I don't know the pitch of the spindle thread, but let's say for argument that the diameter is 1", and one revolution displaces the nut axially by 1/20". I.e., the nut displaces that 1/10" for every 3.14" of tangential displacement. At 45 ft-lb torque applied, the axial force will be 2826 lbs. In other words, you have just used a precision 1.5 ton mechanical press to seat the inner and outer bearing races into near perfect parallel planes while you spin the wheels.

If someone would like to measure the actual diameter and pitch, this could be refined.

Steve

__________________
'91 MB 190E 2.3
'08 RAV4 Ltd 3.5
'83 Lazy Daze m'home 5.7

Sounds to me like the out bearing were replaced but not the inners.

Rick Hall, did you replace both?

If you did the inners, how did you replace the inner races? If they are slightly crooked the bearings will beat themselves out of adjustment.

Haasman

__________________
'03 E320 Wagon-Sold
'95 E320 Wagon-Went to Ex
'93 190E 2.6-Wrecked
'91 300E-Went to Ex
'65 911 Coupe (#302580)

Oops, problems with posting too close to sleepitime - forgot torque-to-force conversion, among others. Tangential force at 1/2" radius from 45 ft-lbs is 1080 lbs, then tangential-to-axial conversion factor of 62.8, and we actually have 34 tons of pressure. That will seat about anything.

Steve

__________________
'91 MB 190E 2.3
'08 RAV4 Ltd 3.5
'83 Lazy Daze m'home 5.7

Brian, you basically said the same thing I did but perhaps more eloquently.

To everyone alse on this thread:
My # of 45 lbs was an arbitrary #. That # has no meaning other than to tighten the nut a little which would be an arbitrary amount. I'm reminded of the time I bought about 4 cases of oil at one of those super stores. The sign on the end of the cash register conveyer belt said not to put heavy objects on belt. I piled all the oil on the belt. The cashier really cycled through. I told her that to me the cases of oil were not heavy. My point is that if I had tighten the nut to remove excess clearance, someone might have used a small pair of water pump pliers.

As an aside has anyone ever given any thought to how rugged wheel bearings are? A 3600 lb Mercedes is supported by four little surfaces that don't amount to much more area than a U.S. dime. In my opinion wheel bearings take every bit as much punishment as the reciprocating parts of the engine.

__________________
Auto Zentral Ltd.

You're missing the point. You can't "seat" the outer race because it's already squeezed into the hub and stopped against the hub face. You could use 1000 tons and it's still going nowhere.

As far as seating the inner race, it lives on the spindle with a very slight clearance in most cases. So, seating it, axially, requires very little force. So, your desire is to simply take up the clearance between the rollers and the outer race and ensure that the bearing, with the inner race, is sitting on the axial stop within the hub. Adding more force, once the clearance is zero, won't accomplish anything. As your calculations show, you don't need much torque (5 ft.-lb. would be more than sufficient) to remove the clearance in the bearing.

You misunderstood me. You don't use the nut to mount races, only to seat them. I happen to be a bearing engineer and this is the terminology we use. I hope I didn't word it such that it can be misinterpreted. Mounting may not be enough to seat the races. And even if they weren't replaced it's always a good idea to take the extra step and make sure they're seated because of any micromovement that can unseat the races, which can happen for any number of reasons.

__________________
95 E320 Cabriolet, 159K

If the outer race is pressed into the hub and lives there with an interference fit, and if it is seated against the stop in the hub face, how will you cause it to move via adding torque to the nut??

Where, exactly, is it going to go??

Wow, this is gettin interestin. I'm beginning to believe we all know how to do it. We just can't splain it. The newbies must be really gettin confused.


Is there a tribologist in the house?

__________________
Auto Zentral Ltd.

Well, here is where all I can offer is my experience. I have drift tapped in cold bearing races until they stop, carefully tapped the circumference to verify I can accomplish no further seating, packed and re-assembled all, torqued to the spec'd 12 ft-lb, then adjusted to finger tight. Everything seems and sounds and rotates flawlessly. In less than 100mi of driving, there is slop. Re-adjust, and all is hunky-dory forever. This would not be possible if the problem was a fit issue among mating bearing surfaces - it is a basic problem when pressing the inner race.

Now, you can decide I do sloppy work and this should never happen, but luckily for me, I know what I do and am satisfied that I do precision work within the limits of my tools at hand. I do NOT have or use a precision press to install the races. That, if available, should do a better job. Still, seating of the race will be incomplete due to flexing of the race and friction of the fit.

Steve

__________________
'91 MB 190E 2.3
'08 RAV4 Ltd 3.5
'83 Lazy Daze m'home 5.7

Well said, Steve. There's just something about metal and machined surfaces where you don't get things perfectly snug. I too ask myself where's it gonna go when it's mounted and bottomed out?.... but it still moves. Experienced machinists know this when they clamp parts on a fixture for machining, and they develop protocols for minimizing slop or movement during machining. I find it difficult to explain this phenomenon.

__________________
95 E320 Cabriolet, 159K

Not being an expert on this subject and one who has been thinking about replacing my bearings for all four wheels, my head is spinning. Makes me think twice about having this all done by my indy. What would be a fair cost to have all four wheels done, inner and out?

Firstly, do we agree that the inner race is a slip fit onto the spindle........so.........no pressing involved there??

Second, if the procedure that you have used to seat the outer race is not satisfactory, and, as you noted, the bearing actually seates the outer race in the first 100 miles of driving, how much do you believe the outer race can move? I'm going to take a stab at the number and say that it cannot be more than a movement of .0005".

And, when you find "slop" when you return and pull on the hub, how much is this slop? Would it be something around .001" or would it be considerably more than this? If you find .005" axial play in the bearing, I think you would have a very difficult time attempting to make a case that the outer race moved sufficiently to generate this level of play.

I'm not questioning the mounting procedure...........simply suggesting that there is another variable that you are unaware of. It cannot be the seating of the outer race.

450slcguy, why do you want to replace ALL bearings? Automotive wheel bearings usually aren't replaced until there is imminent failure. Though if you're looking to refresh an older car, I would agree or regreasing and replacing seals on all four corners.

Brian, MB specs the bearings at 0.01-0.02 mm axial play for proper installation, yet allows for 0.01-0.05 mm axial play as a check after use.

If I may take a stab at where this extra play comes from, it can be grease or grit that finds its way onto the mounting surface that hasn't been squeezed out or crushed down. Also, there are asperities on the surface that preclude a perfect mount. During use there is some flexing of the components and micromovement of the surfaces that act on these irregularities and increase play. Large amounts of it are called "fretting". This presents itself as polish with perhaps some brown discoloration on the mount surfaces. The brown is iron oxide products from the fretted surfaces.

__________________
95 E320 Cabriolet, 159K

I'm aware of the specs. I like to work in inches. The limit is .0008" axial play during a proper installation.

I agree with all of your statements regarding where the extra play comes from. All of those possibilites will give a variable on the order of .0005" or so. None of them can be of significance to the point where an axial play of .005" would be realized.

So, the question remains. If the axial play increased to approximately .005", where did it come from? It's not from flexing of the components, micromovement of the surfaces, or fretting.

Like Kestas, I have machining experience that has taught me to pay attention to the details of a procedure, and to try to understand when my experience does not match my expectations. When you press in the seat, you might expect that when it 'bottoms' that you have removed all play and slack. Many things can be undermining that. The seat may be contacting at just one point, there may be minute shavings or hard debris that have scraped into the gap, and certainly there will be rebound of the race when you release pressure or impact. This is just the nature of real systems.

As to my experience: I do a final adjust by backing off the nut until I can finger tighten it. Then I go slightly forward or back to the closest cotter pin alignment. When slop develops, you can easily tell by the difference in play shaking the raised wheel at 6 and 12. With my technique, I find the new cotter alignment will be usually one hole beyond previously.

I find actually driving around the block a few times will usually seat the race, followed by the readjust. But, 'over' torquing the hub nut on initial adjust may be as effective.

Steve

__________________
'91 MB 190E 2.3
'08 RAV4 Ltd 3.5
'83 Lazy Daze m'home 5.7