right "C" pillar.

    If the circuit failed open after the invertor, then the voltage at pin 36 would rise to battery voltage (12-14v). Since it stayed low, that confirmed that half of the circuit was operating correctly because it stayed at ground.

    What would be next? What should the circuit voltage be at pin 35 when it is operating correctly?
    If the voltage at that point went high, that would indicate what?
    If it went low that would indicate?

    The next move(s) would be?

    BTW. This is the advantage of using a multiple trace oscilloscope, being able to monitor four channels at a time saves tons of diagnostic time. (I can actually monitor 14 simultaneously with various tools combined.)
  • thecardoc3thecardoc3 Posts: 3,870

    Doc, maybe I'll fly you out to California for this one---the "check gauges" warning comes on randomly on my 1998 Dodge Dakota SLT pickup, and the oil pressure gauge falls to zero. Those are the only symptoms.

    Remove the sender and measure and confirm the oil pressure.



  • guitarzanguitarzan OhioPosts: 733
    edited January 4

    What would be next? What should the circuit voltage be at pin 35 when it is operating correctly?
    If the voltage at that point went high, that would indicate what?
    If it went low that would indicate?

    Pin ID 35 is sinusoidal 3Φ power? I don't know the spec. A cursory google shows 500 Volts output?

    If the voltage is too high that is a result of resistance and can be caused by corrosion/heat/a wire burning through on the power controller side.

    If the voltage is too low, a problem with the supply side. Test the batteries? If they test good then the inverter is toast.
  • Mr_ShiftrightMr_Shiftright CaliforniaPosts: 54,664

    Doc, maybe I'll fly you out to California for this one---the "check gauges" warning comes on randomly on my 1998 Dodge Dakota SLT pickup, and the oil pressure gauge falls to zero. Those are the only symptoms.

    Remove the sender and measure and confirm the oil pressure.



    I already did---I drove it all the way home----JUUUUUST kidding, Doc.

    Sure, I can do that.

    MODERATOR --Need help with anything? Click on my name!

  • thecardoc3thecardoc3 Posts: 3,870
    guitarzan said:


    Pin ID 35 is sinusoidal 3Φ power? I don't know the spec.

    It comes from pin 32 of the Power management controller, 12v signal.

  • thecardoc3thecardoc3 Posts: 3,870
    edited January 5
    Looking at the schematic and the linked trouble tree, the interlock circuit has several connectors and two, one ohm resistors inside of the invertor. Looking at the Power Management ECU you see the power supply for the circuit and a resistor in the circuit, placing that resistor in series with the ones inside the invertor assembly.

    When the circuit is operating there would be current flowing through the circuit to ground and that current flow would be controlled by the circuit's total resistance. As the current flows through each resistor the voltage level will drop. The trouble tree shows the open circuit voltage at pin 35 to be 12v. But what is the voltage supposed to be when it is connected and operational? How much current should be flowing in the circuit?

    In every diagnostic there is often information that makes analyzing the system easier and the answers to those last two questions are what you need to know. Try to find that information.
  • guitarzanguitarzan OhioPosts: 733
    The open circuit voltage of 12V at pin 35 means no current is flowing in the controller, indicating that the power supply is also 12 V. Then connecting the circuit allows a total current I = 12/(1+1+Rps), where Rps = the Power Supply resistor.

    The voltage drop across each inverter resistor during normal operation is (I)(1) = I volts.

    The operational voltage at pin35 = 12 - (I)(Rps).
  • thecardoc3thecardoc3 Posts: 3,870
    While the base algebra represents the equation, the question is; "How much current is flowing and what is the value of that resistor in the controller? "

    Having that information in hand makes any diagnostic with that circuit much simpler, and in this case I had no choice but to measure and figure out the value of that resistor as it was presented. Without a spec however, how is one to know if there isn't some issue present that would be sufficient to have an impact on the circuit's performance but short of causing it to code?

  • thecardoc3thecardoc3 Posts: 3,870
    edited January 7
    How many times do you see someone report their vehicle has a problem, but its a random failure and even though they have had it looked at numerous times, nobody can find the cause. The Prius was one of those cars, and the owner had been told the next thing to do would be to replace the invertor, but they couldn't guarantee that would fix the problem.

    As described, when the failure occurred Invertor pin 36 was at 0v, the thing is pin 36 should be at 0v (or at least very close to it) because that is the ground side of the interlock circuit. If someone disconnected the service interlock, pin 36 would rise to the source 12v signal. The question that needed to be asked and then answered by measurement is what was the voltage at Invertor pin 35 when the circuit failed and it was a 12v.

    What does that mean, and what is the next step?
Sign In or Register to comment.