Howdy, Stranger!

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





Howdy, Stranger!

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

Did you get a great deal? Let us know in the Values & Prices Paid section!
Meet your fellow owners in our Owners Clubs

A Mechanic's Life - Tales From Under the Hood

1238239241243244289

Comments

  • kyfdxkyfdx Everywhere, USAPosts: 126,124
    stever said:

    Just that I was born twenty years too late?

    20 years too soon, evidently..

    Did you get a good deal? Be sure to come back and share!

    Edmunds Moderator

  • Mr_ShiftrightMr_Shiftright Sonoma, CaliforniaPosts: 64,490
    nobody knows the answer until they do :p
  • steverstever Posts: 52,462
    edited January 2017
    lol, there's that math again.

    I dunno, sounds like a lot of taxi talk in here. This is the way to do it, this is the way that it's done. It won't work any other way, this is how we train you and this is how you'll do it and none of our customers expect anything different. And it's impossible to test for non-continuity (that's a Latka reference I suppose).

    Then one day you look up and you're run over by Uber.
  • Mr_ShiftrightMr_Shiftright Sonoma, CaliforniaPosts: 64,490
    edited January 2017
    and who's to say Uber is a good thing? You may be creating a monster here, Steve.

    You may end up with a car full of Microsoft "bloatware".
  • thecardoc3thecardoc3 Posts: 5,051
    Yep, and nobody would have known the answer without someone who first was willing and able to work both smarter and harder providing it. Isn't it sad how little respect that has earned techs through the years? But anyway Steve provided valuable insight to the situation. Instead of appreciating what really goes into doing the work we do, there is always someone who will just concentrate on trying to minimalize it any way they can. That's the whole purpose of trying to say that all someone had to do is look while ignoring the fact that they didn't have any idea what someone could possibly find nor where it might be.

    But with a plan in place and a collection of clues to guide the way.......
  • steverstever Posts: 52,462
    edited January 2017
    Yeah, Uber has been a big fat failure. Taxis were so much better.

    I need a job at Ford. They get it.

    "In 2004, Ford Motor Co.’s resident futurist, Sheryl Connelly, led a team that imagined what would happen if an economic shock and a rapid increase in the price of gasoline led to a crash in automotive sales."

    Think Like a Futurist to Be Prepared for the Totally Unexpected (WSJ - registration link but a search on the headline may flop up a free version of the article)
  • thecardoc3thecardoc3 Posts: 5,051
    Not following you here, how does someone else's idea account for your not having learned to creatively investigate and analyze a system failure?
  • steverstever Posts: 52,462
    edited January 2017
    You are still in the forest. Here's an example of how it should be:

    What can cause a car to break?

    Hm, miles of electrical wiring exposed to a harsh environment.

    What's the fix?

    Someone not invested in the way stuff currently works won't focus on improving the wiring, they'll figure out how to eliminate the wiring.

    Your profession and most of the upstream engineers are addressing the actual failure, not really working to prevent it in the first place.
  • thecardoc3thecardoc3 Posts: 5,051
    Moore's law worked for a while, but has pretty much run its course.

    https://en.wikipedia.org/wiki/Moore's_law

    Rocks law explains why.

    Making a perfect car might be possible some day, but who would be able to afford it?
  • steverstever Posts: 52,462
    They've been predicting the death of Moore's law for a couple of decades now. Even Moore predicted it. But it's still going. Your Wiki link doesn't say it's dead either, it just that new tech is required to keep up. So we move on from silicon. Sound familiar?

    We can't afford not to expand the possible.

    Anyone wanting to be an auto tech or engineer these days better learn how to code first. Or at least learn how to tell the proggie how they want something coded. :)
  • thecardoc3thecardoc3 Posts: 5,051
    stever said:


    Anyone wanting to be an auto tech or engineer these days better learn how to code first. Or at least learn how to tell the proggie how they want something coded. :)

    Does anyone else see the irony in the fact that just a few minutes ago you brought up thinking like a futurist and how they feel that trying to predict the future isn't what they are trying to do and then you wrote this? What's more there is this quote. " One thing all the futurists I talked to had in common was disdain for anyone willing to attempt to predict the future. In futuring circles, paradoxically, this is the mark of an amateur."

    Meanwhile there was this tidbit. "Actually practicing futurism, even if only for a day, showed me the reason the future is so confounding: Aside from the fact that anything can happen, those unexpected events rapidly compound on one another. This leads to second, third and nth-order effects that can seem completely beyond the realm of plausibility until they happen. Hence the impossibility of predicting financial crises, wars and technological revolutions."

    Which is why just being able to wiggle a harness or simply look at the location of some random failure will fail to reveal the fault more often than it succeeds. But take a logical, disciplined approach and any problem can be solved.
  • steverstever Posts: 52,462
    Thanks for reading the link. Interesting angle eh?

    I suggested a way to fix the problem based on limited personal experience. If the car had been in my garage, I might have been able to crawl around, remove some plastic and find the rub. Other mere mortals may throw parts at it or find the fix on YouTube or a car forum or in a TSB.

    There's more than one way to skin a cat and it can help not to box yourself in. The pro repair sites are full of similar shortcuts that are posted to help other techs - if your '99 Chevy does this, take a look at the whizzenjammer first.
  • thecardoc3thecardoc3 Posts: 5,051
    Lets repeat one of the quotes above. "Aside from the fact that anything can happen, those unexpected events rapidly compound on one another. This leads to second, third and nth-order effects that can seem completely beyond the realm of plausibility until they happen. Hence the impossibility of predicting" that you can find any given failure based on what happened to a different car is the hallmark of an amateur.

    "If the car had been in my garage, I might have been able to crawl around, remove some plastic and find the rub"

    Except for the annoying little detail that you didn't have any legitimate reason to begin looking at any given point on the car. If there was any validity to your assertion you should have recommended doing that a week ago, and what about the next car that comes in with the instrument cluster acting up would you be putting that one in the air and pulling panels off of it because of what was found this time? You may never learn that the short-cuts or silver bullets as they are known are traps that at the very best discourage younger technicians from getting continuing educational training and developing real career skills. At the worst they cause someone to shotgun parts and when that doesn't work they have no-where to turn because they didn't put enough effort into building the skills that they needed to do it the right way the first time.
  • Mr_ShiftrightMr_Shiftright Sonoma, CaliforniaPosts: 64,490
    stever said:

    Yeah, Uber has been a big fat failure. Taxis were so much better.

    I need a job at Ford. They get it.

    "In 2004, Ford Motor Co.’s resident futurist, Sheryl Connelly, led a team that imagined what would happen if an economic shock and a rapid increase in the price of gasoline led to a crash in automotive sales."

    Think Like a Futurist to Be Prepared for the Totally Unexpected (WSJ - registration link but a search on the headline may flop up a free version of the article)

    Uber isn't a failure for its creators--just for the drivers who work for them and for public transit.

    Be careful what you wish for, is the lesson here.
  • kyfdxkyfdx Everywhere, USAPosts: 126,124
    The jury is still out on Uber. At some point, you have to show a profit.

    Did you get a good deal? Be sure to come back and share!

    Edmunds Moderator

  • steverstever Posts: 52,462
    edited January 2017
    kyfdx said:

    The jury is still out on Uber. At some point, you have to show a profit.

    Like GM, lol. The jury is still out on everything. ;)
  • Mr_ShiftrightMr_Shiftright Sonoma, CaliforniaPosts: 64,490
    Apparently you DON'T have to show a profit anymore. Just like Dorothy's red shoes in the Wizard of Oz, you just click your heels and make a wish and everything comes true.

    I don't agree with Steve that some kind of voodoo telemetrics is going to fix cars in the near future, but I do agree with him that some of the systems engineering in modern cars seems very wrong-headed indeed.

    A defect in your power steering harness should not make dash gauges dance. That's just nuts. It should be possible to isolate faults like this from screwing up the entire system. It's like your windshield washer fluid somehow intermingling with your transmission oil through a vacuum leak. What engineer would be proud to have allowed that to happen?






  • thecardoc3thecardoc3 Posts: 5,051

    A defect in your power steering harness should not make dash gauges dance. That's just nuts.

    It is nuts, but it still is also a perfect example of what we have to be able to deal with. I'll spare you the details of what was wrong with the hybrid controls on the 2011 Prius that I worked with this week. The Focus was easy in comparison.


    It should be possible to isolate faults like this from screwing up the entire system.

    Many of the operations that vehicles carry out require cooperation between a number of modules. Just to have the tachometer to work you have at the minimum the PCM and IPC which communicate either directly with each other on one of the data buses or else through another module known as a gateway. Either way, if something disrupts the PCM's ability to send data to the IPC then the IPC will be unable to display that information to the driver.
  • thecardoc3thecardoc3 Posts: 5,051
    The discussion yesterday reminds me of some of the more difficult times that I dealt with decades ago as a young technician trying to find my way in this trade. It also reflects one of the challenges that a lot of technicians still face today and it centers around pressure to be too fast and therefore "cheaper" for the consumer. But pressure on the technicians doesn't necessarily make it cheaper for the consumer and that's the gross error in Steve's position. His taking the solution once it was handed to him and then trying to twist it into something that would otherwise require less talent and therefore be cheaper is all about consumer greed. We have explored this issue in the past and unfortunately dropped the discussion well before ever driving all of the points home. Maybe this time it will be different.

    There are boundaries to this discussion that have to be recognized and the first and most important one is that you don't get to take the solution for any car after it has been provided and attempt to short cut the diagnostic like Steve tried to do. Why? There was no legitimate reason to look at any part of the car for any type of a failure until enough data was collected in order to guide someone to do it. It was only after the observation was made that the fault could be manipulated by changing the drivetrain torque that it became logical to inspect the under-hood wiring harnesses for damage. That relatively simple observation couldn't be made until after the communication voltage waveforms were being measured with the digital oscilloscope.

    No matter how someone tries to manipulate this, there was just one way to efficiently come up with the solution for this vehicle failure and nobody else provided any of it other than me. It's also safe to say that no-one else participating in or just reading these forums has the skill, experience, tools and discipline to have succeeded in analyzing that failure. FWIW, I do know other techs who would have figured this communication problem out too, but they aren't here. This vehicle failure was also shared with them including how the observations and testing progressed and the majority of the opinions from them came down to they were glad that it wasn't them that took this challenge on. That's a pretty normal emotion among top techs because we have all been beat up so many times during our careers that we all know that any one of these problems can eat us alive if we falter even slightly. Nobody ever held training classes to teach us how to do what you just got to see a part of here with that post, we have had to create these kinds of routines completely on our own

    When it comes to O.E. trouble codes and diagnostic charts they only work while the problem is present and are useless when it isn't. The routine you saw here works whether the problem is there all of the time or not provided that it does in fact occur or can be made to occur.

    The next thing that needs to be addressed were some of the arguments presented that didn't have anything to do with solving the problem with Neil E's car. It is totally irrelevant what any other group of people are doing inside their own realms (the futurists for example) in the context of this discussion. Whether cars can be made someday that don't suffer the failure that this one did has nothing to do with Neil being able to get to work on Monday morning. Any attempt of an argument that ignores that consumers immediate needs is a waste of effort in this forum beyond the angle that nonsense like that has been used towards making the career undesirable and less likely that Neil would have someone available to him to actually fix his car instead of being forced to get a different one.



  • steverstever Posts: 52,462
    >There was no legitimate reason to look at any part of the car for any type of a failure until enough data was collected in order to guide someone to do it.

    So all we need to do now is hook the car up to a scanner and have some tech in India diagnose it for $8?

    Works for me. :)

    Meanwhile, I'll keep eyeballing and jiggling stuff, and taking stuff apart and putting it back together, whether it's my car or my bike or my computer. I've fixed a lot of stuff over the years just by "looking at it".
  • thecardoc3thecardoc3 Posts: 5,051
    edited January 2017
    But you would not have fixed that Focus, and you darn sure wouldn't have solved the Prius that was setting the P0A0D. But since you can claim anything and not have to genuinely put up you think you can keep pretending in front of everyone else that you have a hint of a clue. So tell everyone how you would fix it, or else of course you can always come clean and admit that you are just blowing smoke and don't honestly have any idea at all. Tell you what, I'll even help you with a warmer/colder answer if you dare to try, and just so you know it was a random failure of the HV interlock circuit.

  • qbrozenqbrozen Posts: 25,960
    broken wire on the focus that i would have found by inspecting harnesses and connections like I originally said? Yeah, I would have fixed it. But, as I always say, its my car and my time. I wouldn't do it for anyone else because I can afford to be wrong on MY car.

    '18 BMW 330xi; '67 Coronet R/T; '14 Town&Country Limited; '18 BMW X2. 47-car history and counting!

  • steverstever Posts: 52,462
    edited January 2017
    I think we touched a nerve. B)

    People will always try to fix their stuff, whether it's their cars, houses or bodies. And that's why there are millions of YouTubes out there. And there's always more than one way to solve a problem and who's to say they have the one and only right way patented?
  • thecardoc3thecardoc3 Posts: 5,051
    edited January 2017
    qbrozen said:

    broken wire on the focus that i would have found by inspecting harnesses and connections like I originally said?

    Not broken, at least not yet. It was grounding randomly.


  • steverstever Posts: 52,462
    Just like my mower. B)
  • thecardoc3thecardoc3 Posts: 5,051
    stever said:

    I think we touched a nerve. B)

    So what's the first step for the P0A0D? All I see is more smoke......

  • steverstever Posts: 52,462
    edited January 2017
    Turn the car on?

    "According to IBM’s studies, cars are increasingly becoming:

    1) Self-healing: Vehicles that are able to diagnose and fix themselves and even fix other vehicles with issues without human help."

    BMW Group to Start Research With IBM Watson (hpcwire.com)

    How Self-Healing Electronics Could Change Everything, from Smartphones to Space Stations (allaboutcircuits.com)

    "Current generation electric vehicles manage batteries by monitoring individual battery cell performance connected to a central BMS master via cables and wiring harnesses. A new proposal from Linear Technology suggests that future vehicles will manage battery conditions with the help of a wireless BMS."

    BMW i3 Concept Car Relies on Wireless Feedback of Individual Battery Cell Performance (powerelectronics.com)
  • Mr_ShiftrightMr_Shiftright Sonoma, CaliforniaPosts: 64,490
    So will their robot arms tape up that torn harness?
  • thecardoc3thecardoc3 Posts: 5,051
    stever said:

    Turn the car on?

    LOL, this could really be fun. You need to look up the code definition and description and you'll see why you couldn't be colder with that guess.
    stever said:


    "According to IBM’s studies, cars are increasingly becoming:

    1) Self-healing: Vehicles that are able to diagnose and fix themselves and even fix other vehicles with issues without human help."

    BMW Group to Start Research With IBM Watson (hpcwire.com)

    So if you can't dazzle them with brilliance (and you aren't) try to baffle them with BS and you are failing there too.

    Step up, or bow out.....

  • thecardoc3thecardoc3 Posts: 5,051
    edited January 2017
    Remember your Futurist link? The premise is that they come up with an idea and then tear it apart in order to try and figure out what might happen.

    For the moment, the self healing circuits. Do you see how large that is and once broken it doesn't restore full current flow sufficient to fully illuminate a simple LED? Lets say they solve that, what would happen if you put multiple circuits like that side by side by side? Will they try and connect to the other circuits where they don't belong? If they solve the problems of size and cross connections, that would be great, BUT we are only talking about high resistance failures or opens at this point. How will that help anyone when a circuit creates an illegal connection such as the Focus was doing? Then you have the inevitable challenge that reducing the work that the techs do serves to diminish the technical skills. With fewer failures to fix there would be fewer techs to handle any problems that did occur. Keep in mind fewer can go all the way down to no techs, but failures will never be totally eliminated.

    The BMW i3 CONCEPT CAR Sure getting the battery modules to report to the central HV module wirelessly could eliminate some complexity, while it creates yet another entire level of it. How tolerant will the system be if one of the transmitters for modules fails? By design cars have to be very fault tolerant or else warning lights would come on more often then they already do. Lets talk about the receiver transmitters for each battery module. The uC Smart mesh system will essentially be connecting a Galaxy 5 to each battery section, what did that do to the cost and complexity if any part of the system acts up? Warranties aren't free, their cost is figured into the purchase price of the vehicle so trying to suggest that making the manufacturer simply guarantee it for some extended period of time doesn't help the consumer in the end. In fact it shifts the burden of the expense all the way to the front side of the purchase.

    Watson...The best they can do with that is program in historical data and it can retrieve it. The Focus failure now published could be logged in but that isn't a valid use of the information because that failure is very unlikely to re-occur and any time spent just looking to see if it happened again will routinely be wasted. What's more that failure could never be logged in before it happened and can only be logged in AFTER a technician encounters it and publishes it. But you want to get rid of the techs and if you got your way that step in the process would cease to exist meaning it would never have been published and therefore never existed in the first place to become retrievable. When engineers write trouble trees they can write the tree with regard to the CAN low wire going open and staying open. They could also write it with regard to the wire grounding and staying grounded. Those failures could be written into the database that Waston could draw from. But they CANNOT write the tree for the random failure that you saw demonstrated in the video, and they never will be able to do that simply because first an engineer would have to imagine every possible failure for every millimeter of every circuit and then design a circuit with multiple extra wires to allow a module to continually test every segment of the circuit. We had that discussion in the past and your response was something on the order that you weren't talking about adding extra wires but that is exactly what it would take to have self diagnosing circuits. If you tried to do that wirelessly you not only still have additional wires you also added the additional complexity of yet another RF signal in an environment that is already electronically speaking very noisy.

    Back to the Prius. P0A0D. When the failure occurs the car will not turn on. You can get to the ignition on, but not the Ready mode. When the problem occurs while driving, the vehicle can be driven until it is turned off and then won't restart unless the problem self corrects. If the problem didn't self correct it would be easier to figure out what is wrong, but since it does trouble trees are useless. So much for self healing circuits........

    Next????
Sign In or Register to comment.