Continuously Variable Transmission (CVT) cars use a belt-like drive to obtain an infinite number of gear ratios. Sadly, in practice, many have not worked out well.
CVTs or Continuously Variable Transmissions are hitting the mainstream and many people are buying cars with CVTs and not even knowing what they are or how they work - and whether they are reliable or not. In theory, CVTs should provide astounding gas mileage, smooth power flow (with no shifting) and a long service life. In practice, many have fallen short.
The most common CVT you may be familiar with is a snowmobile transmission. Snowmobiles used to have a rubber belt drive with two sheaves (pulleys) that had variable diameters. The variable diameter of the pulley is achieved by having two cone-shaped sides that are spring loaded or otherwise varied in width. As the width of the pulley increases, the effective diameter decreases, as the belt rides lower in the pulley. They work well in simple machines like snowmobiles, where centrifugal action can be used to activate the pulley, and as you accelerate, the ratio automatically changes.
They do eat up belts, over time, due to the side pressure on the belt. But for a snowmobiler, who may drive only a few hundred (or few thousand) miles a season, changing the belt is no big deal (indeed, it almost falls off, at rest) and the belts are cheap.
People have tried to use such drives on cars and other equipment, over the ages. When I worked at Carrier, we used such a drive to provide continuously variable fan speed for air handlers. It worked OK, but I remember the belts making a lot of dust in the motor compartment, as they slowly wore out.
For cars, a rubber belt won't do, and many have tried variations on steel belts. Now imagine a steel belt running on a steel pulley. What could possibly go wrong? Well, as you imagine, all that steel-on-steel contact results in wear, and the belt eventually breaks (even though it may run in an oil bath). BMW tried using a CVT in the Mini, and it did not work out well. The transmissions failed, usually right out of warranty, and people got pissed off and sued. BMW pulled the product from the market and went to a more conventional transmission design.
GM had used CVTs in its Saturn lineup, with similar results - a class action lawsuit. Audi had a similar experience. The track record of CVT cars has not been stellar.
This is not to say that all CVTs are problematic - or indeed that conventional transmissions are bulletproof. Nissan had a suit filed against it for its 5-speed box. And BMW's "dry clutch" automatic has had its share of issues. Perhaps that is the nature of the beast, with modern cars. As they become more complicated and more esoteric technology is applied - in the search for better efficiency, safety, and cost reduction - some problems will inevitably emerge.
CVTs are attractive to engineers, as they have few moving parts, are small and lightweight and cheap to produce. And they provide infinite gear ratio possibilities, which means the car can be programmed to always be at the most optimal gear ratio, and thus be running at its most efficient state. In theory these are all great things.
There are some odd behaviors, that some consumers may be troubled by. When you step on the gas on a CVT car, the engine may rev very high, until the transmission "shifts" higher and the car accelerates. New users sometimes freak out when the engine revs like that, and they take their foot off the gas, and the car sort of lurches, as the transmission is trying to keep up with driver input. "Speed up or slow down? Make up your mind!" The transmission gets frustrated.
But most drivers get used to driving the CVT over time - although it does take reprogramming your brain a bit. Unlike a conventional automatic, there are no "gears" to shift through, so the driver doesn't feel the shift points like in a conventional automatic. Some CVT makers have compensated for this by programming the CVT to stop at certain points in its shifting curve to provide an artificial shifting feel - or allowing the user to "manually" shift the CVT. However, this sort of defeats the main advantages of the CVT - smooth power and optimal engine rpm.
Traditionally, CVTs have been limited to small, low-horsepower cars, as the torque from a larger engine is usually too much for the CVT "belt". However, recently, CVTs are being rolled out on a number of midsize and larger cars. Nissan offers its CVTs in a number of sedans and mini-utes. For some cars, it is the only transmission option.
Nissan is being pretty brave, I think, to offer the CVT across the product line. Given the disaster BMW had with its Mini. But other automakers are offering CVTs (see link above) and in fact, it seems nearly every automaker has offered, is offering, or will offer, a CVT transmission in one of their cars.
Are CVTs the wave of the future? Are they reliable? Tough questions. A lot of technologies that fared poorly at the outset, later on turned into mainstream products. The Chevy Vega did poorly with its high-tectate silica engine block. But years later, BMW adopted similar technology with great success (well, except for the high-sulfur fuel problem in the 1990's).
The interesting thing, to me, is that conventional automatics are keeping up with the times as well. Many are going to dry-plate-clutch transmissions (such as BMW) with conventional gearing. And the number of gears in an automatic has gone from two, to three (in the 1960's) to four (in the 1970's) to five, to six - even up to nine. With so many gears in a "conventional" automatic, the advantage of a CVT may not be so pronounced.
However, I think the CVT is probably here to stay, at least for smaller cars and smaller engines. The small size of the transmission, combined with its light weight and better fuel economy, is too irresistible for manufacturers to pass up. But, they have not been on the market long and it can take 10 years or more for a design to prove itself. Mini owners found this out the hard way, as the cars failed after the car was out of warranty, and the cost of a new transmission was often more than the resale value of the car. Since the cars had a poor reputation because of the transmission, the resale value dropped, making replacing the transmission cost more than the car was worth.
As I noted in another posting, being an "early adaptor" can be costly. If a new technology has teething problems, well, you can end up losing a lot of cash and being frustrated. Consider, for example, the poor bastard who buys a Corvair back in 1965. Ralph Nader writes a book about the car, and resale values plummet. So trades it in for a Chevy Vega. Whoops! That didn't quite work out. So he trades the Vega in for a Chevy Citation. After that, well, Toyota has a customer for life.
The problem with the CVTs today is that it may be impossible to escape from them. All the major manufacturers have played with them or are offering them or are going to offer them. And you may not even know the car has a CVT, unless you ask some pointed questions.
For myself, I guess I will sit on the fence awhile and see how it works out for others. Nissan is selling an awful lot of Altimas with CVTs. We'll know in a few years how that works out. If it works out fine, well, maybe I will buy one. If not, well, I'm glad I waited. It does appear Nissan had some teething problems with the transmissions at product launch. It appears that these may have been worked out, and many of them are related to customers being unfamiliar with how CVTs work. Some very early CVTs models (e.g., 2008) may have had some failure problems as well.
What is interesting to me is that Nissan is putting these transmissions in fairly heavy cars with fairly big engines. To date, CVTs have largely been limited to small cars (Subaru Justy, BMW Mini, etc.) with smaller engines. It will be interesting to see how these CVTs handle torque from larger engines and larger, heavier cars.
What is interesting to me is that Nissan is putting these transmissions in fairly heavy cars with fairly big engines. To date, CVTs have largely been limited to small cars (Subaru Justy, BMW Mini, etc.) with smaller engines. It will be interesting to see how these CVTs handle torque from larger engines and larger, heavier cars.
Ordinarily, I am not a fan of extended warranties. However, if you are buying a car with some new technology that is unproven, it might be worthwhile to investigate a manufacturers extended warranty (third party extended warranties are generally worthless). If you own or are thinking of buying a Nissan, there is good news: Nissan has extended the warranty on their CVTs to 10 years or 120,000 miles - to show their confidence in the new technology (they have some balls there at Nissan, I will give them that!). So, as long as you don't plan on keeping the car longer than that, well, nothing to worry about.
In two to three years, we should know whether we will be living in a brave new world of belt-driven cars, or whether more conventional technology will continue to soldier on. Stay tuned.
P.S. - if you have a CVT car, I would suggest driving it gently. Beating on the car with sudden acceleration and the like, is not going to help matters any.
