Nissan's next-generation steering employs a steering angle sensor at the steering wheel, three ECUs for control, electric motors to power the rack, and a steering force actuator near the driver to retain the vehicle's "steering feel."
Soooo... You compare 2000 microscopic surface mount components some of which contain a million transistors with 100 mechanical components (guestimate) and say that the electronic version is simpler because the 2000 parts are in only THREE boxes.
I LOVE the Infinity styling but I think they have lost their minds in Tokyo.
Note to Nissan: Give the stylists a raise with the money saved by firing the guys that came up with this Idea.
The article suggests that S-B-W needs a lot of electrical power. If the end goal is improved fuel economy, a comparison of power required by S-B-W verus hydraulically boosed power steering would be interesting, maybe HP imposed at the crankshaft output. The comparison could include effects of weight of the system components but I speculate that ends up being a minor effect.
From a FMEA perspective there are two requirements. The first is to get to the side of the road safely (high severity). The second is to be able to drive maybe an hour home, to a dealer, or to a town (low-high severity depending on the situation). If the vehicle charging system fails, it sounds like the battery can't support S-B-W for more than a few minutes. So the mechanical backup will always be necessary unless being stranded at the roadside is acceptable from a safety and marketing perspective.
"Big, Bulky, and complex"? I think not. The complexity is in the electronics, not the mechanicals. And aircraft have multiple backup systems in their fly by wire systems. I don't see that happening in automobiles unless you want a price tag similar to a 747.
Not necessarily. We build airplanes today that are completely dependent on fly-by-wire systems, and would fall right out of the air if those computers were to fail. And making those systems "fail safe" is essentially a solved problem.
On a normal car, the steering column is a major structural limiting factor, driving a large number of constraints for the surrounding vehicle. And it's a big, bulky, complex assembly. Electrically driven steering would make the steering system more modular, more flexible in physical layout, and allow things that are currently not practical, like swapable bodies on common frames, or handing the steering wheel to the passenger without having to pull over or swap seats (anyone remember the GM Hy-X?).
Over the past few decades, we've been steadily finding ways to move complexity from mechanicals to electronics, software, and firmware. It's been demonstrated that this can be done while also increasing reliability and decreasing cost. There's no reason to assume that automobiles are an exception to this.
That having been said, I don't think I'll be lining up to buy the first model year with drive-by-wire steering. :)
I agree. Where's the improvement in this? Aircraft control systems use double and triple levels of redundancy and go through years of testing to please the controlling agency (FAA) before they're allowed to go forward.
This almost crosses the line into the catagory "Technology in Desparate Search of a Use", like keyless ignition, voice commands, digital bathroom scales, etc. Maybe Design News can start a new blog of that title to go along with Sherlock Ohms and Made by Monkeys.
IMHO this is just another monumentally bad idea from carmakers, who seem to believe today that electronics make everything better. Wrong!
There already exists electric power steering systems in many cars at various price points, but it's for assist only (or for automated parallel parking). The low end cars I've driven with this ( a Saturn Coupe we used to own ) had practically zero steering feel and was miserable to drive. When it failed the steering was way over boosted to one side and not at all to the other, creating a very unsafe situation. BMW, according to a recent Car and Driver article, has produced a system that may be subjectively better than hydraulic assist, but it took years for them to get the algorithms right and they can charge what it costs to get good performance. I can only assume there is a similar failure mode.
Auto designers are forgetting that driving is inherently risky and should not be too comfortable, or too easy, or too much like riding your recliner while playing with your latest electronic gadget. I dread the day I have to replace my '04 RSX, because any new car available by then and in my budget will not be worth driving due to the excess electronic control interfaces.
The reasons in the article for switching to steer by wire are:
1. Elimination of mechanical components (what's the advantage of this, other than weight savings?)
2. Weight savings
3. Elimination of vibration from the road surface (no, you will still feel vibrations, and mechanical steering could be dsigned to "eliminate" vibration, too)
4. Minute adjustment of tire angle to compensate for wind and sloped roads (I suspect that the driver will still have to do at least some of this)
5. Enable drivers to feel the road (we can with a mechanical system)(this conflicts with 3, above)
6. Enable auto manufacturers to readily change the design/configuration of cars (e.g. change a left-hand drive car to a right-hand drive car)
There are other motivations for changing to steer by wire. These may include ease of implementation of driverless cars in the future, or cars that drive themsleves when necessary to avoid accidents, and the desire by manufacturers to add gimmicks that will inrease the prices of cars and increase manufacturers' revenues.
A simple mechanical system, even without power steering, is reliable, simple, and inexpensive. This is what I want in my car, not a lot of expensive-to-replace-when-it-fails electronics.
I agree, the clutch would be electrical. But for fail-safe, the clutch would have to be held 'disengaged'. Meaning if a power fail occured the clutch releases, in theory. But, this does not cover failures other than power fail.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.