Another runaway reported in New York

It's amazing !! ..these cars go wide open throttle right into the lawyers office .. Prius first .....Then all cars !! ..... Al Gore will be so proud !..Soooo... I'm going to just get a LEAR like him instead ..................!!

 

As a software developer I don't find the idea that there might be some sporadic unhandled fault state in an IC at all improbable. Hell the company I work for produces many of the chips and sensors that go into the Prius and virtually every other car made in the world and I've heard some whoppers from the engineers. My favorite was a linear controller that was never tested for linearity. It made it's way all the way to first silicon before they figured out that it was useless.
What people don't seem to appreciate is the layered nature of these systems. At its root is the 4:45 problem. As a developer you invariably reach 4:45 Friday. You can spend another 2 hours making sure it's right or you can say F it and drink beer. Beer usually wins. That one potential bug is no big deal, but every new generation of system inherits all the 4:45s of the preceding generation. All the 4:45s in the compiler, all the 4:45s in circuit design software, all the 4:45s in the lithography layout package etc. Sporadic problems are usually the result of the interaction of 4:45s.
That said, in the current climate ANY reported failure could be more a product of hysteria than engineering. Additionally, Toyota may not even be the party at fault if the problem is electrical. They are buying parts and assemblies from other parties. The root problem could be in silicon 5 suppliers down the supply chain. Toyota doesn't produce their own silicon.

 

I'm not sure it is worth the time to respond to such an ignorant post. For the record, my 2006 Prius experienced unintended acceleration on several occasions. There are over 500 incidents filed with the NHTSA. The US Congress is investigating. This is a real problem.

 

Your manual does mention a recorder (black box). It only stores a few seconds of data, which is enough in an accident. Toyota has refused to release the data in these black boxes after accidents. The state of Washington passed a law requiring them to provide this information to car owners. Toyota has still refused.

Some of us (at least me) offered to let Toyota put a data recorder in my 2006 Prius which experienced unintended acceleration. They didn't take me up on my offer. I have since traded the car for a 2010 which so far has shown no issues with braking or unintended acceleration.

 

The 2010's have a black box. Yours might too. Black boxes in both planes and cars only record a few seconds of data then get overwritten. That won't help in most cases. It *might* help in the NY case.

 

FWIW, (not much) black box's in planes record 30+ minutes of data from several hundred sensors.
Any vehicle with airbags has an event recorder.

 

I too am in the software business. Fortunately for everyone, I don't write code. No code is perfect. It is impossible that every possible twist an turn in complex systems can be fully tested. Because I experienced unintended acceleration in my Prius and because I took the time to check the gas pedal (I was stopped at a light and when I started releasing the brake, the car started to take off. Pressing the brake again stopped it). I checked everything I could. I am convinced this isn't mechanical. Since it is sporadic and doesn't occur under any particular conditions, it's reasonable to consider that some component of an ECU can be defective and "fail" some of the time. I don't know.

What I do know beyond any doubt is that this problem is real. I also agree with some that if it happens to you, there are things you can do to minimize your risk. I don't consider turning off the car a great idea. First of all, you have to press and hold the power button for at least 4 seconds to force shutdown. That's a long time when you are careening out of control. When you do power off, you lose any steering assist and only your front brakes work. Without the power assist, those brakes take a lot of force to manage.

In my case, each time I started to release the brake (note: I slowly release my brake. If I just took my foot off of the pedal, I would have it the car in front of me. That means you can reduce your risk by doing the same thing). Interestingly, in my case, when I pressed the brake again, the engine "calmed down" and returned to idle. Starting to release the brake made it rev and try to run away.

I think that there are multiple causes to the uninended acceleration problem. Every single car company has some cases of this every year. I bet that in a few (very few) cases, the floor mat really was at fault. In other cases maybe it was a sticky gas pedal. In my case, it was certainly not either of those causes.

 

With so many people blaming cars for everything the manufactures have come up with some very sophisticated black boxes, Some of the newer ones record all kinds of information beyond the obvious stuff even the octane of the fuel , percentage of alcohol in the fuel, outside temperature, even if you were using the cell phone.

BC

 

Anyone notice that this problem is isolated to America.

Have any of you heard of a Run Away Prius in Japan, Mexico, Canada, France, England, Spain, Germany, Italy, Norway, Belgium, Iceland, Czech Republic, Australia, Philippines to name a few of Toyota's markets?

They are all built in the same factories in Japan so this problem should be world wide.

They're not paying attention and wreck the car. The first thing coming out of the drivers mouth is "STUCK ACCELERATOR", BAD BRAKES".

Once in a while it will happen but not every week in only one country.

 

I'm glad to see some of the engineers checking in on this one. I do this kind of work too but thankfully not where safety is important. If you CD's start playing backwards, that would likely be one of my bugs

There is only one thing that I worry about: What if the program crashes, and just ignores all my inputs? That is a scary thought, and the continuing stories worry me--can they all be wrong?

This illustrates my point, that if the CPU is still in control, then I trust the firmware to do the right things and take care of itself, and also give me an assortment of "workarounds" just in case. The Golden Opportunity would be if you can "show it to me while it's doing it". Imagine getting to sit at the intersection for half an hour with your foot on the brake while the Toyota Engineering Swat Team makes an emergency response!

One last thought about real CPU crashes--there wouldn't be any log data either. (Unless they are smart enough to use a separate controller--but then you have a way to make an escape mechanism at the same time).

 

This is a real concern, which is why microcontrollers have watchdog timers and other independent safety systems.

Regardless, though, all systems will fail when enough accumulated failures overwhelm the safety systems. When this happens people die.

The good news is that the current crop of electronic controllers for cars are safer than the mechanical systems they replace. We must not lose sight of the fact that mechanical systems also fail. I've had my mechanical brakes fail in another car, I've had a throttle cable stick, I've had my flaps fail on an airplane, I've had an altimeter read the wrong altitude, and I've lost the radio in a plane. All of these were mechanical failures, except for the radio.

Looking at boats, I've been dismasted, popped hoses, had a stove fire, leaks, rigging failure, dead batteries, broken wires, parted lines, fouled anchors, and engine failures.

No system is immune to failure. As engineers, we try to balance costs against the probability and severity of a failure. Anyone that thinks all systems must be 100% reliable regardless of cost is naive.

Tom

 

True, but if you watch local news, especially morning news ime, it's not at all uncommon to hear about car vs. building. At least that is my experience from living in three major metro areas.

Check out how many of these we aren't talking about.

car hits building - Google News

 

Found this link on another thread on PC today

Toyota Prius driver who knocked down and killed wife blames 'faulty' car | Mail Online

We have our own alleged unintended acceleration.

 

The guy drives a Corvette too since he had a
'Corvette Owners Association" jacket on...

 

I have not heard in EU, JP or elsewhere except the US, cases of unintended accelerations. And mats getting stuck. And people have car accidents because of any of the above.
Moreover, I have never heard yet that any other Toyota model has ever had an accelerator stuck.
The EU is 10+ states. Each has its own vehicle approval body and there are EU wide checks on vehicles. If something so common, as it appears to be in the US, happened, we would know. And yes, in the EU and elsewhere in the world there are less Priuses, but at the speed at which these unintended accelerations happen, one would have happened here already.
But, let's assume there is really a problem. What is the difference between the US and the rest of the planet?
The cars/Priuses are all built at the same Japanese plant. So the cars are all the same in the US and the rest of the planet. The only thing that comes to mind that could affect the electronics, is some EMI interference.
The only powerful electronics in a car that could affect its functioning, that are not the electronics of the car itself, are mobile phones which in the US work on frequency bands completely different from the those of EU and JP and the rest of the planet. Namely 800MHz, 1900MHz - the most common ones. There are more, but you get the idea.
So the only real difference between a US and an, say, EU driver, is the type of frequency the mobile phone works. All else being, hopefully equal.
The question is then, does a mobile phone in the US affect the electronics of the Prius that bad??? I don't know, it seems to me unlikely.

For the time being, the only thing I know is that only on US cars I have driven, I have seen more stickers, warnings and "obvious" messages all over the car, than anywhere else on the planet. The only place where on the side rear view mirror itself it is written "objects may look nearer as they seem" - the first time I tried to read that I almost did an accident to try to understand what was that dirt preventing me to look at the back on the mirror (I am exaggerating).
The EU Prius user's manual is literally full of all sorts of weird and honestly speaking, dumb, remarks of what you should not do when around, operating, in the car. Honestly speaking I ended skipping a lot of those remarks - 90% of them were *so* obvious I could only think there were to cover Toyota's nice person in the US in case a driver banged the car someplace while drinking very hot coffee or a burger and burning him/herself and then suing both Toyota and MacDonalds (US are famous for the lady who got damages paid because the coffee was hot in the paper cup....).
Again, if the statics are raising so quickly or where there all the time, something would have happened in the EU or elsewhere already, or something peculiar to the operation of Toyota cars in the US is making the cars behave weirdly/unexpectedly in the US only. BTW: in Canada? is something happening there or not? I haven't heard anything coming to that side of the fence.
I am not defending Toyota - if there is an issue it has to be fixed. But it is really weird that everything is happening in the US only. (or at least it gets to the news only in the US).

 

The guy can't even pull into a parking spot w/o scraping the car next to him. I'm keeping an open mind on almost all the claims against Toyota, but this one sure seems like someone hitting the wrong pedal.

 

Different reaction in Japan.

Toyota Owners Report Problems in Japan to No Avail - NYTimes.com

 

Sorry, but anything written on the US media about it for me has almost no value anymore. There are too many interests in this US-Toyota game, that anything coming from the US media for me might is not the best source.

Moreover, since when Police have interest in telling a 64 year old woman to sign a statement where she admits to be at fault by pressing the accelerator? I don't buy it. Police report statements of people involved - they don't invite people to state something different from the facts that can affect negatively those who suffered the most consequences.

BTW: I haven't searched deeply, but this story has not made the news both here in Germany (and they would love it) or in Italy newspapers.

 

I think PC is a pretty large community now. Is there any thread where people are complaining and describing in detail this electronic failure? While as engineer I agree that any system can fail, it worries me when this is happening systematically.

On the other hand people can find many times easier to hide their mistakes and blame Toyota. I don't know if this is the case (and as a Prius owner I hope it is), but in my home country it is known that the probability of pressing the gas pedal instead of the break is pretty high for some people. And this happened to my girlfriend once. So hopefully this happened in the NY case too.

 

Has there been any runaway Gen III's?