hypocritical of edmunds to try and have a contest to try and find something that does not exist. just another trying to take advantage of the foolish followers of hyped and non-scientific research and evaluation. it is only intended to add to the fear so that you think edmunds is trying to help.
Why would water get into the INTERIOR of car. It is not a GM product? And so what if water got in. It has been tested and causes nothing, as usual.
#314 of 315 Re: Naked APP sensor board free for water contact [srs_49] by nguyenngocbich Mar 12, 2010 (5:56 am) Save | Reply
Replying to: srs_49 (Mar 12, 2010 3:34 am) I used worked with electronic boards for long period that I believe this PCB is coated only printed board, soldered connections of componets with printed board are not coated. I just have found other error of the gas pedal when watching video "“An important update: 02.11.2010” from www.toyota.com, that a metal back lid of the gas pedal was assembled without gasket! That means moisture or water may enter the gas pedal easily.
I have see on TV the Onstar add.Big sister comes on and says "Mr. Jones I see you had an accident"." Are you hurt and do you need any Medical assistance or and a tow truck" ? I have Bluetooth in my car and It makes me nervous that it can be hacked into. Do you think some messed with Toyota's software? Onstar can stop your car in its tracks and on you 24-7 when your driving. Car makers ca'nt afford redundant systems. Although we might get a black box mounted some where on vehicle.
If the engine rpm increases by 200rpm at idle, to compensate for a starting radiator fan. How big of an rpm increase will occur if that fan motor starts shorting out and creating a much higher load on the alternator? Did they put a rpm top limit for all current loads? Just wondering....
The problem with the 'unintended acceleration' or the Toyotas is that there is no problem. It is simply driver error and old people are smashing the gas instead of the brakes.
I'm sure most car manufacturers have an in house test for Radiated Susceptability. Check with the IEEE in Detroit. So what? who cares who wrote/approves the test? The most thorough is MIL-STD-461. This specifies an EUT (Equiptment under test). An EUT can be a relay, a radio, a radar a gizmo du jour or a CAR. I personally use modified MIL-STD-461 testing when I trouble shoot these EMI (electromagneti interference) problems. Years ago I contributed to the writting of 461. And of course a horse designed by committee looks like a camel.
Here's another link to a news story talking about how GM is capitalizing on Toyota's uncontrollable acceleration woes. Talking in terms of nature, that would be a classic move for many predators in the wild - inflict a serious wound, then step back until the victim weakens and begins to flounder, then move in for the kill, and the feast. I thought the most interesting aspect of the story was where it was mentioned that the other major car makers like Ford, Honda, and Nissan, have not involved themselves in this battle. Very interesting that one specific US car manufacturer jumped at the opportunity to increase it's own sales based on the "misfortune" of another car maker, while other car makers, both US and Asian, did not attempt to do the same thing. If anything, if a problem truly does exist, the safety of the consumer should be the ultimate concern of all auto manufacturers. And, all the major automakers should be working together if necessary, to get to the bottom of this issue, if again. this issue truly does exist at all. This is getting really interesting, and the pieces all seem to be falling into place. It's gonna be a hoot to see what ultimately all comes out in the wash.
Now I'm no expert, but here is the first thing that I thought of when I heard of this problem: So a car can't run without air being sucked through the intake right? It doesn't matter what is wrong with the engine, no matter what it won't run if the throttle is closed all the way, thus no air for combustion. For the engine to accelerate, more air HAS to be allowed to flow into it. IF the gas pedal is fully depressed then you have to assume the the throttle is only partially open, and no matter how much you mess with engine it won't accelerate dramatically because of the small amount of airflow. So the problem has to be with the throttle...So I think the aerodynamics of the throttle body/ intake are causing the throttle "flap" to open more causing the engine to accelerate regardless of whether the gas pedal is being pushed. Just my thought.
Here's another story concerning the uncontrolled acceleration incidents that are being reported in the news. Goes along with what I've been offering as a possibility from the beginning. Does anyone else hear the the distinct sound of a ticking clock like I do?
The ideas that UA is a scam by various unrelated people, or just problems of the old and simple hitting the gas instead an be refuted by a few simple questions:
1) If it were possible to run such a scam and profit from it, why wouldn't people be running this sort of scam every year? why was it only Audi 30 years ago? and now it's restarted mainly with Toyota?
2) why if it's a scam is Toyota' s UA rate much higher than any other manufacturer?
3) if it were so easy to run a UA scam, I'm sure the former Big 3 would have wanted to do something like that 10-20 years ago, when they were losing market-share to the Japanese.
4) if UA is a problem with older people, then you should see the same statistical rates of UA across all brands. You don't - you currently see Toyota much higher.
This is not to say that out of all the reports that there aren't a few people trying a scam or a few people who did get the pedals mixed up. But this does not account for the statistics showing Toyota to be significantly higher and different.
Here's the link. As you can see although Toyota does have some issues with UA, many other manufacturers have an astoundingly higher rate - almost triple the rate for Toyota (so much for the assertion by other posters that Toyota's UA rates are much higher than other auto manufacturers). And, in some recent years, a US manufacturer, Ford, even surpassed Toyota for this issue (possibly the reason they have kept a very low profile in terms of jumping on the Toyota bashing bandwagon). But, where is the news story about those other manufacturers.
But, if you can see, most of the auto manufacturers that have the astounding rate of UA, although large manufacturers in their own right, mostly can be considered niche market auto sellers. So essentially, they really don't offer much of a threat to "the big 3" as it was put. The only manufacturers that offer any true threat to the big 3 in terms of sales to the general mass automobile market are, the Asian automakers, Toyota most specifically. And, just to clarify, the fact that the big 3 have been recently driven to bankruptcy or bailout, speaks for itself as to why no "scam" as it was put, or for that matter, any legitimate plan was implemented "10-20 years ago" to deal with loss of market share to Asian automakers - US automakers were either just too fat, rich, and lazy, or just simply asleep at the wheel.
Professor Gilbert’s Theory on what is causing the unintended acceleration is basically correct.
The problem is WITHIN the LSI Driver IC chip that controls the engine. No, it has nothing to do with the design of the chip from an electronic design standpoint. That's why everyone at Toyota is scratching their heads over this problem.
I have seen this type of problem in development of a new product in the past. The LSI chip was the culprit. The engineers did not believe me when I showed them how I could make the problem appear with just using basic electronic troubleshooting location isolation.
It would not happen at exactly the same time, but depending on the LSI chip, I could make it happen over and over again in the lab. After the chip manufaturer was made aware of how we could duplicate the problem, they changed their chip manufacturing process and some additional resistors and capacitors were added to the chip legs on the defective LSI Chips to compensate for the problem preliminary LSI chips we had received.
So the $1,000,000 answer to the question is as follows.
1. What is causing the problem? Answer: The process that is used to close the IC chips alows mosture into the chips. When the right tempature is reached, it basically rains inside the LSI Chip. This will cause shorts and resistance across the interior of the chip. That in turn makes the chip act in the manner currently being observed - unintended acceleration. Turning off the car and turning it back on again will reset the LSI chip and it will look normal again. NO TROUBLE FOUND!
2. How can I make the problem appear? Answer: Use a HEAT GUN and a can of FREEZE MIST to track down the defective chip. That's how I did it. When I hit the defective LSI Chip with the FREEZE MIST to cool it - WHAM! The chip turned on the motor and heating it did nt stop the chip. It was in a runaway condition, until you turned off the power.
3. How can this be fixed? Answer: The defective chips must be replaced in ALL TOYOTAS. But you can't use the current LSI chips being manufactured because they are all being manufactured the same way.
The Manufacturing of the LSI IC Chips must me changed and an inert gas must be used to seal the IC chips during the manufacturing process. Also, additional design changes need to be put into the LSI chips to prevent the runaway condition from happening. Since this LSI information is all propriatery to Toyota, I can not give you any additional information on the design changes needed to the LSI Chip.
Bottom Line - The LSI IC Driver that controls the acceleration is defective. There is no way to tell how many defective chips have been installed. I would say ALL need to be replaced. They ALL Have the potential to act in the same way and there is no way to prove, without a shadow of a doubt, that you don't have one in your Toyota. I have a 2009 Toyota RAV4, so I am very concerned.
I would suspect that the defective IC chips were all made in the same plant, using the same process to seal them.
LG 71' B.S. Electical Technology New York Institute of Technology, Old Westbury Campus
So, what exactly was the problem you were tracking down - the one that was caused by the LSI Driver IC? Was it the UA problem, or something else?
If that is what the problem is - moisture inside the chip cavity, it should be a simple matter to remove the suspected part from a vehicle that has experienced UA, open it up or remove the encapsulant, and visually inspect the die for moisture related defects. They are pretty easy to see, especially for someone who does that sort of post-mortem analysis for a living.
The problem I observed was with a new LSI designed chip in a new daisy wheel printer being developed. During the pilot stage and prototype demos being placed in the field, we heard that the printweels were being sheered off. Two of the demo units were brought back to see if we could observe the problem. Sure enough, after printing on the lab workbench it happened.
The print hammer (activared by a relay closure action) would just stick on and the printwheel motor would come on full speed. Spinning pettles went flying everywhere. The only way to stop it was to turn off the power. When you turned the printer back on.... it functioned normaly for awhile, no trouble found. Very strange behavior. It would happen at random for each demo printer. IT WAS VERY DIFFICULT to determine the real cause.
The people from the Engineering Department did not believe it was the LSI Chip. I had to bring them in, along with the engineers from the company who manufactured the LSI chip to prove I could make it happen.
The problem LSI chip was located on the Power Board. So it was responsible for turning the power on and off for the printer hammer (relay power) and the printwheel motor power control. At random, the power to both would come on full voltage!
SO DOES THAT SOULD LIKE THE SAME PROBLEM? YES IT DOES!
The basic report back from the Engineering Department put the cause of the problem in the sealing of the LSI chip at the manufacturers plant. Now, this is in Silicon Valley where there is moisture in the air. They stated that it was basically raining inside the LSI chip. The chip manufacturer had to change the process and insert an inert gas to fix the problem.
Since these chips are being manufactured in countries that have high moisture in the air, it is reasonable to assume that the chips are being closed and not evacuated prior to sealing them.
Since I was not a witness to the opening of the LSI Chip or the evaluation process at the chip manufacturing plant, I can not tell you more. I suspect the problem with the Toyota chips is the same issue, moisture inside the chip. Thus it is random and I suspect a temperature activated related problem. It depends on where the mosture, when condensing, comes in contact with the traces within the LSI chip.
That is where I would focus the investigation.
I sent a message to the Director of Engineering at Toyota about a month ago that I would consult with them on the problem. They would pay me if I was right and it would not cost them anything if I was wrong. They never responded.
The next thing I heard on the news was that they had a fix... the shim on the gas pedal. So I thought I may have ben wrong about the problem. Now it looks like I may have been correct in my theory of the problem.
When the news broke that night (nationally) about James Sikes and the "apparent" runaway Prius, my wife and I almost simultaneously agreed he was a complete liar as we watched the "spectacular" news unfold, in San Diego, on the cell, Toyota reeling out of control. Wow, what a coincidence.
I posted previously that media hype over Toyota is going to turn into a can of worms - and not for the right reasons. Now if you screw up and run a red light while driving your Toyota.... Well, you can guess what the low-lives are going to start claiming.
Nice post on data saying unintended accelerations are nothing new, This Toyota thing has GOT to be put into perspective!
Once again a warning to folks - if you're in a car and the driver seriously does not know where or what neutral does, get out of that vehicle!
Skimming through the NHTSA reports for these other manufactures you are going to find far less pronounced UA problems, such as increases from 60 to 65mph while in cruise-control mode, but not really the kind of wide open throttle problems being reported against Toyota, not to mention the braking problems with the Prius.
Now, this is in Silicon Valley where there is moisture in the air
Actually, Silicon Valley is pretty dry. I believe it's classified as semi-arid (San Jose, 14.4 in rainfall/year).
Since these chips are being manufactured in countries that have high moisture in the air, it is reasonable to assume that the chips are being closed and not evacuated prior to sealing them
Not likely. All of the chip vendors I'm familiar with manufacture their devices in environmentally controlled facilities, class 10K or better clean room facilities. The chips themselves are probable never exposed to ambient air. Dry nitrogen is the usual gas constituent in the areas where bare chips are handled. Dry nitrogen is almost guaranteed to be used where the final packaging occurs - what you call the sealing. So it is highly improbable, though not 100%, that moisture is being sealed in with the chips on a regular basis.
Now, once the chips are packaged, unless they are in a ceramic, hermetic package (not likely because of costs involved), is is possible for moisture to leak in, particularly where the leads pass through epoxy (plastic) package that contains the die. Even then the probability is remote. Even in military avionics and space applications, the industry is learning how to use plastic encapsulated ICs with good success, something that was unheard of 20 or 30 years ago.
Plus, like I said. Moisture and it's effects inside a package on the die is pretty easy to detect.
Questions concerning "stuck" accelerator incidents: 1) What percentage had "cruise control" engaged or in standby? VS shut-off? 2) The 2- Hall Effect sensors in pedal: Do they have built in signal conditioning providing low output Z or do they provide a high impedance signal to ECM? 3) What type sensor provides feedback from throttle control butterfly and is it's signal conditioned? 4) Can the stepping motor activating the butterfly be electro-mechanically de-coupled or is it integral to the ETB?
Interim Solution: Since intermittent problems are always frustrating and may require a very long time to find, fix and test, the following may be feasible. Brake Switch to disconnect servo-motor electrically (preferably mechanically).
Here's the link. As you can see although Toyota does have some issues with UA, many other manufacturers have an astoundingly higher rate.
Why are you posting data from the1990's? Toyota cars from the 1990's are nothing like the hi-tech cars they put on the road today, where every transmission, ABS, stability control, cruise, and ECU's are so complicated and interlinked electronically. No one's questioning that Toyota had great quality and no significant UA until just recently. All going back 20 years and counting that does, is downplay how bad the recent string is.
Someone put a link here to 2009 reports, and that was bad for Toyota. And I'm sure if 2010 data was available you'd see that now everyone who has a problem is contacting NHTSA, and the numbers are much, much worse.
Again if fraud and running-down Toyota was possible when there was no real problem, people who compete would have done so years ago. There has to be some truth to the UA problem being exceptional, no matter how much you think it is overblown. All companies probably have some UA over many years; something has happened with Toyota vehicles, to make them standout recently.
The problem I had experience with was 30 years ago.
Even though the world has more knowledge about sealing these chips, it does not mean errors in manufacturing can not occur.
As you say -- Moisture and it's effects inside a package may be pretty easy to detect, if you are looking for this issue with moisture on the die.
It's something that needs more investigation to rule out this possibility. Also, you have to look at the chips in the cars having the problem. It may not be all chips, just certain problem LSI Chips that were manufactured overseas.
I lived and worked in the bay area 27 years ago. YES, it does rain there and floods at times. So mosture is in the air, but like you state - the chips these days are made in a controlled room. Then again we are talking in the USA.
When your trying to drive down cost, you have the chips made overseas. We have seen where they can be a bit lax in their controls during manufacturing. Maybe they did not use the correct gas while sealing the chips or the hermetic packaging process was sloppy. We are zeroing in on chips made for Toyota. That may be why we don't see the issue with other LSI Chips in other cars.
It's still worth looking into to rule out this as the problem.
Actually, concerning the issue we're talking about here, the specific type of UA problem with any make or model of automobile, or debating the difference of a percentage point or two, is not the point. The point is that in certain recent years there have been other auto manufacturers (including US manufacturers) whose figures, regardless of the specific mechanical, electrical, or software cause, have neared, equaled, or surpassed Toyota for UA issues. And yet, Toyota has been the one investigated, and who has gotten the media attention for this problem, at least in the public's perception. My question (rhetorically because I have already posted what I believe to be the answer) is what was the catalyst for this singular zeal for finding out about a supposed UA problem with Toyotas, when most auto manufacturers have experienced this issue to varying degrees, and for whatever cause, in past and recent years? I believe Toyota makes a good, sound vehicle, but believe me, I'm not championing it's cause for some great love I have for the company. No, I am motivated by the prize offered, and also to be the one who proposed the ultimate solution for the alleged Toyota UA issues. Let's see what happens?
Doesn't look like the 90's to me, The link I put up was a recent article, and the table of figures listed NHTSA confirmed data gathered from 1990 to 2009 - that was just as recent as last year by my calendar. Now that link you talked about in one of the posts here, was it actual 2009 NHTSA confirmed statistics, or something along the lines of as yet uninvestigated and unconfirmed reports to the NHTSA, or even just media stories, or statistics? Actual NHTSA statistics that were based on confirmed, investigated reports, and statistics based solely on reports that have not had fully concluded investigations, are two separate animals. Actually in this part of your reply, you kind of make one of my own points, you said - "And I'm sure if 2010 data was available you'd see that now everyone who has a problem is contacting NHTSA." Obviously, like I said in my previous post, although there is the possibility of a legitimate issue concerning UA in Toyotas, if there is not, many people might have many reasons for making UA claims against Toyota now that it has had a media blitz that it has. I'll leave it to everyone’s imagination as to why, but, does the adage "blood in the water" ring a bell?
Thank your for information. I hardly believe that water is culprit of most SUA without any evidence, but where water can affect to the electronic equipment to make malfunction to car control? This need to research to find! Due most SUA occurred with CTS gas pedal that I pay more attention to APP sensor circuit board as you see. Do you believe that your product can work correctly in wet condition? This easily to check with an ohmmeter. I think air inside cage where the board is located may have high humidity when assembled in high humidity condition of some manufacturing plant, example in rain day. IC produced in more artificial condition that their have good quality, especially they are made in USA, but assembly APP sensor circuit board to gas pedal may be carried out in open to ambient condition.
If my Toyota starts having a unintended acceleration what will happen to my car if I throw the car in neutral while it is going 70+ mph? Will it cause a crash?
Information from #310 by srs 49: And how do you know that the board is not coated with something? A number of the moisture resistant products out there are very thin and clear, and it is all but impossible to tell from a photo whether a PCB (printed circuit board) is so treated or not. Parylene(tm) coating that are put on by vapor deposition are almost impossible to see, even when you have the board in your hands. Humiseal(tm) is an acrylic that, if sprayed on thinly (as opposed to dipping the board) is difficult to tell if it's there or not from a photo Information from #317 by srs 49: The boards we do, we coat everything except for the connector pins. This is done after final assembly, and before board test.
From the information I can make conclusion that your APP sensor circuit board can not work correctly in wet condition because sprayed parylene could not create a continuous film to coat the board from moisture, especially bottom side of the hall IC and backside of IC legs were not sprayed (that the APP sensor board need to be located in sealed cage on the gas pedal). SUA may occur with gas pedals what have high humidity in confined space of APP sensor board. In vehicle with running air conditioner, when cabin’s temperature reduces to right level, humidity will condensate to water on the APP sensor board, enter underneath the IC (due lower position of IC on the board) and create short circuit between IC legs follow to situation as professor Gilbert’s experiment! After SUA, the vehicle is in rest for time that water on the APP sensor board disappeared to humidity that nobody can identify what caused SUA. Why humidity is exist in gas pedal? If APP sensor board is assembled to gas pedal while ambient air with high humidity (example in rainy weather), air with high humidity can enter into space of APP sensor on the gas pedal and exist in sealed confined space for long time. Measure to eliminate: APP sensor board from recall vehicles will be remounted from gas pedal, dried and coated with thick layer of parylene or epoxy, reassembled in dry condition to the gas pedal and sealed.
From what I've read, Professor Gilbert made modifications to the circuit - rewired it and was able to cause UA. This from Toyota's response to Professor Gilbert's experiment:
The analysis of Professor's Gilbert's demonstration establishes that he has reengineered and rewired [my emphasis] the signals from the accelerator pedal. This rewired circuit is highly unlikely to occur naturally and can only be contrived in a laboratory. There is no evidence to suggest that this highly unlikely scenario has ever occurred in the real world.
Well, what's surprising about that? Rewire the circuit and cause a problem???
Now, if your theory is that water condensing on the circuit card someplace can duplicate Professor's Gilbert's results - that it can mimic his circuit changes -, then yes, that experiment should be run. But, I would be surprised if Prof. Gilbert did not do that himself. What does it take, some water and the circuit card? The fact that nothing of the sort has been reported leads me to believe that the Gilbert did run an experiment with the water (especially if he's as sharp as he and Congress think he is), but nothing out of the ordinary occurred - that no UA was caused.
Don't get me wrong. Gilbert did some good work here. Being able to reproduce a problem in a more controlled, repeatable, laboratory environment is a big step towards understanding exactly what his happening. It's just that he failed to go the next step and demonstrate how his "rewired circuit" could occur in the confines of a passenger compartment.
Thank internet help us to exchange thinkings! I think, professor Gilbert could not repeat phenomena created in real world by water because water as natural superman force can create one phenomena like professor Gilbert’s experiment result from hundreds of short circuits in the APP sensor card. Most of the short circuits ECM can detect and only some times can cause SUA! I try to understand mechanism of SUA and give measures to eliminate but not try to repeat the phenomena because I understand this is impossible!
I work in the forklift/material handling repair business and have so for about 20 years now. I specialize in Electric units. That being said: "I have never personally worked on an electric or hybrid automobile." I do recognize that some of the controllers that have served in our industry have been adapted, to the automobile industry (at least at the local enthusiast level). There has always been an issue with RFI and magnetic shielding. anecdotally there are several foundry customers that will not place these vehicles around a working smelter. Due to the high levels of electrical current and induced magnetic fields. There has been several occasions where I have responded to a foundry to "re-initialize" the controller, by rewriting the model number to the controller"(theoretically - this could be done by a "watchdog checksum" circuit on power-up...That would make it impossible to detect fact settings had been changed by induced voltage/magnetic fields) These same voltages/fields could probably be induced by the common "linear amps" in use by truck drivers. This would be very difficult to determine, in that the strength of the signal would not be known(and thus the effect on the shielding of the controller). I am wondering now... if a "Driver" was useing a linear amp on his CB, while talking it would tend to (possibly) induce small spikes(as this is an analog Frequency Modulation signal). If there was a "reverbe or echo" function on this radio, would this not allow a sustained, STRONG RF signal across a wide frequency(wider the closer you are to the transmitter) to affect this type of controller? This being said it takes much less to affect some of our lift equipment, and an example of the specs would be (protected against 20 gauss, 15kv static discharge,etc.) Just a thought. Whats your opinion?
With memory devices becoming cheaper, what is needed is a blackbox type routine that would monitor and save sensor readings for a certain length of time, in a continous loop. This is the only way they will know what the computer 'saw' during the time of the accident. If it is interference that causes the problem, a black box will indicate what readings got compromised..
As it is now, they have CEL codes. like TPS stuck open, TPS stuck closed.. etc.. But they don't have codes for Unintended Acceleration.. And When the most powerful computer (the driver's brain) reports an Unintended acceleration, the manufacturer will accuse the driver of lying.
Because of cost constraints, this function, if implemented, would probably just be added to the existing ECU's processor. The downside to that is now you've made the ECU and its resident software even more complicated. Instead of doing its primary job, controlling the engine, now the SW will have to log sensor readings (why stop with just the throttle position) and write that data to some sort of non-volatile memory in a circular buffer fashion. When writing to the NVM, the SW will have to manage bad bits/bytes/words/blocks in the memory, make sure write cycles are not being exceeded (flash type memories have limits on the number of times a cell can be written), etc. And it will have to continue to do this at the rate required so satisfy the control laws for the engine's closed-loop functions, which may requires a faster processor, which may require re-compiling and re-verifying the code.
You can look up my previous post about this - it's post 281. Like I said there, if the engine is in high revolution from UA, and then you put the transmission in neutral, there is the possibility of causing damage to the engine. But again, once the drive system has been disengaged by placing the vehicle in neutral, you can begin to bring the car to a stop by applying the brake, out of other motorists way if possible. And, better a damaged engine than a damaged driver, or passengers.
No. I used to shift my V-8 Firebird into neutral many times at 60mph, to save gas coasting down mountain roads. There is no big ruckus from the automatic trans. either going out of drive or reengaging, say at 50mph.
What will happen though if you're experiencing UA is now your engine is no longer slowly gaining rpm's as you gain speed. If you shift into nN with full throttle, your engine is going to hit redline very quickly. I'm not sure if there's a rev-limiter, or if it works during certain UA, but a blown-engine may be the result within 10-30 (?) seconds.
"As it is now, they have CEL codes. like TPS stuck open, TPS stuck closed.. etc.. But they don't have codes for Unintended Acceleration.. And When the most powerful computer (the driver's brain) reports an Unintended acceleration, the manufacturer will accuse the driver of lying. "
You can try to accuse the driver of lying, but there are too many people reporting the same problem. This has to be taken very seriously, lives are at stake.
The error codes will never appear because the point it captures the codes is prior to the point the problem appears. A defective IC Driver chip will turn on the power to the output device, in this case causing the UA. The monitor is looking at the input signals to the LSI chp only. So the computer thinks that everything is working normal. The computer ASSUMES the correct signal is right, which it is because that is whi it is monitoring - the input to the chip - and assumes the output will be correct. The Input signal is normal, however the output of the LSI chip is not!
Because the LSI driver is defective in some way, the output is at MAX and is not being monitored. Thus assumptions are being made the chip is not defective and people are looking at other points for the problem.
Think of it this way. Suddenly your stereo goes on FULL BLAST! The input to the final amp stage is at the correct voltage/db level and so the sound should be normal but it is blasting. Why? A problem with the final IC stage is defective.
In an LSI (Large Scale Intergration) chip, you have many small transistors working together. If one of those transistors shorts - wham - the voltage to the next stage is at MAX. This cascades through the other transistor stages an the output goes to MAX!
Now take that faulty internal transistor and make it flakey with temperature. As the temperature changes there is a Notch effect in some IC's! As the temperature rises, the transistor is good, then goes bad, and then can go good again. This is well know in QA circles as a problem with IC's.
Some companys use hot boxes for final testing of products. This is where they find those defective IC's, that act this way on final circuit board testing. Take out this QA testing and you run the risk of sending out defective products.
Now apply this information to the Toyota problem and you can see the possibility of having a bad IC as the problem causing UA. You need to take of the cars with known UA and do a complete analysis of the electronics (in a heat/cold box with the entire car) to locate the defective part. I believe the enviormentals are a factor in the UA issue.
I too believe the issue is with the Cruise Control system:
I am a software designer and tester, who prides himself on the ability to break software. I have called Toyota AND e-mailed them, asking them to contact me, but I have not heard from them which is why I am on this forum. I guess Toyota has their hands full right now.
This problem has track marks of a software glitch all over it. This is of course is only a theory, but I believe it is one that makes good sound logical sense.
So here is the million dollar question, as it were. What would happen if the car's computer was confused about what state it was in, meaning cruise control ON or cruise control OFF?
So if the cruise control was actually ON, but some of the local variables within these functions considered the system OFF, some very dangerous malfunctions could occur. For example, if the Resume function was engaged, then the car would accelerate on its own. HUGE RED FLAG HERE. There would be a brake function where when you tap the breaks it disengages the Cruise Control. This function might only be called, if the Cruise Control is actually ON. SO, if the braking function thinks the system is OFF, but it is really ON, then the brake function to disengage the Cruise Control would not be run.
This is similar to a Three way switch for a hallway light. Normally, flip the switch up and its on and down and its off. But on a three way switch, it changes depending on what you have done with the third switch and you can't tell whether you are flipping the circuit on or off. This would also be a reason why they are not able to find the problem. Ironically, the system is working as it should, hence no diagnostics would uncover it. The system is working appropriately to the state it THINKS the cruise control is in.
How could the state of the Cruise Control become screwed up like this? Well quite simply, an event happened that the Software designers did not plan for. Perhaps, something physical such as the cruise control was on and the car stalled for some reason. Perhaps, something happened that disconnected power from the battery to the computer, i.e. battery died, cables disconnected, short in the system, etc. Perhaps it was something in the code (a bug) that switched the state of the Cruise control when it shouldn't have. Weird things happen sometimes when unintended actions are performed during run-time. For example, two buttons were pushed at the same time. The number of things that can trigger a software glitch like this are too numerous to count.
If the State of the cruise control IS the problem, Toyota must first test the software by purposely setting the variables to a wrong state and observe the affects. If they trigger the anomaly, then they at least know they have cornered the problem. Then the software engineers have somewhere to look to find what is screwing up the state of the system. Scott Allam
The throttle drive by wire electronics already has various 'safety' measures incorporated. If you have the workshop manual and scan throught the CEL codes, you will see there are ample safeguards already designed. It is difficult for an errant circuit to give a wrong signal without it being detected. It is a closed loop system. If the throttle plate do not open according to expected opening, a CEL will be generated.
If it is true that there was no CEL codes detected: Then the system must have thought it was performing its job pefectly.
So, I googled for those companies offering engine Mods, and how they have their own 'throttle maps' to alter the car's throttle response: Read some of their documentations.
=== The throttle actuator motor movement is controlled by a signal in 'duty-cycle'. Duty-cycle 0 - throttle is closed. Duty cycle 100% -- throttle is wide open. In between 0 and 100% are values 'mapped' into the computer's memory.
There is a map programmed into the system. For a given rpm, and accelerator Pedal position, a value (duty-cycle) is sent to the throttle actuator.
When pedal position is 0, i.e. pedal is released. The duty cycle sent is 0 at all range of engine rpm. This will make the throttle actuator close the plate. Or at least put it in idle position.
The question is : Can I create a map so that at certain rpm, the pedal position 0 will give a duty-cycle of 100%? giving WOT signal to the actuator?
If the answer is yes.. Then a glitch in the map can also cause the same problem.
Supposed Computer sees Pedal is released. But its programming tells it to give a wide open throttle ?.
What would happen if the car's computer was confused about what state it was in, meaning cruise control ON or cruise control OFF?
Weird things happen sometimes when unintended actions are performed during run-time.
These are good points to investigate. Toyota needs to release its software code for investigation by others, including a full set of schematics on the system. All we can do is guess at the posibilities of what is happening with UA.
On the news this morning - The police can not duplicate the problem with the run away Prius this past weekend. They are suggesting a HOAX by the driver! Making assumptions again, based on their general tests of the cars operation.
Yet the brakes tell a different story, they have been worn down. This is no HOAX! There is a real problem and it is intermittent, which is the hardest problem to find when troubleshooting. There is no solution that can be put into place until you discover the true cause of the problem.
There are too many reports of UA to say there is nothing wrong with the car at this point. Under the right conditions it will reappear in this car.
We don't have any facts on what the driver was doing, what electronics were turned on in the car (radio/CD/navigation system), car exact speed, cell phone use, inside and outside temperatures, cruise on or off, etc. Too many combinations.
Toyota is still guessing and in the dark with no true solution at this point. They are too close to the problem and making assumptions that are incorrect. (If you assume a part or IC is working correctly and it is not, you tend to overlook where the real problem lays.) They need to step back and let others investigate.
All power disengagement needs to be mechanical and not electronic for safety reasons.
The question is : Can I create a map so that at certain rpm, the pedal position 0 will give a duty-cycle of 100%? giving WOT signal to the actuator?
All you need is an obscure line of code with a typo, that is used under certain system-conditions - such as a parentheses in the wrong spot; and you could be dividing by 0 !! So now the car wants to infinitely accelerate.
Toyota Failures: My understanding of how the ECM and TPS interact is as follows: The closed loop control system is made up of the Hall effect sensors in the gas pedal which send a voltage "command" to the Electronic Control Module. The ECM then provides pulses to a stepping motor which rotates the throttle butterfly valve and Throttle Positioning Sensor. The TPS provides a feedback voltage to the ECM and when it is in balance with the “command “ signal the motor stops. If the TPS feedback signal were abnormally low the motor would continue to open the throttle valve attempting to balance, thus increasing speed. Agree with Von that the use of only one TPS is dangerous. Also MikeML enlightened me that a "pot" is used as the TPS, not a Hall effect device. As a design engineer at CTS for 7 years, am well aware of how a potentiometer can fail: crack in resistive element, open wiper or high contact resistance, an intermittent at any terminal. The TPS should be a prime suspect, as its failure would open the control loop, resulting in unexpected acceleration. Positive identification of the fault, providing a fix and adequate testing will take quite a while. An interim fix may be to simply add a brake switch contact that disconnects the stepping motor when brake pedal pressed.
Note: When “cruise control” activated the Vehicle Speed Sensor also provides feedback to the ECM but do not know influence of TPS or which sensor-feedback is dominant. Many times the cruise control is forgotten in “Pause” mode after tapping the brake. Accidental (or spurious) activation of the “resume” switch would cause acceleration to set-point speed; however, unaware if cruise was on or in pause during any of the incidents.
Traditional pianos need regular tuning, electric ones do not. If you sit down at either one and don't know how to play and then exclaim "It's the pianos fault - it's electric and something is obviously wrong with it", people are going to give you strange looks. Inevitably, somebody will point out at the wonderful Steinway right next to the digital piano and say "Then, try that one, it's not electric."
If Toyota's drive by wire design involved storing the positions of the gas pedal or other sensors using improperly qualified DRAMs as buffers, alpha particles could cause data corruption. That may explain the intermittent, rare occurrence of sudden acceleration in their cars. Data is soft, DRAM is hard, the problem's considered both software and hardware error.
The auto companies have been giving their customers a foolproof way of killing the power/energy to the car. It's called an ignition switch. I am really surprised by people driving their cars into a ditch or gaurdrail to stop it when all they have to do is turn off the key!!! You will lose power steering, but you can still steer the car. It will be harder to turn. you will also lose power brakes. This is what an emergency brake is for, emergencies!!!
Comments
#314 of 315
Re: Naked APP sensor board free for water contact [srs_49] by nguyenngocbich
Mar 12, 2010 (5:56 am)
Save | Reply
Replying to: srs_49 (Mar 12, 2010 3:34 am)
I used worked with electronic boards for long period that I believe this PCB is coated only printed board, soldered connections of componets with printed board are not coated. I just have found other error of the gas pedal when watching video "“An important update: 02.11.2010” from www.toyota.com, that a metal back lid of the gas pedal was assembled without gasket! That means moisture or water may enter the gas pedal easily.
The boards we do, we coat everything except for the connector pins. This is done after final assembly, and before board test.
I have Bluetooth in my car and It makes me nervous that it can be hacked into. Do you think some messed with Toyota's software?
Onstar can stop your car in its tracks and on you 24-7 when your driving.
Car makers ca'nt afford redundant systems. Although we might get a black box mounted some where on vehicle.
I'm sure most car manufacturers have an in house test for Radiated Susceptability. Check with the IEEE in Detroit. So what? who cares who wrote/approves the test? The most thorough is MIL-STD-461. This specifies an EUT (Equiptment under test). An EUT can be a relay, a radio, a radar a gizmo du jour or a CAR. I personally use modified MIL-STD-461 testing when I trouble shoot these EMI (electromagneti interference) problems. Years ago I contributed to the writting of 461. And of course a horse designed by committee looks like a camel.
http://autos.aol.com/article/toyota-recall-deals
John Alvarez
http://autos.aol.com/article/prius-driver-scam
1) If it were possible to run such a scam and profit from it, why wouldn't people be running this sort of scam every year? why was it only Audi 30 years ago? and now it's restarted mainly with Toyota?
2) why if it's a scam is Toyota' s UA rate much higher than any other manufacturer?
3) if it were so easy to run a UA scam, I'm sure the former Big 3 would have wanted to do something like that 10-20 years ago, when they were losing market-share to the Japanese.
4) if UA is a problem with older people, then you should see the same statistical rates of UA across all brands. You don't - you currently see Toyota much higher.
This is not to say that out of all the reports that there aren't a few people trying a scam or a few people who did get the pedals mixed up. But this does not account for the statistics showing Toyota to be significantly higher and different.
But, if you can see, most of the auto manufacturers that have the astounding rate of UA, although large manufacturers in their own right, mostly can be considered niche market auto sellers. So essentially, they really don't offer much of a threat to "the big 3" as it was put. The only manufacturers that offer any true threat to the big 3 in terms of sales to the general mass automobile market are, the Asian automakers, Toyota most specifically. And, just to clarify, the fact that the big 3 have been recently driven to bankruptcy or bailout, speaks for itself as to why no "scam" as it was put, or for that matter, any legitimate plan was implemented "10-20 years ago" to deal with loss of market share to Asian automakers - US automakers were either just too fat, rich, and lazy, or just simply asleep at the wheel.
http://www.thetruthaboutcars.com/tag/unintended-acceleration-rates-by-make/
The problem is WITHIN the LSI Driver IC chip that controls the engine. No, it has nothing to do with the design of the chip from an electronic design standpoint. That's why everyone at Toyota is scratching their heads over this problem.
I have seen this type of problem in development of a new product in the past. The LSI chip was the culprit. The engineers did not believe me when I showed them how I could make the problem appear with just using basic electronic troubleshooting location isolation.
It would not happen at exactly the same time, but depending on the LSI chip, I could make it happen over and over again in the lab. After the chip manufaturer was made aware of how we could duplicate the problem, they changed their chip manufacturing process and some additional resistors and capacitors were added to the chip legs on the defective LSI Chips to compensate for the problem preliminary LSI chips we had received.
So the $1,000,000 answer to the question is as follows.
1. What is causing the problem?
Answer: The process that is used to close the IC chips alows mosture into the chips. When the right tempature is reached, it basically rains inside the LSI Chip. This will cause shorts and resistance across the interior of the chip. That in turn makes the chip act in the manner currently being observed - unintended acceleration.
Turning off the car and turning it back on again will reset the LSI chip and it will look normal again. NO TROUBLE FOUND!
2. How can I make the problem appear?
Answer: Use a HEAT GUN and a can of FREEZE MIST to track down the defective chip. That's how I did it. When I hit the defective LSI Chip with the FREEZE MIST to cool it - WHAM! The chip turned on the motor and heating it did nt stop the chip. It was in a runaway condition, until you turned off the power.
3. How can this be fixed?
Answer: The defective chips must be replaced in ALL TOYOTAS.
But you can't use the current LSI chips being manufactured because they are all being manufactured the same way.
The Manufacturing of the LSI IC Chips must me changed and an inert gas must be used to seal the IC chips during the manufacturing process. Also, additional design changes need to be put into the LSI chips to prevent the runaway condition from happening. Since this LSI information is all propriatery to Toyota, I can not give you any additional information on the design changes needed to the LSI Chip.
Bottom Line - The LSI IC Driver that controls the acceleration is defective. There is no way to tell how many defective chips have been installed. I would say ALL need to be replaced. They ALL Have the potential to act in the same way and there is no way to prove, without a shadow of a doubt, that you don't have one in your Toyota. I have a 2009 Toyota RAV4, so I am very concerned.
I would suspect that the defective IC chips were all made in the same plant, using the same process to seal them.
LG 71'
B.S. Electical Technology
New York Institute of Technology, Old Westbury Campus
If that is what the problem is - moisture inside the chip cavity, it should be a simple matter to remove the suspected part from a vehicle that has experienced UA, open it up or remove the encapsulant, and visually inspect the die for moisture related defects. They are pretty easy to see, especially for someone who does that sort of post-mortem analysis for a living.
NPS research: http://www.npr.org/templates/story/story.php?storyId=124235858
NHTSA website
The print hammer (activared by a relay closure action) would just stick on and the printwheel motor would come on full speed. Spinning pettles went flying everywhere. The only way to stop it was to turn off the power. When you turned the printer back on.... it functioned normaly for awhile, no trouble found. Very strange behavior. It would happen at random for each demo printer. IT WAS VERY DIFFICULT to determine the real cause.
The people from the Engineering Department did not believe it was the LSI Chip. I had to bring them in, along with the engineers from the company who manufactured the LSI chip to prove I could make it happen.
The problem LSI chip was located on the Power Board. So it was responsible for turning the power on and off for the printer hammer (relay power) and the printwheel motor power control.
At random, the power to both would come on full voltage!
SO DOES THAT SOULD LIKE THE SAME PROBLEM? YES IT DOES!
The basic report back from the Engineering Department put the cause of the problem in the sealing of the LSI chip at the manufacturers plant. Now, this is in Silicon Valley where there is moisture in the air. They stated that it was basically raining inside the LSI chip. The chip manufacturer had to change the process and insert an inert gas to fix the problem.
Since these chips are being manufactured in countries that have high moisture in the air, it is reasonable to assume that the chips are being closed and not evacuated prior to sealing them.
Since I was not a witness to the opening of the LSI Chip or the evaluation process at the chip manufacturing plant, I can not tell you more. I suspect the problem with the Toyota chips is the same issue, moisture inside the chip. Thus it is random and I suspect a temperature activated related problem. It depends on where the mosture, when condensing, comes in contact with the traces within the LSI chip.
That is where I would focus the investigation.
I sent a message to the Director of Engineering at Toyota about a month ago that I would consult with them on the problem. They would pay me if I was right and it would not cost them anything if I was wrong. They never responded.
The next thing I heard on the news was that they had a fix... the shim on the gas pedal. So I thought I may have ben wrong about the problem. Now it looks like I may have been correct in my theory of the problem.
LG
I posted previously that media hype over Toyota is going to turn into a can of worms - and not for the right reasons. Now if you screw up and run a red light while driving your Toyota.... Well, you can guess what the low-lives are going to start claiming.
Nice post on data saying unintended accelerations are nothing new, This Toyota thing has GOT to be put into perspective!
Once again a warning to folks - if you're in a car and the driver seriously does not know where or what neutral does, get out of that vehicle!
Actually, Silicon Valley is pretty dry. I believe it's classified as semi-arid (San Jose, 14.4 in rainfall/year).
Since these chips are being manufactured in countries that have high moisture in the air, it is reasonable to assume that the chips are being closed and not evacuated prior to sealing them
Not likely. All of the chip vendors I'm familiar with manufacture their devices in environmentally controlled facilities, class 10K or better clean room facilities. The chips themselves are probable never exposed to ambient air. Dry nitrogen is the usual gas constituent in the areas where bare chips are handled. Dry nitrogen is almost guaranteed to be used where the final packaging occurs - what you call the sealing. So it is highly improbable, though not 100%, that moisture is being sealed in with the chips on a regular basis.
Now, once the chips are packaged, unless they are in a ceramic, hermetic package (not likely because of costs involved), is is possible for moisture to leak in, particularly where the leads pass through epoxy (plastic) package that contains the die. Even then the probability is remote. Even in military avionics and space applications, the industry is learning how to use plastic encapsulated ICs with good success, something that was unheard of 20 or 30 years ago.
Plus, like I said. Moisture and it's effects inside a package on the die is pretty easy to detect.
1) What percentage had "cruise control" engaged or in standby? VS shut-off?
2) The 2- Hall Effect sensors in pedal: Do they have built in signal conditioning
providing low output Z or do they provide a high impedance signal to ECM?
3) What type sensor provides feedback from throttle control butterfly and is it's
signal conditioned?
4) Can the stepping motor activating the butterfly be electro-mechanically
de-coupled or is it integral to the ETB?
Interim Solution: Since intermittent problems are always frustrating and may require a very long time to find, fix and test, the following may be feasible.
Brake Switch to disconnect servo-motor electrically (preferably mechanically).
Why are you posting data from the1990's? Toyota cars from the 1990's are nothing like the hi-tech cars they put on the road today, where every transmission, ABS, stability control, cruise, and ECU's are so complicated and interlinked electronically. No one's questioning that Toyota had great quality and no significant UA until just recently. All going back 20 years and counting that does, is downplay how bad the recent string is.
Someone put a link here to 2009 reports, and that was bad for Toyota. And I'm sure if 2010 data was available you'd see that now everyone who has a problem is contacting NHTSA, and the numbers are much, much worse.
Again if fraud and running-down Toyota was possible when there was no real problem, people who compete would have done so years ago. There has to be some truth to the UA problem being exceptional, no matter how much you think it is overblown. All companies probably have some UA over many years; something has happened with Toyota vehicles, to make them standout recently.
Even though the world has more knowledge about sealing these chips, it does not mean errors in manufacturing can not occur.
As you say -- Moisture and it's effects inside a package may be pretty easy to detect, if you are looking for this issue with moisture on the die.
It's something that needs more investigation to rule out this possibility. Also, you have to look at the chips in the cars having the problem. It may not be all chips, just certain problem LSI Chips that were manufactured overseas.
I lived and worked in the bay area 27 years ago. YES, it does rain there and floods at times. So mosture is in the air, but like you state - the chips these days are made in a controlled room. Then again we are talking in the USA.
When your trying to drive down cost, you have the chips made overseas. We have seen where they can be a bit lax in their controls during manufacturing. Maybe they did not use the correct gas while sealing the chips or the hermetic packaging process was sloppy. We are zeroing in on chips made for Toyota. That may be why we don't see the issue with other LSI Chips in other cars.
It's still worth looking into to rule out this as the problem.
LG
I agree. But I would be surprised if that has not already been looked into.
Information from #317 by srs 49: The boards we do, we coat everything except for the connector pins. This is done after final assembly, and before board test.
From the information I can make conclusion that your APP sensor circuit board can not work correctly in wet condition because sprayed parylene could not create a continuous film to coat the board from moisture, especially bottom side of the hall IC and backside of IC legs were not sprayed (that the APP sensor board need to be located in sealed cage on the gas pedal). SUA may occur with gas pedals what have high humidity in confined space of APP sensor board. In vehicle with running air conditioner, when cabin’s temperature reduces to right level, humidity will condensate to water on the APP sensor board, enter underneath the IC (due lower position of IC on the board) and create short circuit between IC legs follow to situation as professor Gilbert’s experiment! After SUA, the vehicle is in rest for time that water on the APP sensor board disappeared to humidity that nobody can identify what caused SUA.
Why humidity is exist in gas pedal? If APP sensor board is assembled to gas pedal while ambient air with high humidity (example in rainy weather), air with high humidity can enter into space of APP sensor on the gas pedal and exist in sealed confined space for long time.
Measure to eliminate: APP sensor board from recall vehicles will be remounted from gas pedal, dried and coated with thick layer of parylene or epoxy, reassembled in dry condition to the gas pedal and sealed.
The analysis of Professor's Gilbert's demonstration establishes that he has reengineered and rewired [my emphasis] the signals from the accelerator pedal. This rewired circuit is highly unlikely to occur naturally and can only be contrived in a laboratory. There is no evidence to suggest that this highly unlikely scenario has ever occurred in the real world.
Well, what's surprising about that? Rewire the circuit and cause a problem???
Now, if your theory is that water condensing on the circuit card someplace can duplicate Professor's Gilbert's results - that it can mimic his circuit changes -, then yes, that experiment should be run. But, I would be surprised if Prof. Gilbert did not do that himself. What does it take, some water and the circuit card? The fact that nothing of the sort has been reported leads me to believe that the Gilbert did run an experiment with the water (especially if he's as sharp as he and Congress think he is), but nothing out of the ordinary occurred - that no UA was caused.
Don't get me wrong. Gilbert did some good work here. Being able to reproduce a problem in a more controlled, repeatable, laboratory environment is a big step towards understanding exactly what his happening. It's just that he failed to go the next step and demonstrate how his "rewired circuit" could occur in the confines of a passenger compartment.
I think, professor Gilbert could not repeat phenomena created in real world by water because water as natural superman force can create one phenomena like professor Gilbert’s experiment result from hundreds of short circuits in the APP sensor card. Most of the short circuits ECM can detect and only some times can cause SUA! I try to understand mechanism of SUA and give measures to eliminate but not try to repeat the phenomena because I understand this is impossible!
These same voltages/fields could probably be induced by the common "linear amps" in use by truck drivers. This would be very difficult to determine, in that the strength of the signal would not be known(and thus the effect on the shielding of the controller).
I am wondering now... if a "Driver" was useing a linear amp on his CB, while talking it would tend to (possibly) induce small spikes(as this is an analog Frequency Modulation signal). If there was a "reverbe or echo" function on this radio, would this not allow a sustained, STRONG RF signal across a wide frequency(wider the closer you are to the transmitter) to affect this type of controller?
This being said it takes much less to affect some of our lift equipment, and an example of the specs would be (protected against 20 gauss, 15kv static discharge,etc.)
Just a thought. Whats your opinion?
a black box will indicate what readings got compromised..
As it is now, they have CEL codes. like TPS stuck open, TPS stuck closed.. etc.. But they don't have codes for Unintended Acceleration.. And When the most powerful computer (the driver's brain) reports an Unintended acceleration, the manufacturer will accuse the driver of lying.
What will happen though if you're experiencing UA is now your engine is no longer slowly gaining rpm's as you gain speed. If you shift into nN with full throttle, your engine is going to hit redline very quickly. I'm not sure if there's a rev-limiter, or if it works during certain UA, but a blown-engine may be the result within 10-30 (?) seconds.
You can try to accuse the driver of lying, but there are too many people reporting the same problem. This has to be taken very seriously, lives are at stake.
The error codes will never appear because the point it captures the codes is prior to the point the problem appears. A defective IC Driver chip will turn on the power to the output device, in this case causing the UA. The monitor is looking at the input signals to the LSI chp only. So the computer thinks that everything is working normal. The computer ASSUMES the correct signal is right, which it is because that is whi it is monitoring - the input to the chip - and assumes the output will be correct. The Input signal is normal, however the output of the LSI chip is not!
Because the LSI driver is defective in some way, the output is at MAX and is not being monitored. Thus assumptions are being made the chip is not defective and people are looking at other points for the problem.
Think of it this way. Suddenly your stereo goes on FULL BLAST! The input to the final amp stage is at the correct voltage/db level and so the sound should be normal but it is blasting. Why? A problem with the final IC stage is defective.
In an LSI (Large Scale Intergration) chip, you have many small transistors working together. If one of those transistors shorts - wham - the voltage to the next stage is at MAX. This cascades through the other transistor stages an the output goes to MAX!
Now take that faulty internal transistor and make it flakey with temperature. As the temperature changes there is a Notch effect in some IC's! As the temperature rises, the transistor is good, then goes bad, and then can go good again. This is well know in QA circles as a problem with IC's.
Some companys use hot boxes for final testing of products. This is where they find those defective IC's, that act this way on final circuit board testing. Take out this QA testing and you run the risk of sending out defective products.
Now apply this information to the Toyota problem and you can see the possibility of having a bad IC as the problem causing UA. You need to take of the cars with known UA and do a complete analysis of the electronics (in a heat/cold box with the entire car) to locate the defective part. I believe the enviormentals are a factor in the UA issue.
LG
I am a software designer and tester, who prides himself on the ability to break software. I have called Toyota AND e-mailed them, asking them to contact me, but I have not heard from them which is why I am on this forum. I guess Toyota has their hands full right now.
This problem has track marks of a software glitch all over it. This is of course is only a theory, but I believe it is one that makes good sound logical sense.
So here is the million dollar question, as it were. What would happen if the car's computer was confused about what state it was in, meaning cruise control ON or cruise control OFF?
So if the cruise control was actually ON, but some of the local variables within these functions considered the system OFF, some very dangerous malfunctions could occur. For example, if the Resume function was engaged, then the car would accelerate on its own. HUGE RED FLAG HERE. There would be a brake function where when you tap the breaks it disengages the Cruise Control. This function might only be called, if the Cruise Control is actually ON. SO, if the braking function thinks the system is OFF, but it is really ON, then the brake function to disengage the Cruise Control would not be run.
This is similar to a Three way switch for a hallway light. Normally, flip the switch up and its on and down and its off. But on a three way switch, it changes depending on what you have done with the third switch and you can't tell whether you are flipping the circuit on or off. This would also be a reason why they are not able to find the problem. Ironically, the system is working as it should, hence no diagnostics would uncover it. The system is working appropriately to the state it THINKS the cruise control is in.
How could the state of the Cruise Control become screwed up like this? Well quite simply, an event happened that the Software designers did not plan for. Perhaps, something physical such as the cruise control was on and the car stalled for some reason. Perhaps, something happened that disconnected power from the battery to the computer, i.e. battery died, cables disconnected, short in the system, etc. Perhaps it was something in the code (a bug) that switched the state of the Cruise control when it shouldn't have. Weird things happen sometimes when unintended actions are performed during run-time. For example, two buttons were pushed at the same time. The number of things that can trigger a software glitch like this are too numerous to count.
If the State of the cruise control IS the problem, Toyota must first test the software by purposely setting the variables to a wrong state and observe the affects. If they trigger the anomaly, then they at least know they have cornered the problem. Then the software engineers have somewhere to look to find what is screwing up the state of the system.
Scott Allam
If it is true that there was no CEL codes detected: Then the system must have
thought it was performing its job pefectly.
So, I googled for those companies offering engine Mods, and how they have their own 'throttle maps' to alter the car's throttle response: Read some of their documentations.
===
The throttle actuator motor movement is controlled by a signal in 'duty-cycle'.
Duty-cycle 0 - throttle is closed. Duty cycle 100% -- throttle is wide open.
In between 0 and 100% are values 'mapped' into the computer's memory.
There is a map programmed into the system. For a given rpm, and accelerator Pedal position, a value (duty-cycle) is sent to the throttle actuator.
When pedal position is 0, i.e. pedal is released. The duty cycle sent is 0 at all
range of engine rpm. This will make the throttle actuator close the plate. Or at least put it in idle position.
The question is : Can I create a map so that at certain rpm, the pedal position
0 will give a duty-cycle of 100%? giving WOT signal to the actuator?
If the answer is yes.. Then a glitch in the map can also cause the same problem.
Supposed Computer sees Pedal is released. But its programming tells it to give a wide open throttle ?.
Weird things happen sometimes when unintended actions are performed during run-time.
These are good points to investigate.
Toyota needs to release its software code for investigation by others, including a full set of schematics on the system. All we can do is guess at the posibilities of what is happening with UA.
On the news this morning - The police can not duplicate the problem with the run away Prius this past weekend. They are suggesting a HOAX by the driver! Making assumptions again, based on their general tests of the cars operation.
Yet the brakes tell a different story, they have been worn down. This is no HOAX! There is a real problem and it is intermittent, which is the hardest problem to find when troubleshooting. There is no solution that can be put into place until you discover the true cause of the problem.
There are too many reports of UA to say there is nothing wrong with the car at this point. Under the right conditions it will reappear in this car.
We don't have any facts on what the driver was doing, what electronics were turned on in the car (radio/CD/navigation system), car exact speed, cell phone use, inside and outside temperatures, cruise on or off, etc. Too many combinations.
Toyota is still guessing and in the dark with no true solution at this point. They are too close to the problem and making assumptions that are incorrect. (If you assume a part or IC is working correctly and it is not, you tend to overlook where the real problem lays.) They need to step back and let others investigate.
All power disengagement needs to be mechanical and not electronic for safety reasons.
LG
0 will give a duty-cycle of 100%? giving WOT signal to the actuator?
All you need is an obscure line of code with a typo, that is used under certain system-conditions - such as a parentheses in the wrong spot; and you could be dividing by 0 !! So now the car wants to infinitely accelerate.
The TPS should be a prime suspect, as its failure would open the control loop, resulting in unexpected acceleration. Positive identification of the fault, providing a fix and adequate testing will take quite a while. An interim fix may be to simply add a brake switch contact that disconnects the stepping motor when brake pedal pressed.
Note: When “cruise control” activated the Vehicle Speed Sensor also provides feedback to the ECM but do not know influence of TPS or which sensor-feedback is dominant. Many times the cruise control is forgotten in “Pause” mode after tapping the brake. Accidental (or spurious) activation of the “resume” switch would cause acceleration to set-point speed; however, unaware if cruise was on or in pause during any of the incidents.
The same result obviously.