PDA

View Full Version : R & R Spindle Bearings and Cleaning



JoeLee
05-14-2011, 08:38 PM
I spent part of the day tearing down this spindle as it was getting noisy due to dry bearings. My cleaning method is as follows........ After removing the spindle from the cartridge and taking careful notes as to what I find inside, bearing orientation, shims, spring washers etc. etc. I soak the bearings in parts washer fluid to remove the old caked up grease. This make take several hours of soaking and blowing out the old stuff. When they look about as clean as they can get I then put them in the vibrator soaking in laquer thinner and let them vibrate for about 15 min. I do this three of four times or untill the thinner remains clear. In the first picture you can see how the thinner is yellow as it removes the grease film that the solvent wouldn't touch and also sifts out some small particles. When I'm satisfied that they are clean I then procede to the final cleaning step.
Soaking and vibrating in alcohol. This step removes any film left by the thinner and the bearing are actually squeeky clean. then I pack them with grease and they are ready to install.

JL....................

http://i911.photobucket.com/albums/ac317/JoeLee09/Grinder%20Spindle%20Repair/Image021.jpg
http://i911.photobucket.com/albums/ac317/JoeLee09/Grinder%20Spindle%20Repair/Image022.jpg
http://i911.photobucket.com/albums/ac317/JoeLee09/Grinder%20Spindle%20Repair/Image023.jpg

JoeLee
05-14-2011, 08:42 PM
I should have posted these pictures first but..................

Here is a picture of the quill after I pulled it apart.
I also indicate the spindle to insure proper bearing orientation.

JL..................

http://i911.photobucket.com/albums/ac317/JoeLee09/Grinder%20Spindle%20Repair/Image007.jpg
http://i911.photobucket.com/albums/ac317/JoeLee09/Grinder%20Spindle%20Repair/Image012.jpg
http://i911.photobucket.com/albums/ac317/JoeLee09/Grinder%20Spindle%20Repair/Image014.jpg

Carld
05-14-2011, 09:52 PM
Are the bearings sealed on both sides?

macona
05-14-2011, 11:08 PM
Pretty much the bearings are shot. Once you remove them from a spindle its all over.

JoeLee
05-14-2011, 11:30 PM
No the bearings are not sealed, the end caps of the spindle are the dust shields, and why do you say once you remove the bearings they are shot????? thats not true. I think some people have this misconception about precision spindle bearings that if you look at them the wrong way you ruin them. I've done several of these with excellent results.

JL...................

macona
05-15-2011, 01:20 AM
Its pretty well known once you take them out they will never be the same again. I have tried it, much as you have, and it has never worked in the long run. Once a set of bearings is installed and ran they take a set a wear in during run in and operation. Unless you are very, very careful and get them back in place exactly how they were in relation to the spindle housing and the spindle they will not run.

Check on ebay, you can often find NOS bearing sets for pretty cheap, sometime under $100 a pair.

Forrest Addy
05-15-2011, 03:59 AM
I've re-installed spindle bearings. Many times a spindle has to be disassembled for non-bearing problems (leaky seals, gears, clutches, or shifters, access, whatever.) Some spindle bearings cost a fortune ($30K a pair to take the most expensive in my experience.) It's criminal waste to discard them without reason.

Naturally, they have to be pulled with care and sometimes this requires a specially made puller. The outer race generally has a small clearance so it will come right out with a little pursasion and maybe heat. If you get purchase on the inner race and pull without loading the outer race the bearing for all practical purposes is still in used but unabused condition. Taper roller bearings have comparable requirements but its often very tricky to extract the outer races without damage inless there are extraction reliefs in the bore shoulder; fortunately if there is no problem there's no reason to.

Then there is cleanliness, grease (if used), race orientation, reassembly order, etc. Touch all the bases and use good technique and spindle bearings can go in and out any number of times. There is surgical risk to consider but if everyone is on his game, there should be no problem.

BTW, what do you call a "vibrator?" The gizmo in the picture looks like an ultrasonic cleaner, a deadly enemy of assembled precision bearings because of the zillions of high energy impacts between balls and races.

DougA
05-15-2011, 07:07 AM
Unless the bearings are worth a fortune and would replace with new. Looks like a lot of work taking the spindle apart just to put a old set of bearings in that were running noisey/dry. That sound is the sound of the bearings being worn.

tyrone shewlaces
05-15-2011, 08:29 AM
Here's a thought....

A guy could always try cleaning and re-lubing bearings first. If it works, it could save hundreds of dollars. If it doesn't, then you could always pay for new ones after determining it needs to be done.

Of course if you've got money to burn, then by all means throw as much money as you want to at it. Otherwise, all you've lost is a little time. Unless you're in a production machine shop where lost machine time offsets the cost of precision bearings, then a second round of removing and installing new bearings is worth the learning experience of doing it the first time.

Just my opinion. But I'm right, so...:D

Forrest Addy
05-15-2011, 09:51 AM
People talk about bearings bein "worng." End of life for a rolling elemenrt bearing is failure of the rolling path in the races and/or the rolling elements.

R/E bearings do not suffer failure from loss of material due to mechanical wear. A preloaded bearing will not get "sloppy". The preload may relax somewhat but if the related parts are still in their orginal adjustments the bearing's axis and smoothness of operation will last for some time.

I was going to write one of my famous "things" but while looking up a tedious point I found this which says it better and with pictures:

http://www.vibanalysis.co.uk/technical/contents.html

Carld
05-15-2011, 10:03 AM
JoeLee, that black ring between the inner and outer races looks like a black rubber seal to me. If the bearing has no seals I should see retainers and balls showing in the photo. It appears to me they have a seal on at least one side.

It is almost impossible to get all the grease and any dirt out of a bearing with a seal on one side. I have cleaned, greased and reused bearings before and sometimes it worked and sometimes it didn't. If they were common run of the mill bearings I would clean and reuse them but bearings for a mill spindle or tool grinder would probably get replaced. Unless money was tight or I couldn't find them.

Many times when the grease dries out in the bearing the balls and races get damage from lack of lube so cleaning and re-greasing them is not much help. Especially in high speed bearings.

moe1942
05-15-2011, 12:06 PM
I have reused bearings of all kinds. The USAF reuses wheel bearings on high speed jets. After a good cleaning and inspection lube and they are ready for reuse. One sure way to trash a bearing is to spin a dry bearing with air..

Also they should never be packed completely full of grease. About 3/4 is enough.

lakeside53
05-15-2011, 12:07 PM
On precsion bearings, the bearing manf. will give you the exact CC of grease allowed for a particular IIRC, my surface grinderBarden's are about 40% fill.


Following on from Forrest's post - to look between the balls in the races, I use an Otoscope (one of the thingys the doctor looks inside your ears etc).

Now they are very cheap (non-medical) and use LED light sources. Mine was about $14. Very useful for a host of machining inpsections, including finding metal slivers in fingers.

Mine looks something like this :http://cgi.ebay.com/LIFETIME-Warranty-Dr-Mom-Stainless-LED-Otoscope-/110672920204?pt=LH_DefaultDomain_0&hash=item19c49ebe8c

JoeLee
05-15-2011, 01:31 PM
Carld, the black ring your reffering to is actually brown until it's coated with grease, it's the phenolic bearing cage or seperator or what ever you whant to call it, it does look as if it were a seal in the picture but there is a slight space where when the bearing is cleaned you can see the balls. A visual inspection of the race does tell a lot, if a matt finish is detected where the ball rolls then I would say that the bearing has some wear and should be replaced. If all is bright and mirror like with no detectable tracking and all else checks out then I would call it reusable. Back when I did my first spindle some 15 years ago I was unsure about R & R,ing such precision bearings. I called Fafnir and spoke to one of the engineers about doing this in my grinder spindle and he said as long as your careful taking the bearing out there shouldn't be a problem. He told me as long as I don't beat them off with a hammer or try to press / pull them off with a 1000 lbs. of force I would be OK. For a second opinion I called KO Lee and spoke to one of thier engineers, he said hell we remove new bearings all the time if were not satisfied with the run out. We pull them off and slightly reposition them on the shaft and check the runout again, and do this until satisfied. Altough the spindle is checked for high point and properly aligned to the marking on the bearing doesn't always mean the results will be perfect.
If you can hold tolerance when grinding and the finish looks good then what else could you ask for??
When you remove these bearing just make note of the markings on the races in relation to the housing and spindle and don't switch left and right bearings around either. Worse case is you buy new bearings.

JL....................

J Tiers
05-15-2011, 06:35 PM
The difference would be whether you can remove each race purely by pressure on THAT race.........alone.

If you can, then what is wrong?

If you cannot, then you will be, for instance, removing an inner race by pulling on the outer race, which will risk brinnelling the race by denting it. That's if the pressure is more than the allowable load for the bearing (static). Even if not, it may remove some life by adding a very heavy "cycle" to the fatigue stress.

The mere act of removing a bearing, properly, should not affect it.

Forrest Addy
05-15-2011, 06:58 PM
I just now thunk.

Back in the day, I flushed out bearings with a "solvent lance" a soft drink carbonator pump drawing from a solvent source hosed to a blunted 20 ga hypo needle.

You need a face shield or a booth. 100 PSI solvent splashes EVERYWHERE but it really cleans bearings - even ones shielded on one side. I've thought a small, cheapo, blow molded plastic blast booth would be a good idea.

A foot swich on the pump is a good safety thing.

Carld
05-15-2011, 08:20 PM
Ok JoeLee, I can go along with you after your explanation. I have reused bearing too with good results many times but it's a gamble.

JoeLee
05-15-2011, 09:15 PM
Ok JoeLee, I can go along with you after your explanation. I have reused bearing too with good results many times but it's a gamble.

Yes Carld it is a gamble as is everything else we do with our machines in the shop, but at worst if the results are less than desirable then you just replace the bearings, at least you have figured out how to disassemble the spindle in the process so if you have to put new ones in there is no guess work. The bearings in my spindle will take at least a hundred pounds of static load without distortion or brinnelling of the race etc. they are hardened steel not aluminum.


JL.......................

J Tiers
05-15-2011, 11:00 PM
The bearings in my spindle will take at least a hundred pounds of static load without distortion or brinnelling of the race etc. they are hardened steel not aluminum.


JL.......................


Likely it is a lot more than that which they are capable of, but we don't know the size and so forth.......

The issue with brinnelling is usually not with a static press load, , although it easily can be, but rather with pounding for removal. Pounding is an impact that can put very large instantaneous forces on the race through one or two balls alone.....

JoeLee
05-15-2011, 11:46 PM
Likely it is a lot more than that which they are capable of, but we don't know the size and so forth.......

The issue with brinnelling is usually not with a static press load, , although it easily can be, but rather with pounding for removal. Pounding is an impact that can put very large instantaneous forces on the race through one or two balls alone.....

I'm sure it is a lot more than 100 Lbs. The balls in this bearing are about 3/16" in dia. I haven't counted how many there are. But the size of the ball really doesn't matter because no matter the size it's still point contact, the number of balls lessens the force to each. How many times have you seen someone crash a machine, either a mill or grinder???? Recently watched something someone posted here on a CNC table running up into the spindle, I wonder if they replaced the bearings in that one???
I saw a guy run a wheel into the edge of a block of steel on a grinder once, it took a chunck out of the side of that wheel and the whole head vibrated wildly due to the out of balance conditions until someone hit the power switch as every one scattered when wheel pieces went flying. They put a new wheel on it and it worked just fine. Tough bearings!

JL....................

J Tiers
05-16-2011, 12:10 AM
bearings are tough, abuse reduces precision, adds small bumpiness, reduces life by "using up" some fatigue life, maybe nobody cares.....

Anyone who clobbers precision bearings with a hammer and punch has an issue anyway....... that might work for some auto wheel bearings.

We're "just sayin"....... it won't help, but I don't see the need to replace a removed bearing that is not whanged-on.

This is one case where the "we are not NASA here" is not an insulting condescension, but a reality.

philbur
05-16-2011, 06:32 AM
100lbs on a point contact equals an infinite stress. the Modulus of elasticity allows deformation which results in a contact area. The larger the ball diameter the less load is required to produce the same contact area, resulting in a reduction in stress for a given load.

Phil:)


But the size of the ball really doesn't matter because no matter the size it's still point contact, the number of balls lessens the force to each.
JL....................

moe1942
05-16-2011, 08:47 AM
Yes Carld it is a gamble as is everything else we do with our machines in the shop, but at worst if the results are less than desirable then you just replace the bearings, at least you have figured out how to disassemble the spindle in the process so if you have to put new ones in there is no guess work. The bearings in my spindle will take at least a hundred pounds of static load without distortion or brinnelling of the race etc. they are hardened steel not aluminum.


JL.......................




The Air Force didn't consider it a gamble when we reused bearings for the Nose and main wheels on an F-4 Phantom weighing in excess of 54K empty. When loaded with a max load of ordnance it was real heavy.

If bearings are properly removed, inspected, lubed and reinstalled there is absolutely no problem. The bearing vendors don't like it..

Of course if one doesn't know how to properly remove and install bearings even new bearings won't work.

philbur
05-16-2011, 11:12 AM
I'm not sure that comparing a wheel bearing with a precision spindle bearing is particularly relevant.

Phil:)


The Air Force didn't consider it a gamble when we reused bearings for the Nose and main wheels on an F-4 Phantom weighing in excess of 54K empty. When loaded with a max load of ordnance it was real heavy.

jimmstruk
05-16-2011, 11:31 AM
Comparing a wheel bearing to a spindle bearing is not relevant? Not at all, the wheel bearing supports our family in an automobile weighing twice as much as a milling machine, traveling at 70 MPH on a road filled with many other people. No big deal. JIM

philbur
05-16-2011, 11:43 AM
You possibly missed the word "precision".

Phil:)


Comparing a wheel bearing to a spindle bearing is not relevant? Not at all, the wheel bearing supports our family in an automobile weighing twice as much as a milling machine, traveling at 70 MPH on a road filled with many other people. No big deal. JIM

moe1942
05-16-2011, 03:33 PM
Comparing a wheel bearing to a spindle bearing is not relevant? Not at all, the wheel bearing supports our family in an automobile weighing twice as much as a milling machine, traveling at 70 MPH on a road filled with many other people. No big deal. JIM



They were precision and high quality.. Cost a hell of a lot more than your spindle bearings.. Can't afford a bearing failure on TO or landing at 220kts..

The point I was trying to make is reuse spindle bearings if they pass inspection.. There are very good inspection check lists with pictures available.

But those that don't know what to look for should buy new. Help the economy.

macona
05-16-2011, 07:28 PM
Pass inspection, I doubt if you truly have the tools to inspect a bearing. No one here does.

Bearing in the wheels of a jet or car do not equate to the bearings in a grinding or milling spindle. These bearings are expected to have run out in the range of tens of millionths of an inch. No one cares in a jet or car if you have .001 of run out as you would never notice it.

JoeLee
05-22-2011, 09:47 AM
Update on my findings inside the quill revealed the (X) marks at the bearing ends of the spindle which are supposed to indicate either high point or low point, and I,ve seen it both ways. I guess it depends on the person doing the inspection. Anyway on this spindle the (X) is the high point. the burnish mark on the inner bearing race that indicates the high point was installed at the (X) on each end of this spindle where the (X) is the high point. I guess that is why I indicated .0003 in run out on either end of this spindle before I took it apart. That was a bit more than other spindles I've worked on. The high point of the bearing should be installed at the low point of the bearing surface on the shaft, KO Lee got it wrong, but that doesn't surprise me, they get a lot of stuff wrong. From my findings the bearings were installed about 90 degrees off from that point. I'll go with my mark and see how I end up. I should be able to end up with .0001 or better s I Have with other spindles of this design in the past. KO Lee sucks !

JL.................

http://i911.photobucket.com/albums/ac317/JoeLee09/Grinder%20Spindle%20Repair/Image026.jpg
http://i911.photobucket.com/albums/ac317/JoeLee09/Grinder%20Spindle%20Repair/Image029.jpg

J Tiers
05-22-2011, 10:40 AM
OK, you found what you think is the "high" or maybe the "low" point.......

Now the question is...... "relative to WHAT?"

it looks like you have the spindle on centers....... and it seems to be a grinder spindle, with no "socket". If it HAD a socket, centers would of course be wrong......

But, it seems that you might need to compare the bearing area to the wheel seating area, and it is not clear you have done that.

if your spindle had a taper socket, the reference would be the socket, assuming you had no plans to re-grind it in place.

IF everything else is perfect, THEN putting the low and high spots together would be right, as that would put the center of the spindle closest to center of rotation.... but it might not put the wheel seat centered...

What's the chance the bearings were located to put the wheel seat area closest to concentric?

That might not be least for vibration of the spindle..... but if putting the bearings right vs spindle were done, there could be some vibration due to wheel mass being slightly off center, so.....

JoeLee
05-22-2011, 12:04 PM
What I found was the low point relative to the spindle center. The side shown in the picture has an internal tapered socket, the other side of the spindel (pulley side) has a center mark. I don't plan on grinding the internal taper, altough some people might to get even closer. As far as comparing the internal taper to the bearing surfacce area, the spindle would have to be set on V blocks off the bearing surfaces and checked that way against the internal taper. That is harder to because if the spindle slides one way or the other then the position of the dial point will be changing, thus making it difficult to be accurate. From what teh engineer at KO told me several years ago this is the way they did it.

JL................

JoeLee
05-22-2011, 03:13 PM
J Tiers, now you have me thinking................ ?????? The other spindles I've done were all external tapers, this one is internal. If I check the the internal taper against the bearing surfaces don't you need 60 degree V blocks to do that?????? 45 degree ones won't give an accurate reading.

JL....................

doctor demo
05-22-2011, 03:21 PM
As far as comparing the internal taper to the bearing surfacce area, the spindle would have to be set on V blocks off the bearing surfaces and checked that way against the internal taper. That is harder to because if the spindle slides one way or the other then the position of the dial point will be changing, thus making it difficult to be accurate. From what teh engineer at KO told me several years ago this is the way they did it.

JL................

The easiest of all the ways to check a situation like You describe is to put a ball in the center hole of the shaft and butt it against an angle plate or stop of some kind so the shaft doesn't move while measuring the tapered bore .

Steve

JoeLee
05-25-2011, 09:23 PM
Well, here is what I found out today. I talked to an engineer at Barden Bearings, he said the proper way to indicate this spindle is just as I've done, it doesn't matter weather the taper is on the inside or the out side of the spindle. Looking back at my notes from the other KO Lee spindles I've done in the past, and all have been outside tapers on one end the (X) has always been the low point of eccentricity and the burnish mark of the inner bearing race has always been installed at that mark. On this spindle the (X) is the high point of eccentricity and the burnish mark on the bearing was installed at this mark. I have come to the conclusion that the (X) mark on this spindle is marked incorrectly. Looking at my sketch of both ends of this spindle is what I find as the low point of eccentricity. Both ends of this spindle are off by about 90 degrees. I've double checked and rechecked and flipped the spindle around and checked again and my mark of the low point repeats every time. I think who ever was doing the inspecting at KO Lee botched this spindle. I'll post an update on the results after I assymble this thing.

JL.....................

http://i911.photobucket.com/albums/ac317/JoeLee09/Grinder%20Spindle%20Repair/Image030.jpg

J Tiers
05-25-2011, 09:49 PM
Out of curiosity.......

You seem to have it on centers, but if it has a socket, what is the center in at that end?

If your centers are true, that will put the low or high relative to the center holes, but with the socket, is the center in the socket? Is there something in the socket that has a center hole?

If you want the socket to run true, maybe you SHOULD check it instead of just the OD vs centers..... If you will re-grind it, as is often done, then you would get the spindle the best, and then get the socket to match by grinding.

But to get the existing socket the best, high and low may be with respect to it. Might be worth a check to see if that could be the case. It may be that there is a pretty good reason why the mark is where it is.

The alternative is that the folks at the OEM were not even as good at this as we are, which I would suspect isn't exactly true on average.

I am always very suspicious when it starts looking like "the idjits who built this did it all wrong". Much of the time that really means that I simply don't understand what they were doing, and some more investigation shows they were right. Or it may show that some idjit messed it up before I got to it.

JoeLee
05-25-2011, 10:21 PM
Mr. Tiers, thanks for staying with me on this..........this is the first time a spindle has got me thinking this much. The left end of the spindle is a socket, I guess it coiuld be called that. the hole at the end of the spindle is about .500 and the internal length runs about 2.5" I don't know what the taper is off hand. There really is no other way to indicate the bearing surfaces as was confirmed by the engineer I spoke to at Barden, it has to be done on centers. The pulley side of the spindle has a small center drill spot and the arbor end has the internal ground taper, it's just a bigger hole. There is no way to mount the spindle on centers and at the same tiime indicate the the inside taper. That can only be done when assembled. I did check for run out before I pulled this apart and I had .0002 on the inside of the taper and .0005 on the pulley end, I can live with the .0002 but wasn't crazy about the .0005. My surface grinder dialed in at .0001 on either end and they are both the same spindle and quill with the one exception being the inside taper. Worst case senario if I end up worse the before I just unscrew the spanner nuts and reposition the bearing back to factory mark. I also become suspicious when I start thinking that the idiots at the factory may have messed up and that is why I've taken this investigation as far as I have. You just don't find too many spindle people out there to question. If you could see some of the stuff I've seen when I've pulled these machines apart you gotta wonder what the hell were they thinking when they made this. I can go on and on about them, and I've had the good fortune to have spoken to a couple people that worked at the KO plant and they told me that there was a time when....... if they could get paint to stick to the machine it went out the door, and they had school kids making machining parts. So thats why I have to question the mfg. in this case.

JL........................

J Tiers
05-25-2011, 10:45 PM
Well, if there is good reason to doubt them............ then maybe.

Thing of it is, when you get to the "high spot" stuff, is when you DO want the 0.0001 runout..... you don't need it for a half thou or more.... So ......

If the socket runout was 0.0002, you are in the territory, and combining runouts makes a difference......... The runout at the tail presumably may affect pulley vibration, dunno what to tell you there.

So how are you holding the socket end? I'm not clear on whether there is a "plug" in it or what. I don't generally trust the mouth of a taper, even though it might fit a center......Unless it has an obvious centering surface cut on it. And even then they get dinged and distorted.

JoeLee
05-26-2011, 12:01 AM
The socket end is being held at the mouth of the taper which is ground, no plug, as I would have to make one. If I could make a master plug that I could bet my life on it being perfect then that would be the way to go but in doing so I may crearte more error than exists. I can't use the arbor that goes into the socket as it's drawn in the spindle socket by a long SHCS and the threaded end is silver soldered so the bolt can't come out of the arbor.

JL.........................

JoeLee
06-03-2011, 12:44 PM
Just finished up this spindle yesterday. Here is what I ended up with going by my indication of low point of eccentricity of the spindle.

Taper end .00015 and pulley end .0002
Before I took this apart the run out was .0002 on the taper side and .0003 on teh pulley side. I think I did better than KO Lee. I knew they messed up. It wasn't a lot of difference but I didn't expect much more since were talking tenths, but it's still better than the factoy specs call for.

JL........................