How Should These Spindle Bearings Be Lubricated? Also, What Bearings to Buy?

[ChandlerJPerry]

Some of you may have seen my unicorn milling machine floating around the site. Herter branded knee mill, made in Japan. Previously owned by an unknown individual who will be named "Blockhead" in all future references due to some repairs and maintenance choices I've found on this machine.


In a fit of boredom tonight I decided to remove the spindle and inspect it due to the general filth level of the machine when I got it, suspecting that it would be similar. The spindle has run quietly, and I put an indicator on it once but I couldn't tell you what the runout was, which must mean it wasn't enough for me to worry about it. Either way, I'm going to replace the bearings just for peace of mind.

The spindle/quill is dead easy to remove on this machine. After you remove the pointer for the quill depth scale and unbolt the quill lever assembly from the side of the head, the whole thing drops out as a unit.

Anyway, I found the spindle bearings to be packed full of old grease. It was still providing lubrication but struck me as odd, since I know spindle bearings usually run in oil.

So here's the question, grease or oil? I have no way of getting a service manual for this machine or know if anyone has been in here before me. Although I do suspect it based on some tool marks etc I noticed while taking it apart.

The only oiling hole on the head is this one on the side, located above the quill lever and mechanism:


Here is a view of it from the inside of the head with the quill/spindle removed:


With the quill installed, the side of it actually covers this oil port, and the quill has no oil passages itself. This leads me to think that this oil passages is just for lubricating the sliding of the quill against the bore. I don't see any way to get oil into the spindle bearings reliably.

Also, the bearings are completely open with no felt etc to trap oil. Any oil that gets to them would simply run out through the bottom of the spindle onto the work since there are no sealing measures.



With all that in mind, I'm inclined to say that grease is actually the best choice for these bearings, as strange as that seems. Although I am interested to hear what others think that have seen the inside of other machines, this is the only mill I've ever taken apart. And what grease to use for that matter if it is the best choice?

I am also wondering what I should be looking for in replacement bearings. The bearings are 7307B and 7208B angular contact bearings, as well as a 6208 ball bearing in the middle of the quill, seemingly just for extra spindle support. NSK brand. I don't see any additional markings on them denoting precision etc.

 

[SouthernChap]

I could be wrong about this, but ball/roller bearings are generally lubricated with grease as they are often somewhere reasonably well protected from foreign object (in this case, swarf, chips, etc.) ingress or are sealed or both. They're not expected to need frequent lubrication.

If you can find quality, dimensionally equivalent, sealed angular contact bearings, I'd say those would be the best choice for replacements for the angular contact bearings (sealed single row deep groove ball bearing for the middle bearing too); install them and forget them.

 

[tailstock4]

To answer your question, I would say grease. I would also recommend using Kluber NBU 15. This grease is expensive but has some desirable characteristics in applications like this. The only downsides are price and compatibility with other lubricants. By the way, I usually order not only enough for what I’m working on but also keep some on hand as it is a good solution for problematic lubrication situations you may encounter in the shop.

You only need about 30% fill on the bearing. Over greasing can cause heating.

Interesting that the bearings themselves are single row, angular contact bearings. They are probably ABEC 7 or 5. I try to use the higher precision bearings as you gain a little tighter tolerance and a little smoothness. The contact angle on these bearings is 40 degrees. This is unusual. Most milling machines will have 15-degree contact angle with an inner and outer spacers to regulate preload and to also achieve stiffness. The 40-degree contact angle can take a much higher axial load.

One machine that I’ve worked on that had this type of spindle setup was a Cincinnati Bickford Super Service drill. However, instead of having a center ball bearing it had a long, bronze bushing to stabilize the center of the spindle. The bearings themselves were angular contact ABEC 7 P4 bearings with 40-degree contact angle. This set up can handle a higher axial load that would be encountered in drilling, but it gives up some higher speed capability. The preload was adjusted by a single nut tightened enough to remove most of the deflection but not so tight so that it developed heat. Heat in a quill will cause it to expand and can cause stiffness or even seizure of the quill if it is close enough tolerance.

I would probably go with NSK, a good quality Japanese bearing. These days you want to pay attention to a good brand and order from a reputable source. I often use EZ Bearing and talk to Darrel. Motion Industry would also work.

Good luck

 

[ChandlerJPerry]

One machine that I’ve worked on that had this type of spindle setup was a Cincinnati Bickford Super Service drill. However, instead of having a center ball bearing it had a long, bronze bushing to stabilize the center of the spindle. The bearings themselves were angular contact ABEC 7 P4 bearings with 40-degree contact angle. This set up can handle a higher axial load that would be encountered in drilling, but it gives up some higher speed capability. The preload was adjusted by a single nut tightened enough to remove most of the deflection but not so tight so that it developed heat. Heat in a quill will cause it to expand and can cause stiffness or even seizure of the quill if it is close enough tolerance.

This sounds exactly like how this is setup. Both of the angular contact bearings are housed in each end of the quill with their thrust directions pointed towards each other. The center ball bearing is located maybe half an inch above the lower angular bearing. There is a threaded section on the spindle with a single nut to adjust preload on the angular contact bearing inner races, and a flat tabbed washer to be bent that retains the nut in position. The high speed compromise seems reasonable given that this mill tops out at about 2,000 RPM on the spindle.

Based on your comments it seems that the 40 degree angle of the bearings may actually be a superior choice in terms of the axial load bearing ability of the spindle? I am curious about why they would be spec'd over a more "typical" spindle configuration like you described.

 

[FOMOGO]

Those bearings are not inexpensive, are you sure they need to be replaced? Have you checked the run out/adjusted? Mike

 

[Ischgl99]

That's a really strange setup, I don't see the point of having a radial ball bearing in-between the two angular contact bearings. Is it possible the previous owner installed them incorrectly? Most mills have the ACBs mounted back to back with a short spacer between them, and then a radial ball bearing at the top of the quill for radial support of the top of the spindle. For the designs with an ACB at the spindle nose and the other end of the quill, I have never seen another bearing in between. I would expect to have 2x 7208 bearings near the nose and then a 6207 at the top. Do you have a picture of the spindle?

A 40 degree contact angle makes sense for a machine primarily used for drilling, but for a milling machine, something in the 15-20 degree angle is preferred since it gives you more radial rigidity needed for cutting tools like end mills. I'm thinking the previous owner replaced the original spindle bearings with standard ACBs since they are cheap, if they were precision class, they would be marked that way. NSK always marks their precision class bearings, the way yours are marked indicates they are off the shelf general purpose bearings. Since it is running well, even though it may not be correct, I would probably leave it for now since new bearings in the proper precision class is probably going to run you $800-1,000. I would not suggest getting bearings off of Ebay, way too many counterfeits out there to take a chance.

 

[ChandlerJPerry]

That's a really strange setup, I don't see the point of having a radial ball bearing in-between the two angular contact bearings. Is it possible the previous owner installed them incorrectly? Most mills have the ACBs mounted back to back with a short spacer between them, and then a radial ball bearing at the top of the quill for radial support of the top of the spindle. For the designs with an ACB at the spindle nose and the other end of the quill, I have never seen another bearing in between. I would expect to have 2x 7208 bearings near the nose and then a 6207 at the top. Do you have a picture of the spindle?

Unfortunately I do not have a picture of the spindle fully assembled. However I can confirm what you're saying which is also leading me to believe that Blockhead got into the spindle as well, because the spindle nose is set up exactly as you described with two bearings, and a short spacer between them. But as I explained, the upper bearing in the nose is a standard ball bearing rather than angular contact.

I will try to get some more grease off the bearings when I am home tonight to see if I missed any precision markings, but this could explain why I was recently having some issues cutting 1/4" thick mild steel and had to switch to a roughing endmill .

 

[tailstock4]

This is the first time I’ve seen a milling machine set up that way. One reason for it might be (and it’s only a guess) that this design would eliminate the need for precision spacers and matched bearings. It would also provide a simpler spindle design, taking less machining and less grinding. In short, cheaper.

One of the challenges in working with older machines is figuring out whether something is original or something someone has done to save either time, money or both (or just didn’t know.)

 

[ChandlerJPerry]

Since this discussion is getting really interesting, to make sure that we're all on the same page I've created an Artist's Interpretation of the spindle/quill as I recall it. I'll have to get more actual pictures when I am at home.


Here is one picture I do have of the spindle. Don't mind the inner race of the 6208 sitting on it, the bearing separated and the outer race stayed in the quill when I took the spindle out. I did get it extracted successfully, so at minimum even if I leave the angular bearings, 6208 will get replaced.

 

[Ischgl99]

I’m at a loss as to what that radial bearing is doing, it will have internal clearance, so it can’t be offering much if any radial support. Since there is a snap ring, there is no way for it to be used in any way other than a glorified spacer ring, and putting an angular contact bearing there won’t do a thing, it still needs one at the top where the lock nut is. There must be a reason the DGBB is there, the only thing I can think of is what tailstock mention and they had resonance issues with the spindle and needed the DGBB to dampen it.

When you think you have seen everything, this pops up lol