Active suspension from Bose 2

[HDS]

From

http://TheCarConnection.com

Bose Sounds Off on Suspension

While it's become one of the auto industry's Tier One suppliers, Bose Corp., is normally a name one associates with in-car entertainment. But a not-so-new project championed by the company's founder, Amar T. Bose, could make the Framingham, Mass., firm a key player in automotive suspension systems.

It's not easy to keep secrets in the auto industry, but Bose has been quietly working on "Project Sound" for more than 24 years. The intentionally misleading codename "had nothing to do with sound," Dr. Bose acknowledged with a broad smile during a recent background briefing. What the company unwrapped was, in fact, a prototype of an active suspension system. The project is not as out-of-place as it might first seem. Some of Bose Corp.'s first products were power amplification systems. Audio technology followed, eventually becoming the Massachusetts firm's public face. But Bose has continued to produce the sophisticated recuperative power hardware at the heart of its suspension design.

Conventional suspensions use springs and shocks to compensate for bumps and turns. Active suspension technology use magnets and motors to react to road inputs and driving forces, and on paper, such systems could yield a smoother ride while enhancing vehicle control. Active suspensions have been around for awhile. Infiniti offered a version on its first Q45 sedan back in 1989. But so far, the technology has fallen far short of expectations. Part of the problem is the amount of power required to operate the motors and electromagnets. But Bose uses its recuperative technology to recapture energy, much the way hybrid cars reuse energy normally lost during braking and coasting. According to Dr. Bose, the system requires barely a third the power of an auto air conditioner.

While Bose would not provide a vehicle for independent testing, it staged a demonstration of its suspension system, using a retrofitted Lexus LS400 sedan. Even when driven at moderate speed over a severely bumpy course, the cabin of the vehicle remained virtually motionless. Compared to a conventional LS400, the modified vehicle was notably more stable in aggressive slalom maneuvers. To drive home the system's capability, the Bose car literally crouched and leapt across a piece of lumber blocking its path on a parking lot course.

"We hope to achieve the benefits of a luxury car and a sports car with the same vehicle," noted Bose engineer Larry Knox. Company officials declined to say exactly how long before the system might be ready for production, though Dr. Bose suggested it was fast approaching that phase. "Within the next six months, our intent is to take 50 percent of the weight out, and significantly reduce the cost." He hinted the firm he founded is now beginning talks with the auto industry. And TheCarConnection.com has been told by a Big Three source - who asked not to be identified by name or company -that preliminary testing is now underway. That automaker would clearly expect that a marketable version would need cost less than the $5000 price for the Q45 active suspension. -Paul A. Eisenstein

http://www.bose.com/controller;jses...e_suspension.jsp&pageName=/learning/index.jsp

 

[PTwizz]

The Bose system, in common with other active suspension systems, has a fundamental problem. The failure modes are all bad, and probably unnacceptable to today's industry and market.
One can imagine how a sensor or connector failure might cause the system to become unstable, resulting in a car that leaps about uncontrollably. A system power failure would leave the car on the bump stops, while a single corner power failure would result in an asymmetric three-wheeler.

Pete.

With thanks to John Miles for his experiences with active suspension systems.

 

[GregLocock]

Don't you think those strut units are big enough to contain a spring? I think they are. I can't see any great virtue in eliminating mechanical springs (whether rubber or steel) from the system, and then relying on some huge DC current at all times.

Nothing else that Bose has ever done, apart from the marketing of his noise making devices, has impressed me quite as much as the claimed performance of this system.

I am intrigued about the cooling system for his electronics - on a rough road a suspension will be absorbing several kW. One sneaky way to absorb this energy is to 'pump' the car higher in the short term, but you can't do that for long, it upsets the handling, and you run out of strut stroke.

By the way PTwizz, I worked with Milesy on the SID car, and was on the design team from the word go, just about.

Cheers

Greg Locock

 

[MAPirc]

Concerning failure modes - first, the system in its current envisioned configuration, uses, or would use, torsion bars to provide the nominal spring force. Second, I'd imagine that should elctronic control of the corner motors fail, that they would operate default in shunt so as to provide electromagnetic self-damping that would approximate the action of shock absorbers. There's more information here (here's hoping the link works...)

http://www.bose.com/controller;jses...e_suspension.jsp&pageName=/learning/index.jsp

Best,
MAPirc

Disclaimer: I work for Bose as a transducer designer, but haven't been closely associated with Project Sound.

 

[GregLocock]

But a shock absorber dissipates something like 4 kW. What sort of motor can handle that sort of heatload?

Using a mechanical spring is a good idea, I don't know why Lotus tended not to.


Cheers

Greg Locock

 

[MAPirc]

I'd ask the engineers directly, as they're personal friends of mine, but the company is internally (let alone externally) extremely tight-lipped about development projects. Communication is strictly on a need-to-know basis; that's why I had to hazard a guess as to how the system would operate with an electronics control failure to the motors.

Back to speculation about the possibility of dissipating 4kW of thermal power: 1.) the motors have lots of surface area and high emissivity, and 2.) the motors probably don't exceed very roughly 200 deg C or so internally. Whether they can hold to this limit without an auxiliary cooling system of some sort is unknown to me, and probably beyond that need-to-know threshold (sincerest apologies about that.)

Best,
Mark

 

[JayMaechtlen]

operating in failure mode, why would the energy need to be dissipated in the motors?

If they are still functioning as motor/generators, the system could fail a big resistor across the leads, either water cooled (this is a car, after all!) or mounted securely to the structure of the car...


Jay Maechtlen

 

[MAPirc]

Again, I need to emphasize that a shunt-action failure mode is only my speculation about how the system would react. My *guess* is that should the system fail, that it needs to function in a way that approximates a passive suspension system, where the shock absorbers, to first order, apply a resistive force whose magnitude varies in direct proportion to velocity. A shunted motor applies an electromagnetic drag term whose magnitude, likewise, varies in direct proportion to velocity, and so it should do a good job of approximating the action of shock absorbers. The higher the motor's force factor, the higher the damping action, and therefore, the higher the thermal power expended in the motor per unit velocity. An analogous response can be elicited in shock absorbers by using a higher-vicosity fluid, or by reducing the area of the bypass path.

One can always reduce the power by reducing the force factor, thus "weakening" the motor, or by deliberating increasing the resistance in the failure-mode shunt path, as you're saying, Jay. The problem is that as the resistance is increased, average total thermal power is decreased, but so is damping action. On the narrow basis of the parameters I've presented here, then, it's a matter of trading-off operating temperature with damping force.

Best,
Mark

 

[NormPeterson]

If the passive function of the motors as shock absorbers is ony a failure mode, then it isn't really necessary for them to provide as much damping as brand-new shocks. In fact, it may be preferable that they do not. Somewhat less damping would tend to provide sensory information that something was amiss, quite possibly before the driver noticed an illuminated warning lamp. I'll stick my neck out a little and suggest that most shocks don't get replaced until their thermal dissipation is far less than 4kw, so maybe 1 or 2 kw would be sufficient.

Norm