Lathe Spindle Bearing Preload

[Antman]

Hi Guys,
I searched HMEM for bearing preload but the search came up empty. I think I am using the right term. My Chinese 250mm X 550mm lathe is a very basic machine. On the drive end of the spindle it is threaded M27 X 1.5mm and fitted with 2 nuts (the kind you use a C spanner for ) which put pressure on the spindle pulley which presses against the 40t gear which in turn bears against a thrust collar which preloads the 2 taper roller spindle bearings. The pulley, gear and thrust collar are a sliding fit on the spindle, the pulley and gear keyed to the spindle. I suspect that many lathes have a similar arrangement.
Of course when I removed the pulley to try for a slower speed I upset the preload as delivered. Now I need to know just how tight or loose must the spindle tension be, and how do you measure it? Do you use the same preload under all conditions of speed, depth of cut, mass of workpiece, etc?
Thanking you again and hoping you don’t tire of my questions,
Ant

 

[rklopp]

On my old Jet 1024P, I'd just snug the bearings up and then monitor temperature. After running at the highest speed for ten minutes, if the bearing housings were too hot to hold a finger on for a second or two, the preload was too high. You do want some preload on these, just not too much.

 

[gbritnell]

Hi Ant,
Although there are methods to measure load on a tapered bearing for your purposes it's more seat of the pants feel. By that you naturally don't want the bearing loose but then how tight do you make it. When I pull the spindle out of my Enco mill/drill, that's the only way to adjust the bearings, I tighten up the load nut until there is a little resistance felt when turning the spindle. Now when you tighten the jamb nut in a lot of cases it changes the load on the first nut so you have to work between the 2 nuts to maintain that light load on the spindle. About the only way you can tell if you're in the ball park is by running the spindle and if it doesn't get too warm you should be good to go. The bearing will expand slightly when it gets warm so you don't want too much initial pressure on it. Like I said it's something you just have to get a feel for.
gbritnell

 

[Antman]

Thanks so much RK and George, I think that you have given me the info I need, with one proviso. Isn’t heating up what plain bearings do, not roller bearings? My lathe has a pair of substantial looking taper roller bearings. They look big enough to drive a truck on. True, I’ve only looked at the bearing on the drive side but the manual gives the same part number for both. Won’t roller bearings with too much preload suffer damage to the rollers and/or races long before they run hot, the damage from too much pressure of the rollers on the races?

George have you solved your hall effect sensor reliability issue yet?

Ant

 

[Blogwitch]

Ant,

The reason for the preload isn't anything really to do with the bearings, but with the spindle.

Lathes with preloaded tapered bearings should be allowed to warm up before use for about 10 to 20 minutes. During this time, as the bearings get warm, the heat is transferred into the spindle, and so the spindle extends by about 0.002", so effectively taking the preload off the bearings and end up with a nice smooth running spindle. It is the same reason surface grinders are warmed up before use. That is why, if you don't allow the warm up period, you can get bad surface finishes and unstable measurements.

Lathe manufacturers usually give in the manual, either a chuck 'pull off' figure, measured with a spring balance, or a degree measurement for tightening the lock nuts on the back of the spindle. Getting the preload too far out will cause rough running and almost certain damage to the bearings.

Another couple of things usually forgotten are, the preload should be released if the machine isn't going to be used for any length of time (say a month), because the rollers will eventually form indentations in the races (brinelling) if left preloaded in a standing position.
The other is that normally, tapered bearings should be oil lubricated, but if grease has to be used, the bearing should be filled no more than about 50% full. If they are overfilled with grease, they can suffer hydraulicking between the races and rollers, which normally causes the edge of the rollers to flake off.

So I would suggest you look into it a little deeper before doing a 'near enough' job on the preloading.

The attached pic is just a small part of the preload instructions for a lathe like mine.


Bogs

 

[Niels Abildgaard]

Hello Antman
Am member (it is free) of Yahho Groups for Boxford Lathes. Their files section has the official Boxford spare part list and it says break-free torque shall be two pound inches.It has a describtion how you measure.I did it according to book first time disassembling and then let lathe run at highest speed 1300 rpm for twenty minutes.Stopped it and put a finger in the spindle (no chuck) and it was close to uncomfortable to touch where the bearing inner ring must have been.
The next two times I did not care to measure but just used my finger to get the same temperature.Lathe is still turning round parts quite well.

 

[Antman]

Bogs, thanks for your patience and taking the time for a very informative answer to my question. Doing something by understanding what to do is a far more valuable experience for me, I mean it adds to the fun.
Thanks Niels I will look at the Boxford info also.
Ant

 

[rklopp]

so effectively taking the preload off the bearings and end up with a nice smooth running spindle.

Another couple of things usually forgotten are, the preload should be released if the machine isn't going to be used for any length of time (say a month), because the rollers will eventually form indentations in the races (brinelling) if left preloaded in a standing position.

Sorry, I disagree with both points. You do not want the preload to come all the way off a set of tapered roller bearings. This would be impossible in a Colchester lathe with Gamet tapered roller bearings anyway. The front two bearings are back-to-back, so you'd need an inordinate temperature increase to get enough thermal expansion to occur over such a short distance. Same with the Aciera F4 mill vertical spindle. The upper bearings are high-precision (P4) tapered roller bearings mounted nearly back to back, but with a gap and a nut to drive in some preload. Being so close together, one could not expect the preload to decrease much as the spindle warms. The preload is simply accommodated by the springiness (compliance) of the assembly.

Sorry, but the second point is nonsense. Otherwise, Colchester lathes and Aciera mills and all sorts of machines using preloaded tapered roller bearings (or ball bearings, for that matter) would be useless after removal from storage, and they are not. Do you realize what a hassle it would be to have to back off the preload on a Colchester front bearing? There is an issue called "false Brinelling" which results in indented bearing races. However, the cause is not race indentation due to long-term stresses. Instead, it is due to tiny vibrational motion of the bearing which results in fretting corrosion. The result is that the rolling elements mine indentations into the races. This notably has happened when shipping machinery via rough railroads.

 

[Blogwitch]

rklopp,

I am sure you could go on all day extolling the virtues of specialist setups for taper bearings, as I am sure I could as well.
There are so many auto anti backlash and preload setups for these types of bearings, springs, shims and the likes, that a whole book could, and most probably has been devoted to it.

If you look at the original posting, rather than just the answers, you will see why I gave such a brief description.

Antman was looking at how to set up a basic spindle and a two taper bearing setup.

I try to moderate my explanations to laymans terms rather than a load of bullsh spouted straight from a book.

Maybe I should have worded it slightly different, instead of "effectively taking the preload off the bearings" it should have read 'effectively taking most of the preload off the bearings', as a small amount is still required to keep the bearings in a stable non-skid situation, and this is where the correct manufacturers set ups come in, rather than using a rule of thumb setup. There are many of these 'rule of thumb' types, and I have used them myself on all sorts of machinery, but only after I have made sure the correct information is not available elsewhere in a reasonable amount of searching.

If people would prefer to read pages and pages of technical jargon and bullsh, then I can surely do so, but I still think people would prefer to have it laid down in easy to understand laymans wording, as I am sure not all members on here have the same technical background and understanding as yourself.

With regards to removing preload from these basic taper bearing setups when stood for any length of time. I can assure you, that it does, and will happen to this type of bearing setup. I have forgotten the number of times I have had to replace bearings or have the races reground to cure this age old problem.


Bogs

 

[Maryak]

Ok guys,

A difference of opinion has been aired. Please let's not turn this into a pi$$ing contest. I hope ant has, from all your efforts, found the answer to his question, and we remain on topic.

Best Regards
Bob

 

[Dan Rowe]

Interesting topic.
Here is a basic overview for preload for different bearing setups:
http://www.promshop.info/cataloguespdf/kbc10.pdf

This one is photos of taper bearing dammage which includes false and true brinelling:
http://www.timken.com/en-us/products/maintdiag/Documents/6347.pdf

Dan

 

[Davo J]

Hi Ant,
I think it has been mostly covered above but I will ad this.
One way to know when you have taken up the slop and start preload is, to put a dial indicator on the front of the spindle and a bar clamped in the chuck. Tighten the bearing a little, then push and pull on the bar to see how much movement you have on the dial indicator, when you have zero movement you have a light preload.

Bogs
If you did read the "pages and pages of technical jargon and bullsh" you would know that most all manufactures recommend 30% grease in taper roller bearings not 50%.

Quote from NTN Bearings web page

How much grease is too much?
For general applications a grease fill of 30% of the free space in the bearing seems to be accepted as standard by most manufacturers. Slow speed applications in harsh environments (such as agricultural machinery) can sometimes require a larger fill. High speed applications (ie. hand tools) can require a lesser fill. Check with NTN engineering for any special requirements. http://www.ntnamerica.com/faq.htm

Dave

 

[Dan Rowe]

Dave,
To be fair the quote said "most manufacturers" so that does not mean Bogs is wrong stating 50%. Also unless the instructions state how much grease to add by weight or volume the calculations of available space will most likely be a seat of the pants call so it is unlikely that two different workers will add the same amount of grease no matter which set of instructions are used.

Dan

 

[Blogwitch]

Again, people not reading posts in the correct context.

How did it get from this,

If people would prefer to read pages and pages of technical jargon and bullsh, then I can surely do so, but I still think people would prefer to have it laid down in easy to understand laymans wording

To this,

Bogs
If you did read the "pages and pages of technical jargon and bullsh" you would know that most all manufactures recommend 30% grease in taper roller bearings not 50%.

How did it get from how I write my posts, to a direct question about me reading about procedures?


Bogs

 

[Stan]

I find the comments on percent fill in tapered bearing quite interesting. Since the invention of the automobile, tapered bearings have been used in front wheels. Originally hand packed to 100% fill with tough, stingy long fiber wheel bearing grease, technology advanced to power packers using a more fluid wheel bearing grease which also fills the bearing 100%.

This 100% fill probably did damage the bearing and shorten their useful life but since most of them outlasted the life of the car (or truck) it was a moot point. Especially on large trucks, it is standard practice to check the hub temperature, after a few miles of driving, after the bearing nut has been moved, for any reason. Warm is good, hot is too tight.

 

[Ned Ludd]

Hi Davo,
For what it's worth, my Colchester handbook says adjust play till there is one tenth of a thou play endfloat. Still, I suppose to some people that would be Zero.
Ned

 

[Antman]

Thanks for all your replies and interesting background information. I now have the lathe’s preload set up so it only gets warm after 20 min at full tilt and I have no movement of the spindle except rotation. And I know what to do when I juggle the pulleys again.

Thanks also to Bogs and RK for the entertainment stickpoke ,and there was me thinking the bearings were half to 70% empty to save a couple of yuan.