Virgil

Gosh, darn it, there are too many topics I could blog about. I could blog about the Oprah segment on the life inside a convent of Dominican Catholic nuns. What a wonderful group of women who really have it together in the modern world. You can read the transcript and see some clips from the show here: http://www.oprah.com/oprahshow/Lisa-...de-a-Convent/1.

Or I could blog about the death today of former Congressman Charlie Wilson, who led the effort in the 1980’s to channel equipment and weapons to the Mujahedeen fighting the Soviets in Afghanistan. The defeat of the Soviets in Afghanistan was a critical event in the demise of the Soviet Union, a subject near and dear to my heart. He was the subject of the movie a few years ago called “Charlie Wilson’s War.” I never did see the movie but perhaps I should do so in his honor. Rest in peace. You can read about Charlie Wilson here: http://www.cnn.com/2010/POLITICS/02/...ion=cnn_latest.

But I think I’m going to blog about the Toyota gas pedal problem. I assume many have heard of the problem – sudden acceleration without pressing on the gas pedal – to several of the most popular of the Toyotas models. It’s a fascinating story, especially since I’m an engineer and I think I can lend a little insight into the mentality of what was probably going on, not that I know specifically the technical problem, but that I can relate to the engineers and corporate reactions. First let me state that Toyota is the damn best car company in the world, period. They have led the way in quality engineering for decades now, and when we get lectures on designing quality into our products, Toyota is typically the model that is always referenced. I own two Toyotas myself. I’ve owned American and European cars, but nothing has impressed me like Toyota. And this pedal issue hit home, since I own one of the models in the recall. I must say I’ve been skeptical of the whole issue. I certainly haven’t experienced it. I wonder what the statistics are. But it was this article from the Wall Street journal that really piqued my interested.

Secretive Culture Led Toyota Astray
By KATE LINEBAUGH, DIONNE SEARCEY and NORIHIKO SHIROUZU
On Jan. 19, in a closed-door meeting in Washington, D.C., two top executives from Toyota Motor Corp. gave American regulators surprising news.
Evidence had been mounting for years that Toyota cars could speed up suddenly, a factor suspected in crashes causing more than a dozen deaths. Toyota had blamed the problem on floor mats pinning the gas pedal. Now, the two Toyota men revealed they knew of a problem in its gas pedals.
The two top officials from the National Highway Traffic Safety Administration "were steamed," according to a person who discussed the meeting with both sides. As the meeting closed, NHTSA chief David Strickland hinted at using the agency's full authority, which can include subpoenas, fines, and even forcing auto makers to stop selling cars.
Toyota had known about the gas-pedal problem for more than a year. Its silence with U.S. regulators, and other newly uncovered details from the crisis enveloping Toyota, reveal a growing rift between the Japanese auto maker and NHTSA, one of its top regulators. Regulators came to doubt Toyota's commitment to addressing safety defects, according to interviews with federal officials and industry executives, and accounts of Toyota and NHTSA interactions the past year
The heart of Toyota's problem: Its secretive corporate culture in Japan clashed with U.S. requirements that auto makers disclose safety threats, people familiar with the matter say. The relationship soured even though Toyota had hired two former NHTSA officials to manage its ties with the agency. [Snip]

Read the rest here: http://online.wsj.com/article/SB1000...toryCollection

First of all this notion that anyone would be surprised about “a secretive corporate culture” is ridiculous and rather melodramatic. Every corporation is secretive to some extent. They’re dealing with competition and a litigious society. They’re not interested in putting out their designs or their dirty laundry. Frankly every organization, not just corporations, I’ve known are “secretive” (not wishing to air their dirty laundry or advertise their foibles) to some extent: school systems, churches, government offices, politicians. I don’t see anything Toyota did that anyone else wouldn’t have done as well.

Second, the notion that “Toyota had known about the gas-pedal problem for more than a year” also chafes at my engineering sensibility. Who is Toyota? Toyota is made up of thousands of people and all sorts of engineers, all with opinions and hypotheses. Whenever some technical issue is sporadic or intermittent, whenever it affects a small population of the total output, whenever the root cause remains undetermined, if not even mysterious, then there will be competing theories within the very product design team itself, let alone any corporate management who are getting all sorts of numerous and varied possibilities. Notice that Toyota went to the expense of recalling millions of cars earlier in the year thinking it was a problem with the floor mats and its possible interference with the pedal. Why would the company spend millions of dollars on a recall if it didn’t think it was the solution? Or even more pertinent, if it didn’t think it would make a difference?

Let me give you an insight into a personal engineering issue I experienced. I was the lead technical engineer on this item (the actual product is not important and not available to the general public, so don’t worry about what it is) that went through development for four years and went into production in 2003. One of the elements to that item is a rather complex set of parts – four parts, of three different materials, interconnected, precisely dimensioned. The unit is also a significant load bearing element, meaning that a large force gets transmitted through it. We analyzed the dickens out of it. We knew it was a potential problem. When you have different materials in contact and absorbing load, it’s a dicey situation, since each material absorbs the load at a different rate (modulus of elasticity) and each material changes dimensionally at a different rate for temperature changes given different thermal expansion coefficients. So how the parts actually share the load could vary with different temperature, and we knew that. We tested across the whole temperature environment at extreme loads, testing several hundred. We modeled the heck out of it too (and when they tell you they prove global warming with modeling I just laugh). We came out of development thinking we nailed that design perfectly. And I handed the product over to the production team and I moved on to another project. Well, wouldn’t you know it, the very first year in production we started to get failures. I don’t recall the statistics now, but it was something like one out of a hundred would break. They called me in to try to figure out what was wrong. We looked over the manufacturing process; when you go into production, you strive to make things cheaper. We believed the problem was in some of the newly introduced assembly processes; certainly it could not be the design. We had gone through an extensive qualification testing. We made a few changes to the manufacturing process and out with another batch for testing. And boom, we got failures again. So management put together a “red team” to investigate the whole thing, and they purposely excluded me or any other person who had worked on the development. We had biases. I was ticked, because (1) I knew that design inside out and I thought I should be included and (2) they were questioning something I believed we had extensively tested and verified. To make a long story short, and this did go on for almost a year, the red team looked over all the manufacturing data, both from development and production, and found that the failures would occur with certain dimensional combinations. Without getting too technical, parts are designed to a toleranced dimension, not an exact dimension. For instance a dimension could be 100.54mm +/- 0.10mm. That +/-0.10mm is the tolerance a part can be machined. It so happened that when there was a certain combination of actual dimensions for the parts (all within the tolerance band) at a certain temperature, the load path changed so that a weaker part received a greater share of the load and it snapped. We had done extensive testing in development, but we never hit on that friggin combination of dimensions. To my credit, that product has never had another issue since, but it just goes to show that engineers become convinced of their results as being comprehensive.

Back to Toyota. I can imagine the varying opinions that went on at Toyota. I can imagine the pride the engineers had in their work, at being considered the most reliable car company in the world and at the shock of being challenged. Toyota earns that title of the most reliable car company through their design and quality methodology. Denial is quite a human emotion here, and the engineers are human. And then when confronted with the indisputable evidence I can imagine the finger pointing between departments, not because of pushing off blame, but because they honestly have hypothesized the flaw. Here’s an interesting paragraph from the article:

Toyota's woes have roots in 2002's redesigned Camry sedan, which featured a new type of gas pedal. Instead of physically connecting to the engine with a mechanical cable, the new pedal used electronic sensors to send signals to a computer controlling the engine. The same technology migrated to cars including Toyota's luxury Lexus ES sedan. The main advantage is fuel efficiency.

So the gas pedal is electronic, a new innovation, and certainly when you are dealing with electronics and software a sense of mystery predominates. Is it the electronic design, the connections between the electronic parts, a short perhaps, a bad electronic component, the mechanical assembly, or a software logic problem? If it were obvious, Toyota would have fixed it from the beginning. The fix seems to be to replace a plastic part in the pedal assembly with a metal one. I’m not sure why that would cause acceleration issues, but possibly because of wear or distortion which might give a different signal to the sensors.

But notice why Toyota had gone to this computer controlled sensor system: “The main advantage is fuel efficiency.” Surely Toyota has to take responsibility for the problems they caused with their designs. And part of the outrage at Toyota is warranted. But does society take any responsibility for pushing products to an extreme complexity for its chimeras?