Sudden hard acceleration starts ICE, but it won't turn off - why?

Discussion in 'Clarity' started by bbct, Apr 16, 2019.

To remove this ad click here.

  1. bbct

    bbct New Member

    Thanks for all the testers out there. @The Gadgeteer, good idea testing on an empty charge as well as full. It appears that perhaps there is reasoning behind the behaviour, we just may never know what it is.

    I also noticed the bug where the arc remains white after fast acceleration, even if the engine does turn off. Be nice if they fix that soon.

    The other day, fully charged (I just unplugged), I accelerated quickly less than a minute into a short ride to the store. The engine came on and stayed on the full 5 minute drive, even continuing to run at a red light. Said it was charging the battery on the display. Interestingly when coasting and the ICE running, the display shows charging from the movement of the wheels, but not from the engine even though it was running. The engine never went above an idle speed, but didn't turn off either when coasting or stopped at a light. When I got to the store though, and pulled into the parking spot, it actually turned off the ICE (before I put it into park). So I guess it wasn't going to run "forever"...

    I don't think any of us can really guess right at what the algorithm is, but it seems there is one. Perhaps it is cooling the battery after that surge of amps for 3 seconds of fast acceleration. Perhaps there is some other purpose, who knows. It does not seem to harm anything, and it seems if you wait long enough (whatever that is) it will go back to its "normal" mode of operation.

    I'm just going to chock it up to a single slightly annoying feature that is unlikely to adversely affect the car or its efficiency in the long run. When it first happened I couldn't find any mention of it, so I'm glad I started this thread, at least it is now documented, though we don't have much clarity about why. (Sorry had to throw in the pun :) ).
     
  2. To remove this ad click here.

  3. KentuckyKen

    KentuckyKen Well-Known Member

    FWIW, the manual on p 13 is the only place I have found that addresses the issue about the white line replacing the blue line on the power gauge.

    It says that during a System Check the curved blue line will not appear until after the System Check is completed even though the EV may be on. So if it only does this intermittently when punching it, it could be a System Check. But if it’s doing it all the time then I would agree it’s a bug that needs an extra line of code or something to correct it. BTW, I saw the same thing when I got what I assumed was the limited regen protect from overcharging by starting the ICE scenario when battery is full(would someone please come up with a one word name for that!). It drove normally like everyday EV but with no curved blue line until I restarted at my destination.
     
  4. Tim66

    Tim66 Active Member

    OK, I decided that maybe the reason the ICE was not shutting off was because I wasn't letting the ICE warm up enough, I wasn't gripping the steering wheel properly or I wasn't chanting I believe, I believe. So I left the house with the battery 1/2 full, in EV mode and 3 minutes later on the freeway I pushed the pedal past the detent and the ICE came on. 1 1/2 hours later it was still on even though the display said ECO mode, I had not touched any buttons and had not gone past the detent again. I pulled into my destination, turned the car off, immediately restated it and the ICE was not running. On the return home I did the same experiment and the ICE never shut off. I talked to the Honda service rep and he has no idea what the problem is nor have any other customers had the same issue. But he said to bring it in. So I will once again. I miss my Volt.

    Tim
     
  5. MNSteve

    MNSteve Well-Known Member

    Here is my informed and uninformed opinion on this issue. It's informed because I have a few decades of experience working with computers. It's uninformed because I have no visibility into the algorithm that controls the car and the parameters that drive that algorithm.

    Computers are deterministic. They are 100% controlled by the program that runs them. These programs can use hundreds of inputs to determine their actions. Since the system can be very complicated it may appear not to be deterministic. But it is.

    I believe there are two factors at work here. First, it's a complicated system that uses lots of inputs, and we can only speculate at what they are. I joke about relative humidity affecting operation; in this case it just might. But consider all the items we know - state of charge, accelerator position, run time of the ICE during the past unknown time interval, temperature, driving history - and then consider that there are others that we haven't deduced. With all these inputs it is very difficult to predict behavior with no actual knowledge of the algorithm in use.

    But second, this is version 1 of the software that controls the car. It is a given that there will be bugs; we've seen patches for a few of them in the form of service bulletins. Some of them may have the effect of corrupting history data, which would explain why rebooting the car sometimes has such a huge effect on the way that it runs.

    For me, the bottom line is that I accept that the operation of the car, including when the ICE is triggered, is not something I will ever be able to predict. For example, it remains a mystery to me how I can drive the same highway at the same speed in the same weather conditions and observe wildly different behavior of the gear icon and intermittent drops into EV on different trips. I have come to terms with it; it is what it is, and as long as it doesn't materially affect the operation of the car, I'm not going to fret. If I had symptoms like some have reported where the ICE comes on for long periods, I might fret. In this case I would invest in the reboot; maybe I missed it in this thread but I have not seen any of these specific reporters take this step and tell us if it affected the unexpected behavior.
     
  6. Tim66

    Tim66 Active Member

    Steve, the material "affect" of the ICE running constantly for me is the use of gas instead of electricity to power an electric motor. I have rebooted the car twice. Once to fix the stuck HV Range display and once to try and fix the ICE always on issue. The HV Range reboot worked for a while but then it didn't and I got a second software update from the dealer. That has worked for about two months. The reboot to fix the always on ICE didn't work. The dealer allegedly reinstalled software to fix that problem and it didn't. Of course I drive efficiently and try and avoid accelerating quickly. That keeps me in EV mode with the ICE occasionally coming on for who knows what reason. When that happens the ICE shuts off on it's own. It is only when I accelerate quickly (crossing the detent line of no return) that the ICE will not shut off. It seems to me that Honda Clarity PHEV system was designed by a committee with the goal being to make its operation a mystery and counterintuitive. All they had to do was look at the Volt's system for ease of use and transparency.

    BTW my ideal car would be the Clarity with the Volt's motor and operating system.
     
  7. To remove this ad click here.

  8. Robert_Alabama

    Robert_Alabama Well-Known Member

    @Tim66, I hate this for you. Hopefully Honda can get this sorted out. I think that the vast majority of the owners of the Honda Clarity aren't experiencing anything like this. When I manage to get the ICE on due to acceleration or regen, it will run for approximately 5 min and shut down. I like the Volt operating system as well, but with the ICE coming on so fast when I really need/want it, I'd take the Honda over the Volt (but I am not plagued with an ICE that doesn't want to go back to sleep).
     
  9. MNSteve

    MNSteve Well-Known Member

    I understand. And I had missed your attempt to address the problem with a reboot. I am sorry it did not work.

    Here's another data point. Just now I managed to trigger the ICE by accelerating just a tad too aggressively on an entrance ramp; I saw the pin dip into the white area just briefly and cursed under my breath. The ICE turned off in four minutes.

    So, no, I do not have any real suggestions for how to address your issue. As others have said, I rather wonder if different cars have different versions of the control software and Honda is collecting data from us to fine tune it.
     
  10. MPower

    MPower Well-Known Member

    Since the theory is that once triggered the ICE will run until the engine warms up and then turn off, I wonder if the problem with the cars that won't turn of is a bad thermostat or sensor that can tell the chief gremlin that it's warm enough.
     
    MNSteve and insightman like this.
  11. The Gadgeteer

    The Gadgeteer Active Member

    Too bad Honda cannot do over the air updates
     
  12. To remove this ad click here.

  13. insightman

    insightman Well-Known Member Subscriber

    I prefer to know when my car's software is being updated--and to choose whether or not it happens.
     
  14. thecompdude

    thecompdude Member

    He is saying that when the ICE turns on for other reasons, it shuts off on its own. Only when it turns on due to fast acceleration, that it does not turn off.

    Not sure if he has tried this - Maybe he can try the acceleration test in the other modes - normal and sport, and check if the behavior is the same.
     
    Tim66 likes this.
  15. Mark W

    Mark W Active Member

    CT
    They could make the OTA updates so that it prompts you, and you have to accept the updates.
     
    insightman likes this.
  16. ClarityMTL

    ClarityMTL New Member

    Regen Overcharge Protection (ROP)?
     
    KentuckyKen likes this.
  17. KentuckyKen

    KentuckyKen Well-Known Member

    We finally have a winner!!!
    I vote we use this from now on or ROPES (Regen Overcharge Protection Engine Start). Either way all props to @ClarityMTL!
    What say you all?
    PS: If approved by the Clarity Brain Trust, PM me your mailing address and I will send your prize out.
    I’ll going to start a poll on this in a new thread so go vote and we’ll finally have a name for this.
     
    Last edited: May 13, 2019
  18. Tim66

    Tim66 Active Member

    I probably have but I will intentionally do it in sport mode and normal mode. I'll let you know.

    Tim
     
  19. insightman

    insightman Well-Known Member Subscriber

    The problem with creating ad-hoc acronyms is you can't google them. I tried to come up with a combination word that encompassed everything going on in this strange process but couldn't concoct a single, self-explanatory word of manageable length. Now that I believe the Clarity is doing what it's supposed to do, my old term, "premature enginulation," doesn't seem appropriate anymore.
     
    tom kirshbaum likes this.
  20. tom kirshbaum

    tom kirshbaum New Member

    LOL. I think IM (InsightMan) is right about non-Googleable acronyms, including those that get fifty Google results. As a non-engineer, non-techie, non-ev buff, I’d love a buck for every half-hour I’ve spent searching for the meaning of an incomprehensible (to me) acronym which makes me feel like an outsider — which I am. To be fair, nobody is ever flamed for ignorance on this forum and when I ask, albeit feeling like a dork, someone always answers kindly. But I wish there were a less embarrassing way to find out.
     
    KentuckyKen, insightman and Mark W like this.
  21. ClarityMTL

    ClarityMTL New Member

    Check out the thread below to cast your vote! I have also made the alternative one-word suggestion of "regenginate" to describe the action of the engine coming on due to regen breaking. As a bonus, this forum is the only thing that comes up if you Google it (so far).

    For example: "The damn Clarity is regenginating. I forgot to stop charging at 95% again! Sometimes I wish I lived in the valley instead."

    https://www.insideevsforum.com/community/index.php?threads/a-poll-on-what-to-%E2%80%9Cofficially%E2%80%9D-call-the-limited-regen-to-protect-the-battery-when-fully-charged.5703/
     
  22. insightman

    insightman Well-Known Member Subscriber

    Still not googleable. Do we need a 1st-page permalink to a list of acronyms and made-up words the hard-core members want to bandy about to save a few keystrokes?
     
    tom kirshbaum likes this.
  23. bill_m

    bill_m Member

    I agree with you that it is a complicated system, but "algorithm?" I just googled automotive realtime control systems and automotive communicating sequential processes. And there is fascinating stuff out there! it appears that we are dealing with perhaps over 70 real time processors, some autonomous, some communicating over the same bus, some communicating over different buses, all programmed in different ways. Consider the realtime requirements of whatever receives the radar inputs which must have priority in talking to whoever controls the braking/regen system which must talk to the traction motor etc. So some of the techniques must involve interrupt service routines. But all we seem to be given to know is that there are the two control units that are diagrammed in the high level honda charts, associated with hybrid operations, etc.

    I am suggesting that this is a complicated system to debug! It would be really helpful if we could get some input from folks in this forum . who have had some experience in real time operating system design.
     

Share This Page