Run-away I-PACE, really?

[Tony Hoyle]

Brake sensor failure normally means no _brake assist_. It just feels like no brakes when you're used to it.

My wife has twice had sensor panics in our Volt where the car slipped on light snow and then all brake assist was lost, and it took a lot of force to stop, so felt like no brakes. (2nd time she knew what to do, and how to get it to reset.)

The leaf used to release the brake assist if you went over a sufficient pothole. Felt like the brakes had failed.. There's a place here where I could trigger it easily... quite a busy bit of road too so nearly went into the back of someone the first time it happened.

Apparently it was a design feature something to do with ABS, but I've never had any other car do it..

 

[rincewind]

Some years ago a neighbor of mine put her car in reverse by mistake, pressed the go pedal, panicked and pressed it even harder. Rammed into the front of my car so hard it shoved it in to the car behind. It happens. A lot.

 

[tsh2]

Throttle by wire typically uses two sensors, producing two distinct voltages in opposition, running on two separate circuits, with a failsafe mode of 'off'.

You didn't understand the Toyota situation then. Sensors were fine, but the software stopped checking the sensor information, and the failsafe didn't work. Pulling the handbrake is assumed to have been a workaround, this should have triggered an independant failsafe, but nobody would expect drivers to know this. Automotive safety design works on an acceptable rate of failure, not absolute guarantees.

 

[Dangerous Fish]

The leaf used to release the brake assist if you went over a sufficient pothole. Felt like the brakes had failed.. There's a place here where I could trigger it easily... quite a busy bit of road too so nearly went into the back of someone the first time it happened.

Apparently it was a design feature something to do with ABS, but I've never had any other car do it..

Many modern ABS systems on all types of car will have issues like that in specific circumstances. If the wheels momentarily lose grip the system can get confused and the driver will suddenly find the brakes aren't working in an expected way.
There's a very good example and explanation of this here:

It's also a reason why every driver should take their car to a skid pan at least once. They will learn a lot about how their car behaves and feel how the brakes work (or don't) when there's little or no grip. A lot of drivers don't know what ABS is, how it works and how the brake pedal feels when it engages. When it does, they think something's wrong with the car.

 

[init6]

Some years ago a neighbor of mine put her car in reverse by mistake, pressed the go pedal, panicked and pressed it even harder. Rammed into the front of my car so hard it shoved it in to the car behind. It happens. A lot.

In one of the Vision videos they mentioned that AP prevents this from happening in 40 incidents per day. I'm pretty sure there will be a lot more that AP didn't catch and that's just one maker.

 

[Roy B]

In one of the Vision videos they mentioned that AP prevents this from happening in 40 incidents per day. I'm pretty sure there will be a lot more that AP didn't catch and that's just one maker.

That sounds like Obstacle-Aware Acceleration. It goes hand in hand with Automatic Emergency Braking. Most new cars have it now, I think, for the forward direction. I have a feeling Tesla might implement it now in reverse, too, which is pretty unusual, I think.

 

[pgkevet]

I'm agreeing with you even if it didn't come across like that. My main point is that we often read about cases where something has gone wrong with the car and it leads to something unexpected happening which the driver, for whatever reason, can't cope with and then calls the car "undriveable" when in fact it's not as long as you have the ability to deal with the situation.
Those sorts of comments then make other owners fearful that that the same thing will happen to them and they won't be able to cope, when in fact a lot of them will be able to if they understood the situation and were presented with the facts ahead of it happening.

A good example is a wheel speed sensor failure on a Tesla. That throws up a load of errors, reduces power steering and disables ABS and TC. To some drivers, that renders the car completely undriveable and they'd abandon it in the street. To others, they'll just drive it to the service centre to get it fixed (or in my case, fix it myself because I can't be bothered to sit for 4 hours in a SC for a 10 minute job and a £25 part).

Another one is the often quoted unintended acceleration in a Tesla, which has never been proven to have happened. Ever. One reason it won't happen is that there is a safeguard built in which reduces power and warns you if you press both pedals at the same time. So you will always be able to brake the car to a halt if you press the brake pedal, even if something is pressing on the accelerator pedal and you're not very strong. The other reason it's always implausible is because there are two opposing potentiometers (sensors) on the accelerator pedal itself so if one were to fail or go short circuit, the other would be the failsafe. If any of the signals controllong the car's power delivery were faulty, it's going to error out and reduce or cut power anyway.

Yes, there will be circumstances particularly in older cars where something fails and you have no brakes or can't stop the car, but that's not what this thread is about. This is about a modern EV with lots of redundancy and failsafes. If JLR come up with an explanation which proves there is a failure mode where an iPace cannot be stopped I'll hold my hands up as being wrong, but a report on the BBC and a comment from Z-list ex-Big Brother celeb. doesn't make it a fact.

It all still remains weird. Driver panic etc is easy to understand but the original article states 25 minutes to police bringing car to a halt (quicker response than needing an emergency ambulance for a sick relative ). Surely enough time for even the most panicked driver to have figured things out. Even weirder to have managed to thread the car through traffic for that long...

 

[init6]

That sounds like Obstacle-Aware Acceleration. It goes hand in hand with Automatic Emergency Braking. Most new cars have it now, I think, for the forward direction. I have a feeling Tesla might implement it now in reverse, too, which is pretty unusual, I think.

They did say both forward and reverse, but it does go to show that it's a common occurrence if manufacturers feel the need to implement it.

 

[Regulo]

There are plenty of videos on YouTube of cars crashing through shop fronts, smashing up parking lots, driving through garage doors etc. to show that getting pedals and/or gears in a muddle isn't too rare an occurrence. In my early driving experience, I nearly did exactly that. In an auto gearbox car I was letting it roll on tickover into a parking bay. As I reached the limit of the bay, I hit the brake - except my foot was over the accelerator. Luckily, there was an empty bay ahead, so no damage to anything except my confidence. Quite shook me up, actually.

 

[Pink Duck]

People seem to misunderstand the classical Obstacle-Aware mitigation, which simply cuts motor power to half or so when car thinks impact is imminent, for the purposes of reducing the amount of body shop effort. The new occupancy network up until FSD v12 is the real anti-crash system, though oddly seems to be ignored in the latest videos in favour of human driving NN training.

I’ve also experienced some terrifyingly close and actual impacts when expecting to go forward but car software gear selection glitching with rear camera display off hiding the fact that rapid reverse is about to happen. If sight is lost of gear selector on screen (easily done on S/X) then all bets are off.

 

[FatFalcon]

You didn't understand the Toyota situation then. Sensors were fine, but the software stopped checking the sensor information, and the failsafe didn't work. Pulling the handbrake is assumed to have been a workaround, this should have triggered an independant failsafe, but nobody would expect drivers to know this. Automotive safety design works on an acceptable rate of failure, not absolute guarantees.

I know this was a theory, but I don't believe this was ever proven. There was a $1M prize for anyone who could replicate the Toyota issue, went unclaimed. There were various news reports which later turned out to have been staged. Do you have a link?

 

[tsh2]

I know this was a theory, but I don't believe this was ever proven. There was a $1M prize for anyone who could replicate the Toyota issue, went unclaimed. There were various news reports which later turned out to have been staged. Do you have a link?

No links, except the presentation I linked earlier. A broken software path is sufficient, demonstrating it is not a necessary step from my point of view. The software is closed making reverse engineering and injecting ram faults kind of tricky. We know that logic faults occur in CPUs, people put a lot of effort into detecting this in safety critical scenarios - but even then there is no guarantee to detect everything. The only unknown is how often an unrecoverable fault occurs per million miles driven.

 

[Dangerous Fish]

Interview with driver:

A lot of it doesn't add up to me, but you be the judge if you are bored enough to listen to it all.

 

[Obie]

of course... reached 120 mph and stopped by other cars at 90 mph.... could not put into neutral but then could. car going to jaguar dealership and not R&D for investigation )))))

but what is most interesting is...
that he talks about this happening 3 AM... and BBC article is about stopping it at 2 PM.

It also happened like week ago, and he talks about Christmas.
it is either 2 different cases either BS. and I am sure it is

Agree, after 30 minutes of driving he manages to get it in neutral just before he is about to ram the police car!

 

[Tony Hoyle]

He can't keep his story straight. I expect after an investigation he'll be quietly charged with dangerous driving.

 

[Pink Duck]

I-PACE on-finance young owner Nathan contacted: Police, Jaguar, Insurance, Finance, Press
Psychological state: Denial “Weird”
Two instances: First time in December 2023 (before 25th), second time March 2024

Dec 03:00 work→home motorway near Trafford-Centre, off M56, happened within 1 minute, little traffic
> Car started to speed up, pressing brake to floor, “wasn’t happening”, phoned police 17–18 mins, looked forward and steered
> Bad wind, chucking it down with rain, pedal to floor acceleration, 120 mph peak vs 2 BMW estate motorway cops flashing blues
> Police operator helpful, asked about handbrake, asked about neutral via button (not lighting up), hit harder (lit→neutral→slowing), before contact box
> Unsure of distance, perception of hours, 20–25 minutes admitted, ended between hard-shoulder and first lane, Jaguar I-PACE ‘conked out’ before stop
> Motorway cop heard of this happening before through work colleagues (Scotland MG)
> Insurance company recovered the vehicle and him, disconnected 12V as precaution, Jaguar Bolton (3–5 visits with battery SoC-loss problems), unable to find fault after 3 days (or on previous occasions) with “no care”. 1st day no action, only 1 electric technician (Bolton/Manchester/Rochdale). Lost charge on way home excessively. Started to trust the car again, 230 mi from 100%, Sport range>Eco claimed (cooling related?)

March 14:30–15:00 work→home, traffic building
> Car fine in morning (1st day of new job), Wednesday home-bound, indicated to lane 3, car shunted forward full, brake pedal having no effect
> Phoned police again immediately, advised moving to hard shoulder, lane 1 if blocked
> Matrix police cars with 'Follow me' indicated to exit off motorway to lay-by/off-network, contact with police car in front made (5-series estate)
> 90–100 mph, other drivers staring/beeping, “couldn’t think”, operator asked about range (50–60 mi in morning), 20 mi remaining
> Police opt to keep on motorway until out of charge, remained in front throughout, limp mode expected but inactive, ran out in lane 3/4
> Button to open recessed door handles worked, given water to calm down, advised to get on safe side of barrier, 7–8 police vehicles surrounded
> Unable to see damage because car wedged into lead police vehicle
> He was taken off-network, police took vehicle to impound, Jaguar asked to release/investigate, police refused (probably diagnostic evidence recovery)
> Made no reference to attempting to put car in neutral
> Garage couldn’t believe it happened again, “were shocked”, alleges problem not logged in diagnostics last time
> Jaguar provide new Range Rover Sport fossil during
> Police ask to forward speed camera offences to them to cancel
> He couldn’t remember top speed reached

 

[Fursty Ferret]

> Police ask to forward speed camera offences to them to cancel

How... convenient.

 

[Dangerous Fish]

What's the betting he wanted to get rid of it because of various problems and bad service. Tried the runaway story before Christmas but didn't get enough media attention so had another go and no gets the clicks he wanted.
Would anyone here honestly get back in a car which truly did:
1. Accelerate to 120mph without driver input
2. Have complete brake failure
3. Disable shift into neutral

No, of course not, because it didn't and doesn't happen.
Also, not being able to open the doors from the outside is called 'central locking'. It's been around a while.

 

[yessuz]

I am sure that this will be a combination of wanting to get rid of the car and avoid fines.

if it sounds like BS, looks like BS and smells like BS... it is a duck..

 

[Dangerous Fish]

I am sure that this will be a combination of wanting to get rid of the car and avoid fines.

if it sounds like BS, looks like BS and smells like BS... it is a duck..

Could be a bull....