Why Sealed Bearings Fail Prematurely

[SeniorCitizen]

If we look closely at the picture we can we see that the spacer is proud of the outer race seat. The reason we know the spacer is longer than the distance between the lower and upper bearing bosses is because the spacer is resting on the lower bearing inner race that I pressed in until the outer race was stopped by the bearing boss ( seat ). The amount it is proud really doesn't matter unless it protrudes less than about 0.001" but you'll never see that . The one pictured is more near the norm and in fact it protrudes above the bearing boss 0.060" . Could it be shorter by 0.060" ? It could be and that would be just as bad, but you'll never see one shorter.

So what's the problem Senior Citizen ? It must be correct because it's factory.

This is the problem.

When the factory engineer BUBBA made this one he designed it for both bearings to be pressed in ( not true in all instances ). Now the assembly man/woman, we'll call her BUBBA too, presses these bearings in, the first one goes in by ( hopefully ) pressing on the outer race with that BIG OLE Press and boy that puppy is fast. That thing sure socked it toit. Next step is to place the spacer in there followed by the second bearing and press operation.

Now here is where the problem begins. When the inner bearing race just reaches the spacer did BUBBA stop pressing. No he didn't because nothing feels solid at this point and BUBBA engineer said until it feels solid. He continues to press until the outer race reaches its seat and all feels good and firm. By continuing to press he has put an u-due lateral pre-load on both bearings. In the case of the pictured assembly that pre-load could be as much as 0.060" on one bearing or some on it and the remainder on the lower bearing, depending whether the first bearing installed outer race moved. The end result is a shortened bearing life. Ball bearings excel with radial thrust but preform poorly with lateral thrust.

This is all the same whether Bubba did it at the factory or you/we do it at our shop with a big hammer and big socket to drive the bearing in.

So what's the solution.

Some manufacturers have been building so one of the bearings is a push fit. When the press fit bearing is installed first until it seats, the spacer is installed then the second push fit bearing with our fingers / thumbs is installed, that bearing is free floating and that solves the problem of lateral thrust.

When I find this problem, rather than attempting to bore the housing larger by a thousandths or two, modifying the outer race diameter until the push fit requirement is satisfied takes only a few seconds.

 

[gfp55]

What brand did that come off of? It looks like a deck spindle, is it? Tell me more about part that is in the picture.

 

[ILENGINE]

Its a spindle from a Murray mower.

 

[SeniorCitizen]

Its a spindle from a Murray mower.

ILENGINE is absolutely correct. Bearing and spindle housing from a 1980's something 11/36 Murray lawn tractor.

The brand isn't of much consequence because most all use this configuration in some way if they incorporate sealed bearings. My personal experience is only with Murray, John Deere and Ranch King ( MTD ). I have two old Craftsman table saws from the mid 1940's and they have the same set up so this system has been around nearly as long as I have.

Pictured in the box is parts of a JD LA-135 and the second pic is the old Craftsman table saw parts. Although a different array of parts, in the end they all do the same job of causing the inner races of the two bearings to compress on the two ends of the spacer sleeve. And the point of this thread is to show how the balls could be pulled from the bearing if the outer race of one bearing isn't a push fit and can float and go with the flow so to speak. The severity depends. This damage can be during assembly or when the blade nut is torqued if one bearing is not allowed to float.

On the john deere the blade adapter on the lower end and the piece just below the pulley on the top end compress the inner races to the sleeve.

On the table saw shaft on the blade end, the raised face is incorporated to do its thing on that side and the V pulley on the opposite end does its handy work.

In all instances the same thing is accomplished, but each manufactures has done it in different ways and even Murray has changed over the years.

 

[gfp55]

So what way would you go about designing an assy that would work and last longer? What style bearings would you use?

 

[Carscw]

The bearings in the table saw pic are not sealed bearings.

I have study bearings and spindles to get them to last longer.

I have spent hours at dalton bearing talking about bearings.

The problem with the floating bearing is it will wear done the ridge it sits on. That's way when you have wobble and change the bearings you still have wobble.

If you have ever used a industrial mower you would see they do not use a seals bearing. They use a roller bearing like on the front axle of a car.

The number one cause of sealed bearing failure is lack of grease.

I repack my bearings once a month. For a home owner that would be every 2 years. Or every time you pressure wash the deck.

 

[SeniorCitizen]

So what way would you go about designing an assy that would work and last longer? What style bearings would you use?

From post #1

QUOTE: So what's the solution.

Some manufacturers have been building so one of the bearings is a push fit. When the press fit bearing is installed first until it seats, the spacer is installed then the second push fit bearing with our fingers / thumbs is installed, that bearing is free floating and that solves the problem of lateral thrust.

When I find this problem, rather than attempting to bore the housing larger by a thousandths or two, modifying the outer race diameter until the push fit requirement is satisfied takes only a few seconds.
********************************************************************
There isn't anything wrong with the sealed bearings if they are good quality and not damaged during installation. Sure there are better non sealed- grease-able but not for the price being paid for bottom of the barrel lawn tractors. But even some of the high dollar tractors even have sealed bearings with the same basic set up. The purpose of the thread was to discuss why sometimes sealed bearings fail prematurely.

The LA-135 housing allows the top bearing to be pushed in by hand - impossible to damage the bearing unless I be an idiot.

The Craftsman saw bearings are both push ( with the fingers ) fit. No pressing , hammering or driving.

The Murray pictured have both bearings a press fit by a thousandth or two. I press or drive the bottom bearing in and grind 0.001"-0.002" off the outer race of the top bearing to get the push fit.

 

[gfp55]

These spindles are rebuildable with the same bearing used to rebuild trailer axles, very cheap $11.00 per spindle. The races are just taped in like you do front wheel bearing or an older car/truck. They have grease zerks so I can keep the dirt and water out by adding grease about every 50-100 hours. Most of the stuff you get today are throwaway unless you install your own zerks in the aluminum spindle housing. I have never tried to rebuild them because the shaft or housing was damaged and it just isn't worth my time to do it. Cheap junk. I do know that adding a grease zerk makes the cheap made junk not cheap priced spindles last much longer.
The design that Senior Citizen is talking about is a poor design unless you do what he does to fix and overcome it short comings, but most folks don't see the flaws in the design and think thats how long the bearings are supposed to last before repairing or replacing the spindle.

 

[SeniorCitizen]

Here is one that ran dry from the lack of grease within 2 hours of operation. It was a case of a mower repair shop failing to replace the spacer during rebuild. When the blade nut was tightened the same scenario happens as if a press was used to install both bearings, undue lateral thrust at its maximum.

 

[gfp55]

Here is one that ran dry from the lack of grease within 2 hours of operation. It was a case of a mower repair shop failing to replace the spacer during rebuild. When the blade nut was tightened the same scenario happens as if a press was used to install both bearings, undue lateral thrust at its maximum.

I may be wrong and tell me if I am, but a lot of the bearing failures are due to lack of knowledge of the person repairing the spindle and poor design front the start. My belief is that manufacturers design products so they last just so long. Design obsolescence. I believe this is used for profit as opposed to product improvement.

The flaw in the design that you showed is the need for the spacer. If the shaft was designed with a ridge instead of the spacer like the shaft in my picture it would fix the problem with preload, thrust, and all the stuff you said about how the bearing fit in the spindle housing so that is no need for all the stresses and forces applied to make the spindle assy. Aren't modern design machines grand?