Regreasing a Jones and Shipman 540 "heavy duty" 4-bearing spindle

[ballen]

Mottrhed, I'm continuing think about the internals, and wanted to run something by you.

I know (from pushing on the pulley as described above) that the rear spacer slides back and forth with hand pressure in the cast iron bore of the wheelhead. So, wouldn't it be better for me to push out the spindle assembly from the wheel side rather than from the pulley side? This way, the front bearings are loaded the correct way, and the rear bearings should not experience much force since I know that the rear spacer slides easily.

Also, once the rear spacer has come out by (say) 25 mm, I can pull it out from the rear. I'd make up a (say 25mm thick) aluminium plate bored precisely for the spacer OD, drill and thread the plate for a clamping bolt, split it on the bandsaw, and use that to pull out the rear spacer.

This would put the correct pressure on both bearings. Also, once the assembly is out I could make up two such clamping rings and use them to push apart the two spacers, thus always loading both tandem bearing pairs in the correct direction.

Cheers,
Bruce

 

[mottrhed]

With a one-piece phenolic or brass cage, the only way to get the balls into the bearing is by pressing the balls and shells together...

There is probably more chance of damage if the outer shell is being constrained by a tichtly fitting housing.

This is absolutely not true, at least not in high precision bearings.

Precision bearings are assembled by heating the outer race (usually induction) until it falls over the balls and inner race. This causes no damage to the bearings.

 

[mottrhed]

Mottrhed, I'm continuing think about the internals, and wanted to run something by you.

I know (from pushing on the pulley as described above) that the rear spacer slides back and forth with hand pressure in the cast iron bore of the wheelhead. So, wouldn't it be better for me to push out the spindle assembly from the wheel side rather than from the pulley side? This way, the front bearings are loaded the correct way, and the rear bearings should not experience much force since I know that the rear spacer slides easily.

This could work, lowering the risk to simply just overloading, depending on the fit. Frankly I don’t usually have to worry about this since if it’s in my shop bearing’s are being replace anyway lol.

Also, once the rear spacer has come out by (say) 25 mm, I can pull it out from the rear. I'd make up a (say 25mm thick) aluminium plate bored precisely for the spacer OD, drill and thread the plate for a clamping bolt, split it on the bandsaw, and use that to pull out the rear spacer.

This would put the correct pressure on both bearings. Also, once the assembly is out I could make up two such clamping rings and use them to push apart the two spacers, thus always loading both tandem bearing pairs in the correct direction.

Again, hopefully not over loading.

The only risk I see with this strategy is pushing the front bearings through the entire housing. There could be a small step, there could be a burr where the set screws come through, who knows, but once you push the bearings that way (bearings will have a bigger od than the spacers) if it grabs a burr or starts dragging some material, it will score the whole housing. Just a thought

 

[ballen]

The only risk I see with this strategy is pushing the front bearings through the entire housing.

Yeah, I don't like that either. If I push the bearings out the front, they only have to go 32mm, not 300. But was has me worried about pushing out the front bearings by pushing from the pulley side: I am pretty sure that if I push the shaft from the pulley side, then front bearings immediately separate. I can not think of any other reasonable explanation for the motion that I see. So depending upon the "spring gap" between front and rear spacers, the front bearings will be quite separated when I am pushing from behind. Hopefully that spring gap is small, just a few mm.

Could it be that the front bearings can be separated by a few mm without damage (as long as the balls and cage don't fall out)?

There could be a small step, there could be a burr where the set screws come through, who knows, but once you push the bearings that way (bearings will have a bigger od than the spacers) if it grabs a burr or starts dragging some material, it will score the whole housing. Just a thought

You're right, though I would like to believe that the people making something of this quality and precision would have taken a minute or two to put internal chamfers on the screw holes and then another minute to two to deburr them.

The front spacer can be smaller OD than the bearings, but the rear spacer needs to be a nice piston fit in the housing, right? Otherwise it will move around with the bearings inside. So pushing the rear spacer all the way through the housing also runs the risk of raising a burr.

In the end, when I take this apart, I'm going to have to base this on "how it feels" along with my (hopefully now correct) understanding of how it all works.

 

[mottrhed]

Both the front and rear spacers will likely be a little smaller than the bearings, while your thought process is right, the rear spacer must also slide in the bore with only the spring pressure-likely 50-80lbs so it won’t be real tight. Going out the front is safer in my opinion.

 

[ballen]

Any thoughts on my previous question? If the front bearing is separating by a few mm but the balls and cage do not fall out, is this damaging it? In the normal assembly, how much space do you think there is between the two spacers?

I ask this because post 19 shows a drawing of the two-bearing 540 spindle, and there, it appears that there are just a few mm of gap. Post 50 shows a drawing of a 4-bearing spindle from a similar machine (J&S 1400) and there also, the gap seems small.

So perhaps the front bearings are a design that can be separated by a few mm without damaging them. In that case, by keeping the spacer gap smaller than a few mm, the designers provided a simple route for disassembly without damage. Is that possible? Or just a fantasy on my part?

(From SKF literature: "The bearings are non-separable and the bearing rings have one high and one low shoulder.")

PS: the SKF "angular ball bearing installation" video made me cringe. I guess this is NOT how one does precision bearings, at least not the hammer approach.

 

[mottrhed]

Ok, you are right about pushing from the back, you are pushing the inner races through the outers on the front. You are unloading them them for sure, but damage only happens when you get to the other side of the free play in the unloaded bearing. This would likely happen before you completely collapse the spring pack, the bearing will move less than a mm before potential damage occurs. I don’t have a good guess on the space between the spacers.

 

[eKretz]

It is a tricky question, and I don't have much to add that hasn't already been noted - other than this: it doesn't matter how finely made and carefully deburred the bore is if one piece of stray grit, chip or the wrong piece of dust gets into the path that you're pushing the bearings along. One unlucky stray particle can start a gall. Clean very carefully and thoroughly if you try to go the long way.

 

[ballen]

Clean very carefully and thoroughly if you try to go the long way.

Not possible for disassembly, because one can not reach the reach the bore until the bearings and spacers have been pushed out. If the people who put it together kept it clean, and nothing got inside, it should still be clean.

 

[eKretz]

Not possible for disassembly, because one can not reach the reach the bore until the bearings and spacers have been pushed out. If the people who put it together kept it clean, and nothing got inside, it should still be clean.

Yep, didn't even think about that. Fingers crossed.

 

[ballen]

Yeah, you can see why the pros just toss the bearings and put in new ones. It wouldn't be so hard to design a spindle so that the bearing spacers could be locked together, so that the bearings could be removed without damage. But I guess that the extra expense and complication don't justify it.

 

[Peter Neill]

I replaced the bearings in my early 80s J&S540 a couple of years ago - the grease in the original front bearing had gone rock hard.
Whilst your heavy-duty spindle is a little bit more involved than mine, for me the easiest way to change them was to remove the entire wheelhead from the machine after loosening the relevant spindle nuts.
I then left the wheelhead on the bench in front of a small fan heater for an hour, and the entire spindle assembly then just slid out with very little effort at all. Heated the casting up the same way for re-assembly.
The spacers and springs did have a little corrosion on them which would have stopped them moving perfectly smoothly, but took very little effort to clean up.
Incidentally, I got my bearings from Misumi, and they worked out a little over half the price that Andmar or Jubillee Mac wanted for the same thing.

 

[ballen]

For me the easiest way to change them was to remove the entire wheelhead from the machine after loosening the relevant spindle nuts.
I then left the wheelhead on the bench in front of a small fan heater for an hour, and the entire spindle assembly then just slid out with very little effort at all. Heated the casting up the same way for re-assembly.
The spacers and springs did have a little corrosion on them which would have stopped them moving perfectly smoothly, but took very little effort to clean up.

Once you got the spindle assembly out, would it have been possible to remove the bearings without destroying them?

Incidentally, I got my bearings from Misumi, and they worked out a little over half the price that Andmar or Jubillee Mac wanted for the same thing.

For people who might be reading this in the future, who are replacing the bearings in two-bearing spindles, could you please provide the exact part number for the bearings that you got? That factor of two savings is a big motivation to DIY

 

[Peter Neill]

Once you got the spindle assembly out, would it have been possible to remove the bearings without destroying them?


For people who might be reading this in the future, who are replacing the bearings in two-bearing spindles, could you please provide the exact part number for the bearings that you got?

From what I remember (it was January '21...) everything came apart very easily once it was hot, I don't remember having to use any force at all, so I'd say yes.

I have a picture of the front bearing in the box, shown below - It's an NSK 7206ATYNP4. Can't find the number for the rear bearing though....but I've only had a brief look.

If your spindle is built the same way as mine you'll also need a puller with a jack screw and a pin spanner for the jam nuts on the back bearing. I ended up making some on the fly for mine as the lathe was only a couple of feet away from the grinder - you can buy these from Andmar/Jubilee but I felt the cost was a bit much for something I'll probably only do once.

 

[ballen]

From what I remember (it was January '21...) everything came apart very easily once it was hot, I don't remember having to use any force at all, so I'd say yes.

That's good.

I have a picture of the front bearing in the box, shown below - It's an NSK 7206ATYNP4. Can't find the number for the rear bearing though....but I've only had a brief look.

I think for the 2 bearing spindle the two bearings are identical. How about the D=-4 and d=-2 which I assume are OD and ID offsets in microns. Did you just "get" these values or did you ask for them?

If your spindle is built the same way as mine you'll also need a puller with a jack screw and a pin spanner for the jam nuts on the back bearing. I ended up making some on the fly for mine as the lathe was only a couple of feet away from the grinder - you can buy these from Andmar/Jubilee but I felt the cost was a bit much for something I'll probably only do once.

Same for me, my lathe and mill are right there, so no problem making pullers as needed.

View attachment 405442
View attachment 405443

 

[ballen]

That's good.


I think for the 2 bearing spindle the two bearings are identical. How about the D=-4 and d=-2 which I assume are OD and ID offsets in microns. Did you just "get" these values or did you ask for them?



Same for me, my lathe and mill are right there, so no problem making pullers as needed.

 

[mottrhed]

D is OD and the number is how far from nominal, not an offset.
d is ID, same same.

Some vendors will look through their stock for you, but either way youll be limited to whats on the shelf. For this application, dont sweat those numbers, everything in the class will work fine right off the shelf.

 

[ballen]

D is OD and the number is how far from nominal, not an offset.
d is ID, same same.

So suppose that the nominal OD = 62mm, and D = -4. Does that mean that the actual OD is offset by -4 microns from the nominal value, and thus is 61.996mm? Similarly if the nominal ID = 30mm, does d=-2 mean that the actual ID is offset by -2 microns from the nominal value, and thus is 29.998mm?

Some vendors will look through their stock for you, but either way youll be limited to whats on the shelf. For this application, dont sweat those numbers, everything in the class will work fine right off the shelf.

Suppose that the vendor was willing to look through their stock. How would you know what specific offsets to look for?

 

[mottrhed]

So suppose that the nominal OD = 62mm, and D = -4. Does that mean that the actual OD is offset by -4 microns from the nominal value, and thus is 61.996mm? Similarly if the nominal ID = 30mm, does d=-2 mean that the actual ID is offset by -2 microns from the nominal value, and thus is 29.998mm?

exactly

Suppose that the vendor was willing to look through their stock. How would you know what specific offsets to look for?

You would have your spindle entirely apart and measure all journals and bores. With those sizes you would make a determination on what you want to target for a total interference or clearance, do the math and try to source a bearing with sizes as close as possible.

Again, you dont need to do that here, any of the "right" bearings will work in this application. Only thing you need to worry about is if your journals or housings are in spec, if they are worn or damaged you have bigger problems.

 

[Mark Rand]

Note that at the 62mm OD of a 7602, 4µm is equivalent to the expansion that you'd get by heating the bearing by about 6.5°C.