STEVE says:

All the new “safety features” in new cars, such as ABS, Traction Control, Stability Control and Cruise Control, acting through common hardware elements controlled by computers are potentially death traps waiting to grab somebody. And considering the complexity of these issues, with the faults not necessarily confined to Toyota, the only realistic way to study these issues is the installation of data collectors on vehicles in question that will allow real-time collection of data in large quantities over long periods of time from large numbers of sensors. The sensors are already on board as parts of all the Emissions Controls and Safety Features. In addition one needs to capture the command and feedback signals from all the actuators, that is, what they were commanded to do and why, and how they actually responded. The other manufacturers are already largely in compliance with such rules as best can be determined by public information but Toyota is holding out to the bitter end, not saying how much of this data is already being collected and they are not sharing anyway, making the pathetic excuse that there is only one “special computer” in this country capable of accessing and evaluating this data.

Toyota claims that their systems save only several seconds of data and that only in response to an event forceful enough to trigger the air bags. But then they prove themselves liars by showing 20 minutes of data from the wild 30 mile ride of James Sipes, allegedly showing very rapid pumping of the brakes and full throttle application. But only when it seemed convenient. They could just as easily have destroyed the data if it had proved inconvenient. Toyota is not doing much to advance its own cause. David Strickland has declared them untruthful. I would add the terms deceitful and deceptive. Toyota needs to be brought into line and not with data collection systems of their own design.

Memory for these proposed systems is dirt cheap and getting cheaper, with common formats such going for about $3 per GB, in sizes up to 32 GB. Such data collection has been going on in the aircraft industry for a long time with the familiar black boxes capturing data up to 17-25 hours. Such data collections could be loaded into personal computers via a USB cable with the systems showing up just as another external device such as a digital camera or taken off in a memory card or USB thumb drive.

This data coming out in pure data streams can then be loaded into spreadsheet programs such as Microsoft XCEL which has been around for a long time, is on large numbers of computers as part of the Student and Teacher Package for about $150 if nothing else. The program is well understood by many people and is easy to work with, by those with computer backgrounds. Nice graphics displays can be used to compare relationships among data streams, relating brake applications or throttle applications to other events to prove what actions are commanded or un-commanded. Aircraft crashes are commonly solved by such methods. What comes readily to mind are the series of Boeing upsets due to un-commanded rudder deflections which were solved with great difficulty due to lack of adequate data collection early on.

In the Harrison NY case Toyota and NHSTA showed up and seemed to conduct a joint investigation and had what looked like a joint news conference with the local chief of police stating at a news conference that the evidence presented by Toyota that the brakes were never applied and that there was constant application of full throttle was “black and white.” But there was no way to know what Toyota was measuring and they made no additional comment before leaving.

For example adequate examination of the throttle application demands analysis of the

1. signal from the gas pedal
2. signal arriving at the ECU
3. signal leaving the ECU
4. signal arriving at the throttle actuator
5. feedback signal from the actuator to the ECU
6. similar signals to and from the cruise control.

So there we have a multitude of signals just to check on the integrity of the throttle control, not to mention anything else.

Things are no simpler with the brake control. There is a small piston up high on the brake pedal with a rubber boot or bellows around it which can only be seen by crawling around on the floor and shining a light up behind the steering wheel or simply feeling with the hand. None of the Toyota salesmen seem to know anything about this. This piston sends hydraulic pressure to something called a “stroke sensor” which sends a signal to the “stroke simulator” which issues further commands through the system. Pressure in the brake lines comes from a “pressure accumulator.” The master cylinder itself does not have to get involved in the normal course of business, with valves opening in the master piston to let pressure pass through. Toyota’s schematic shows at least six pressure sensors in the ABS unit but it is hard to tell just how accurate these are. Toyota’s schematics for its flexible fuel cell in the rigid outer shell in years 2004-2009 for the Prius are highly deceptive and this makes other Toyota diagrams problematic.

So getting back to the question, where did Toyota get its data to indicate how the brakes and throttle were applied or not applied in the Harrison NY case. Impossible to tell.

Note that the data streams would be available to the public and subject to analysis with Toyota nowhere around. NASA has been called on board and perhaps they will have the stature to force needed changes Organizations such as NASA and NTSB National Transportation Safety Board have led the way in organizing such studies in the airlines, but the airlines have been willing participants such all this data helps them better monitor the performance and maintenance of their airplanes. Toyota is not part of a similar culture. Much of their participation in publicized cases reeks of tampering with and destroying evidence and obstruction of justice. And NHSTA has not done much about these things. The recent $16 million fine seems mostly designed to put some distance between Toyota and NHSTA. If NHSTA had any fortitude all along this would not have been necessary.

Toyota has recently advised that they will make a few more “special computers” available to NHSTA and others But with proper third party systems and un-encrypted data, Toyota’s failure to cooperate would be non-material. Hopefully Toyota won’t stand in the way forever.

With all the bundles of wires under the hood there seem to be lots of places for third parties to break in to install their systems to collect dozens of data streams.

Proper third party systems would make Toyota non-material in the studies proposed before.

So let us just imagine how we might approach some of the problems discussed in the Prius groups.

TC Traction Control and ABS act by establishing the “optimum slip angle” at each wheel, which means allowing the wheels to slip not too much and not too little but just right. If the wheels involved were fully locked or freely turning the system would have no way to evaluate what it was doing.

In the case of Traction Control the system attempts to keep the drive wheels from slipping by applying just enough brakes intermittently to keep the wheels slipping just right and by reducing engine power. Some drivers report very little ability to go on slick surfaces and the simplest most plausible explanation of this is an overly aggressive TC preventing the wheels from doing anything useful, keeping the cars from climbing even mild grades.

Some drivers report loss of brakes as James Sipes and the Laotian man in Minneapolis did. Nothing can reach out and steal your brakes away completely faster than the ABS and leave no trace afterward. Toyota claims this is impossible but adequate data collection and analysis could prove them wrong. Toyota declared that Sipe’s story was inconsistent with the facts because the front brakes were severely worn while the back brakes were in good shape. The best and simplest explanation here is that the brake systems in modern cars are separate front and rear to confine defaults to two wheels when possible. I think this is what happened to Sipes. With proper data the analysis would be a lot better. Toyota is not on board.

Stability Control has the power to apply braking to different wheels individually as required to maintain control. Traction control could do the same, at least for the front wheels. Either would produce the sudden lane changes reported by different Prius drivers. All that is required is an errant command from an over-loaded or confused computer.

Some Prius drivers report loss of brakes for several seconds at a time while driving across rough surfaces with brakes applied. It sounds like one or more wheels are coming unglued and locking up, to which the system responds by calling for an emergency pressure dump on all four wheels. This seems reproducible to some drivers and as such would seem to be a good place to start the studies. Toyota fluffs it off as a hesitation in the system in deciding whether to be in regenerative braking or friction braking, which seems laughable.

All the new “safety features” in new cars, such as ABS, Traction Control, Stability Control and Cruise Control, acting through common hardware elements controlled by computers are simply death traps reaching out to grab somebody. And considering the complexity of these issues, with the faults not necessarily confined to Toyota, the only realistic way to study these issues is the installation of data collectors on all new vehicles that will allow real-time collection of data in large quantities over long periods of time from large numbers of sensors. The sensors are already on board as parts of all the Emissions Controls and Safety Features. In addition one needs to capture the command and feedback signals from all the actuators, that is, what they were commanded to do and why, and how they actually responded. The other manufacturers are already in compliance with such rules but Toyota is holding out to the bitter end, not saying how much of this data is already being collected and they are not sharing anyway, making the pathetic excuse that there is only one special computer in this country capable of accessing and evaluating this data and

Toyota claims that their systems save only several seconds of data are saved and that only in response to an event forceful enough to trigger the air bags. But then they cross themselves by showing 20 minutes of data from the wild 30 minute ride of James Sipes, when it seems convenient to the, allegedly showing very rapid pumping of the brakes and full throttle application. Frankly Toyota is not doing much to advance its own cause. Indeed David Strickland has declared them untruthful. I would add the terms deceitful and deceptive. Toyota in my view needs to be brought into line and not with data collection systems of their own design.

Memory for these proposed systems is dirt cheap and getting cheaper, with common formats such going for about $3 per GB, in sizes up to 32 GB. Such data collection has been going on in the aircraft industry for a long time with the familiar black boxes capturing data up to 17-25 hours. Such data collections could be loaded into personal computers via a USB cable with the systems showing up just as another external device such as a digital camera or taken off in a memory card or USB thumb drive.

This data coming out in pure data streams which can then be loaded into spreadsheet programs such as Microsoft XCEL which has been around for a long time, is on large numbers of computers as part of the Student and Teacher Package for about $150 if nothing else. The program is well understood by many people and is easy to work with, by those with computer backgrounds. Nice graphics displays can be used to compare relationships among data streams, relating brake applications or throttle applications to other events to prove what actions are commanded or un-commanded. Aircraft crashes are commonly solved by such methods by showing whether control movements are intended or unintended.

In the Harrison NY case Toyota and NHSTA showed up and seemed to conduct a joint investigation and had what looked like a joint news conference with the local chief of police stating at a news conference that the evidence presented by Toyota that the brakes were never applied and that there was constant application of full throttle was “black and white.” But there was no way to know what Toyota was measuring and they made no additional comment before leaving.

For example adequate examination of the throttle application demands analysis of the

7. signal from the gas pedal
8. signal arriving at the ECU
9. signal leaving the ECU
10. signal arriving at the throttle actuator
11. feedback signal from the actuator to the ECU
12. similar signals to and from the cruise control.

So there we have a multitude of signals just to check on the integrity of the throttle control.

Things are no simpler with the brake control. There is a small piston up high on the brake pedal with a rubber boot or bellows around it which can only be seen by crawling around on the floor and shining a light up behind the steering wheel or simply feeling with the hand. None of the Toyota salesmen seem to know anything about this. This piston sends hydraulic pressure to something called a “stroke sensor” which sends a signal to the “stroke simulator” which issues further commands through the system. Pressure in the brake lines comes from a “pressure accumulator.” The master cylinder itself does not have to get involved in the normal course of business, with valves opening in the master piston to let pressure pass through. Toyota’s schematic shows at least pressure sensors in the ABS unit but it is hard to tell just how accurate these are. Toyota’s schematics for its flexible fuel cell in the rigid outer shell in years 2004-2009 for the Prius are highly deceptive and this makes other Toyota diagrams problematic.

So getting back to the question, where did Toyota get its data to indicate how the brakes and throttle were applied or not applied in the Harrison NY case. Impossible to tell.

Note that the data streams would be available to the public and subject to analysis with Toyota nowhere around, a welcome relief from their constant presence now. NASA has been called on board and perhaps they will have the stature to force needed changes through suggestion and public pressure. Organizations such as NASA and NTSB National Transportation Safety Board have led the way in organizing such studies in the airlines, but the airlines have been willing participants such all this data helps them better monitor the performance and maintenance of their airplanes. Toyota is not part of a similar culture.

Toyota has recently advised that they will make a few more “special computers” available to NHSTA and others but that only the raw data would be available to owners. With proper third party systems and un-encrypted data, Toyota’s failure to cooperate would be less crucial. Hopefully Toyota won’t stand in the way forever.

With all the bundles of wires under the hood there seem to be lots of places for third parties to break in to install their systems.

Considering the gravity of the situation Toyota is in, one would hope that they would see fit to bring their older cars up to speed in this critical investigation.

NEXT ISSUE

“””Driver error won’t totally excuse it. In a majority of fatalities related to recall scandals, you find driver error somewhere in the vicinity. With proper tire inflation and strict observance of payload limits, for example, Ford Explorers wouldn’t have had so many blowouts. Skilled drivers didn’t flip their Explorers after blowouts. And it’s the same thing for runaway acceleration. What does it say about the state of driver training in this country when an off-duty police officer’s passenger has time to phone 9-1-1 after their Lexus takes off unexpectedly, but the trooper doesn’t get it together to shift his car into neutral before it crashes, killing four?””””

I want to comment on several parts of your story. So Toyota puts out all this stuff to the press and on YouTube about how to stop a Prius. Much of it is either bogus on its face or does not hold up to strict examination.

Now let’s get things straight with Toyota. They have no interest joining the search for truth about all their vehicles. With $$ billions on the line why should they? They run around the country with what they claim is the only example of a “special computer” which is required to get any data from the Toyota Event Recorders (Black Boxes), which they claim require special programs to interpret anyway, and it’s all experimental and unproven, which is why they have not released any of this stuff. And most importantly that only a few seconds of data are retained in any event and only then into response to air bag deployment. Just to be sure I checked the Toyota web site recently and their official message remains the same. All of this makes one wonder what exactly those black boxed are good for. Well actually a lot more than Toyota has been letting on, if they can be believed in the James Sipes case, where they reconstructed the 20 minutes of data for the 30 mile ride. When it was convenient for them. There is a Laotian man in Minnesota sitting in jail since 2006 after being convicted of vehicular homicide since his Toyota raced out of control while he was claiming it was UA and even the jury is discussing how they think they convicted an innocent man, but I don’t see Toyota rushing up there to help out and all the prosecutor says is that she would allow the defense to examine the car again, when the first examination showed no defect. What about Toyota? They should not be allowed anywhere near another accident car. That means some independent parties should be given the codes to get data out of the black boxes and allowed to evaluate it somehow, possibly with Microsoft XCEL. What is going on in nothing less than a travesty.

(1) For example advice is given to simply press down on the parking brake. They only go to the rear wheels but they are completely mechanical, activated by cables. One only needs to sit in a Prius and play with the parking brake to detect the severe short-comings. It is clearly labeled to engage it with one press of the brake pedal and to disengage it with next push. So if one gives the parking brake a sharp clean shot with his foot, it goes down and stays down. But if he is nervous or stressed and wriggles it a little bit to make sure it is fully applied it comes unlatched and follows his foot up, something the driver might not notice. Or if it is applied and he checks and presses it again to make sure, perhaps because some police operator like in the James Sikes case in San Diego tells him to, well it come unglued. The lady talking to James Sipes was a “nervous Nelly” type anyway who seemed incredulous that the Sipes Prius was out of control in the first place. Not the type to inspire much confidence. The CHP along in his big Crown Vic had a much more commanding presence, demanding some respect at least.

(2) So next the driver is supposed to shift into neutral. So go sit in a new Prius in the showroom and check that one out. The Prius shift lever has a short throw and a vague feel, like a flight simulator game controller, that depends upon context for any meaning. But that is the whole idea, the Prius is sort of like a big video game much of the time, with all the video playing out across the dashboard, with many Prius owners especially the so-called hyper-milers trying to post big mileage numbers, so they can post them along their bio data on the Prius boards. The little tag on the steering column clearly states that the shift lever must be HELD IN NEUTRAL after shifting it there. I had not seen that one before. Besides shifting to neutral is “fly-by-wire” just like pressing on the brake pedal or gas pedal. The driver merely issues a request or a suggestion to the system. So nobody has really tested things out to see what would happen in the real-world, if somebody did actually shift a Prius into Neutral under WOT and failed brakes at high speed. Toyota won’t even admit that UA and failed brakes even occur and here they go telling us what to do about it
(3) Maybe Saylor the CHP cop did stand on the emergency brake and shift the little lever into neutral and neither did any good? How would we know? He and his family are dead and if there was anything on his EV Event Recorder, that indicated the true cause of the UA or loss of brakes or anything else incriminating to Toyota, Toyota could not be depended upon to help in its recovery and while pretending to attempt its recovery, would be in the perfect position to destroy all evidence and claim it never existed, actions which in the normal course of business would be called destruction of evidence and obstruction of justice. Again, Toyota should be forced to cough up everything having to do with those Black Boxes.
.
(4)So the Government is not trying to hurt Toyota. Right on. Toyota and NHSTA until quite recently were going around the country conducting what passed for joint investigations and giving joint news conference and putting out news releases that NHSTA had bought into their story that everything had been checked out and nothing was involved besides driver error. Until it got so bad especially after the Saylor case in California, that NHSTA told them to stop it, that they were tired of being made the patsy like when NHSTA had to go before Congress and expose their incompetence to the world. Toyota frankly had been going around acting like NHSTA had been bought and paid for and they could use NHSTA as they pleased.

NEXT ISSUE

The problems with the GX460 Stability Control System recently reported by Consumer Reports and actually acknowledged by Toyota simply have to be in the software, operating through various hardware components already on board. More and more responsibilities have been placed on these basic hardware items, mediated through computers, or as Toyota is careful to say, “electronics” and Toyota simply has not handled the integration of all these items very well and now many Toyota products, particularly the Prius, are beginning to look more and more like SOUTHERN ENGINEERING, beneath whatever slick veneer Toyota has managed to place around them. So Toyota is now going around the country with a certain laptop computer, which they claim is the only computer in this country with the software program capable of downloading and reading all the data in the Toyota electronic systems and that it is not offered more broadly or publicly since they claim the programs are experimental and unreliable and still under development etc, and lots of verbiage about how only several seconds of data are captured anyway and only if the airbags are deployed etc. Yet they show up all proud and pompous to study the Sipes Prius in San Diego and the Prius in Harrison NY and pull miraculously all sorts of data in both cases including the whole 20 minutes from the Sipes Prius and in the Harrison NY case they seem all anxious about how much data will be retrieved and then seem tremendously gratified in the second case about how much data was retrieved, indicating to skeptical observers that they were prepared to gloss over any data in either case that did not fit their agenda.

Toyota in its defense of the Prius has always maintained that there is no defect in the electronics, and by extension in other vehicles in their lineup including the Lexus, since their testing has shown no defects. But this is clearly bogus and hardly anybody with any knowledge of computers has bought into it. If Toyota itself believes any of it, it would indicate MAGICAL THINKING on their part, that merely thinking something is perfect makes it perfect, or that thinking some problem has been solved makes it go away.

“Safety Features” as used below indicates (1) ABS or anti-lock braking to shorten stopping distances and maintain steering control on slick surfaces. (2) Traction Control to reduce drive wheel spin on slick surface and improve steering control. (3) Stability Control to prevent dangerous situations related to “under-steer” or “over-steer” and skidding or tendency to over-turn during turns and other maneuvers.

The hardware items involved in the “safety features” include (4) valves in the ABS brake control unit to increase or decrease or hold braking forces in each wheel mediated by computer commands (5) speed sensors on each wheel (6) yaw sensors in two planes at right angles to each other detecting G-forces (accelerations/decelerations) in multiple planes (7) steering angle sensors to indicating steering wheel position to help judge driver intent. (8) ECU Engine Control Unit to increase or decrease engine speed. Note that the ECU is often called on by the system to reduce engine power but this necessarily requires that it be able to restore or increase engine power. (9) special radar unit in the Prius acting through hardware and computer programs to maintain proper distance behind other vehicles, based on traffic speed by increasing/decreasing distance as seems appropriate to the system. (10) Computers to integrate all of this, implying that these computers must be spot on to avoid dangerous mistakes. (11) Clearly missing in all this are ways to collect data in sufficient detail to allow careful study of the system over long periods of time, to study each component in relation to all the others whether a fault is suspected or not. This implies good graphics programs allowing easy study of the data.

So a lot of SAFETY FEATURES have developed from modest beginnings, based on a few basic features in hardware, and enabled by increasing computer power. There is nothing wrong with the basic concepts behind these “safety features” if they actually work and if they actually increase safety. It is my view that we should continue to evaluate all these “safety features” in real time to evaluate their continuing usefulness. For example some studies indicate that overall ABS anti-lock brake systems do not improve overall collision and fatality experiences, with improved safety confined to multi-vehicle accidents with less safety in single vehicle accidents and this leads to the suggestion that perhaps ABS systems lead to more aggressive driving. But suppose that we subject that to critical analysis. ABS units are commonly installed along with Traction Control Units and Stability Control Units, all operating through the same basic hardware and software. Random faults in any one of three systems could be causing accidents without calling specific attention to themselves. Suppose that we had sophisticated Data Recorders capable of recording perhaps 100 channels of data, each at twice per second with increased capture rates in burst mode occurring with sudden changes in any parameter. These would be comparable to the units in jet transports capable of capturing over 700 channels of data at comparable rates but the ones in automobiles would be much cheaper due to less requirements for crash survivability and due to economy of large scale production etc. These Data Recorders could hold large amounts of data in internal memory and plugging a laptop computer into them via a USB cable would have them show up as just another external device from which data could be downloaded, just like digital cameras are handled. Or they could in addition store data on data cards such as the CF Compact Flash cards. These CF cards now cost about $3 per GB (up to 32 GB) and are quickly becoming cheaper and larger. There would be universal programs which would allow various displays of the data, such as graphics displays of all driver inputs along with responses of the brakes, throttle etc. These sets of data could be accessed at random moments for routine study or after suspicious events. Completely unsuspected inappropriate responses of brakes and throttle etc, causing no symptoms, might be detected just on casual observation of the video. Plus allow careful study of any suspicious events. This would be enhanced by good graphics programs, simplified by the vastly increased power of modern computers with high speed graphics cards. Data could be easily shared with other enthusiasts. There is no evidence that Toyota is engaged in any of this and their constant claims that there is not possibly any defects in their electronics comes across as decidedly anti-intellectual.

For example consider that both the Traction and Stability Control Units in the normal course of events go around casually applying brakes to one front wheel or the other as deemed appropriate, but if this somehow occurs unexpectedly with nothing suggesting that it is about to happen, it can lead to an uncontrollable turning motion, a force irresistible even if the driver knew it was coming, leading to swerving into another lane of traffic or completely off the road for example, especially dangerous if occurring out of the blue or during some other maneuver. The driver might not even recognize that such an event had occurred and police investigators would not even believe him even if he claimed it had occurred. Note that numerous Prius drivers have claimed this swerving motion happened to them and they all seem quite mystified by it, and seem not to have considered the basic feature of their electronic systems that would have made it possible. This may have happened to drivers of other brands of automobiles as well, but the Prius groups spend more time going into all sorts of mystifying things that have occurred to them, yet they seem to resist inputs from non-Prius drivers and people casting doubts about the Prius seem to have their rights to log on and comment quietly cancelled.

Many of the faults of which Prius owners complain online seem to be software related, but reading through lots of these posts reveal little if any attempt to define what the mechanism of the fault is. One would like to develop the simplest theory which explains all the known facts and to make predictions which could be tested by experimentation.

The Federal government requires that the newest jet transports monitor a minimum of 88 definite parameters in their “black boxes” but actually the manufactures have gone far beyond this, monitoring over 700 channels of data in the latest models, including lots of data from the engines, positions of controls in the cockpit, position of control surfaces, lots of data at intermediate points etc. In general these systems are designed to sort out intended from unintended (un-commanded) events (among other things.)

Data recorders are not mandated in automobiles for two more years but the Big 3 are generally installing them anyway and making readouts easily available to all interested parties whereas Toyota is holding on until the bitter end and will not have them until 2012. Flight data recorders in airplanes monitor over 700 channels of data typically several times per second, with much higher rates in burst mode if sudden changes are detected and typically record 17-25 hours of data. The requirements for automobiles are much less rigorous and not well described online. NHSTA has mostly been standing quietly by, waiting to see what the car manufacturers would do on their own. Good luck NHSTA, we are proud of you.

If Toyota really wanted to prove all its assertions about its electronics it would install data recorders capable of storing data for many hours from all significant sensors already on board as well as all commands issued to the hardware items. This would include all the signals (commands) to the brakes (pressure increase, decrease and hold) signals to the throttle and how these commands correlated with driver inputs from the gas pedal and the brake pedal etc. One should keep in mind that all commands to the throttle and brakes are merely “requests” since all systems are “fly-by-wire.” To repeat, there is no mechanical connection of the gas pedal or the brake pedal to the respective modules in the engine compartment.

Good data collection is increasingly feasible since digital memory is dirt cheap and getting cheaper, as well as coming in larger and larger capacities. Various “safety features” such as stability control will be mandatory in 2013, so it seems reasonable to establish ways to ensure that it is actually doing something for us. Perhaps supposed increased safety on the roads is due to over-zealous police action rather than any safety improvements in vehicles. This means that lots of problems caused by electronic “safety systems” can easily go undetected amid all the noise. Meaningful analysis requires breaking out the actual numbers, to place blame or credit where it belongs.

The ABS has yaw sensors to detect skidding sideways & speed sensors on all four wheels to prevent brake lockup, improving both stopping and steering . The ABS operates on slippery surfaces by using valves in the ABS brake control unit to increase and decrease braking forces at each wheel up to 20 times per second, yielding its familiar pulsation. The ECU simultaneously operates to reduce engine power as needed and this requires the ability to restore or INCREASE engine power when appropriate.

So next consider TRACTION CONTOL, which acts by detecting the drive wheel that is spinning the most and applying a braking force to that wheel, sending more power to the other wheel with less spin due to better traction and reducing engine power as part of the total effort. Note again that the system must be able to INCREASE (restore) as easily as it reduces engine power,

STABILITY CONTROL operates through the
Further, Toyota in its Prius model has this cute little feature that uses radar to help control the distance from the vehicle in front. Again one can see that the system must have the capability of either decreasing or INCREASING engine power and thus vehicle speed to get this done.

So we have all these neat safety features which depend on many different things operating perfectly in concert, controlled by computers, which must be operating without error. Notice that all of these features have the requirement of both decreasing and increasing engine power and vehicle speed. This provides the perfect mechanism to explain all the episodes of unintended acceleration UA that Toyota continues to deny, except to invoke bogus explanations like gas pedal entrapment under the floor mats and issues of stuck gas pedal which can be cured by inserting a magic little shim. These two explanations for UA meet with little acceptance outside of Toyota itself and those who seem to be their fan boys.

So my point is that all these software issues now interact with all the basic hardware features already on board to produce the PERFECT STORM for Toyota. So there was the well-publicized San Diego case of James Sipes, with the wild 20 minute 30 mile chase down the freeway, followed shortly by the case in Harrison NY where a Prius underwent UA from a standing stop and struck a stone wall. These two cases, the first at high speed and the second from a standing stop simply indicate to me that the PRIUS IS UNSAFE AT ANY SPEED.

ISSUES PRIUS BLADDER FUEL TANK

http://priuschat.com/forums/gen-ii-prius-technical-discussion/30593-gas-bladder-exposed.html

Summary: There needs to be a general recall on all Prius gas tanks prior to 2010 because of potential for emissions control system failures.

David, this is a follow-up on our conversation at Boulder Toyota several days ago regarding the Prius fuel tanks prior to introduction of the conventional tanks on the 2010 Prius. .

So this is the so-called bladder gas tank of the Prius. The whole thing is a nightmare. The top and bottom are molded resin rather than metal as I had stated. It is not clear how much if any the bottom flexes to accept gasoline. It would have helped if he had shown what I assume is the bellows action that must occur along the sides as the tank fills, that is how the tank gets deeper. In any event the pictures shown in Toyota’s manuals are clearly and highly deceptive and arguably meant to obscure the issues. Frankly I can see how Toyota would not want to admit just how poorly designed this tank is. You are a sales and leasing representative for Boulder Toyota and presumably highly trained but you had trouble believing my description of the tank and for good reason. “Highly implausible” is how one might describe the design of this tank and Toyota should never have approved the design anyway.

There are lots of posts online of users having trouble getting the tank to take on acceptable amounts of fuel, the tank spitting back after filling, of tanks being ruined due to presumed over-filling leading to replacements costing $1000, of users unexpectedly running out of gasoline, being stranded in dangerous situations, of expensive calls for tow trucks etc. Various manuals go through drawn out procedures for testing various performance issues with the tank that seem highly subjective to me, with high probability of not solving anything. There seems to be a strong tendency for Toyota to blame the user for over-filling and refusing to take any responsibility whatsoever for anything.

There are lots of valves and controls and everything seems confusing. I would not blame Toyota for not wanting to be involved and taking the path of least resistance and declaring everything to be “normal.”

Ventilation, by manifold vacuum, of the space between the rigid metal outer shell and this resin bladder, is not constant but occurs only upon a “duty cycle” driven by the ECU. So if gasoline is leaking into this space due to a leak in the bladder, ventilation might not be adequate to keep it cleaned out. Gallons of gasoline accumulating in this space would then lead to reduced fuel tank capacity. I would not be impressed by any of Toyota’s claims that this would be easily detected or impossible in the first place. All the steps described in diagnosis show just how uncertain Toyota itself is in the whole process. Such leaks are a constant concern to Toyota as they discuss the issues. Gasoline leaking into this space and being ventilated away clearly would help explain many of the low mpg problems discussed online.

If this could be developed into an issue of failure of the Prius emissions controls, Toyota could and should be forced into a massive recall to replace the tanks with the new more conventional tanks.

So there is a valve run by some sort of detector that is supposed to allow air in the space to be expelled to the outside as the bladder fills and failure of this valve and/or detector would account for lack of ability to accept gasoline and spitting back etc.

Knowing where everything is, one should simply put the car up on a hoist and drill a small hole in a dependent part of the tank, using a battery powered drill to avoid sparks which could cause ignition of the gasoline inside the rigid shell . This would allow any gas to drain out. One should have a can to catch it in or at least be prepared to place a finger over the whole to stop the flow of gas while somebody fetches one. Remember that there might be as much as five or six gallons of gas. This would solve any problems related to retained gasoline and/or air. One could then see if the tank performs better.

Further study would consist of drilling other small holes to allow sounding with metal rods to detect any “tilting” of the tank at near-empty. Any such tilting could result in significant amounts of unusable fuel, since the drain in the middle would simply not be in the most dependent position. Large fuel gauge errors would also result. This probing with metal rods with the tank filled to capacity would detect issues with the bellows action not allowing full expansion.

Any holes in the bottom of the rigid shell could be easily repaired with small self-expanding rubber plugs commonly used to mount roof racks etc. Removing one or more of these plugs periodically might possibly but not necessarily allow one to detect gasoline leaks from the resin bladder. But in any event these leaks would be corrected by the general recall which I propose.

Another diagnostic approach would be to drill one hole big enough to insert a bore-scope, such which a mechanic might use to look into the combustion chamber of an ICE internal combustion engine. A proper bore-scope would be mounted on a small digital camera to take pictures for documentation.

Suppose that one does detect a significant leak in the resin bladder and returns to Toyota with the evidence, it would seem that Toyota would be now be more anxious to play ball and replace the tank in or out of warranty, just to suppress the news.

It might be better to keep the tank for awhile though to do further research. For example one could mount an after-market fuel drain on in the most dependent part of the tank, operated by a cable with the knob behind the fuel filler door, with the drain tube coming out at the driver’s feet when he is adding fuel. Making a big deal out of catching the gas running out toward the ground could add a lot to a person’s day, provided enough people were watching.

Steve