Interesting Article by Popular Mechanics:
http://www.popularmechanics.com/automotive/how_to/4347704.html
http://www.popularmechanics.com/automotive/how_to/4347704.html
Anatomy of Toyota's Problem Pedal
- Thread starter SuperBusa
- Start date
- Status
Interesting. I like the reference to the Alien tractor beam...
Also, the Mustang that was used in the last article must have been a hoot to drive. Guess that is a perk of the job (and probably why they choose that car in their tests in the first place).
Also, the Mustang that was used in the last article must have been a hoot to drive. Guess that is a perk of the job (and probably why they choose that car in their tests in the first place).
" The voltages used in the system are DC voltages—any RF voltages introduced into the system, by, say, that microwave oven you have in the passenger seat, would be AC voltages, which the ECM's conditioned inputs would simply ignore."
PM mixes fact with fiction here:
Fact:
RF can affect a DC voltage through rectification in a solid state device.
While having a powered and RF leaky microwave in the passenger seat is highly unlikely;
Commercial/public service two way radio, CB radio, electromagnetic radiation from electrical power distribution etc are commonplace.
"Toyota Recall: Scandal, Media Circus, and Stupid Drivers - Editorial"
I can't believe that C&D would use the word stupid in this context
For just this one time I wish the Moderators would look the other way.
C&D deserves a good tongue lashing, a trip to the woodshed, and so much more for that "stupid" comment.
IMHO both magazines writers are drinking and enjoying the Kool Aid.
Very biased and mis-informative writing indeed.
Yellow journalism.
PM mixes fact with fiction here:
Fact:
RF can affect a DC voltage through rectification in a solid state device.
While having a powered and RF leaky microwave in the passenger seat is highly unlikely;
Commercial/public service two way radio, CB radio, electromagnetic radiation from electrical power distribution etc are commonplace.
"Toyota Recall: Scandal, Media Circus, and Stupid Drivers - Editorial"
I can't believe that C&D would use the word stupid in this context
For just this one time I wish the Moderators would look the other way.
C&D deserves a good tongue lashing, a trip to the woodshed, and so much more for that "stupid" comment.
IMHO both magazines writers are drinking and enjoying the Kool Aid.
Very biased and mis-informative writing indeed.
Yellow journalism.
Amazing that the Audi problem was ALL operator problems ! I quote, "In the end, the U.S. government determined that every single so-called unintended acceleration accident was the result of driver error."
Good articles and interesting comments from people after them.
Good articles and interesting comments from people after them.
Last edited:
Originally Posted By: Rickey
" The voltages used in the system are DC voltages—any RF voltages introduced into the system, by, say, that microwave oven you have in the passenger seat, would be AC voltages, which the ECM's conditioned inputs would simply ignore."
PM mixes fact with fiction here:
Fact:
RF can affect a DC voltage through rectification in a solid state device.
I wonder if car manufactures use twisted wires, or shielded wires in critical control circuits like this. They should IMO.
" The voltages used in the system are DC voltages—any RF voltages introduced into the system, by, say, that microwave oven you have in the passenger seat, would be AC voltages, which the ECM's conditioned inputs would simply ignore."
PM mixes fact with fiction here:
Fact:
RF can affect a DC voltage through rectification in a solid state device.
I wonder if car manufactures use twisted wires, or shielded wires in critical control circuits like this. They should IMO.
Originally Posted By: Rickey
" The voltages used in the system are DC voltages—any RF voltages introduced into the system, by, say, that microwave oven you have in the passenger seat, would be AC voltages, which the ECM's conditioned inputs would simply ignore."
PM mixes fact with fiction here:
Fact:
RF can affect a DC voltage through rectification in a solid state device.
While having a powered and RF leaky microwave in the passenger seat is highly unlikely;
Commercial/public service two way radio, CB radio, electromagnetic radiation from electrical power distribution etc are commonplace.
"Toyota Recall: Scandal, Media Circus, and Stupid Drivers - Editorial"
I can't believe that C&D would use the word stupid in this context
For just this one time I wish the Moderators would look the other way.
C&D deserves a good tongue lashing, a trip to the woodshed, and so much more for that "stupid" comment.
IMHO both magazines writers are drinking and enjoying the Kool Aid.
Very biased and mis-informative writing indeed.
Yellow journalism.
[NERD TALK]
That depends on the analog input device's sampling rate, and sampling method (averaging method) and the frequency of the interference.
In a analog to digital conversion, a device's sampling rate has to be at least twice of that of the source. Let's say the source is a 900 MHz GSM cell phone, the device has to read 1,800,000,000 times per second, to notice the added AC component, in addition to the DC offset.
Now, this is different than the buzzing noise that your cell phone makes when it's close to your stereo. It's amplifying the analog noise with no digital sampling.
However, it is possible that they use a very sorry averaging method, and the interference's RMS voltage is large compare to the real DC signal. The system has to 1) get a chance to read the peak of the sine wave, then 2) be dumb enough to use this value as input and disregard future inputs (by probability, if one averages AC RMS + DC offset over samples, and take the limit of the average as the number of sample increase, the average approaches DC offset.
It is far more likely for the processors in the ECM to suffer a complete melt down from poor shielding, then any RF interference in the analog input channels - i.e. the interference causes the processors to not being able to read enough bits (beyond error correction can compensate), and lose track of what it is doing. I do not know what Toyota does when this happens, but most of the embedded controllers I worked on would perform a self-reset.
[/NERD TALK]
" The voltages used in the system are DC voltages—any RF voltages introduced into the system, by, say, that microwave oven you have in the passenger seat, would be AC voltages, which the ECM's conditioned inputs would simply ignore."
PM mixes fact with fiction here:
Fact:
RF can affect a DC voltage through rectification in a solid state device.
While having a powered and RF leaky microwave in the passenger seat is highly unlikely;
Commercial/public service two way radio, CB radio, electromagnetic radiation from electrical power distribution etc are commonplace.
"Toyota Recall: Scandal, Media Circus, and Stupid Drivers - Editorial"
I can't believe that C&D would use the word stupid in this context
For just this one time I wish the Moderators would look the other way.
C&D deserves a good tongue lashing, a trip to the woodshed, and so much more for that "stupid" comment.
IMHO both magazines writers are drinking and enjoying the Kool Aid.
Very biased and mis-informative writing indeed.
Yellow journalism.
[NERD TALK]
That depends on the analog input device's sampling rate, and sampling method (averaging method) and the frequency of the interference.
In a analog to digital conversion, a device's sampling rate has to be at least twice of that of the source. Let's say the source is a 900 MHz GSM cell phone, the device has to read 1,800,000,000 times per second, to notice the added AC component, in addition to the DC offset.
Now, this is different than the buzzing noise that your cell phone makes when it's close to your stereo. It's amplifying the analog noise with no digital sampling.
However, it is possible that they use a very sorry averaging method, and the interference's RMS voltage is large compare to the real DC signal. The system has to 1) get a chance to read the peak of the sine wave, then 2) be dumb enough to use this value as input and disregard future inputs (by probability, if one averages AC RMS + DC offset over samples, and take the limit of the average as the number of sample increase, the average approaches DC offset.
It is far more likely for the processors in the ECM to suffer a complete melt down from poor shielding, then any RF interference in the analog input channels - i.e. the interference causes the processors to not being able to read enough bits (beyond error correction can compensate), and lose track of what it is doing. I do not know what Toyota does when this happens, but most of the embedded controllers I worked on would perform a self-reset.
[/NERD TALK]
Originally Posted By: SuperBusa
Originally Posted By: Rickey
" The voltages used in the system are DC voltages—any RF voltages introduced into the system, by, say, that microwave oven you have in the passenger seat, would be AC voltages, which the ECM's conditioned inputs would simply ignore."
PM mixes fact with fiction here:
Fact:
RF can affect a DC voltage through rectification in a solid state device.
I wonder if car manufactures use twisted wires, or shielded wires in critical control circuits like this. They should IMO.
I agree they should be shielded and probably are.
However given resonance effects; shielding and/or twisted pair is no panacea.
The hall sensors undoubtedly have buffer amplifiers(s)and here is one possible place for RF to wreak havoc.
There are many possibilities, and in fault diagnosis one of the first steps is to identify then eliminate possibilities.
And I am not suggesting that RF is the answer by any means.
I do however believe that it should be on the list of possibilities.
C/D was thoroughly ignorant in attempting to eliminate RF effects through their faulty logic.
My gut tells me that this is a voltage regulation problem in the ECM.
That would be consistent with the reported symptoms and their relative rarity.
If I had to bet on something this would be it.
It would be a low confidence bet however.
Originally Posted By: Rickey
" The voltages used in the system are DC voltages—any RF voltages introduced into the system, by, say, that microwave oven you have in the passenger seat, would be AC voltages, which the ECM's conditioned inputs would simply ignore."
PM mixes fact with fiction here:
Fact:
RF can affect a DC voltage through rectification in a solid state device.
I wonder if car manufactures use twisted wires, or shielded wires in critical control circuits like this. They should IMO.
I agree they should be shielded and probably are.
However given resonance effects; shielding and/or twisted pair is no panacea.
The hall sensors undoubtedly have buffer amplifiers(s)and here is one possible place for RF to wreak havoc.
There are many possibilities, and in fault diagnosis one of the first steps is to identify then eliminate possibilities.
And I am not suggesting that RF is the answer by any means.
I do however believe that it should be on the list of possibilities.
C/D was thoroughly ignorant in attempting to eliminate RF effects through their faulty logic.
My gut tells me that this is a voltage regulation problem in the ECM.
That would be consistent with the reported symptoms and their relative rarity.
If I had to bet on something this would be it.
It would be a low confidence bet however.
Jonny Z, what you posted is all very accurate.....when an analog to digital converter is functioning normally.
However RF rectification goes outside the normal operating parameters for any active device unless the circuit is designed for the primary purpose of RF rectification.
So RF at the input of any active (Vs
assive) device may result in a wide variety of symptoms on it's output.
The symptoms will depend on RF level, frequency(s), modulation, Etc.
And what I am saying is that this scenario is not very likely which is consistent with the rare percentage of occurrence.
And having built the case for this possibility I still don't believe that RF is it, just still a possibility.
See my other post.
However RF rectification goes outside the normal operating parameters for any active device unless the circuit is designed for the primary purpose of RF rectification.
So RF at the input of any active (Vs
assive) device may result in a wide variety of symptoms on it's output.The symptoms will depend on RF level, frequency(s), modulation, Etc.
And what I am saying is that this scenario is not very likely which is consistent with the rare percentage of occurrence.
And having built the case for this possibility I still don't believe that RF is it, just still a possibility.
See my other post.
Last edited:
As a matter of principle, I no longer read Car & Drivel, but I did skim that link and see it is still a shill rag.
Anyone who has played with Ham Radio for any length of time can attest to how easily RF signals can be rectifed. A flaky connection between two pieces of metal, can, and has countless times in the past, acted as a diode. There are myriad other ways that rectification can occur. And anyone that thinks it is easy to accurately measure small amounts of DC in the presence of AC, and vice versa, has never tried to do it.
So the same U.S. Government that determined that one magic bullet wounded President Kennedy and Governor Connally also determined that every incident of UA in Audi's was driver induced? Well, heck, it must be true then.
Anyone who has played with Ham Radio for any length of time can attest to how easily RF signals can be rectifed. A flaky connection between two pieces of metal, can, and has countless times in the past, acted as a diode. There are myriad other ways that rectification can occur. And anyone that thinks it is easy to accurately measure small amounts of DC in the presence of AC, and vice versa, has never tried to do it.
So the same U.S. Government that determined that one magic bullet wounded President Kennedy and Governor Connally also determined that every incident of UA in Audi's was driver induced? Well, heck, it must be true then.
Originally Posted By: Win
So the same U.S. Government that determined that one magic bullet wounded President Kennedy and Governor Connally also determined that every incident of UA in Audi's was driver induced? Well, heck, it must be true then.
I'm pretty sure that
A. "The government" didn't determine that; it was just some nutcases.
B. We've had several different governments since then.
But hey, what do I know...
So the same U.S. Government that determined that one magic bullet wounded President Kennedy and Governor Connally also determined that every incident of UA in Audi's was driver induced? Well, heck, it must be true then.
I'm pretty sure that
A. "The government" didn't determine that; it was just some nutcases.
B. We've had several different governments since then.
But hey, what do I know...
Originally Posted By: SuperBusa
I wonder if car manufactures use twisted wires, or shielded wires in critical control circuits like this. They should IMO.
Ford used twisted wires in the not-so-critical control circuit between the head unit (tape player) and the rear chassis unit(tuner/amplifier). One would expect they'd do so for critical circuits as well.
I wonder if car manufactures use twisted wires, or shielded wires in critical control circuits like this. They should IMO.
Ford used twisted wires in the not-so-critical control circuit between the head unit (tape player) and the rear chassis unit(tuner/amplifier). One would expect they'd do so for critical circuits as well.
Originally Posted By: Win
Anyone who has played with Ham Radio for any length of time can attest to how easily RF signals can be rectifed.
Anyone whose ever talked on a phone near an AM station knows, too. Even keying up a CB radio is enough to blast the transmitted audio to every phone in the vicinity.
Anyone who has played with Ham Radio for any length of time can attest to how easily RF signals can be rectifed.
Anyone whose ever talked on a phone near an AM station knows, too. Even keying up a CB radio is enough to blast the transmitted audio to every phone in the vicinity.
At the 2005 Dayton Hamvention, before GM went belly up, they had one of their vehicles on display and some people from their Electromagnetic Compatibility lab present.
I didn't pay any attention to it at the time, wish I had now. Apparently they thought EMI/RFI was important enough to showcase their work to a bunch of RF enthusiasts.
I didn't pay any attention to it at the time, wish I had now. Apparently they thought EMI/RFI was important enough to showcase their work to a bunch of RF enthusiasts.
Originally Posted By: d00df00d
But hey, what do I know...
You're probably not the first to call the Warren Commission "nutcases", but they were very much the government. Earl Warren was the sitting Chief Justice of the United States, and chaired the commission by appointment of President Lyndon Johnson.
But hey, what do I know...
You're probably not the first to call the Warren Commission "nutcases", but they were very much the government. Earl Warren was the sitting Chief Justice of the United States, and chaired the commission by appointment of President Lyndon Johnson.
Originally Posted By: Popular Mechanics
... would be AC voltages, which the ECM's conditioned inputs would simply ignore.
By "conditioned" they are presumably referring to the electronic term, that is the components the ECU designers have chosen to filter each throttle signal, perhaps something as simple as a small capacitor across the signal at the ECU end, or complex as digital filtering. I don't see PM's statement as being inaccurate.
In the unlikely event that it is an RF issue, the only susceptible location is the Hall coil or its IC, where smaller currents and voltages are present. The amplified signal leaving the pedal assembly will be too strong and the wiring designed to long-established and proven RF-resistant conventions.
... would be AC voltages, which the ECM's conditioned inputs would simply ignore.
By "conditioned" they are presumably referring to the electronic term, that is the components the ECU designers have chosen to filter each throttle signal, perhaps something as simple as a small capacitor across the signal at the ECU end, or complex as digital filtering. I don't see PM's statement as being inaccurate.
In the unlikely event that it is an RF issue, the only susceptible location is the Hall coil or its IC, where smaller currents and voltages are present. The amplified signal leaving the pedal assembly will be too strong and the wiring designed to long-established and proven RF-resistant conventions.
Originally Posted By: Kiwi_ME
Originally Posted By: Popular Mechanics
... would be AC voltages, which the ECM's conditioned inputs would simply ignore.
By "conditioned" they are presumably referring to the electronic term, that is the components the ECU designers have chosen to filter each throttle signal, perhaps something as simple as a small capacitor across the signal at the ECU end, or complex as digital filtering. I don't see PM's statement as being inaccurate.
In the unlikely event that it is an RF issue, the only susceptible location is the Hall coil or its IC, where smaller currents and voltages are present. The amplified signal leaving the pedal assembly will be too strong and the wiring designed to long-established and proven RF-resistant conventions.
I've seen lots of "conditioned" inputs (P/S rails and outputs too for that matter) that were not immune to certain types of RFI.
I know about these things because I've had to modify quite a few circuits in my career to cure same.
Until a solution is found; taking RFI off the table without independent rigorous proof testing is a gross mistake.
PM needs to stick to selling advertising and quit being a misguided cheerleader.
And I will repeat again that IMHO the real culprit is the equivalent to the ECM having a seizure.
In the "seizure" scenario some internal fault or data glitch would cause the ECM's CPU to process data in an "illogical" manner resulting in bizarre control outputs.
This could result in all devices on the car's data buss misbehaving.
Included in this might be ABS, transmission, throttle ETC.
This scenario is more or less consistent with several of the reports that I have read.
Originally Posted By: Popular Mechanics
... would be AC voltages, which the ECM's conditioned inputs would simply ignore.
By "conditioned" they are presumably referring to the electronic term, that is the components the ECU designers have chosen to filter each throttle signal, perhaps something as simple as a small capacitor across the signal at the ECU end, or complex as digital filtering. I don't see PM's statement as being inaccurate.
In the unlikely event that it is an RF issue, the only susceptible location is the Hall coil or its IC, where smaller currents and voltages are present. The amplified signal leaving the pedal assembly will be too strong and the wiring designed to long-established and proven RF-resistant conventions.
I've seen lots of "conditioned" inputs (P/S rails and outputs too for that matter) that were not immune to certain types of RFI.
I know about these things because I've had to modify quite a few circuits in my career to cure same.
Until a solution is found; taking RFI off the table without independent rigorous proof testing is a gross mistake.
PM needs to stick to selling advertising and quit being a misguided cheerleader.
And I will repeat again that IMHO the real culprit is the equivalent to the ECM having a seizure.
In the "seizure" scenario some internal fault or data glitch would cause the ECM's CPU to process data in an "illogical" manner resulting in bizarre control outputs.
This could result in all devices on the car's data buss misbehaving.
Included in this might be ABS, transmission, throttle ETC.
This scenario is more or less consistent with several of the reports that I have read.
Originally Posted By: Win
I didn't pay any attention to it at the time, wish I had now. Apparently they thought EMI/RFI was important enough to showcase their work to a bunch of RF enthusiasts.
Have you ever seen the factory grounding straps that the P71 Crown Vic comes with? Trunk lid bonded to the frame on both sides, ground strap around the weatherstrip to make contact with the frame, hood bonded to the frame, even the exhaust is bonded to the frame.
I didn't pay any attention to it at the time, wish I had now. Apparently they thought EMI/RFI was important enough to showcase their work to a bunch of RF enthusiasts.
Have you ever seen the factory grounding straps that the P71 Crown Vic comes with? Trunk lid bonded to the frame on both sides, ground strap around the weatherstrip to make contact with the frame, hood bonded to the frame, even the exhaust is bonded to the frame.
Originally Posted By: Rickey
And I will repeat again that IMHO the real culprit is the equivalent to the ECM having a seizure.
In the "seizure" scenario some internal fault or data glitch would cause the ECM's CPU to process data in an "illogical" manner resulting in bizarre control outputs.
This could result in all devices on the car's data buss misbehaving.
Included in this might be ABS, transmission, throttle ETC.
This scenario is more or less consistent with several of the reports that I have read.
I tend to agree here ... and I'm HOPING that Toyota and whoever else is looking into the electronic gremlin theory is instrumenting and testing actual cars that have reported "sudden acceleration" reports in order to capture a "seizure" event so someone can make some progress on a real root cause if it hasn't been fully discovered yet.
And I will repeat again that IMHO the real culprit is the equivalent to the ECM having a seizure.
In the "seizure" scenario some internal fault or data glitch would cause the ECM's CPU to process data in an "illogical" manner resulting in bizarre control outputs.
This could result in all devices on the car's data buss misbehaving.
Included in this might be ABS, transmission, throttle ETC.
This scenario is more or less consistent with several of the reports that I have read.
I tend to agree here ... and I'm HOPING that Toyota and whoever else is looking into the electronic gremlin theory is instrumenting and testing actual cars that have reported "sudden acceleration" reports in order to capture a "seizure" event so someone can make some progress on a real root cause if it hasn't been fully discovered yet.
- Status
