Searching \ for '[TECH] Crash Data Suggest Driver Error in Toyota A' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/mems.htm?key=data
Search entire site for: 'Crash Data Suggest Driver Error in Toyota A'.

Exact match. Not showing close matches.
PICList Thread
'[TECH] Crash Data Suggest Driver Error in Toyota A'
2010\07\14@193529 by Vitaliy

face
flavicon
face
tinyurl.com/28fctt3

"The U.S. Department of Transportation has analyzed dozens of data recorders
from Toyota Motor Corp. vehicles involved in accidents blamed on sudden
acceleration and found that the throttles were wide open and the brakes
weren't engaged at the time of the crash, people familiar with the findings
said.

The early results suggest that some drivers who said their Toyotas and
Lexuses surged out of control were mistakenly flooring the accelerator when
they intended to jam on the brakes."

This part is especially interesting, from a psychological POV:

"The initial findings are consistent with a 1989 government-sponsored study
that blamed similar driver mistakes for a rash of sudden-acceleration
reports involving Audi 5000 sedans."

2010\07\15@092114 by Olin Lathrop

face picon face
Vitaliy wrote:
> "The U.S. Department of Transportation has analyzed dozens of data
> recorders from Toyota Motor Corp. vehicles involved in accidents
> blamed on sudden acceleration and found that the throttles were wide
> open and the brakes weren't engaged at the time of the crash, people
> familiar with the findings said.

But then you have to believe the same computer that might be at fault in the
first place.  Suppose the problem is that the accellerator pedal (it's a
long way from the throttle these days) position is misreported, and the rest
of the vehicle was doing what it was supposed to?


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\15@094252 by Herbert Graf

picon face
On Wed, 2010-07-14 at 16:34 -0700, Vitaliy wrote:
> http://tinyurl.com/28fctt3
>
> "The U.S. Department of Transportation has analyzed dozens of data recorders
> from Toyota Motor Corp. vehicles involved in accidents blamed on sudden
> acceleration and found that the throttles were wide open and the brakes
> weren't engaged at the time of the crash, people familiar with the findings
> said.
>
> The early results suggest that some drivers who said their Toyotas and
> Lexuses surged out of control were mistakenly flooring the accelerator when
> they intended to jam on the brakes."
>
> This part is especially interesting, from a psychological POV:
>
> "The initial findings are consistent with a 1989 government-sponsored study
> that blamed similar driver mistakes for a rash of sudden-acceleration
> reports involving Audi 5000 sedans."

Unfortunately for Toyota, "the public" tends to believe the most
"interesting" story, whether or not it's true. I don't think anyone here
is very surprised, but your average person will simply reject the
evidence and believe the fantasy.

I guarantee 10 years from now people will still be 100% convinced that
Toyotas have throttle problems.

Humans can be very illogical.

TTYL

2010\07\15@094500 by Alan B Pearce

face picon face


> -----Original Message-----
> From: spam_OUTpiclist-bouncesTakeThisOuTspammit.edu [.....piclist-bouncesKILLspamspam@spam@mit.edu] On
Behalf Of Olin
> Lathrop
> Sent: 15 July 2010 14:21
> To: Microcontroller discussion list - Public.
> Subject: Re: [TECH] Crash Data Suggest Driver Error in Toyota
Accidents (WSJ)
>
> Vitaliy wrote:
> > "The U.S. Department of Transportation has analyzed dozens of data
> > recorders from Toyota Motor Corp. vehicles involved in accidents
> > blamed on sudden acceleration and found that the throttles were wide
> > open and the brakes weren't engaged at the time of the crash, people
> > familiar with the findings said.
>
> But then you have to believe the same computer that might be at fault
in the
> first place.  Suppose the problem is that the accellerator pedal (it's
a
> long way from the throttle these days) position is misreported, and
the rest
> of the vehicle was doing what it was supposed to?

But the same computer reports that the brake pedal isn't actuated at
all. Are brakes done 'fly by wire' at all? I would have thought it would
'just' look at the brake light sensor, and record if that is on. It
could be that the computer has two faults that affect these two pedals,
or it could be that the chip used has a problem with certain code
patterns, but somehow I doubt it. If the brake pedal was being actuated,
I would have thought the computer would record that, but in all cases it
is reported as not being actuated. In view of this I would expect
someone has done some pretty deep forensic digging in the code to see if
there is a code fault that could cause this, but the same conclusion was
come to some years previously with a similar 'problem' with some Audis.
It really does sound like the pedal layout is such that it is all too
easy to put ones foot on the wrong pedal.
--
Scanned by iCritical.

2010\07\15@094955 by Herbert Graf

picon face
On Thu, 2010-07-15 at 09:21 -0400, Olin Lathrop wrote:
> Vitaliy wrote:
> > "The U.S. Department of Transportation has analyzed dozens of data
> > recorders from Toyota Motor Corp. vehicles involved in accidents
> > blamed on sudden acceleration and found that the throttles were wide
> > open and the brakes weren't engaged at the time of the crash, people
> > familiar with the findings said.
>
> But then you have to believe the same computer that might be at fault in the
> first place.  

Nope.

The "black box" in most cars is a separate box from the ECU (it's
usually the ABS controller, completely separate, often located under the
drivers seat).

The sensors then are "suspect". While it's possible the throttle sensor
was reporting incorrectly, it is very doubtful that in EVERY case both
the throttle sensor and brake sensor were wrong (incorrect "no brake,
full throttle" readings).

> Suppose the problem is that the accellerator pedal (it's a
> long way from the throttle these days) position is misreported, and the rest
> of the vehicle was doing what it was supposed to?

Then why wasn't the brake depressed? If your car starts racing the FIRST
instinct of pretty much every driver is press the brake, preliminary
evidence shows that in almost EVERY crash (I think one was the odd one
out) the evidence states both WOT and no brake, a classic "hitting the
wrong pedal" mistake.

This is all consistent with a multitude of prior incidents where people
simply press the wrong pedal (my mothers car was hit by a person backing
into a spot who hit the wrong pedal, mounted a curb, went across a grass
lawn, travelled about 50 feet and smacked into the back of my moms car;
and no, this car didn't have a computer controlled throttle).

TTYL

2010\07\15@100146 by Olin Lathrop

face picon face
alan.b.pearce@stfc.ac.uk wrote:
> It really does sound like the
> pedal layout is such that it is all too easy to put ones foot on the
> wrong pedal.

But that doesn't explain all the symptoms.  As I heard it, at least in
several instances someone was driving down the highway and the car suddenly
accellerated.  In that case they wouldn't be on the brake at all, and they'd
know if they had stepped harder on the gas.  The immediate reaction would be
to take the foot off the gas, which isn't subject to pedal confusion.  The
second reaction would be to step on the brake, which I suppose could
accidentally be stepping on the gas in some cases.  But wouldn't you realize
you stepped on the wrong pedal quickly?  I haven't driven these cars, but
all the cars I have driven have a very different feel between the gas and
brake pedals, not to mention different placement.

I don't know what the right answer is, but something still doesn't smell
quite right.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\15@100955 by M.L.

flavicon
face
On Wed, Jul 14, 2010 at 7:34 PM, Vitaliy <piclistspamKILLspammaksimov.org> wrote:
> http://tinyurl.com/28fctt3
>
> "The U.S. Department of Transportation has analyzed dozens of data recorders
> from Toyota Motor Corp. vehicles involved in accidents blamed on sudden
> acceleration and found that the throttles were wide open and the brakes
> weren't engaged at the time of the crash, people familiar with the findings
> said.
>
> The early results suggest that some drivers who said their Toyotas and
> Lexuses surged out of control were mistakenly flooring the accelerator when
> they intended to jam on the brakes."
>
> This part is especially interesting, from a psychological POV:
>
> "The initial findings are consistent with a 1989 government-sponsored study
> that blamed similar driver mistakes for a rash of sudden-acceleration
> reports involving Audi 5000 sedans."
>

I've heard through numerous outlets that this "Toyota embedded system
defect" apparently discriminates against older drivers:
<http://www.theatlantic.com/business/archive/2010/03/how-real-are-the-defects-in-toyotas-cars/37448/?rss=37448>

--
Martin K.

2010\07\15@101024 by Olin Lathrop

face picon face
Herbert Graf wrote:
> This is all consistent with a multitude of prior incidents where
> people simply press the wrong pedal (my mothers car was hit by a
> person backing
> into a spot who hit the wrong pedal, mounted a curb, went across a
> grass lawn, travelled about 50 feet and smacked into the back of my
> moms car;
> and no, this car didn't have a computer controlled throttle).

Yeah, I got hit by a driver doing that too.  I was going straight down a
main rode, and suddenly another car smashes head onto into the right side of
my car.  The other car come out of side street.  Fortunately the girl
driving the other car was heard to say to the cop "I'm so sorry, I meant to
hit the brake and hit the gas instead".  However, she was very young and was
in fact illegally driving on a learner's permit.  In this case there wasn't
enough time for her to realize she pressed the wrong pedal, move the foot to
the brake, and stop in time.

But I thought at least some of the report I heard about Toyotas had to do
with going for miles at high speed down a freeway.  Wouldn't you think that
the driver would figure out the wrong pedal was pressed in that amount of
time?  Of course there could be deliberate fraud at work too, which clouds
the issue.

Then there was the case where a guy said he managed to get the car to slow
down somewhat by pressing really hard on the brake, but eventually the
brakes faded and the car sped up again.  Checking the brake linings would be
a good way to decide what really happened there, but I never heard if that
was done or its outcome if it was.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\15@101124 by Carl Denk

flavicon
face
In most OBDII cars, there is a "Throttle Position Sensor" (TPS) that is
among the most replaced items of the environmental sensors. Just Google
the " ". From one of the pages are the following symptoms:
>
>     * Bucking and jerking of the car
>     * Idle surging
>     * Sudden stalling of the car engine
>     * Hesitation while the driver of the car is trying to accelerate
>     * Sudden surge in car's speed while driving on the highway
>
I think most people would complain of the first 4 reasons and  drive it
in to a service facility and complain the car is sick or not acting
right. The last one, most drivers probably would panic, even though, the
simple act of moving the gear control to either a neutral or a lower
gear, or pushing the clutch in would solve the immediate situation.
Turning the key off, sounds good, but most vehicles would result in a
locked steering wheel. In the case of disconnecting the engine from the
remainder of the drive line, that same computer is supposed to limit the
engine speed. On my Cougar, the tach redline is 6500 RPM, but no load,
the computer limits it to 3500 RPM. The vehicle apparently doesn't have
the usual 106 mph limiter, since it has been to 110 mph several times,
no problem.

   * Bucking and jerking of the car
   * Idle surging
   * Sudden stalling of the car engine
   * Hesitation while the driver of the car is trying to accelerate
   * Sudden surge in car's speed while driving on the highway


On another message the brake sensor is questioned. On the Cougar, I was
hoping to use the reverse switch to indicated when I applied a load the
the engine, and was surprised to see no action. The car drives totally
normal with backup lights functioning in a normal manner. I will have to
add the brake switch data, and see what happens. Sometimes these sensors
are there for an obscure reason. One of the functions of the brake
switch is to turn off the cruise control. :)

If the brake sensor should fail, or maybe even the disconnect signal to
the cruise control, and the vehicle speed sensor reads low, the computer
would open the throttle to maintain speed (just forget to turn the
cruise off on slippery pavement and see what happens), and the normal
push the brakes would not disconnect the cruise. You would have a
runaway vehicle. At the same time, the intake manifold vacuum would be
low, and 99% of power brakes (and what doesn't have power brakes today!)
source of power is the vacuum. The power assist to the brakes would be
nil, and who do you think that would have enough leg muscle to stop the
vehicle??  Yes, this is a series of events, probably low probability of
happening, but... This is why it is important to pay attention to the
vehicle talking to you, and fix complaints promptly. Besides, it's
easier to troubleshoot one item at a time.

On 7/15/2010 9:21 AM, Olin Lathrop wrote:
{Quote hidden}

2010\07\15@102042 by RussellMc

face picon face
> "The U.S. Department of Transportation has analyzed dozens of data recorders
> from Toyota Motor Corp. vehicles involved in accidents blamed on sudden
> acceleration and found that the throttles were wide open and the brakes
> weren't engaged at the time of the crash, people familiar with the findings
> said.
>
> The early results suggest that some drivers who said their Toyotas and
> Lexuses surged out of control were mistakenly flooring the accelerator when
> they intended to jam on the brakes."

Presumably and hopefully there is a bit more analysis behind the
stated conclusion than is evident from that report.
Great care needs to be taken in understanding and interpreting
information such as that above. It MAY be that this level of analysis
has already been carried out and that this effectively summarises a
proper deeper technical understanding. One would hope this is the
case.

BUT it may be that the sensor and logged information is being taken at
or close to face value without an adequate analysis of the
implications. While this sounds almost inconceivable, far worse
examples occur. eg STS51-L aka Space Shuttle Challenger was destroyed
because people kept moving the goalposts and then drawing new targets
around the new positions, and then turning a blind eye to the fact
that the specification had crept far away from where it has been
originally when based on sound engineering principles.*  If its good
enough for a Space Shuttle it may apply to Prius's as well.

Consider:

What is "the throttle" in this context.
Is it the commanded acceleration state, or the sensed position of the
floor pedal or ???.

If it is the commanded acceleration state, is this what you'd expect
if the system caused the acceleration independent of user action?

If it is the sensed floor pedal position, is there a prospect that
this is wrongly sensed and that the user action is not what is
reported.

ie - What degree of confidence is there that what is being reported is
well correlated with what is happening?
Depending on the answer, this could be an indication that there is no
problem, or a clear example of the problem in action.

And, note that the first paragraph specifies "dozens of data
recorders" and implies that all cases analysed indicated user error.
It is immediately followed by a statement involving "some people".
Does  "some people" indicate the same number of people or cases as
"dozens of recorders" and if not why not / if so why is it put that
way. Is "some people" a subset of the total number of accidents
reported and if so, what proportion, or the total number. Same for
"dozens of cases".

None of this necessarily implies sloppy inexact analysis but that's
certainly a possibility based on this report. Hopefully its just
populist reporting of a far more precise underlying analysis. One can
hope.



 Russell

* http://en.wikipedia.org/wiki/Space_Shuttle_Challenger_disaster -
especially "Prelaunch conditions and delays".

2010\07\15@103947 by Michael Watterson

face picon face
 On 15/07/2010 14:49, Herbert Graf wrote:
>
> Then why wasn't the brake depressed? If your car starts racing the FIRST
> instinct of pretty much every driver is press the brake, preliminary
> evidence shows that in almost EVERY crash (I think one was the odd one
> out) the evidence states both WOT and no brake, a classic "hitting the
> wrong pedal" mistake.
>
>
Are these Automatics?

Have people with automatics got a bad habit in some cases of leaving
right foot on Acc. and using left for break?

(I only drove an automatic once a long time ago, I have no idea if
people do this, but throwing out idea for consideration)

2010\07\15@104913 by Herbert Graf

picon face
On Thu, 2010-07-15 at 15:39 +0100, Michael Watterson wrote:
> On 15/07/2010 14:49, Herbert Graf wrote:
> >
> > Then why wasn't the brake depressed? If your car starts racing the FIRST
> > instinct of pretty much every driver is press the brake, preliminary
> > evidence shows that in almost EVERY crash (I think one was the odd one
> > out) the evidence states both WOT and no brake, a classic "hitting the
> > wrong pedal" mistake.
> >
> >
> Are these Automatics?

In the US? Mostly yes. Much of the North American automarket is
unfortunately automatics.

> Have people with automatics got a bad habit in some cases of leaving
> right foot on Acc. and using left for break?

Mostly no. People are taught to use only one foot.

The only exception to this is certain situations in extreme weather
(i.e. trying to get unstuck in snow), and for those situations you're
already stopped and stuck.

Race drivers (i.e. rally drivers) do use two feet, but I don't think
they count! :)

TTYL

2010\07\15@105542 by Carl Denk

flavicon
face


> Great care needs to be taken in understanding and interpreting
> information such as that above.
>    
I'll second that. In particular, without knowing what the programming
logic is, to understand what sensors affect what outputs in what manner
makes troubleshooting difficult. I can picture the manufacturers not
releasing that info for proprietary reasons.

> What is "the throttle" in this context.s".
>    
I would assume, unless it could be proved, that the mechanical throttle
(pedal) linkage was stuck in a position, they are talking of the
throttle position sensor readings stored.

I haven't got very deep into it, but the Cougar computer does store more
than a few operational parameters, including Maximum speed in a gear for
one.

2010\07\15@105543 by Carl Denk

flavicon
face
2 items:
1: Where does the black box get it's data? Maybe a CAN bus connection,
and is just a receiver of data for storage :~)
I don't recall any of the Fords I have worked on, having a separate
black box. I would have thought, it would have caught my eye by now.
Over the years, I probably have owned (and used) at least a half dozen
Ford shop manuals, including all the emissions and wiring diagrams.

2: Check my Cougar PID's, there are 2 items:
    1: Brake pedal Switch Input status
    2: Brake: On/Off

Sounds like the first one checks for power to the following, and I have
to assume, would set the Check Engine light, ABS light , or something.

The Cougar wiring diagram shows the stoplight switch as SPST NC. One
contact is from "Hot in start or run" through a 15 amp fuse. The other
contact to the usual stoplights, ABS (Anti Lock brakes) control module,
vehicle (cruise) speed control, PCM (Power train control
module)(computer), and instrument panel interface.

2010\07\15@111506 by Olin Lathrop

face picon face
Herbert Graf wrote:
> Much of the North American automarket is
> unfortunately automatics.

You are rushing to judgement before considering all the angles.  Up until my
current car, I have always had manual transmissions because I liked them
better and they were also generally more efficient than the same car with a
automatic.

However, now computers are controlling more and more low level details in
cars.  In my current car (2008 Honda Civic Hybrid), there is so much for the
computer to do and tweak in real time, I would have no chance of doing as
good a job manually, and would be overloaded if I tried.  The efficiency
situation is now flipped.  Since the computer can monitor many more inputs
many times a second, these advanced cars are now more efficient than they
would be if you stuck a manual transmission in there and therefore had to
dumb down the rest of the drive train to accomodate it.  I reliably get at
least 50 miles/gallon on the highway (this morning it was saying I got over
55 MPG over the last 500 miles, and it usually estimates the milage low), so
it's rather hard to argue with whatever the computer is doing in there.
This car doesn't come in with a manual transmission since it wouldn't make
any sense.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\15@112955 by Alan B Pearce

face picon face
> In most OBDII cars, there is a "Throttle Position Sensor" (TPS) that is
> among the most replaced items of the environmental sensors. Just Google
> the " ". From one of the pages are the following symptoms:
> >
> >     * Bucking and jerking of the car
> >     * Idle surging
> >     * Sudden stalling of the car engine
> >     * Hesitation while the driver of the car is trying to accelerate
> >     * Sudden surge in car's speed while driving on the highway
> >

I have an interesting problem with my European Ford Focus Diesel at present. It starts and runs quite happily, but after warming up it idles fast (around 1200rpm instead of around 750) and runs very lumpy, the best description I could give the garage was it sounded like it has a race cam in it. They reckon the problem is that there is a exhaust recirculation valve that recirculates the exhaust gases (or a proportion thereof) back through the inlet to re-use unburnt fuel. I'll live with it for the moment, as they are talking in terms of around £500 to replace it as it comes fitted to a complete inlet manifold! But a by-product of this symptom is that when taking foot off the accelerator the engine doesn't slow down, but continues to drive the car. A most unnerving situation if you don't expect it, when slowing to traffic lights or similar without braking, just expecting to use engine braking.

> At the same time, the intake manifold vacuum would be
> low, and 99% of power brakes (and what doesn't have power brakes today!)
> source of power is the vacuum. The power assist to the brakes would be
> nil, and who do you think that would have enough leg muscle to stop the
> vehicle??  Yes, this is a series of events, probably low probability of
> happening, but... This is why it is important to pay attention to the
> vehicle talking to you, and fix complaints promptly. Besides, it's
> easier to troubleshoot one item at a time.

In all vehicles I have driven there is a vacuum tank that provides some reservoir in this situation. If the car is out of gear, and I turn the engine off, then put foot on brake it takes around 5 seconds for the brake pedal to harden up as the vacuum is depleted. In most cases a stamp on the brake under these conditions would provide quite a load shock to the engine, even if the vacuum didn't get replenished at the normal rate.

I remember my fathers 1936 Plymouth car having a vacuum operated windscreen wiper. The faster the car went the slower the wiper went, but for it to actually stop was real rare. I doubt that one would really get to the situation of having no power brakes at all.
--
Scanned by iCritical.

2010\07\15@120508 by Picbits Sales

flavicon
face
Just my 0.16 pence worth here .....

I drive an automatic in the UK - I always left foot brake and use my right
foot for the accelerator in an automatic. Always have done and always will
do. Often the right foot says "go go go" but the left foot says "be
sensible" and takes over.

Now ..... my wife has a brand new VW something or another (usually a Golf
TDI) every 6 months. They usually have a manual gearbox (I have to remember
the extra pedal and adjust my left foot braking accordingly !).

They are always drive by wire throttle. I've had a couple of them get
slightly confused near some overhead power lines (at exactly the same place
interestingly enough) and although the throttle hasn't surged, its become
temporarily unresponsive (around 0.5s to 1s)

Two memorable occasions were as follows:

Once when driving to the local shopping center, I pushed the clutch in and
blipped the throttle. The engine went crazy hitting the limiter even though
my foot was off the accelerator pedal - this was in a VW Golf GT TDI Sport
with drive by wire throttle. After a couple of seconds it corrected itself
but it still makes me nervous.

Another occasion was when I was walking my stepson to school when I heard a
diesel car revving and hitting the limiter. I looked back just at the right
time to see another VW Golf shoot across the pavement before being abruptly
stopped by the neighbours brick wall. Unfortunately in a terraced set of
houses, the front walls tend to be mechanically linked to each other via
gates etc so it took out 3 houses worth of wall. After running up the street
to see if I could help, the shocked driver told me this wasn't the first
time the throttle had gone haywire and she'd also had some very random
problems with other electrics (wipers/windows/heaters) in the car in the
previous couple of weeks.

Now I'm not singling VW out in any shape or form - we've had a lot of them
and they have been very good cars (apart from a Jetta which gave my wife a
seriously bad back). We've had very few problems or glitches but it does
show that all this drive by wire isn't infallible.

Dom

2010\07\15@122703 by Carl Denk

flavicon
face
Then why don't you run, not walk to obtain vehicles to drive that are
not throttle (and other basic systems) by wire?? You have enough
personal experience to warrant not even riding in someone else's similar
vehicle. The repeatability of the power line situation should make the
manufacturer sit up straight and try to avoid publicity! In Ohio we have
a lemon law, it's 3 tries and you get a new car or refund. Here, the
local TV station would jump on that repeatability. It is your duty to
warn others of this dangerous situation.

On 7/15/2010 12:05 PM, Picbits Sales wrote:
{Quote hidden}

2010\07\15@122926 by Michael Rigby-Jones

flavicon
face


> -----Original Message-----
> From: .....piclist-bouncesKILLspamspam.....mit.edu [EraseMEpiclist-bouncesspam_OUTspamTakeThisOuTmit.edu] On
Behalf
> Of Picbits Sales
> Sent: 15 July 2010 17:05
> To: Microcontroller discussion list - Public.
> Subject: Re: [TECH] Crash Data Suggest Driver Error in Toyota
Accidents
> (WSJ)
>
> Just my 0.16 pence worth here .....
>
> Now I'm not singling VW out in any shape or form - we've had a lot of
them
> and they have been very good cars (apart from a Jetta which gave my
wife a
> seriously bad back). We've had very few problems or glitches but it
does
> show that all this drive by wire isn't infallible.

I think it's one of the worst ideas to be implemented on modern cars.
Mushy throttle response coupled with far worse reliability and higher
chance of non failsafe modes.

They top my personal list of automotive hates which also include
electronic handbrakes and stupid electronic indicator stalks.

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

2010\07\15@124704 by Picbits Sales

flavicon
face

----- Original Message -----
From: "Carl Denk"

> Then why don't you run, not walk to obtain vehicles to drive that are
> not throttle (and other basic systems) by wire?? You have enough
> personal experience to warrant not even riding in someone else's similar
> vehicle. The repeatability of the power line situation should make the
> manufacturer sit up straight and try to avoid publicity!

The power line thing never seems to repeat itself when you are expecting
it - repeatability for test purposes appears to be zero !!! Plus its not a
"dangerous" failure - its seems fail safe in fact by cutting power briefly.
Its only happened twice despite maybe 50+ trips past the same lines.

As far as obtaining alternative vehicles, my wife gets given what is
available - she's not in the position to request alternatives.

We've had two other brands of car which have both been diesels and have also
been drive by wire - there seems to be no escaping it in the UK these days.

I drive a 1998 "drive by cable" car. Everything is loose and sloppy and its
a pig to drive but cost very little and because I've sworn not to replace it
until the wheels fall off it keeps soldiering on just to spite me.

Dom

2010\07\15@143236 by Vitaliy

face
flavicon
face
Olin Lathrop wrote:
>> "The U.S. Department of Transportation has analyzed dozens of data
>> recorders from Toyota Motor Corp. vehicles involved in accidents
>> blamed on sudden acceleration and found that the throttles were wide
>> open and the brakes weren't engaged at the time of the crash, people
>> familiar with the findings said.
>
> But then you have to believe the same computer that might be at fault in
> the
> first place.  Suppose the problem is that the accellerator pedal (it's a
> long way from the throttle these days) position is misreported, and the
> rest
> of the vehicle was doing what it was supposed to?

AFAIK the "black box" is an independent module.

Vitaliy

2010\07\15@144055 by Vitaliy

face
flavicon
face
Olin Lathrop wrote:
> But I thought at least some of the report I heard about Toyotas had to do
> with going for miles at high speed down a freeway.  Wouldn't you think
> that
> the driver would figure out the wrong pedal was pressed in that amount of
> time?  Of course there could be deliberate fraud at work too, which clouds
> the issue.

IIRC there was a report recently where a guy pressed both pedals
simultaneously to simulate a "stuck" accelerator.

However, if you had minutes to react, wouldn't your instinct be to shift to
neutral?

Vitaliy

2010\07\15@150330 by Vitaliy

face
flavicon
face
Michael Rigby-Jones wrote:
>> show that all this drive by wire isn't infallible.
>
> I think it's one of the worst ideas to be implemented on modern cars.
> Mushy throttle response coupled with far worse reliability and higher
> chance of non failsafe modes.
>
> They top my personal list of automotive hates which also include
> electronic handbrakes and stupid electronic indicator stalks.

You have to compare apples to apples, I think you'll agree that it's
unreasonable to expect the systems to be 100% safe. I know that
drive/brake-by-wire systems include multiple redundancies and should in
theory be at least as safe as traditional throttles and brakes. Based on the
best currently available information, despite millions of DBW cars on the
road, most accidents are caused by mechanical failures.

Vitaliy

2010\07\15@151648 by Herbert Graf

picon face
On Thu, 2010-07-15 at 11:31 -0700, Vitaliy wrote:
> Olin Lathrop wrote:
> >> "The U.S. Department of Transportation has analyzed dozens of data
> >> recorders from Toyota Motor Corp. vehicles involved in accidents
> >> blamed on sudden acceleration and found that the throttles were wide
> >> open and the brakes weren't engaged at the time of the crash, people
> >> familiar with the findings said.
> >
> > But then you have to believe the same computer that might be at fault in
> > the
> > first place.  Suppose the problem is that the accellerator pedal (it's a
> > long way from the throttle these days) position is misreported, and the
> > rest
> > of the vehicle was doing what it was supposed to?
>
> AFAIK the "black box" is an independent module.

Not really, usually the "black box" functionality is part of the airbag
module, so while independent from the ECU, it isn't usually it's own
box.

TTYL

2010\07\15@151806 by Herbert Graf

picon face
On Thu, 2010-07-15 at 12:02 -0700, Vitaliy wrote:
> Michael Rigby-Jones wrote:
> >> show that all this drive by wire isn't infallible.
> >
> > I think it's one of the worst ideas to be implemented on modern cars.
> > Mushy throttle response coupled with far worse reliability and higher
> > chance of non failsafe modes.
> >
> > They top my personal list of automotive hates which also include
> > electronic handbrakes and stupid electronic indicator stalks.
>
>  You have to compare apples to apples, I think you'll agree that it's
> unreasonable to expect the systems to be 100% safe. I know that
> drive/brake-by-wire systems include multiple redundancies and should in
> theory be at least as safe as traditional throttles and brakes. Based on the
> best currently available information, despite millions of DBW cars on the
> road, most accidents are caused by mechanical failures.

Agreed, "drive by wire" scares people, yet I've seen mechanical throttle
cables seized in the WOT position, and yet people aren't worried about
that.

TTYL

2010\07\15@153834 by Carl Denk

flavicon
face
On the Cougar, probably typical Ford, but  others could be different,
the airbag diagnostic monitor has a data link connection to the computer
module, but there doesn't seem to be any data storage with the airbag
module. There are other modules on the data link including ABS, Security.
>> AFAIK the "black box" is an independent module.
>>      
> Not really, usually the "black box" functionality is part of the airbag
> module, so while independent from the ECU, it isn't usually it's own
> box.
>
> TTYL
>
>    

2010\07\15@155236 by Bob Blick

face
flavicon
face

On Thu, 15 Jul 2010 15:38:30 -0400, "Carl Denk" said:
> On the Cougar, probably typical Ford, but  others could be different,
> the airbag diagnostic monitor has a data link connection to the computer
> module, but there doesn't seem to be any data storage with the airbag
> module. There are other modules on the data link including ABS, Security.
> >> AFAIK the "black box" is an independent module.
> >>      
> > Not really, usually the "black box" functionality is part of the airbag
> > module, so while independent from the ECU, it isn't usually it's own
> > box.

The black box in the Mustang I used to have is the airbag module and it
is bright blue, mounted on the hump behind the radio.

Bob

--
http://www.fastmail.fm - Does exactly what it says on the tin

2010\07\15@160637 by Olin Lathrop

face picon face
Vitaliy wrote:
>> But then you have to believe the same computer that might be at
>> fault in the
>> first place.  Suppose the problem is that the accellerator pedal
>> (it's a long way from the throttle these days) position is
>> misreported, and the rest
>> of the vehicle was doing what it was supposed to?
>
> AFAIK the "black box" is an independent module.

It's not about the box itself being independent, but whether it is
independently measuring the signals.  If the gas pedal sensor broke and
reported fully pressed, would the black box just record that or does it have
access to a truly independent measurement?


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\15@163233 by Carl Denk

flavicon
face
Agreed on both, it's not the separate module, but,   if somewhere the
data is saved in a similar way as an aircraft black box. :)

All the throttle position sensors I have seen, have only 2 wires going
to them and are nothing more than a variable resistance  (Pot), but
there might be some that use optical encoders or some other method.
There are some rough checks, though, the Mass Airflow sensor (MAP)
should be close to how much air is going into the engine. If throttle
closed, a minimum,  along with manifold pressure (vacuum). Probably
along with RPM and other sensors input, there could be an algorithm, to
check on the TPS.

On the Cougar, I did check the sensor for roughness, with engine off,
slowly open and close the throttle. Captured the output voltage readings
with the scanner, imported to Excel, and plotted a chart. Got close to a
straight line with no wild points. Could do same thing with a voltmeter,
but to break into the wires, I preferred not to, to preserve the
moisture resistance of the harness, but could have made some short
jumper wires. The scanner was easy to do.

On 7/15/2010 4:06 PM, Olin Lathrop wrote:
{Quote hidden}

2010\07\15@164425 by Olin Lathrop

face picon face
Herbert Graf wrote:
> Agreed, "drive by wire" scares people, yet I've seen mechanical
> throttle cables seized in the WOT position, and yet people aren't
> worried about that.

Yes, this has actually happened to me in a 1972 Ford Pinto long ago.  I was
going up a hill and had to floor the gas, as was usual for going up a hill
in that car.  When I tried to slow down at a stop light, I of course took my
foot off the gas and was expecting the hill to mostly slow the car.  Instead
the car kept going.  It does take a little time for this to sink in since
it's not a common occurence.  I had some room in front of me so I turned off
the ignition, stepped on the clutch, shifted to neutral, and eventually
stepped on the brake.

Stopping wasn't the problem.  Getting started again was, since the engine
got a little flooded and the throttle was stuck wide open.  There were too
many cars around to try to roll backwards to the side of the road, so I
popped the hood in the middle of the street.  It didn't take long to
discover that a linkage had come loose from the automatic choke.  This
allowed a cam to dangle free.  If everything bounced just right when the
throttle was wide open, the automatic choke cam would get wedged and hold
the throttle in the wide open position.

The throttle sticking open in this way happened at least one other time
while I owned the car.

I doubt drive by wire systems are inherently and more dangerous.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\15@165229 by Olin Lathrop

face picon face
Carl Denk wrote:
> All the throttle position sensors I have seen, have only 2 wires going
> to them and are nothing more than a variable resistance  (Pot), but
> there might be some that use optical encoders or some other method.
> There are some rough checks, though, the Mass Airflow sensor (MAP)
> should be close to how much air is going into the engine. If throttle
> closed, a minimum,  along with manifold pressure (vacuum). Probably
> along with RPM and other sensors input, there could be an algorithm,
> to check on the TPS.

It sounds like you are assuming that the foot pedal still mechanically moves
a throttle, and you are looking for evidence whether it was open or closed.

I thought the cars in question were drive by wire.  So the question is how
is the foot pedal position measured, how redundant is that, and does the
black box record a completely independent measurement of the foot pedal
position.  If not, a single point of failure in the foot pedal position
sensor can not be distinguished from the driver flooring the pedal.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\15@170638 by Carl Denk

flavicon
face


On 7/15/2010 2:39 PM, Vitaliy wrote:
>
> However, if you had minutes to react, wouldn't your instinct be to shift to
> neutral?
>
>    
I don't know, there is a fairly widely circulated video of a full size
SUV, might have been a Tahoe or expedition sort vehicle. On a wide
(maybe 4 lanes one way) and light traffic, not sure exactly why, but
maybe a sudden lane change to the left, the rear wheels started sliding
to the left (dry pavement). We are all told to steer in the direction of
the skid. The rear slide continued until the vehicle was perpendicular
to the direction of original travel and the lanes, at which point it
rolled over with the front wheels still pointed vehicle forward. There
was no appropriate steering action. If there had been even a moderate
appropriate action (reaction), there would have been no accident.
Undoubtedly, officially the vehicle got blamed, but, I believe that was
driver error.

The point being, today driver training is just the start of it, but
there are available today for the younger drivers, classes where
defensive and not routine driving conditions are practiced including a
skid pad. :)  I have discussed with more than a few, what to do with a
stuck open throttle. Many (most?) were surprised to hear the put it in
neutral thing. Many would have turned the key off, which would have
locked the steering wheel.

2010\07\15@180042 by Olin Lathrop

face picon face
Carl Denk wrote:
> Many would have turned the key off, which would have
> locked the steering wheel.

Of all the cars I've driven that had a steering wheel lock, there was always
a position that turned off the engine before the lock was engaged.  Usually
you have to press a button or do something special to remove the key and
lock the wheel, probably for exactly this reason.

Back in school we'd routinely turn off the engine going down a hill.
Stearing was still fine.  Pressing the button and removing the key would
have locked the wheel, and that definitely would have been bad, but it never
happened.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\15@180333 by Sean Breheny

face picon face
On Thu, Jul 15, 2010 at 5:06 PM, Carl Denk <cdenkspamspam_OUTwindstream.net> wrote:
> The point being, today driver training is just the start of it, but
> there are available today for the younger drivers, classes where
> defensive and not routine driving conditions are practiced including a
> skid pad. :)  I have discussed with more than a few, what to do with a
> stuck open throttle. Many (most?) were surprised to hear the put it in
> neutral thing. Many would have turned the key off, which would have
> locked the steering wheel.

I think that most automatics will not let you put the key in the LOCK
position unless the shifter is in Park. While you are absolutely right
that it would be bad to actually turn the key to the normal off
position (which is really LOCK), it may be OK to turn it to the actual
position marked OFF which will not lock the steering wheel. However,
it will still result in loss of power steering and eventually loss of
brake assist.

I think that driver training is atrocious in most of the US. I got my
license with very little real training - basically just reading the
drivers manual and spending about 15 hours driving with an instructor,
just really to get used to the mechanics of driving under normal
conditions as well as parallel parking. My drivers test was even
worse, it really was no test at all.

I think that most people do not have any idea what to do in situations
where the road is wet or icy, or if they need to stop the car in the
minimum distance safely, or if a tire blows out, etc. One way to tell
this is how much the phrase "lost control of the car" gets thrown
about as if cars just suddenly become uncontrollable. I think that
skid pad simulation of loss of traction and oversteer/understeer is a
must, as well as practicing minimum distance panic braking on both wet
and dry roads. That said, I have not done too much of this myself.
Sometimes in the winter when I see an unplowed parking lot I take the
opportunity to do some simple practice but nothing more than that.

Sean



>

2010\07\15@181043 by Veronica Merryfield

picon face

On 2010-07-15, at 1:32 PM, Carl Denk wrote:

> All the throttle position sensors I have seen, have only 2 wires going
> to them and are nothing more than a variable resistance  (Pot), but
> there might be some that use optical encoders or some other method.
> There are some rough checks, though, the Mass Airflow sensor (MAP)
> should be close to how much air is going into the engine. If throttle
> closed, a minimum,  along with manifold pressure (vacuum). Probably
> along with RPM and other sensors input, there could be an algorithm, to
> check on the TPS.

Whilst the sensor, TPS or otherwise, may only be a pot or other resistive element, there is a lot that can be done at the raw signal processing stage that can identify a number of fault modes, including open circuit and short circuit (both usually arranged not to be co-incident with full open and full closed peddle position), rate of rise or fall of signal and so on. Sensors and wiring are known to be weak points in the system and handled accordingly. I very much doubt Toyota do not have this sort of thing in place. 15 years ago when I last worked in that industry, racing and commercial, these sorts of signal integrity tests were common place.

Further, most ECUs log the signals, raw and processed, not forever, but long enough to see what happened prior to an incident. I am not saying this is as good as a black box with independent measuring ability, but in most cases, the logging of what is happening in the ECU is kept separate and is tested. Nobody has yet mentioned a vehicle manufacturer that I did not have some contact with during my time working in ECU system simulation testing - i.e. simulating engines, vehicles and all the fault modes that had ever been reported. All that report data does make it back to the engineering teams. Some very good people have the unenviable task of sifting through all that to produce system test specs, which are weighty tomes, and I and my team had the equally unenviable task of producing test scripts to simulate those failures. IIRC, the short form automatic system test for most manufactures took 2 weeks to run. We had the overnight build and system test and then on the regular r!
elease schedule, the system tests were run. Release candidates were taken from these periodically and put through more simulated and real testing. And before anyone says anything, as well as fully scripted, there were repeatable random scripts too, or rather results forms that contained enough to repeat a pseudo random stream.

FWIW, that system, although it's been through a number of hardware upgrades, is still being used by many engine makers in commercial and racing markets. Last I heard, the bulk of the simulation code and script code is still the same, just the UI has changed and the scrips have increased.

Whilst Toyota may have had some sticking pedal mechanics, which I think they owned up to, I highly doubt they had a spate of ECU problems. They are more likely to have mechanical issues from pedal to pot or driver errors.

My 2pw

Vrnc

2010\07\15@181752 by Veronica Merryfield

picon face

On 2010-07-15, at 2:06 PM, Carl Denk wrote:
> The point being, today driver training is just the start of it, but
> there are available today for the younger drivers, classes where
> defensive and not routine driving conditions are practiced including a
> skid pad. :)  I have discussed with more than a few, what to do with a
> stuck open throttle. Many (most?) were surprised to hear the put it in
> neutral thing. Many would have turned the key off, which would have
> locked the steering wheel.

I test this on my vehicles and I know that I can turn off the ignition without locking the the steering. Further, if I leave it in gear, I still have vacuum assit brakes and power steering. However, I also check the braking and steering without being in gear and rolling so I know the feel and handling. Doing these sorts of things in a controlled manor may help later. I drive manuals and not automatics.

FWIW, the winter after getting my license, I found a large car park one evening and practised slipping and sliding. I also put myself through other advanced driver training courses.

Vrnc


2010\07\15@183229 by Jake Anderson

flavicon
face
On 16/07/10 06:32, Carl Denk wrote:
> Agreed on both, it's not the separate module, but,   if somewhere the
> data is saved in a similar way as an aircraft black box. :)
>
> All the throttle position sensors I have seen, have only 2 wires going
> to them and are nothing more than a variable resistance  (Pot), but
> there might be some that use optical encoders or some other method.
> There are some rough checks, though, the Mass Airflow sensor (MAP)
> should be close to how much air is going into the engine. If throttle
> closed, a minimum,  along with manifold pressure (vacuum). Probably
> along with RPM and other sensors input, there could be an algorithm, to
> check on the TPS.
>
> On the Cougar, I did check the sensor for roughness, with engine off,
> slowly open and close the throttle. Captured the output voltage readings
> with the scanner, imported to Excel, and plotted a chart. Got close to a
> straight line with no wild points. Could do same thing with a voltmeter,
> but to break into the wires, I preferred not to, to preserve the
> moisture resistance of the harness, but could have made some short
> jumper wires. The scanner was easy to do.
>    
I believe (at least in the better ones) cars that do drive by wire use a
pair of hall effect sensors to read their position.
The sensors have to agree within a certain % or otherwise the car turns
off basically.

pots basically being regarded as "pretty crap" in general

2010\07\15@183631 by Jake Anderson

flavicon
face
On 16/07/10 08:03, Sean Breheny wrote:
> On Thu, Jul 15, 2010 at 5:06 PM, Carl Denk<@spam@cdenkKILLspamspamwindstream.net>  wrote:
>    
>> The point being, today driver training is just the start of it, but
>> there are available today for the younger drivers, classes where
>> defensive and not routine driving conditions are practiced including a
>> skid pad. :)  I have discussed with more than a few, what to do with a
>> stuck open throttle. Many (most?) were surprised to hear the put it in
>> neutral thing. Many would have turned the key off, which would have
>> locked the steering wheel.
>>      
> I think that most automatics will not let you put the key in the LOCK
> position unless the shifter is in Park. While you are absolutely right
> that it would be bad to actually turn the key to the normal off
> position (which is really LOCK), it may be OK to turn it to the actual
> position marked OFF which will not lock the steering wheel. However,
> it will still result in loss of power steering and eventually loss of
> brake assist.
>    
on my cars the switch sequence goes
"off," where you can pull the key out
acc, turns the accessories (lights radio etc) on and stops the steering
lock.
on, which as you can guess lets the engine run
start, you can probably guess this one.

knocking it back from on to acc wouldn't cause any major problems, it'd
be a pain to steer but I have at least 3 brake applications before
running out of vacuum (that's mandated by law here, its a bit of a pain
in the ass for the electric car people)

{Quote hidden}

>> --

2010\07\15@183703 by Vitaliy

face
flavicon
face
Veronica Merryfield wrote:
> Whilst the sensor, TPS or otherwise, may only be a pot or other resistive
> element, there is a lot that can be done at the raw signal processing
> stage that can identify a number of fault modes, including open circuit
> and short circuit (both usually arranged not to be co-incident with full
> open and full closed peddle position), rate of rise or fall of signal and
> so on. Sensors and wiring are known to be weak points in the system and
> handled accordingly. I very much doubt Toyota do not have this sort of
> thing in place. 15 years ago when I last worked in that industry, racing
> and commercial, these sorts of signal integrity tests were common place. <


Besides checking for shorts or opens, the ECU also check the TPS voltage at
idle. These three generic diagnostic trouble codes describe the three
possible conditions:

P0121 Throttle/Pedal Position Sensor/Switch A Circuit Range/Performance
P0122 Throttle/Pedal Position Sensor/Switch A Circuit Low
P0123 Throttle/Pedal Position Sensor/Switch A Circuit High

Vitaliy

2010\07\16@041337 by Picbits Sales

flavicon
face

----- Original Message -----
From: "Carl Denk" <KILLspamcdenkKILLspamspamwindstream.net>
To: "Microcontroller discussion list - Public." <RemoveMEpiclistTakeThisOuTspammit.edu>
Sent: Friday, July 16, 2010 1:03 AM
Subject: Re: [TECH] Crash Data Suggest Driver Error in Toyota Accidents
(WSJ)


{Quote hidden}

Don't get the TPS (Throttle Position Sensor) confused with a fly by wire
Accelerator (or Gas) pedal sensor - two totally different things.

The TPS tends to be a potentiometer mounted on the throttle body in the
engine compartment while the DBW Accelerator sensor is at the pedal.

2010\07\16@054045 by Alan B Pearce

face picon face
> FWIW, the winter after getting my license, I found a large car park
one evening and
> practised slipping and sliding. I also put myself through other
advanced driver
> training courses.

I had an Austin 1300 at the time I was a leader at a summer camp. The
camp had a field where the long grass had dried out and had been
polished by the childrens activities. I went out there a couple of
evenings with my car and could have the hand brake on and with front
wheel drive could do all sorts of sliding around. The experiences there
were to serve me well later in situations where I could otherwise have
got into a bad situation without knowing how the car was going to
handle.
--
Scanned by iCritical.

2010\07\16@073912 by Michael Rigby-Jones

flavicon
face


> -----Original Message-----
> From: spamBeGonepiclist-bouncesspamBeGonespammit.edu [TakeThisOuTpiclist-bouncesEraseMEspamspam_OUTmit.edu] On
Behalf
> Of Herbert Graf
> Sent: 15 July 2010 20:18
> To: Microcontroller discussion list - Public.
> Subject: Re: [TECH] Crash Data Suggest Driver Error in Toyota
Accidents
> (WSJ)
>
> On Thu, 2010-07-15 at 12:02 -0700, Vitaliy wrote:
> > Michael Rigby-Jones wrote:
> > >> show that all this drive by wire isn't infallible.
> > >
> > > I think it's one of the worst ideas to be implemented on modern
cars.
> > > Mushy throttle response coupled with far worse reliability and
higher
> > > chance of non failsafe modes.
> > >
> > > They top my personal list of automotive hates which also include
> > > electronic handbrakes and stupid electronic indicator stalks.
> >
> >  You have to compare apples to apples, I think you'll agree that
it's
> > unreasonable to expect the systems to be 100% safe. I know that
> > drive/brake-by-wire systems include multiple redundancies and should
in
> > theory be at least as safe as traditional throttles and brakes.
Based on
> the
> > best currently available information, despite millions of DBW cars
on
> the
> > road, most accidents are caused by mechanical failures.
>
> Agreed, "drive by wire" scares people, yet I've seen mechanical
throttle
> cables seized in the WOT position, and yet people aren't worried about
> that.

That's largely a maintenance issue though, throttle cables don't
magically seize when you drive under a pylon...

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

2010\07\16@084824 by Carl Denk

flavicon
face
Sorry, there seems to be 2 sub threads going, yeas of course, the TPS is
on the engine, and the other is on the pedal assembly. But all vehicles
could have a TPS, that would be a check on the actual throttle plate of
the engine position.

On 7/16/2010 4:13 AM, Picbits Sales wrote:
> {Original Message removed}

2010\07\16@101757 by Herbert Graf

picon face
On Fri, 2010-07-16 at 12:39 +0100, Michael Rigby-Jones wrote:
> That's largely a maintenance issue though, throttle cables don't
> magically seize when you drive under a pylon...

What's your point? A throttle cable can stick at ANY time, regular
maintenance doesn't preclude that.

The fact is anything mechanical can break, it might just happen when you
"drive under a pylon", whatever that means.

Again, I find it astonishing that people in general put so much more
trust in something mechanical then electronic. I understand some of the
psychology (people are afraid of what they don't understand), but there
is no reason to perpetuate half truths.

TTYL

2010\07\16@112727 by Michael Rigby-Jones

flavicon
face


> -----Original Message-----
> From: RemoveMEpiclist-bouncesspamTakeThisOuTmit.edu [piclist-bouncesEraseMEspam.....mit.edu] On
Behalf
> Of Herbert Graf
> Sent: 16 July 2010 15:18
> To: Microcontroller discussion list - Public.
> Subject: RE: [TECH] Crash Data Suggest Driver Error in Toyota
Accidents
> (WSJ)
>
> On Fri, 2010-07-16 at 12:39 +0100, Michael Rigby-Jones wrote:
> > That's largely a maintenance issue though, throttle cables don't
> > magically seize when you drive under a pylon...
>
> What's your point? A throttle cable can stick at ANY time, regular
> maintenance doesn't preclude that.
>

Of course it does, a throttle cable will never just randomly stick for
no reason.  Usually it's because it's frayed, or has been routed
incorrectly and kinked or melted on a hot component.  These things just
don't happen if you look after your car.


> The fact is anything mechanical can break, it might just happen when
you
> "drive under a pylon", whatever that means.
>

A snapped throttle cable is usually fail safe (though not always),
albeit inconvenient; the throttle will close.  However, this comes down
to maintenance again, a throttle cable doesn't snap unless something is
wrong with the rest of the system (usually a throttle stop incorrectly
adjusted causing the cable to be stressed at the end of travel).

> Again, I find it astonishing that people in general put so much more
> trust in something mechanical then electronic. I understand some of
the
> psychology (people are afraid of what they don't understand), but
there
> is no reason to perpetuate half truths.

I am stating the facts here.  I can guarantee that a drive by wire
system is less reliable than a very simple mechanical one, and certainly
far more difficult and costly to diagnose faults.  Add software
reliability into the mix and things look even worse.  This isn't just
speculation, I have been quite involved in the motor trade and drive by
wire components often cause faults e.g. faulty pedal sensors, faulty
throttle bodies, chaffed wiring or corroded connectors.

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

2010\07\16@113706 by Gary Crowell

picon face
I'm wondering if the whole thing might be a combination of problems.
Perhaps a car has an intermittent software/hardware problem that causes
short surges in the throttle.  Then in response to one of those surges, the
driver stomps on the 'brake', and floors the accelerator.  Then you would
have a 'history' of uncommanded acceleration, and a crash record of no
brake/full throttle.

----------------------------------------------
Gary A. Crowell Sr., P.E., CID+
http://www.linkedin.com/in/garyacrowellsr

2010\07\16@130758 by Olin Lathrop

face picon face
Michael Rigby-Jones wrote:
> I can guarantee that a drive by wire
> system is less reliable than a very simple mechanical one,

You can guarantee it?  Really?  That's a extraordinary statement that
requires extraordinary proof.

I think in fact there is no such proof since the underlying statement is not
true.  Both mechanical and electrical systems can fail.  Moving mechanical
systems are subject to stresses and friction.  In other words, they wear out
with use.  Electrical systems largely don't wear out with use, but usually
fail due to stresses caused by thermal cycling and vibration, and corrosion
at interconnect points.  Declaring either one to be inherently more or less
safe is pretty much impossible because it's a apples to oranges comparison.
To be fair, you'd also have to compare the costs of the systems.  Just about
any design, mechanical and electrical, can be made more reliable if it is
allowed to cost more.

I have personally had a mechanical throttle system fail on me at least two
times (I think three, but can only remember two indicents for sure).  That's
too small a sample to extrapolate from, but it does prove it can happen.
Mechanical systems are not inherently safe.

And then you have to look at the bigger picture, even if you are only trying
to measure safety.  One of the big advantage of electrical systems is that
you can get lots of logic complexity cheaply, in a way that would be
impossible from a mechanical system for a reasonable cost, weight, size,
power consumption, etc.  This allows electrical system to do things
mechanical systems can't.  If this capability is harnessed properly, then it
can enhance safety by dealing with human factors, such as alleviating tasks
that could be distracting, repetitive so that attention is lost, etc.

Of course there are crappy electrical systems just like there are crappy
mechanical systems.  You have to compare systems that perform similar
functions, have similar costs, and were both competently designed.  These
things are very hard to arrange.  Real proof would be very hard to come by,
and then would only apply to the particular designs that were measured
anyway.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\16@134613 by Gordon Williams

flavicon
face

----- Original Message -----
From: "Olin Lathrop" <EraseMEolin_piclistspamembedinc.com>
To: "Microcontroller discussion list - Public." <RemoveMEpiclistEraseMEspamEraseMEmit.edu>
Sent: Friday, July 16, 2010 12:08 PM
Subject: Re: [TECH] Crash Data Suggest Driver Error in Toyota Accidents
(WSJ)


> Michael Rigby-Jones wrote:
> > I can guarantee that a drive by wire
> > system is less reliable than a very simple mechanical one,
>
> You can guarantee it?  Really?  That's a extraordinary statement that
> requires extraordinary proof.
>
>Declaring either one to be inherently more or less
> safe is pretty much impossible because it's a apples to oranges
comparison.
>

More unsupported definitive statements ....

The reliability of different types of systems (apples and oranges, to use
your terminology) is examined and calculated all the time.  You need to look
at "fault tree analysis" and "failure modes and effects analysis".

The way it works is that each component or subpart of a system is assigned a
failure rate based on past experience of similar types of components or
subsystems.  These failure rates are then combined to get the overall
failure rate.

These types of analyses are done all the time in the nuclear and aerospace
fields and I would guess in the automotive field as well.  It is not rocket
science but it does require access to good data and a dose of common sense.

Gordon Williams

2010\07\16@154814 by Olin Lathrop

face picon face
Gordon Williams wrote:
> The reliability of different types of systems (apples and oranges, to
> use your terminology) is examined and calculated all the time.  You
> need to look at "fault tree analysis" and "failure modes and effects
> analysis".
>
> The way it works is that each component or subpart of a system is
> assigned a failure rate based on past experience of similar types of
> components or subsystems.  These failure rates are then combined to
> get the overall failure rate.

But that only works for specific instances.  I was reacting to the
ridiculous blanket statement that mechanical systems in general are more
reliable than electronic ones.

You can use data to show that a particular mechanical throttle system, for
example, has been more reliable than a particular alternative electronic
throttle system.  You can even use past data to make some educated
predictions on the likely reliability of particular new systems.

That's very different from saying mechanical systems are more reliable than
electronic systems.  I've had a HP11C calculator for 28 years and it's still
working fine.  During that time I've had to get rid of three cars due to
mechanical systems wearing out.  So electronic systems are obviously more
reliable than mechanical ones, right?  This is what I mean by apples to
oranges comparison.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

More... (looser matching)
- Last day of these posts
- In 2010 , 2011 only
- Today
- New search...