6 ms·
>Toyota models, including Lexus, have figured in almost half of the carbon monoxide fatalities and injuries identified by The Times. Toyota says its keyless ign
by schiffern 8y ago
>Toyota models, including Lexus, have figured in almost half of the carbon monoxide fatalities and injuries identified by The Times. Toyota says its keyless ignition system “meets or exceeds all relevant federal safety standards.”
That's quite significant, and I wonder why that is?
I have a hypothesis:
Personally I've had problems with Toyotas (specifically, Prius GenIII and Camry XV50) where trying to shut off the car too fast, eg using "muscle memory," results in the car not shutting off.
At first glance (and according to the manual), the steps to shutting off the Prius / Camry appear to be:
1. Shift the car into Park.
2. Press the ignition button.
3. Exit the vehicle.
Simple, right?
Well, it's not. If you don't wait ~1 second between step 1 and step 2, the car has a race condition where it doesn't realize that you already put it in Park, so it displays a "Shift to Park" message and fails to shut down. To fix, you have to shift the car into Reverse (or really, Anything-But-Park), shift it back into Park, and press the button again.
This sequence is extremely confusing, especially if you don't notice the error message. Neither conventional shifters nor conventional key-based ignitions make you wait -- when you put it in gear, it's in gear!
Toyota's system has hidden state that's related, but separate from the physical movements of the shift lever and the start button (unlike a conventional key and shifter).
So if you want to reliably shut off the car, here are the actual steps to shutting down a Prius / Camry:
1. Shift the car into Park.
2. Wait a beat (for the vehicle to realize it's in Park).
3. Press the ignition button.
4. Exit the vehicle.
The proper way for Toyota to do it would have been to buffer the inputs in order, so the same sequence of actions always results in the same result, and the mere act of speeding up one's muscle memory doesn't result in unexpected operation (ie, acting just like the old, more-easily-predictable physical controls).
- ataturk 8y agoBMW: Pull up and park, press stop button. Get out, lock door, and walk away. Easy. The car puts itself into park, etc. The only problem is when I get into other cars that don't do this. I honestly don't like the all-electronic shifters. I feel like the car could end up in state in which shifter inputs get ignored and then I would be unable to shift into neutral or otherwise stop the transfer of power to the wheels. I also happen to think that intelligence agencies know how to assassinate people by taking over their cars and causing runaway conditions that the driver is unable to negate. A mechanical kill switch like on a motorcylce would be a great feature, as long as non-assisted brakes and steering would still function. I know that modern aircraft separate all these linkages electronically now, but I also know that modern aircraft have triple redundancy and that cars don't.
- mrguyorama 8y agoAre you saying it's impossible to shut off the car while not in park? I find that odd, and will certainly have to try it on my mother's Toyota when I get the chance
- singingboyo 8y agoYeah, that seems odd to me. I drive a Nissan, not a Toyota, but it will shut off if I hit the button before going into park, then beep like there's no tomorrow until it's in park. Though it occurs to me that the old Dodge Caravan I drove before may have required being in park to turn off, though it used the key for ignition, so that made sense.
- schiffern 8y ago>Are you saying it's impossible to shut off the car while not in park? Yes! The vehicle faithfully ignores you when you press the Start button, beeps softly, and displays a message: "SHIFT TO P POSITION."[2] As I said, in my experience there's a ~1s delay between shifting to Park and being permitted to turn the car off. You get the same error message, but it's already in Park. The result? Performing the same sequence of actions at a different speed results in a very different outcome. And the Park button (yes I know) is very close to the Start button, so it's easy to perform the two actions in quick succession. Classic race condition. Btw, you can't just press Start again. Even though it's now "in Park." It's in some strange, poorly defined, in-between state, not Park but not Not-Park. To recover, you have to press the Park button again, and then press the Start button again. From a software perspective, this makes me uneasy. It would be easy to miss such the subtle cues indicating an error (soft beep/message/IC doesn't down). Especially for older people who are hard of hearing, and just pulled into the garage from a sunny drive (bright pupil adaptation takes longer to dilate for older people). From the Prius GenIII manual[1], pp549 > * Stopping the hybrid system >Set the parking brake, shift the shift position to P and press the “POWER” switch as you normally do when stopping the hybrid system. Setting the parking brake is optional, but shifting to Park is not. There's also a confusing chart with all the warning messages that should explain this behavior, but doesn't. It's on pp525, and it seems to have evidence of incomplete revision: >The driver’s door has been opened with the shift position in a position other than P and without first turning the “POWER” switch is OFF. There is no power switch, just a button, and the grammar is wrong. Hmmm... Curiouser and curiouser. [1] https://priuschat.com/threads/2012-2015-prius-owners-manual-combined-pdf.149300/ https://priuschat.com/threads/2012-2015-prius-owners-manual-... [2] Btw, googling for "shift to P position" indicates that both Hyundai and Kia have this interlock. Whether they're subject to the same race condition is unknown.
- pwinnski 8y agoWeird. With my 2014 Prius, I pull into a parking spot, brake, and hit the power button. It shifts into Park for me. At least I think that's what I do. Now I'm doubting myself and will check later today. But I'm pretty sure.