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

engine failure

Discussion in 'Gen 2 Prius Care, Maintenance and Troubleshooting' started by bobarnold, Nov 12, 2004.

  1. jayman

    jayman Senior Member

    Joined:
    Oct 21, 2004
    13,439
    641
    0
    Location:
    Winnipeg Manitoba
    Vehicle:
    2004 Prius
    Unfortunately, you are correct in that all Prius owners are "beta testers" in the Real World. Much like my condo neigbor with the year old BMW and that insane I-Drive system, it's ALWAYS broken. I do a lot of embedded programming and saw a similar development in the Industrial Automation world when it "matured" from analog 4-20mA instruments to the first networked/bussed instruments. As an example, the Honeywell TDC3000 process control computer with LCN (Local Control Network) was durable and predictable, the process interface (Hiway Gateway) also durable and predictable. About the only scary thing was OS210.M1 with "LCN Reconnect" which would turn every operating node into an isolated virtual node. That was good for a laugh if the day was going very slowly! The first industrial instruments based on HART/MODbus, DeviceNet, ControlNet, Foundation FieldBus and, God Help You, Siemens ProfiBus DP, introduced entirely new complexities to industrial control. For the first time, outright hardware failures (Which are predictiable based on FRACAS and Weibull Distribution analysis of discrete MTBF) were superseded by software failures, which are unpredictable. Suddenly, the ratio of outright failure went up 10x! Since the software failures predated CMM (Capability Maturity Model, as developed by Carnegie Mellon for U.S. DOD), we were forced to deal with bizarre process reactions to something as "simple" as a bad punctuation mark in +1,500 lines of code. That's why you will NEVER see the bussed devices in charge of something critical, like a Chlorination plant that could spew a fatal green cloud over an entire city if the software decided to ABEND. Since almost every new car uses CANbus and LIN, we'll see more of these odd failures that are software in nature. Since iteration testing takes time and MONEY, I doubt the situation will get better, so we'll continue to be Beta testers, "secure" in the knowledge that our Black Boxes will do a stack dump when the Red Triangle Of Doom turns on and the ICE shuts off in BFE.
     
  2. DaveinOlyWA

    DaveinOlyWA 3rd Time was Solariffic!!

    Joined:
    Apr 13, 2004
    15,140
    611
    0
    Location:
    South Puget Sound, WA
    Vehicle:
    2013 Nissan LEAF
    Model:
    Persona
    im sorry but i have to disagree with the last two posts...

    i dont think the Prius is experiencing anywhere near enough problems to be considered "bleeding edge" technology. if they are bleeding edge then what is a honda accord? (in the next few weeks, they will be announcing a 550,000 vehicle recall concerning a CRTICAL safety defect)

    The Prius has had only relatively minor problems. there were also 3 other recalls in the news this weekend that are set to start within the next few weeks. so does that mean that they are all "beta" cars?? Saturn Ion?? Honda Accord??

    nope i dont think so. The Prius has been extensively tested and is good technology. realize that we, because we are first to drive the new model will be under the microscope of the automotive world. there are many out there that are praying for our failure. That, i believe, has made Toyota extra careful in their product updates for the car. (although they seem to be extra careful about most of their products and it shows. in the latest CR study, Toyota/Lexus was named in every major category for reliability gathering 80% of the top spots. there were not named in any category for worst reliability. the Prius was given top spot for hybrid reliability which as we all know is not as of yet a crowded field. I would have liked to see how the Prius stacked up against reg. ICE 's but...)

    although Toyota doesnt officially admit it, they do do beta testing in a very general sense when they do limited releases of products in japan. but even those vehicles are pretty much have been throughly tested.
     
  3. jayman

    jayman Senior Member

    Joined:
    Oct 21, 2004
    13,439
    641
    0
    Location:
    Winnipeg Manitoba
    Vehicle:
    2004 Prius
    I have been a licensed driver for 25 years, and this is the first new car that has DIED on me and left me - briefly - stranded. I have driven my fair share of junky vehicles, old and new, and although having a door handle repeatedly fall off on a brand new 1984 Ford F-150 is annoying as all h***, the vehicle is still driveable. The fact is, if prototyping and iteration testing had been done, these sort of software failures would have become apparent and, more importantly, corrected before the fleet hit the road. I'm still p***** off with GMC for my Sierra having problems with the AutoTrac 4WD system that would randomly fail, default into 2WD, and leave me stuck on the hill at my vacation property. In that situation, the amber words "SERVICE 4WD" in the Driver Information Center was of little comfort to me, so I would drive the old primitive Ford F-150 with manual lock hubs and manual transfer case (I finally fixed the interior door handle problem by welding a chunk of rebar to the actuator, it looks ugly but works). Again, I have encountered the same issues in industrial process control where the manufacturer touts their latest-n-greatest bussed system, then us pions have to try to figure out what caused the kernel panic that allowed the valve to open and dump chemicals all over the shop floor. I hope we don't end up using our Bluetooth cell phones to hook up to the Toyota Prius Update Center once a week to download hotfixes, that would get old in a hurry.
     
  4. DaveinOlyWA

    DaveinOlyWA 3rd Time was Solariffic!!

    Joined:
    Apr 13, 2004
    15,140
    611
    0
    Location:
    South Puget Sound, WA
    Vehicle:
    2013 Nissan LEAF
    Model:
    Persona
    why would downloading hotfixes get old??

    the process could be automated so you never even know its happening. if you chose to do it manually, it would probably take no more than 2 minutes of your time.

    although i doubt that the process would use your cell phone since it would only be covering a small part of the Prius population and Toyota would have to negotiate with several companies to do this. (i wouldnt wish that task on my worst enemy!!)

    people are funny in that they hate to do anything to maintain their electronic toys. i know people who spend a minimum of 2-3 hours every week pampering their cars, but wont spend 5 minutes to update their virus definitions for their computer or update the software on their cellphones.

    then the next day they are complaining about dropped calls and excessive roaming... well cellphone companies update the software that runs their cell towers every 3 months or so. to make your cellphone work as good as it can, the software in the cellphone must be updated also. things like towers added to the network, new contracts with 3rd party cell tower owners and whatnot will not be automatically recognized by the cellphone if you do not.

    these updates take about 45 secs, its a free call, and is done right from your handset...with all that, it still amazes me how many people think that its a hassle.
     
  5. bbangert

    bbangert New Member

    Joined:
    Jul 15, 2004
    7
    0
    0
    <div class='quotetop'>QUOTE(DaveinOlyWA\";p=\"52296)</div>
    I don't think you're keeping in mind the fact that the 2004 Prius has more computerized parts talking to each other than in any other car on the road. Saturn Ion and Honda Accord do not have regenerative braking and the dozens of other highly computerized functionality our Prius vehicles do. So no, they aren't beta cars.

    I'm sure Toyota does do testing, which works out obvious bugs. As a software developer myself, I'm familiar with bug testing of software. Testing is to work out critical and major bugs and if done well can usually get most of them. However, as anyone using a computer must know by now, all software has bugs in it.

    Before highly computerized cars, car failure revolved around mechanical failure, and design failure (probably more, I'm generalizing). Now and in the future, there will be electronics failure and software failure to throw in the mix as well. It should be obvious that putting in more systems that can fail, will increase the possibility of failure.

    Also, you cite testing the vehicles in Japan. I would be surprised if people driving in Japan experience the same road conditions available in the U.S. The fact of the matter is that you cannot account for every scenario in testing.

    I currently have no intention of abandoning my 2004 Prius, I enjoy it greatly. I do plan on taking some precautions on where I drive though, and planning for the worst case scenario. For example, if I'm doing a cross country trip, I'll make sure I don't stray too far from a Toyota service center for too long.

    I have high hopes that Toyota will continue to be the leader in reliability, especially with cars as computerized as the Prius due in part to their lead in this area. Being the first into a new technology generally qualifies it as "bleeding edge".

    New models will continue to have new "bleeding edge" technology in them. How many car buyers go to a dealer and say, "Well, I want a car exactly like last years, but with all the fixes you added along the way to address the remaining bugs that came up." I'm going to guess none. Every year they add features, components, doo-dads, etc. to make it more appealing than last years model.

    While you would think that starting with what they knew after the prior model will help (and it does a bit), the problem with software is that new additions, expose new bugs that either never occurred, or were never observable before the new addition. This is in addition to whatever new bugs the new additions have as well. This is the nature of software, its very complex, and its interactions extremely difficult to predict.

    I look at this is something we all might as well get used to, since every new model I see continues to promise more "integration".
     
  6. bbangert

    bbangert New Member

    Joined:
    Jul 15, 2004
    7
    0
    0
    <div class='quotetop'>QUOTE(DaveinOlyWA\";p=\"52314)</div>
    I wouldn't mind doing this. As cars have more software, they'll need updates. Makes sense to have them delivered as soon as possible, no point inconveniencing the owner by making them take the car into a dealership everytime it needs an update.

    At the same time, please keep in mind that on multiple occasions, patches released to address a problem have sometimes created new ones as well. Plus, once you have some sort of way for Toyota to feed software into your vehicles computer automatically, you're vulnerable to Toyota's computers being compromised and bad software automatically installed. Not a promising future in my opinion.
     
  7. DaveinOlyWA

    DaveinOlyWA 3rd Time was Solariffic!!

    Joined:
    Apr 13, 2004
    15,140
    611
    0
    Location:
    South Puget Sound, WA
    Vehicle:
    2013 Nissan LEAF
    Model:
    Persona
    that is why updates will most likely continue to be done at the dealership since the average owner has no easy way of downloading the software off the net and doing their own updates, but i see that as the next major change.

    i think in a few years we might see the option to buy the option that allows downloads from the internet. the technology is already here.

    also on my post about the other cars currently experiencing recalls. i am not comparing them in anyway to the Prius. i brought it up to illustrate that nothing is perfect and problems will happen and that the Prius is still new but has met the challenge very very well. they are much more complicated than nearly every other car on the road and still has managed to keep major problems to a minimum.

    now that does cut both ways. the ion and the accord have been on the road much longer in much larger numbers. the Prius' record of no major problems may simply be biding time until a defect is found. after all, the new version just had its first birthday. in the ion's case, they are looking at defects from 3 years ago.

    who knows, after 3 years, the 2004 Prius may have revealed a major defect also. i hope not and things are looking good, but up until this past weekend, things were looking good for the accord too.
     
  8. jayman

    jayman Senior Member

    Joined:
    Oct 21, 2004
    13,439
    641
    0
    Location:
    Winnipeg Manitoba
    Vehicle:
    2004 Prius
    Well, what set me off on this topic is the very real problem I had when my one month old Prius, with around 3,000km on the odometer, DIED on me. I have been a licensed driver for 25 years and have never had a new vehicle outright DIE on me.

    I'm not clueless when it comes to software development, as I have a lot of embedded programming experience with industrial process control, Class 5 phone switches, and TDMA & GSM cell systems, in addition to CMM methodology.

    There are two ways to ensure delivery of absolutely reliable systems: 1. Do extensive prototyping and iteration testing, which is time consuming and very expensive; 2. Let the automated test tools catch the "obvious" traps and let the end user catch the rest.

    Last time I checked, I paid for my Prius, I'm NOT being paid to be a Beta tester. It's ludicrous to have to worry about how far you are from the nearest authorized Toyota Prius dealer. In my case it was a 10 minute drive, but what if I had been in the middle of BFE? I no longer trust my Prius. Given Toyota's reputation, I had expected far better.

    The Prius-certified mechanic at my local Toyota dealer has no problem admitting he is quite intimidated by the Prius, and feels that the drivers are "guinea pigs" for driving them. That's not something I would expect to hear. Judging by other posts, the dealer lack of understanding is a very real concern and very troubling as well.

    If the similar learning curve I witnessed in Industrial Process Control is any indication, the problem will get worse before it gets better. Basic reliability testing, either of discrete components or entire systems, should be predictable according to Weibull Distribution Analysis and FRACAS, and CMM requires understanding of these topics to meet CMM +4 levels.

    The only other system I can think of with the same intimidating level of integration is a fly-by-wire civilian aircraft, such as the Boeing 777 or the Airbus family. Airbus has had a few accidents that, upon close examination, were caused by the pilot, such as CFIT. I'm unaware of an actual software or hardware problem causing an accident, which is credit to Honeywell for their stringent testing.

    Given the current state of RTOS development with Sun's Chorus, Wind River VXWorks, or QNX Neutrino, one would expect very high standards and not expect something like the Red Triangle Of Doom.
     
  9. IsrAmeriPrius

    IsrAmeriPrius Progressive Member

    Joined:
    May 27, 2004
    4,333
    7
    0
    Location:
    Southern California
    Vehicle:
    2005 Prius
    jayman, paragraphs are good and make reading the posts easier.
     
  10. jayman

    jayman Senior Member

    Joined:
    Oct 21, 2004
    13,439
    641
    0
    Location:
    Winnipeg Manitoba
    Vehicle:
    2004 Prius
    ?? I did have paragraphs. Why is it one solid paragraph - let me try to edit it again.
     
  11. Frank Hudon

    Frank Hudon Senior Member

    Joined:
    Jan 11, 2004
    4,147
    19
    0
    jayman tell exactly what happened just prior to getting the triangle of death. Running rough? just a clean break? or any other thing that come to mind. Level road, rain, just washed the car in the last month or so, just filled up? girl friend burped what ever.
     
  12. jayman

    jayman Senior Member

    Joined:
    Oct 21, 2004
    13,439
    641
    0
    Location:
    Winnipeg Manitoba
    Vehicle:
    2004 Prius
    The car gave absolutely no warning whatsoever. The first start in the morning was smooth, and the motor has NEVER run rough.

    After the red light, the ICE started normally at around 35 km/h (It was imperceptible as usual). I was following city traffic flow of around 45 km/h when the car gave a lurch, a tone went off, and the following dash indicators came on:

    1. Check Engine
    2. Hybrid problem
    3. Master Warning "Triangle of Doom"

    In addition, the MFD displayed the icon for Hybrid Problem, with the word "PROBLEM" beside it.

    I was about 450 meters from the next intersection, so I pulled off on HV alone with the battery at 2 bars. I tried the power off / power on sequence four times and the ICE refused to restart, so I called Roadside and the rest is history.

    Oddly enough, after waiting almost half an hour for the flat bed wrecker, the ICE did restart when we had to move the car to clear a parked semi. This is something the Toyota tech cannot explain and he has received NO help from Toyota about why the motor would decide to restart after half an hour. He did verify it had NOT overheated, and the inverter temps had been normal.

    At the time the tank was around 1/2. The car has run on Mohawk ethanol 90 from new, and Toyota immediately ruled that out as a problem. I even suggested they pull a sample from the tank to verify that, and later on went to the Mohawk station to see if they have had any complaints. Again, the local Toyota dealer said Mohawk was NOT to blame.

    I washed the car in the underground parking wash bay about two weeks before. We had a nice break in the weather and the car had stayed clean. I asked the tech about water in the connectors and although he said he would check all the connectors, he doubted it was that. The connectors are WeatherPak sealed.

    To sum up, it was the same driving cycle under the same approximate conditions. I do know my local tech is very frustrated over this, especially with Toyota.

    My local Toyota dealer had hoped, as I was their first Prius customer, to get a lot of positive comments and interest in the car. To be sure, the Prius attracts attention wherever it goes. However, when somebody asks the inevitable "Is it reliable?" question, I honestly have to relate what happened to me.

    I'm planning a drive over the Xmas period to some good friends in North West Ontario. For some reason, this is REALLY worrying me, as cell phone coverage is very spotty and primitive. I may end up renting a car or pickup for the trip.
     
  13. DaveinOlyWA

    DaveinOlyWA 3rd Time was Solariffic!!

    Joined:
    Apr 13, 2004
    15,140
    611
    0
    Location:
    South Puget Sound, WA
    Vehicle:
    2013 Nissan LEAF
    Model:
    Persona
    too bad they didnt actually test the gas in your tank. contamination of gas isnt all that unusual. i wonder if the Prius might not be a bit more sensitive to contamination
     
  14. jayman

    jayman Senior Member

    Joined:
    Oct 21, 2004
    13,439
    641
    0
    Location:
    Winnipeg Manitoba
    Vehicle:
    2004 Prius
    Well, after I picked up the car, I drove the tank down to one bar, with no problems whatsoever. I refilled at the same Mohawk station, and no problems.

    The nice but frustrated local Toyota tech took the car on a 20 km test drive before he turned it over to me, just to make sure it was working ok, he even had his scan tool with him.

    He told me if it had been bad gas, the ICE would have died before he left the dealer lot, as he has had experience with bad gas.