We Won't See CVTs PDQ Three letters that spell better transmission efficiency

Automotive customers are steadily becoming more shiftless. No, not in their general deportment. In their attitude about manipulating transmissions themselves.

Ward's data marks 1997 domestic passenger-car installation rates for manual transmissions at only 13.6%; light trucks came in a bit higher at 16.2%. Nearly nine out of 10 vehicles produced domestically have an automatic transmission, and installation rates for import passenger cars in the U.S. have doubled from 1982's 36.8% to 74.6% in 1996.

So U.S. buyers don't want to shift. Europeans, too, are indicating a similar predilection. But talk of drastically increasing the fuel economy of passenger vehicles flies directly at odds with the use of automatic transmissions. The reason: Despite relatively efficient modern designs, automatics are energy wasters when compared to manual transmissions.

Or CVTs - Continuously Variable Transmissions .

Mike Kluger, manager-automatic transmission technology section at Southwest Research Institute (SwRI) in San Antonio, TX, says CVTs can be up to 90% efficient, which compares favorably with the 89% maximum efficiency of today's best electronically controlled automatics and even the 97% efficiency of a manual transmission. CVTs offer an overall efficiency leap, says Mr. Kluger, because they are much more adept than conventional automatics at always matching the speed of the driven wheels to the optimum power-producing engine speed. So if you accept the notion that few drivers can be expected to regress to shifting their own gears, from a drivetrain "systems" standpoint CVTs are destined to be the carnivores of the next-generation transmission food chain.

Mr. Kluger is one of the industry's foremost automatic transmission experts. His work at SwRI encompasses research, testing and evaluation of advanced transmissions, many of which are highly secret. CVTs fall under his jurisdiction because they are fully automatic - and that's fortunate. It makes CVTs one of the few high-efficiency technologies that falls in step with current customers' buying preferences.

Instead of the three or four fixed gearsets found in a conventional automatic transmission, a CVT can "select" from, effectivley, an infinite number of gear ratios within a given range.

The most common CVT design is the belt-driven configuration, most closely associated with Van Doorne's Transmissie bv, the company that developed the critical metal-banded belts that transmit drive torque. The belt-type CVT locates the belt between two pulleys, one on the crankshaft side of the CVT, the other on the driveshaft side. Each pulley's diameter is variable by mechanically squeezing together the sides of the pulley; as the pulley sides open or close, the "trough" in which the belt rides is widened or narrowed, varying the diameter of the pulley itself. The width of each pulley, then, determines the effective drive "ratio" at each end of the CVT.

Here's the rub: early belt-type CVTs could handle only a middling amount of engine torque - limiting potential applications to sub-2L engines. But improvements in belt technology and electronic control enhancements have elevated belt-type CVTs to new horizons of larger-engine application.

At Ford Motor Co., for example, Walter K. Muench, principal research engineer-transmission and engine systems department, assures that Ford (with help from Van Doorne's) has a belt-type CVT capable of handling the torque of a stock Duratec 3L DOHC V-6 that develops 200 hp and 200 ft.-lbs. (271 Nm) of torque. Mr. Muench's "assurance" comes in the form of a Taurus CVT-motivated prototype he's happy to let you drive. It works marvelously.

He adds that CVTs can improve efficiency over 4-speed automatics by 15% to 20%, yet the CVT is simpler and costs less than adopting a 5-speed automatic to achieve similar gains.

Asked if he believes Ford's CVT is ready for volume production in a vehicle like the Taurus, Mr. Muench confidently claims "I'd love to do it. We're ready." The technology might be ready, but the auto-industry manufacturing complex isn't. Today's market environment doesn't justify that sort of move, even if CVTs can deliver "a guaranteed 5% fuel economy improvement for any car," by Mr. Kluger's estimate.

The only CVT currently available in the U.S. is a belt-type design offered by Honda Motor Co. Ltd. in the Civic HX. Volumes are low, but the Civic HX's fuel-economy numbers are high: it's the only automatic-transmission vehicle to make the U.S. Environmental Protection Agency's 1998 list of top-10 fuel-efficient vehicles sold here.

Right now, Japanese automakers are the primary motivators in CVT development and the Big Three all are taking a look, too. Nissan Motor Co. Ltd., for one, has a variety of CVTs; its Hyper CVT and Hyper CVT-M6 are belt-type units incorporating high-pressure hydraulic controls and a torque converter that helps the CVT "mimic" some of the more esoteric driving characteristics of common automatics.

Nissan and a number of other interests also are developing toroidal-design CVTs. The toroidal units use discs and roller elements that, while arranged in a fashion generally similar to a belt-type CVT, transmit drive via friction forces between the discs instead of friction between a belt and pulleys. Toroidal designs can handle larger-displacement engines and serious torque.

Virtually nobody in the industry disputes the advantages of CVTs. But increased fuel economy simply isn't enough to drive carmakers to invest ... just yet. SwRI's Mr. Kluger says "The 'need' just isn't there."

But he quickly adds, "If somebody's got to bite the bullet really hard, they'll do it."