Technical Spindle bearing preload

I believe that if you could be inside the hub at operating temperature and speed, you would find that while the body of the 'grease' doesn't move... The oil in the grease does...
Grease is about 40% or more oil, the rest is additives and thickeners...
This isn't perfect, but think of the thickener as a sponge. As the unit operates the sponge allows the oils to be released and reabsorbed.... Adding extra grease to the hub center provides a reserve of oil to the bearing, adding extra grease to the cap does the same, as well as creating a seal if needed.

Also a lot of things that are perfectly correct in the automotive world are huge no-no's in the industrial world... In industry you almost never find a bearing with freeplay or packed 100% with grease..

 

I was taught to spin the hub and use the weight of the big pliers to set the pre-load. It works out pretty well for me and I've only lost one bearing in 45+ years of doing them. That was probably washed with solvent as was pointed out as not a good thing to do. If the bearing is that gritty, replace it.

I worked in one shop and one guy would stuff the cap with grease. This was the same idiot that would do drum brakes 1 side at a time and run back and forth looking at the completed side for reference. Same guy who would apply disc brake quiet to the pad surface.

 

I couldn’t get an A spindle bearing to the couple of thou endfloat I like even when tightened up fully , only to realise the spindle outer bearing area was worn quite a few thou.
I cut a .003” shim(as the bearing area was worn .006”) to wrap around the worn area on the spindle so the bearing was a good slip fit then was able to get it correct on the end float.
A bit of mucking around but worked fine.
Good A spindles aren’t lying around down here and this fix worked well.
Anal or rough,call it what you will.

 

One of the bearing manufacturer engineers, Timken or SKF probably, discussed the objective when setting tapered wheel bearings in an article. The idea is actually zero preload, neither too tight nor too loose. All of the different procedures described or taught are simply methods to achieve a "close enough" without benefit of special tools. Even the Model T operator's manual describes a pretty clear description.

I think you'll find anytime there is a set of tapered wheel bearings on a wheel w/ a hub cavity they always specify to fill the hub with bearing grease. Maybe not industrial, because they don't operate outside? Aircraft manuals - landing gear. Fill the hub.

I get why people don't want to do things, but they at least should be able to explain why it really isn't necessary and they are (I guess) smarter than the people who designed the equipment. Nome sayin'?

 

My opinion , and why I put extra grease in the cap and hub, is that if you experience bearing failure, it will cause heat, melting the grease and letting it flow into the bearing area.
On hot rods this will problaly never happen , as you would hear/ feel the problem long before that. But it could help on your trailer. 95 % of the wheel bearing failure I have encountered was due to lack of grease!








Bones

 

The way it was explained to me was ideally a bearing should be adjusted with zero preload or zero free play. Since it is virtually impossible to measure “zero”, it is better to err towards some free play. I try to set them to the low side of the prescribed.001-.005.

 

Right on the money. In heavy trucks this has long been a topic on the correct way to set wheel bearings. If you talk to enough mechanics it gets down to a "feel" thing. Bearing manufacturers agree that a slight pre-load is ideal. However, measuring pre-load is very difficult. You can't say a specific torque will give you the correct pre-load because of different threads, lock nuts and manufacturing tolerance. It is relatively easy to check end play with a dial indicator, so the 0.001 - 0.005 spec is the industry standard.

On heavy truck we seat the bearings by torquing the adjusting nut to 200 ft lb while turning the wheel. Then back off the adjusting nut and re-torque to 50 ft lb while turning the wheel. Back off the adjust about an 1/8 turn and check the end play with a dial indicator. Then install the locking device and re-check the end play. If it remains 0.001 - 0.005 you are good. You should realize that the bearings, spindles and locking nuts on heavy trucks are much larger than those on cars so the torque numbers are much higher.

 

Weellll, I'll throw my 2 cents in....a good friend of mine, happens to be a mechanical engineer and was a Vice President of a glass bottle company ( Wheaton industries) found that tapered/opposed bearings were substantially more durable when they had a SUBSTANTIAL amount of preloading on them in big and small machines in the glass industry....when I do wheel bearings however I still take the slack out so they don't wiggle and tighten to the next cotter pin hole....

 

I don't get why people don't want to do things, I guess some people like to do the same job twice? Just about every manufacturer has a variation of this theme: install outer bearing races in hub, pack bearings install inboard bearing and seal, slip hub on spindle, install outboard bearing and retainer(s), take up slack while rotating resistance is felt, back-off some degree...

My earlier post shows ideally you want 0.001-0.008" free play. One weekend when you are board, get a dial indicator out and check your free play; I'd bet that "feel" (that you have learned to be acceptable) is probably on the low end of that range, mine usually is.

How full you pack a bearing housing has some to little to do with environmental considerations.

Partially filling the hub with grease provides a reserve of lubricant (oil) to the bearing as stated previously.

Packing the housing full leads to grease churning, which causes the unit to heat up, which would increase the load on the bearing (greater thermal expansion) while simultaneously degrading the lubricant.

Not packing the hub would likely lead to starving the bearing of lubricant.. on short trips this probably isn't an issue, but load the vehicle heavy and spend hours on the interstate....

Most heavy trucks run oil in the hubs for the above reason; the bath can be maintained at a level which reduces friction and wear while carring heat away from the bearing.

Look around post #28, SKF agrees with the preload... The problem is, as I mentioned previously re industrial applications, industrial apps generally operate a very predictable speeds, loads and environmental conditions... Something you will almost never see on an automobile.

 

As a son and grandson of GM - New Departure Ball Bearing Div. employees, I can say that you are correct, Sir!

 

At work, all of our roller bearings are preloaded (and by robots).

We're just a bunch of two-bit hacks, though. We only sold ~7,700,000 vehicles last year.

 

Since you need to back the nut off to where the cotter pin lines up,I don't think it is overly critical.

 

That's how I was taught as well. Work's great. Spinning the wheel and checking clearance is the key. Too tight and you get to do it all over again with new bearings....

Sent from my Pixel 3a XL using The H.A.M.B. mobile app

 

Well said as long as it isn't too loose.

Sent from my Pixel 3a XL using The H.A.M.B. mobile app

 

Hardly ever board, fir some reason.

 

The chart defeats the argument that Taper Roller Bearing should be setup with free play, as it shows the bearing life expectancy peaking at circa 0.003" pre-load.

For many years to now, part of my work has been rebuilding CNC machine tool spindles. In the majority of cases, the bearing set up is Angular Contact Bearings arranged Back to Back, or Font to Front, with the pre-load built into the bearings. However, on some US built machines, the spindle bearings were opposing Taper Roller Bearings. These were pre-loaded via a spacer set between the bearings with the pre-load adjustment made by grinding the spacer to a length that would result in the correct bearing pre-load, measured in terms of the torque required to turn the spindle with nothing attached to it.

With regards to lubrication, the majority of CNC machine tool spindle bearings are a Grease Once for Life and the rule of thumb for the amount of grease to use is 10% of the available air space within the confines of the bearing. Accordingly, for Taper Roller wheel bearings, I pack the bearings with grease and that's it. In the many years I've been servicing E Type Jaguar cars, we have never had a problem, or return with this procedure.

A CNC machine, particularly a Machining Centre, running 24/7, the spindle bearings will rev in a couple of month, far more than a typical driver car will in its expected life.

Whether the bearings are better set up in pre-load or end float for a motor vehicle depends greatly on the axle/bearing arrangement. Many modern cars use Double Row Angular Contact bearings that have a split inner race with pre-load built in and when two halves are pressed together when the bearing is secured, the correct pre-load results.

With many older cars that use opposing Taper Roller Bearings, the ID of the bearing is a close running fit on the shaft. All Jaguars, perhaps excluding very current models used this system and in fact, the front stub axle of all E Type Jaguars used the same bearings and axle dimensions as a typical, small box trailer. Every Jaguar where I have dismantled the front hubs, ALL show a wear band as wide as the bearing, between 4 and 8 o'clock on the shaft; clear evidence that the inner race of the Taper Roller bearing has been spinning on the axle. And this is with the bearings set with recommended end float.

The reason for the running fit of the bearings on the axle is for ease of maintenance. If the bearings were an interference fit, particularly the inner bearing, when removing the hub for any reason, the inner bearing and seal will be left on the shaft, requiring another operation involving a puller of some description.

Given that the bearings can spin on the shaft with end float, by setting this opposing Taper Roller bearing arrangement with pre-load will simply guarantee the bearings to spin on the shaft from the get go.

There is ample evidence and text indicating that Taper Roller bearings prefer to run in pre-load than end float. Accordingly, to achieve this and to assure that the bearings can't spin on the axle, I use a spacer between the opposing bearings on the shaft and obtain the correct pre-load via shims. This arrangement allows for the axle nut to be torqued up relatively tight so that the inner and outer bearings are clamped between the nut and and the stub axle carrier and can't spin on the axle.

Regards,

Bill

 

Hi Bill, welcome to the HAMB.

Yup, that's exactly what happens when you pull the front hub off a really old MG. They have a precision spacer between two ball bearings, and they're a light press fit on the spindle. The spindle doesn't wear out, but other forms of maintenance mayhem ensues, from this arrangement.

The funny thing about those awfully designed front tapered wheel bearings with end play, is that they still work, decades later. Sometimes the cheap design is good enough for the job at hand.

 

Depends on how you define "critical" I suppose, this technique is more or less a field expedient, going all the way back to at least the Model T days. It works really well from the standpoint that no special tools are required.

If you've turned wrenches for a living, you well know the pressure is always on to work just about as fast as possible. Not really "cut corners" but... Well. Nobody is likely to be busting out the dial indicator ( if they even know how). Time is money.

The main thing, so long as the spindles don't get torched on the drive away from the shop, everybody including (and especially) the customer, is happy.

 

This is a huge topic in the heavy truck industry. Tapered roller bearings on a wheel will last the longest with a slight amount of pre-load. However, measuring pre-load on a wheel bearing isn't easy. Therefore, the industry standard procedure for heavy trucks is to set up the bearings for 0.001 - 0.005 inch end play. You can easily measure this with a dial indicator.

With that said, on my cars I snug down the bearings pretty tight, maybe 25 ft lb (never really checked), then back off until I get near zero end play. Been doing that a long time.

Back in the 70's I worked with a guy who miss-read the Ford manual. There procedure was to tighten to 50 ft lb while turning the wheel, then back off and re-tighten to 25 ft lb. After that you back off the adjusting nut two spots and put the cotter pin in. Chester didn't see the final back off and he left the bearings with the 25 ft lb pre-load. He only did this on cars with disc brakes and I never saw one come back with a wheel bearing problem.

 

Maybe they didn't want to take it back to the same place that f&@$ked it up to begin with? /jk

"Just right" adjustment means the bearings are supporting the load as designed, about half of the rollers, instead of just a couple. They will last longer if installed correctly.

 

Hello Jim,
Thanks for the welcome.

When using a spacer between opposing Taper Roller Bearings there is no requirement for the bearings to be an interference fit on the shaft, not that you would want a big clearance fit either, but a close running fit is just fine. In terms of alignment, the intimate contact of the Cone and Cup of a Taper Roller Bearing assembly when in pre-load makes for very good alignment. That, combined with the close running fit with the shaft and the rigidness of the assembly imposed by being able to torque the axle nut up tight, is the equal to having the bearings an interference fit with the shaft. In fact, a stub axle with the typical wear margin from 4 to 8 o'clock can be given another lease of life by using the spacer and bearings in pre-load. I don't do this often, as we make our own stub axles, but a moderately worn shaft can be rotated 180degs to put the unworn section of the shaft at 6 o'clock and with the use of the spacer, it's quite an acceptable alternative to a new shaft.

Over many years, we have been monitoring different approaches to the issue of the use of opposing Taper Roller Bearings and because we manufacture Stub Axles, have tried all possible combinations, from close running fit of the bearing to heavy interference (where we were able to put the bearings in pre-load without using a spacer). The clear conclusion to me, is that the close running fit of the bearing with the shaft and the use of a spacer between the bearings is the best solution. I've not encountered a downside, nor other maintenance issue with this arrangement.

The picture below is of a stub axle we manufacture for S3 E Type Jaguar.

Regards,

Bill

 

No, but the MG uses ball bearings, to make things interesting.

 

That's fine, but the stub axles/spindles we use here are not a separate piece.

 

Hello Jim,
I'm not overly familiar with the MG arrangement, but I am with a wide range of bearing types. If using ball bearing races, they would have to be angular contact bearings arranged how I described they are in Machine Tool Spindles, otherwise there would be little or no axial load bearing capacity. The spacer in between the inner races of the bearing still apply a clamping force on the inner races and an Interference Fit would not be necessary.

Regards,

Bill

 

Well setting the bearings up with end float on worn spindles isn't going to stop them from spinning and you would have a far better chance of success with a centre spacer irrespective of whether you aimed for end float or pre-load. In each case, the bearings are locked tight against the vertical link and are prevented from turning.

If the spindle is stuffed and its one piece with the vertical link, you replace the whole arrangement if there is no work around.

Regards,

Bill

 

Jesus, 100's of thousands of trouble free miles...now I have a problem I never knew I had.

 

Rough anal?

Yous guys are going off on a tangent again...

 

Many of the clients I have don't have a clue the bearings are spinning on the axles when they bring their cars in for work to be done.

As Jim stated a few Posts up, "The funny thing about those awfully designed front tapered wheel bearings with end play, is that they still work, decades later"; motor vehicles are rather forgiving devices and go for a remarkably long time being abused, or with poorly adjusted or designed components.

 

Sounds like excellent workmanship, Elcohaulic.