Technology is a fickle friend, nudging us forward with one hand while charging exorbitant tolls with the other. *Modern engines that impress us with power and efficiency are being decoupled from the manual transmissions that help them sing. Stout body structures guard us from peril in a crash but are so heavy and hard to see out of that we’re more liable to bump into hazards easily avoided by truly wieldy cars. Electronic stability and traction aids are wonderful except when there’s no way to disable them.
Electrically assisted power steering (EPS) is the latest technological cross we bear. Replacing hydraulic assist with a computer-controlled electric motor seemed like a reasonable idea when it first surfaced. Someday every car control will be by-wire; today’s EPS looks like a step in that direction. But in the past decade of driving EPS-equipped cars, we’ve found them lacking in feel, poorly tuned, and sometimes simply weird in comparison with the hydraulic-assist setups that have benefited from more than half a century of development.
This matters because steering is the driver’s main line of communication with the car; distortion in the guidance channel makes every other perception more difficult to comprehend.
Hydraulic Power Steering
The steering gear's internal cavity is divided into two chambers by a sealed piston attached to the rack. Applying pressurized hydraulic fluid to one side of the piston while allowing fluid to return from the other side to a reservoir provides steering assistance. A valve attached to the pinion shaft controls the hydraulic-fluid flow.
1) Hydraulic Control Valve 2)Pinion Gear 3) Hydraulic Pressure/Return Lines
4) Hydraulic Piston 5) Rack Housing
Electric Power Steering
To provide steering assistance, an electric motor mounted to the side of the rack housing drives a ball-screw mechanism via a toothed rubber belt. The screw engages a spiral cut in the outside of the steering rack. A torque sensor attached to the pinion shaft signals a control computer when to provide assistance.
1) Pinion Shaft 2) Steering Torque Sensor 3) Rack-and-Pinion Housing 4)Electric Motor 5) Ball-screw Mechanism 6) Steering Rack 7) Drive Belt
Makers are moving aggressively to EPS because eliminating an engine-driven hydraulic pump increases gas mileage by about 1 mpg. Before hydraulic power steering (HPS) goes the way of the buggy whip, though, we wanted to understand the differences so we concocted this test. BMW conveniently offers both types of assist on 5-series models equipped with four-wheel drive. (HPS survives here because the slightly *bulkier EPS unit fits in only the four-cylinder model.) A $61,125 BMW 528i xDrive served as our EPS lab rat while its 535i xDrive ($69,695) sibling stood up for HPS. Beyond the type of steering assistance, the notable differences are six cylinders in the 535i xDrive versus four in the 528i xDrive, 143 pounds more weight on the more expensive BMW’s front axle, and 19-inch 40-series tires (535i) versus 18-inch 45-series rubber on the 528i. The basic run-flat, all-season tire design and 245-mm *section width are common to both.
We divided our test into two segments—a hands-on, purely subjective driving evaluation, followed by instrumented proving-grounds tests. For the first phase, a steering committee of 11 editors with many decades of combined experience drove the BMWs—without knowing what type of steering assist was at work—over our favorite southeast Michigan test loop: 13.8 miles of whoops and hollows, sweeping lefts, and decreasing-radius rights; essentially our own North American Nordschleife. Each driver completed one 10-item ballot per car. Then the two BMWs were handed over to colleagues at Cayman Dynamics, where engineers fit an array of sensors and recorded for analysis the results of various steering-response tests.
Several of our findings were predictable; some surprised us. The first discovery was that neither BMW excels in steering feel and feedback, an observation consistent with the mediocre reviews we’ve given every 5-series, including the M5, since the sixth-generation design arrived three years ago.
After griping about EPS for years, the shocking revelation is that C/D’s editorial staff preferred BMW’s electric system over its hydraulic assistance. Total votes in seven out of ten rating categories favored EPS by two to eight points each. Hydraulic shined in only the three Feedback categories where it won the on-center comparison by four points and tied with EPS in our middle-of-maneuver and at-cornering-extremes performance ratings.
The objective measures suggest our team knows its stuff: The steering-effort plot compiled by Cayman Dynamics confirms the subjective Linearity of Effort Change ratings we collected from the steering committee. At parking-lot speeds (below 0.24 g), both BMWs demand nearly the same effort. As lateral acceleration increases, the 528i’s effort climbs very slowly but linearly while the 535i’s steering torque rises by a factor of two before leveling off between 0.60 and 0.70 g. Beyond 0.68 g, effort drops off on final approach to the 535i’s adhesion limit. Theoretically, the ideal steering effort would be a blend of the 528i and 535i characteristics—embodying the linearity shown by the 528i and the steeper slope in the effort curve evident in the 535i.
Cayman Dynamics also produced steering-response curves by slowly accelerating each BMW on a 400-foot circle. The 528i with EPS requires substantially less lock (wheel turning) as lateral acceleration rises. Less heavily loaded front tires favor the 528i and support our subjective feeling of slightly better agility. In addition, the four-cylinder BMW benefits from a 5-percent-quicker steering ratio, which backed up the editors’ Quickness scores.
To study steering characteristics in the crucial on-center zone, a Cayman tester recorded steering torque versus lateral acceleration while turning the steering wheel left then right in a smooth, sinusoidal fashion at various steady speeds. This was repeated in four steps from 30 to 75 mph at 0.1-g, 0.2-g, and 0.3-g levels of lateral acceleration to produce a matrix of 24 plots.
We learned from these measurements that the rise in steering effort immediately off-center is significantly quicker and higher with HPS, giving the 535i a tighter, more connected feeling in straight-ahead driving. Also, all 12 of the 535i traces (only two of which are shown) depict significantly less wander and deviation than is evident in the 528i plots. In other words, the buildup and decay of steering effort during back-and-forth maneuvers is more consistent with HPS, an advantage useful during passing maneuvers. Higher efforts do not equate to better feedback, but at least the 535i’s on-center characteristics lend a more secure sensation on the highway.
Cayman Dynamics's instruments measure steering-wheel torque and angular position. Other sensors feed velocity, drift angle, and four-corner ride-height info to an onboard recorder. Cayman conducted these instrumented tests on a wide straightaway and a 400-foot-diameter circle.
While parking efforts don’t really fac*tor into dynamic performance, Cayman Dynamics also looked into this area, seeking differences between electric and hydraulic assist. The 535i with HPS showed a consistent four newton-meters of steering-wheel torque needed in both left and right turning with the car static and between three and four newton-meters once the car started moving. In contrast, the EPS-equipped 528i showed higher and less-consistent effort requirements. Results ranged between six and ten newton-meters with the car static, with even higher efforts—from eight to eleven newton-meters—as soon as the wheels were rolling.
If there’s a moral victor here, it’s EPS. This 5-series confrontation proves that BMW has already learned how to make a computer-controlled electric motor that can perform about as well as the traditional hydraulic system it will soon supplant. But not every carmaker is as focused on dynamics as BMW, and EPS still has a way to go.
When hydraulically assisted rack-and-pinion steering arrived 57 years ago, it was a godsend. The combination of a fast steering ratio with reasonable effort and precise (for the day) feel eliminated sweat and elbow waving during parking maneuvers while providing an agile, in-command feel at speed. In 1969, we called Jaguar’s new XJ6 equipped with Varamatic hydraulically assisted rack-and-pinion steering “one of the world’s best-balanced sedans.” New low-profile radial tires; four-wheel disc brakes; accurate, 3.5-turn lock-to-lock, power rack-and-pinion steering; and astute tuning made for a very sporting four-door.
Now manufacturers can’t wait to ditch HPS, and for several reasons beyond the aforementioned 1-mpg fuel-efficiency gain. With no pressurized fluid coursing through rubber hoses under the hood, there’s no chance a leak will erupt, eliminating related service stops and warranty claims. Engineers add that EPS is easier to install during a car’s assembly. It’s also more tolerant of out-of-spec alignment settings and crowned road surfaces; the software that controls the assist motor is programmed to compensate, keeping the car on the beam during straight-ahead cruising. Finally, computer-controlled, motor-driven steering enables features such as lane keeping, cross-wind correction, and auto parking.
EPS is not all sweetness and light, though. Engineers are well aware of its *inherent shortcomings, some of which prompt our negative bias toward this type of power assist. These include far more friction and inertia than HPS, shortcomings that can make a car feel numb and lackadaisical unless the software governing EPS is tuned to compensate. For any feedback to travel from the road to the driver’s hands, EPS’s spinning electric motor must be slowed, stopped, and/or reversed. This is where the electronic controller comes into play. An array of chassis sensors—monitoring velocity, yaw rate, lateral acceleration, and other dynamic variables—keeps the controller apprised so it can issue appropriate commands to the electric-assist motor. Measuring what’s going on with the tires and the suspension indirectly amounts to intelligent guesswork, but it’s the best we have today to minimize EPS’s limitations. No wonder we found hydraulic better than electric assist in our subjective Feedback analysis.
Most of EPS’s friction comes from the ball-screw mechanism that converts the motor’s rotation into the linear force used to move the steering rack left and right. Ironically, this mechanism was also used in the recirculating-ball steering gears that were supplanted by rack-and-pinion systems ages ago.
There have been many attempts to cut EPS’s inertia. The Powertronic system introduced by TRW in 1986 was a third-generation design using a hollow-core motor, which allowed mounting the electric helper concentrically around the steering rack. That system was more compact than today’s side-saddle arrangement, but the hollow feature moved the heavy rotor windings farther from the motor’s center, adding to its inertia. (VW currently manufactures concentric EPS systems for use on some Audis.) Proving that there are many ways to skin this cat, TRW and others also have attached the motor to the pinion shaft (which works acceptably well for light cars) and to a second pinion gear in mesh with the rack. New arrangements may arise in the future, but for now the side-saddle design that BMW selected for several of its models is the most popular form of EPS.
While EPS was initially more expensive than HPS, suppliers quickly rose to the cause by cutting cost as demand grew for this new technology.
Rack-and-pinion won out over recirculating-ball steering gears decades ago because it was a significantly more direct means of guiding the front wheels with the minimum number of links at work. Fewer links meant lower inertia, minimal deflection under load, and less chance that friction would filter out useful feedback from the road. For EPS-assisted rack-and-pinion to be a genuine success before it is eventually replaced by steer-by-wire systems, it needs to strive to match what we consider the ultimate achievement in this car-design discipline: The Lotus Elise and Exige provided superb feel and feedback without pumps, motors, fluids, magnets, or artificial assists of any kind.
Download Test Data Here: http://blog.caranddriver.com/wp-cont...mance-Data.pdf
BMW 528i xDrive
VEHICLE TYPE: front-engine, 4-wheel-drive, 5-passenger, 4-door sedan
BASE PRICE: $49,875
AS TESTED PRICE: $61,125
ENGINE TYPE: turbocharged and intercooled DOHC 16-valve inline-4, aluminum block and head, direct fuel injection
DISPLACEMENT: 122 cu in, 1997 cc
POWER: 240 hp @ 5000 rpm
TORQUE: 255 lb-ft @ 1250 rpm
TRANSMISSION: 8-speed automatic with manual shifting mode
STEERING: rack-and-pinion with electric power assist, 3.0 turns lock-to-lock
STEERING RATIO: 17.1:1
CURB WEIGHT: 4069 lb
DISTRIBUTION: F: 50.3% R: 49.7%
TIRES: Goodyear Eagle LS2 RunOnFlat, 245/45R-18 100V M+S
BMW 535i xDrive
VEHICLE TYPE: front-engine, 4-wheel-drive, 5-passenger, 4-door sedan
BASE PRICE: $55,695
AS TESTED PRICE: $69,695
ENGINE TYPE: turbocharged and intercooled DOHC 24-valve inline-6, aluminum block and head, direct fuel injection
DISPLACEMENT: 182 cu in, 2979 cc
POWER: 300 hp @ 6250 rpm
TORQUE: 300 lb-ft @ 1200 rpm
TRANSMISSION: 8-speed automatic with manual shifting mode
STEERING: rack-and-pinion with hydraulic power assist, 3.0 turns lock-to-lock
STEERING RATIO: 17.9:1
CURB WEIGHT: 4212 lb
DISTRIBUTION: F: 52.0% R: 48.0%
TIRES: Goodyear Eagle LS2 RunOnFlat, 245/40R-19 98V M+S