CV Joint Diagnosis and Repair Alternatives

If you think FWD is a relatively new idea, you're way off. It dates back to 1877 when New York attorney George B. Selden filed a patent application for a car with a transversely-mounted, three-cylinder, two-stroke engine driving its front wheels. Although his "road engine," as he called it, wasn't actually built until 1905, and even then couldn't be called a useable vehicle, it's still historically important because the licensing association formed under its patent had a profound effect on the early growth of the auto industry.

The potential advantages of pulling instead of pushing have intrigued vehicle designers ever since. As automotive pioneer Andre Citroen said, "Driving the front wheels is a natural idea. After all, the horse doesn't push the cart." That was fine in theory. The trouble was that channeling torque to wheels that have to both steer and rise and fall with the suspension required a component that hadn't been invented yet. The ordinary Hooke or Cardan universal joint simply couldn't handle the angles involved.

In 1928, Alfred Rzeppa patented what he called a "ball-type universal joint." Not only could it transmit twisting power through unbelievable bends (modern versions have to tolerate 20 deg. up and down, and over 40 deg. side to side), it also eliminated an unsettling characteristic of the regular U-joint: When operated at a sharp angle, output shaft rpm fluctuates even with steady input speed. So, Rzeppa's creation was eventually given the name "constant velocity joint." Later came the tripod (sometimes called "tripode") and tulip type, a French invention, that did essentially the same thing in a different way.

But there were serious durability problems with early joints. They needed much more development and better lubricants before they could be considered dependable.

Improved technology and the genius of Sir Alec Issigonis came together in 1960 with the introduction of the first truly successful FWD automobile: the Austin-Morris Mini, which stands as a landmark in engineering. It had a transverse engine, a lot of interior space for its size, and, with modern CV's and grease, its front wheels were able to continue transmitting power to the pavement for a reasonable number of miles.

As far as significant numbers are concerned, the FWD revolution in the U.S. began in the early seventies. Volkswagen was a heavy contributor with the Dasher, then the benchmark Rabbit, and there were Audis, Saabs, Subarus, and Fiats. Of course, we had our own home-grown examples in the Toronado and Eldorado.

Although some highly-respected automakers -- Mercedes, BMW, Porsche, Volvo, et al -- are sticking with the cart-before-the-horse arrangement, FWD cars dominate new vehicle production today. There's just no denying this layout's advantages in terms of space and weight savings, traction, and forgiving handling traits.

But there's never a rose without a thorn. Simply stated, FWD complicates serviceability. It'll be the rare do-it-yourselfer who's bold enough to attempt CV joint
replacement.

Although some specimens just keep going and going -- I had a VW diesel in my shop recently that's still running on the original joints after 300,000 miles -- premature failures aren't uncommon.

Whenever you're worried about strange noises or vibrations in an FWD car, do a thorough road test. Accelerate hard, decelerate, cruise straight, turn right and left slightly while maintaining a constant speed, then try some tight turns.

A clicking noise when turning that sounds like marbles in a hubcap indicates a problem in an outboard joint. Clunking on accel/decel suggests that an inboard unit is in trouble.

Inner CV's should also be suspect if there's vibration while cruising, but it'll be different from that caused by tire imbalance because joint trouble usually shows up only at very specific speeds. At 49 mph, for instance, you might feel like you're getting a mechanical massage, but it'll smooth out at maybe 51. The same is true if the problem discloses itself at certain points during acceleration, but not all the way through. Because of the plunging action, noise or vibes will change as you go over bumps. A whining sound that's sensitive to small steering inputs is usually caused by a bad wheel bearing.

Unfortunately, in many cases you still won't feel sure of what's wrong. Some technicians like to take the car to a deserted parking lot and start driving backwards like a crazy man. This is supposed to exaggerate the symptoms because it shifts the stress points inside the CV's. Do a tight slalom in reverse and note when the problem is at its worst. Because of the way torque is applied, the wheel on the inside of the curve will complain the most.

Take your mental notes back to the garage and get the car safely up in the air for a visual exam. The first thing to look for is a damaged boot, by far the most frequent killer of CV's. If there are no obvious rips or punctures, but you see evidence of grease on the rubber or around the boot on other components, there's a hole in it somewhere.

Next, go hands-on. Support the lower control arms with stands so that the driveline isn't forced into an unnatural angle, get the brake pads away from the rotor, then feel and listen for play and roughness in the wheel bearing as you rock and spin the tire. You might uncover a loose hub or play in a suspension bushing or ball joint while you're at it. Then, put one hand on a CV and the other on the shaft and see if you can detect lash in the joint (on outboards, don't confuse this with the clearance between the inner race and the splines).

I hope the next step isn't often necessary because it sure is dangerous. With the suspension still supported, have a helper start the car, put it in gear, run it up to no
more than 30 mph, and steer back and forth while you watch and listen. Just keep your wits about you because all that spinning rubber and steel could really beat you up. Use a stethoscope at the knuckles if you suspect the wheel bearings.

Since outers have to deal with much sharper angles than inners, their failure rate is far higher. And those on the right typically have a shorter life span than those on the driver's side.

A bad knuckle bearing can fool you, so I'd better expand on the subject. As one successful independent shop owner once told me, "We do many FWD wheel bearings, especially on Chrysler products. The big problem with this work is finding out which side is bad. You can't depend on weight shift -- steering back and forth -- to give you a sure diagnosis. It seems that every time I say it's the left, it's really the right, and vice versa. If you try to run it on the lift, there's no weight on the wheel, so the same sound might not be there. I handle this with the customer two ways. One, I say 'You tell me which one is bad, and I'll replace it.' Or, two, we change 'em both. You can always come down on an estimate, but you can't go up."

Negative scrub radius contributes to the difficulty of zeroing in on the side with the problem. If you were to disassemble a failed unitized ball-type bearing, you'd probably find that the inside ball/race is shot and the outside is okay.

Another shop owner said, "One problem is noise transmittal all across the front end of some new, tinny cars. And if you put it on the lift and listen, the brinnelling is
probably on the top, so without the weight on the wheel you might not hear it. There are also plenty of instances where both sides ARE bad -- maybe 50% to 60%."

With a little luck you should have zeroed in on the offending component pretty well by now. But leave your thinking open-ended in case you find that more than one item is FUBAR (Fouled Up Beyond All Repair) upon disassembly. One shop owner I know uses a good approach. Whenever he finds a damaged CV, he explains to the customer that the others may be in trouble too, and that the best course of action at this point is to remove both shafts, replace or repair the bad joint, clean, inspect, and relube the others, then install four new boots as insurance against another big bucks repair.

Direct inspection is interesting, and in some cases the only way to be sure which joint is actually defunct -- they may all have sounded noisy. Naturally, start with the
side of the car and the joint that your investigations so far have led you to suspect. Once you've got all the grease out, look for unusual wear patterns. On the ball type, you'll generally see similar scoring, pitting, etc. on every other load point on the balls and races. An even, dull finish is normal, and don't worry about a discolored housing -- that's from the heat treating. Frequently, the ball cage will be cracked, rippled, or just plain worn out. With the tripod type, the rollers and needle bearings are the weak points.

Before you begin your exam, think about keeping the parts in order in case the joint is reused. Remove, inspect, and reinstall each ball individually before going on to the next. Index a tripod and its housing so you can put the rollers back into their accustomed slots.

If you've found a ripped boot, but the only symptom so far is a slight humming sound and the grease looks fine and feels okay between your fingers, there's a good
chance you can save the joint by cleaning it thoroughly, packing it with the proper lube, and installing a new boot.

Although some technicians regard rebuilding CV's with the same distaste as putting new lifters on an old cam, it can be a successful repair. Major joint makers are
offering parts kits, which typically include a cage and balls, or a tripod assembly, a boot, clamps, and a package of lube. New inner races are available separately.

Caveats apply. If the housing displays any damage, you can forget about rebuilding. And, as one expert told me, "When you rebuild a joint, you're pretty much guessing as far as fit and tolerances go. An experienced guy might do okay, but it's sure not something for somebody who's not too familiar with CV's to try. He'd be lucky if the job lasted 20,000 miles."

Add the substantial reduction in the price of a CV over the last few years, and you may find that installing a new joint is a more attractive alternative than refurbishing the old one.

While you may think of it as extravagant, replacing the whole shaft assembly is the ultimate in speed and convenience, and you won't have to deal with vast quantities of sticky grease or worry about getting that joint locked on or securing boots properly. Other advantages to this repair route include a low disaster quotient and no potential mix-ups when different joint styles were used in the same year. In those cases where the outboard is difficult or impossible to remove from the shaft (such as on various Hondas, Toyotas, Subarus, etc.), it's the only practical fix.

And the economics aren't too bad anymore. For example, a brand new outer joint for an '84 Caravan can cost $240, whereas a complete remanufactured shaft goes for around $300. On the same vintage Escort, the difference is a mere $13. Since you really should be installing a new boot on the other joint, deducting its price brings the two alternatives even closer together.

Different grades of shaft assemblies are available. You can buy an all-new unit, but you'll save money with rebuilt or remanufactured. With the former, you'll get a fresh cage, balls, and maybe an inner race, or a new tripod, installed in an inspected housing. To qualify as remanufactured, the outer race will have been reground for the installation of oversize balls (up to .015 in. larger in diameter than stock, which still leaves over half of the heat treat). Companies that manufacture O.E. joints generally don't bother with regrinding. If a housing's bad, they just replace it. There's usually a good warranty with brand-name remans -- limited lifetime with labor reimbursement at 50% of the aftermarket flat rate, for example.

Situations vary all over the map, so I can't tell you which repair to choose for a particular car. But I can say that talking to your parts man about the alternatives he offers (the track record of returns is especially important) and following the ads in automotive magazines will help you decide.

A few generic tips and I'll let you go:

The grease supplied with CV's and boot kits is special stuff. Use it all, and don't even think about substituting ordinary lube.
During service, never allow the weight of the halfshaft to bear on a joint because it'll jam it at an unnatural angle and probably damage it. Suspend the shaft from a wire.
Frequently, a boot will be collapsed and distorted after installation, which can make it fail quickly. It'll pop back to its normal shape if you insert a narrow, rounded-off rod or a dull screwdriver into the small end to relieve the vacuum.
Taping the splines will protect that new boot.
Not securing a boot clamp properly because you don't have the right tool or instructions is asking for trouble.
Don't try to clean tripod joints with regular solvent. It doesn't really dissolve CV grease, and you won't be able to get it all out of the needle bearings. Use an evaporating spray cleaner such as the type made for brakes.
There are so many variations in R&R procedures out there (Do you have to take out the speedo cable or drain the transaxle? Is there a roll pin somewhere?) that you should always refer to the manual.