TMC is an independent, primarily volunteer organization that relies on ad revenue to cover its operating costs. Please consider whitelisting TMC on your ad blocker or making a Paypal contribution here: paypal.me/SupportTMC

Speculation: is torque sleep the cause of reduced power safety failures in .167?

Discussion in 'Model S: Driving Dynamics' started by yobigd20, Feb 27, 2015.

  1. yobigd20

    yobigd20 Well-Known Member

    Joined:
    Oct 28, 2012
    Messages:
    5,793
    Location:
    Skaneateles, NY
    So now that there's been many widespread reports of the frequent lost power/pull over now that seem to be tied to P85D's in range mode on .167, and service centers seem to be swamped with affected cars (also no loaner vehicles available) and tesla seems to have confirmed the fault exists given their responses to owners inquiries ... now the question remains what is the real problem?

    I think it's a valid assumption that this is all due to a flaw in the new torque sleep. If I recall I think .167 had some new torque sleep features (given the pics someone posted somewhere about how range mode on enables torque sleep). They probably tried to rush this out too fast since it was a hot topic and in high demand. However, the big question that remains for me is whether or not the flaw is in the software and can be corrected, or is it a more serious hardware design flaw in that the front motor that it's just not up to the task. Or perhaps something like once power is cut to the rear there are issues restarting it. Who knows. Are there any S85D's experiencing this or just P85D's?

    I think it's entirely possible that Tesla pushed this out way to quickly and didn't long term soak test torque sleep and that the front motor itself might not actually be able to handle the power required or its being overheated or something like that.

    If it's really the case then I would think there's two solutions: either torque sleep goes away permanently or Tesla's going to have a major recall to put more powerful motors up front or fix whatever hardware/electrical/chips may be faulty.

    Or maybe none of the above and it's really just firmware related and bad programming by their engineers.

    In any case, I'm very disappointed in the recent software bugs introduced in 6.1.xxx. Brakes, steering, and power safety failures disabling the vehicle. It's alarming and concerning knowing they can screw up critical systems overnight by pushing faulty software to the vehicles. They are accountable for these. I fear the possibility of them accidentally bricking the entire fleet by pushing out their software too fast or of course worse, critical systems failing while in motion. They really need to improve their software process to prevent any of these issues in the future. I'm still just utterly shocked these kinds of bugs made it to customers across the fleet. I know this is new territory for all (continuous software updates in auto) , but there's really no room for error here. Unfortunately it doesn't seem that Tesla understands this yet.

    I know we are all excited here about the car but everyone needs to take a step back here and recognize that human life is involved here and careful steps need to be taken to ensure all of the above never happens again on any future release. I'm reading reports of cars shutting down on busy highways with no room to pull over and stop. This is seriously dangerous. God forbid something horrible happens as a result of a software flaw from an immature or lack of good software processes by Tesla.

    Also, I hope that EVERYBODY who experienced this reports these incidents to be NHSTA. That is exactly why they exist.
     
  2. AmpedRealtor

    AmpedRealtor Active Member

    Joined:
    Jun 30, 2013
    Messages:
    4,688
    Location:
    Buckeye, AZ
    Would you be willing to publicly tweet something to Elon Musk?
     
  3. wk057

    wk057 Senior Tinkerer

    Joined:
    Feb 23, 2014
    Messages:
    4,713
    Location:
    Hickory, NC, USA
    I've stated my feelings about NHSTA reports in another thread. Personally, I don't see the point in this case. They've acknowledged the issue and are working to correct it. If that weren't the case or turns out not to be the case, then perhaps another route should be taken.

    The NHSTA has no ability to get Tesla to fix this issue any faster than they are going to anyway. Their (NHSTA) track record for getting things taken care of generally drags into the hundreds of days time frame, which wouldn't be acceptable here. On the contrary, the time, money, and resources spent by Tesla dealing with the NHSTA on this would be much better spent improving other things. By the time the NHSTA did anything about this we'd all have firmware v7.2 with valet mode, autopilot, the bug fix for these issues in v6.1 (2.2.167), live supercharger status, and more.

    I for one see no reason to cause more trouble than there already is.
     
  4. yobigd20

    yobigd20 Well-Known Member

    Joined:
    Oct 28, 2012
    Messages:
    5,793
    Location:
    Skaneateles, NY
    And say what? "Hey what's up with the random cars shutting down on the latest software?" Or "might want to check on your engineering team?" Or "do you have someone competent in software QA that works for your company? ..might want to put out a few reqs" or "hey noticed that your software engineer salaries on glassdoor are extremely low...maybe that's why there seems to be a lack of talent regarding major bugs making it through into customer vehicles". Or "hey maybe you should focus on increasing software quality rather than embedding James Bond Easter eggs" or "help! I've fallen and I can't get up"

    Seriously though. I'm sure he knows about it already and hope this one's put him over the edge where he starts cracking down and holding people accountable for the faults and making sure processes are put in place to significantly reduce the chance of it ever happening again.
     
  5. Firewired

    Firewired Member

    Joined:
    May 23, 2013
    Messages:
    367
    Location:
    San Antonio, Texas
    As someone that had a similar issue with the Focus EV, I echo the sentiment that NHTSA is a waste of time. I made a complaint regarding the Focus EV when Ford was receiving multiple similar complaints of just loosing power for no reason and Ford would not acknowledge there was a problem. It took the NHTSA over a year of complaints before they issued anything. It would make no sense for Tesla not to want to fix the problem as quickly and make this a priority. By the time NHTSA ever did anything we will be many software updates from where we currently are.
     
  6. dennis

    dennis P85D

    Joined:
    Jul 26, 2012
    Messages:
    1,214
    Location:
    Silicon Valley CA
    While I am not disputing the severity of this problem, your speculation could have been reduced by doing a little research on TMC. Torque sleep was introduced in very early February in release .139, which was quickly followed (for some) by .140. Most P85D owners reported much better efficiency in highway usage on both releases, comparable to their prior experience with the S85/P85/P85+. There was a minor NVH issue a low speeds that some reported. So the problem introduced in .167 was not because "They probably tried to rush this out too fast since it was a hot topic and in high demand." And there is no indication of "a more serious hardware design flaw in that the front motor that it's just not up to the task" since torque sleep was working successfully in both .139 and .140.
     
  7. dirkhh

    dirkhh Middle-aged Member

    Joined:
    Jul 7, 2013
    Messages:
    3,645
    Location:
    Portland, OR, USA
    I don't think we have enough data to answer that question. My guess is that Tesla is mostly busy trying to collect the data that can answer this question.
    The electronics that drive the dual motor setup are very tricky, adding torque sleep (and making it still responsive) has to make it even more so. My guess is that it's an unintended interaction of various pieces there, and whether or not the change that triggered it has anything to do with tuning torque sleep (as that had been available since .139) is not something you can tell with going through the source code and tons of data logs. Not something we can do.
    So while I will openly admit that I am just guessing as well, I don't think your conclusion is justified based on the data we have access to.
     
  8. AmpedRealtor

    AmpedRealtor Active Member

    Joined:
    Jun 30, 2013
    Messages:
    4,688
    Location:
    Buckeye, AZ
    wk057 posted in the other thread that the new .176 and later releases may have the bug fix, so this may all be a moot point in a day or two.
     
  9. yobigd20

    yobigd20 Well-Known Member

    Joined:
    Oct 28, 2012
    Messages:
    5,793
    Location:
    Skaneateles, NY
    Well that's good at least.
     
  10. jewelerdave

    jewelerdave Member

    Joined:
    Feb 8, 2015
    Messages:
    13
    Location:
    Fort Collins, CO
    I for one am glad that any problems that could be induced can also be corrected by a software update. beats having to go back to a service center or a dealership to have work done due to a faulty part. Lines of code are much easier to fix and reload. I dont feel anyone who does any wrighting of software intentionaly puts bugs in the system, things can be corupted or done by error as it is written by people. to think that anyone or that any team cant make a mistake is foolish. If a problem shows up, that is why it is fixed.

    I am sure it is not lost on the programers that peoples lives do depend on them. However ICE cars have failed for decades for unknown reasons or break downs. Often and most times they slow to a stop in the most safe place they can. Any car can have that happen...its worse in an aircraft or space craft however, cars. not so bad. not fun, but not often life threatening.
     

Share This Page