What's new
What's new

Plain Bearing clearances

Fal Grunt

Titanium
Joined
Aug 5, 2010
Location
Medina OH
Have a surface grinder I am rebuilding, and am trying to find information on how to correctly set up the main bearing.

The surface grinder in question has a roughly 6" front main bearing, that has a tapered thread at the nose that you can tighten to reduce clearance. Because the spindle runs in an oil film, I am curious how to set this.
 
On my much less diameter (2.000") I just used a spotting "bar" that was actual size of journal during scraping. The bearing is adjustable via a solid phenolic "wedge" and the fitting involved seeing that it ran without heating much. It warms up to the touch promptly and does not go higher. This is with a drip supply of Mobil Velocite #6, a very light spindle oil

In other words there is no measurable "clearance":D

Thumbnail illustrates performance of this 1947 OD grinder
 

Attachments

  • P1000351sm.jpg
    P1000351sm.jpg
    52.4 KB · Views: 246
Machinery's handbook has a whole section on bearing clearances. As I recall, it's a function of diameter, load, speed etc. Not at all straightforward unless you have all the necessary information.
 
A true Plain Bearing being a shaft spinning in some form of bushing perhaps babbet or bronze might have .001 .002 to the OD clearance when cold, much closer warm...depending on diameter perhaps up to 6" (?), and may have an end play nut adjustment.Heat rise is often the gauge. Guess I would bring to just zero end play and run it to see a little heat at 10 minuets warm up..getting hot=back off.
Other bearings ball roller and the like are often good with just bringing them up to zero end play and let the warm up make them tight. Finding a little heat at 10 minuets is good, with the spindle not getting hot to the hand touch after a half hour..
Good to state the grinder brand and a PM guy may have the specs.
*You can count your threads and know what a 1/16" turn would equal in thousandths.
just thinking that might be about .0002 or so with a 20 thread per inch and 6" diameter...

Duplex bearings make their own pre-load up to a point as to a certain RPM and load range with just being tight together as specified direction..
 
On my much less diameter (2.000") I just used a spotting "bar" that was actual size of journal during scraping. The bearing is adjustable via a solid phenolic "wedge" and the fitting involved seeing that it ran without heating much It warms up to the touch promptly and does not go higher. This is with a drip supply of Mobil Velocite #6, a very light spindle oil

In other words there is no measurable "clearance":D

Thumbnail illustrates performance of this 1947 OD grinder

John, to clarify, the spindle diameter is 1.250, the bearing surface is about 6" in length (guessing here, I don't have it handy).

On one that I use regularly, call it grinder #1, the surface finish has degraded to the point it is not usable. Speaking with my spindle guy (who does not work with plain bearings) he said I should have MAX .0002" play in the spindle. I can currently push the spindle down about .0004", I can pick the spindle up .0035". If I tighten the bearing up further, it heats up FAST and continues to get hotter.

On the other (grinder #2) that I use for roughing and general work, I can pick the spindle up about .0021, and push the spindle down about .0004. I get good surface finishes with this machine, though if I take a heavy cut, or the material is hard I do get a bit of wavy ness if the wheel isn't open and sharp. This machine warms up quick to about 104 and stays there.

These numbers are with a cold spindle that was rotated by hand to distribute some oil.

So then we have grinder #3, which I have been rebuilding. Now that grinder # 1 is basically inoperable, I need to get #3 finished. Or I need to rebuild the spindle on #1 since it is otherwise a good machine.

In regards to the ID of the bearing surfaces, on grinder #3, there are no remaining scrape marks, and it was a good running machine with good finish. Until it fell over when a friend was hauling it. So that leads me to wonder, if I shouldn't chrome the spindle, grind to .001 over and rescrape the ID of the bearing? Given the length, and the small opening, that would be quite a chore.

Sorry, replies faster than I can type!

Machinery's handbook has a whole section on bearing clearances. As I recall, it's a function of diameter, load, speed etc. Not at all straightforward unless you have all the necessary information.

I will check Machinery's Handbook, an old one!

A true Plain Bearing being a shaft spinning in some form of bushing perhaps babbet or bronze might have .001 .002 to the OD clearance when cold..depending on diameter perhaps up to 6" (?), and may have an end play nut adjustment.Heat rise is often the gauge. Guess I would bring to just zero end play and run it to see a little heat at 10 minuets warm up..getting hot=back off.
Other bearings ball roller and the like are often good with just bringing them up to zero end play and let the warm up make them tight. Finding a little heat at 10 minuets is good, with the spindle not getting hot to the hand touch after a half hour..
Good to state the grinder brand and a PM guy may have the specs.
*You can count your threads and know what a 1/16" turn would equal in thousandths.
just thinking that might be about .0002 or so.

Duplex bearings make their own pre-load up to a point as to a certain RPM and load range with just being tight together as specified direction..

These are Abrasive 1 1/2 machines, 1930-1940 vintage as best I can tell.

I went ahead and pulled the front bearing from #3 that in the spindle parts bin, it is 5 3/4" long, and the thread to tighten is 10 pitch.

I have a few minutes, I will do some experimenting.
 
Clearance depends greatly on the oil to be used
I have some documentation from ABA
They say 0.01mm clearance
and when the power is cut off it should come to a standstill in 20 to 30 seconds
0.01mm clearance requiers a very thin oil
Thinner as water even
Also this spindle is equiped with a internal kind of pump to circulate the oil

The problem is scraping your bearing that acurate
The spindle also needs to have a mirror like finisch

Peter
 
No idea how applicable to a grinder. But the rule of thumb I've always seen and used is .001 per inch of diameter.

For turbine bearings (big dumb plain metal bearings), there is some info at work that that says .0005-.0007 per inch of diameter.
 
John, to clarify, the spindle diameter is 1.250, the bearing surface is about 6" in length (guessing here, I don't have it handy).

On one that I use regularly, call it grinder #1, the surface finish has degraded to the point it is not usable. Speaking with my spindle guy (who does not work with plain bearings) he said I should have MAX .0002" play in the spindle. I can currently push the spindle down about .0004", I can pick the spindle up .0035". If I tighten the bearing up further, it heats up FAST and continues to get hotter.

On the other (grinder #2) that I use for roughing and general work, I can pick the spindle up about .0021, and push the spindle down about .0004. I get good surface finishes with this machine, though if I take a heavy cut, or the material is hard I do get a bit of wavy ness if the wheel isn't open and sharp. This machine warms up quick to about 104 and stays there.

These numbers are with a cold spindle that was rotated by hand to distribute some oil.

So then we have grinder #3, which I have been rebuilding. Now that grinder # 1 is basically inoperable, I need to get #3 finished. Or I need to rebuild the spindle on #1 since it is otherwise a good machine.

In regards to the ID of the bearing surfaces, on grinder #3, there are no remaining scrape marks, and it was a good running machine with good finish. Until it fell over when a friend was hauling it. So that leads me to wonder, if I shouldn't chrome the spindle, grind to .001 over and rescrape the ID of the bearing? Given the length, and the small opening, that would be quite a chore.

Sorry, replies faster than I can type!



I will check Machinery's Handbook, an old one!



These are Abrasive 1 1/2 machines, 1930-1940 vintage as best I can tell.

I went ahead and pulled the front bearing from #3 that in the spindle parts bin, it is 5 3/4" long, and the thread to tighten is 10 pitch.

I have a few minutes, I will do some experimenting.
The "plain" bearing is should be treated as a hydrostatic bearing. The clearance will be greater at a static condition than when it is running.

Peter's reply is spot on as having the bearing properly scraped and the properly polished shaft will generate the proper oil thickness to float the shaft.

Having the bearing too tight will not allow the correct oil thickness to form. This will not allow the proper hydrostatic forces to build which really is the bearing.

Setting for zero clearance will not get you where you need to be. This is one the those situations that is counter-intuitive when comparing to ball or tapered bearings.

Assuming that your bearing and shaft are in good shape and you are using the proper viscosity of oil, I would start at .015in. and let it warmup and see where the temp settles at. You do want the rapid warmup with a moderate running temp. (100F). If the warmup is slow,tighten it a little more.

The advice your spindle guy gave you for the clearance is incorrect. Fine for a rotating element bearing but improper for a sleeve type bearing.

If you are getting a wavy finish under heavy load, tighten the bearing slightly. A quarter turn on this type of bearing is a large amount of adjustment.
 
Ok, so these samples are taken from the two running grinders, no wheel, no hub, just the bare spindle nose.

#1#2
minutesspindlecastingspindlecasting
cold71717171
573-8276
1076-8579
15(5)77768882
20(10)85889384
25(15)90929484
30(20)9597/1059587
35(25)9496/1059689

Two things to clarify, the bracket times are when I tightened the bearing nut. It was less than the width of a sharpie marking, it took a fair amount of pressure, and I got two small independant "clicks" when the nut rotated.

Second is the two temperatures, I noticed that the casting at the back of where the bearing sits was hotter than the front. On the other machine, it was consistent.

When I turned both machines off, the spindle stopped turning in less than 10 seconds, probably more like 5 seconds. It was difficult to turn the spindle by hand.

Grinder # 1

Spindle nose had zero run out while running
Spindle nose had .0004" run out when turning the spindle manually

I could pickup the spindle .0025" and push down .0015"
I could push the spindle .0005" and pull the spindle .0002".

Grinder # 2

Spindle nose had .0005" run out while running (:eek:)
Spindle nose had .0004" run out when turning the spindle manually

I could pick up the spindle .0035 and pull down the spindle .0015"
I could push the spindle .0005" and pull the spindle .0001"

End Report
 
I don't think that is much heat rise.

Oil can act like fluid ball bearings...so it is/may be possible to go to a higher viscosity oil to get you running to finish the job. 10wt to 20 or 30 perhaps worth a try..Non auto oil.
 
I don't think that is much heat rise.

Oil can act like fluid ball bearings...so it is/may be possible to go to a higher viscosity oil to get you running to finish the job. 10wt to 20 or 30 perhaps worth a try..Non auto oil.

Currently using Velocite #10, was originally using #8? but found some old manufacturers literature that I cross referenced and found it was closer to a #10.
 
ISO 46 is 20 weight, I would try that if intending to rebuild a PB spindle and not having a grinder to run.
Some quality compressor oil is 20 wt but you have to see that on the packaging because some is 30 wt and 30 wt may not be suitable..Amsoil, mobil, kobalt,3&1 may be a source.
likely one could not find 15 wt(?)

Lubrication/compressor oil /not automotive oil.
 
A true Plain Bearing being a shaft spinning in some form of bushing perhaps babbet or bronze might have .001 .002 to the OD clearance when cold, much closer warm ... Heat rise is often the gauge. Guess I would bring to just zero end play and run it to see a little heat at 10 minuets warm up..getting hot=back off. ... Finding a little heat at 10 minuets is good, with the spindle not getting hot to the hand touch after a half hour. ... [additional good information trimmed]

Years ago at the San Diego maritime museum, my brother-in-law and I toured the engine room of an old steam ship that used to ferry people across San Francisco Bay.

The three cylinder, triple expansion steam engine was a marvel of engineering.

After about fifteen minutes of seeing us examine and discuss each piece, the somewhat bored docent decided we were interested enough that he should get up and talk to us.

He told us many interesting facts about the engine, including that the two foot diameter prop shaft running the length of the ship -- props on both ends -- could be instantly reversed under full power!

He showed us how the engineers would lovingly place the back of their hands on the bearings to check their temperature. Why the back of the hand? "Because if you use your palm you're likely to get your arm broken!"
 
A fellow at Greenfield Village made practice of running his red rag down the steam engine shaft as it was running at perhaps 200/300 RPM. Some how the rag caught and his body was wound tight and apart. Yes a number of people were outside the glass window.
 
Its the performance that counts.....if you can grind a flat piece with large holes in it without the wheel "dropping" into the holes ,then there is no need for less clearance...and the possibility of a seizure.
 
The journal bearings on a grinder are designed to have a very low eccentricity while running. The eccentricity is the offset distance between the center line of the journal and the center line of the bearing. It would be very difficult to hold tolerances on a ground part if the grinding spindle displaced more than .0001" radially when in operation. Journal bearings generate the hydrodynamic pressure to support the journal by trapping oil in the radial gap to form a wedge shaped oil film. A journal bearing running with zero eccentricity has no load capacity.

There is a strong incentive to design grinder bearings with as small a radial clearance as possible between the bore and journal. The minimum clearance is limited by defects in the journal geometry and surface finish, debris in the circulating oil and the need to permit some misalignment in the journal due to a difference in loads on the pulley and the wheel ends of the spindle.

The older grinder designs deal with the last requirement by mounting the bearings on self aligning spherical seats. Other designs use fixed bearings with very large surface areas to keep the working pressures at the journal low.

For an existing bearing design there is a design curve which relates bearing load capacity to journal eccentricity. The curve is initially very steep and it slowly levels out at high loads and eccentricities. Grinding spindles operate at the steep part of the curve where very small changes in eccentricity produce large changes in load capacity. This insures that the incremental radial deflection of the spindle will be small when transitioning from rough grinding to spark out grinding.

The eccentricity of a given bearing drops in half if the oil viscosity is doubled while maintaining a constant load.

The eccentricity is increased by a factor of roughly 10 if the radial clearance between the journal and bearing is doubled with a constant load.

Bearing design consists of selecting the best guess geometry and operating oil viscosity. This is a iterative process. The bearing geometry and loads will determine the heat generated. The heat generated by the rotating spindle will match the heat lost through the spindle housing at equilibrium. The equilibrium temperature will determine the actual oil viscosity. This temperature determined viscosity is then compared to the assumed viscosity and temperature at the start of the design process. If they do not match the design will need to be altered.

In practice, journal bearing designs work with oil operating temperatures between 180 to 200 deg F. The oil viscosity curves are very flat in this temperature range. If a design worked at near room temperature the oil viscosity could change significantly as the shop temperature changed between early morning and late afternoon. This is why oil viscosity charts specify viscosity at 100 deg F and 200 deg F. The 100 deg viscosity is what is used to determine the starting torque required to get the spindle moving. The 200 deg viscosity is what the designer uses to determine the rotating eccentricity.

The old drip fed spindle designs rely on the lost oil to keep the housing temperature low. The operating temperature of the oil within the bearing will be higher than the housing temperature.

In a modern grinder the high spindle housing temperature would be unacceptable. The thermal distortion in the grinding head and base would cause a loss in accuracy. The frame distortion would also prevent a hydrostatic way type grinder from operating. For these machines the spindle oil sumps are equipped with chillers to maintain a fixed temperature compatible with the machine bed design.

Using a more viscous oil will make the journal bearings very sensitive to temperature.

The viscosity of a 20 weight oil at 100 deg F is 71 centistokes. At 200 deg F it is about 8.5 centistokes.
The number 6 spindle oil has a 100 deg F viscosity of 10. centistokes. At 200 deg F it is 2.6. centistokes.
The number 3 spindle oil has a 100 deg F viscosity of 2.6 centistokes. At 200 deg F it is .95 centistokes.

A spindle set up to run with the number six oil would not be able to break free at startup if it were supplied with the 20 weight oil.
 
I think on grinder number one you are going to have to rescrape your bearings. The front to back dimension with very little free play is what causes your bearing to heat when you tighten it. Meanwhile, the up-and-down movement is too loose for a proper hydrodynamic wedge.
 
You might be able to hand lap the spindle bearing with the spindle using Time Saver lapping compound...

otherwise your going to be scraping it.
 








 
Back
Top