Journal Bearing Maintenance

I would like to get some opinions on journal bearing installations and procedures that everyone used when they replace one. Do you just take the old one out and slap the new one in and hit the start button or do you do additional tasks? I have by own opinions and policies, but I would like to see what others do. Please be as descriptive as you can. Thanks.
Original Post
For any journal bearing, I would never just "slap them in" as you call it. Why did it fail in the first place? It's a good bet there is some sort of damage on the journal so at the very least I would check the journal for wear, rubs, pitting, scouring etc. before fitting new shells/liners. I would also check both shaft and shell diameters to make sure they are going to give me the correct clearance.

On larger bearings, probably 2" diameter and above (really depends on application/speed etc.), after cleaning up the journal as required (and this could entail anything from just a clean to complete re-machining work) I would want to "fit" the shells properly myself. I would NOT rely on what the supplier has given me.
This "fitting" would entail "blueing" the journal, fitting the shell/s and turning the rotor a couple of revolutions by hand to transfer the blue to the contact area on the shell/s. With the weight of the rotor on the bottom of the shell as it will be, you will obviously be looking for a nice even blue pattern along the bottom of the shell. Looking at this blue pattern you can see everything about that bearing assembly including alignment. Any "high" spots (climbing up the shell sides for example) will also be very obvious and these shold then be "scraped" to remove. This procedure should be repeated until you do have a nice even contact along the bottom. If you have never done it (scraping)before, you do need to be very careful as you can obviously damage the white metal quite easily with the hardened steel scraper. You might want to practise on an old shell first!
You can check your final clearances by applying "plasti-guage" or similar along the top of the journal before fitting and tightening the top shell.
That should do for now. I'm sure you'll get loads of other good advice.
Tilting pads? Regular sleeve bearing?

Clean, clean & clean. Set bearing and measure to make sure you have the right fit. Measure clearance and document hard copy to file.

I have Elliott compressors. When setting and aligning I do an electrcal by-pass to start the aux oil pump and have flow before turning the shafts.

I was called to do a start-up documentation. Setup the vibration equipment and did a start. (everything had been rebuilt by a contractor and installed). Upon startup I noticed a rub at 0.03 IPS and told them they would crater in 15 minutes from drag-up of seize of the bearing - they shut-down immediately and babbit had drug-up to remove all clearance. So, a save before it went too far. They had made a pour using diametrical clearance for total clearance.
Thank you for your responsed gentlemen. To answere a few questions. The reason for the change was due to high temperature. I will include a picture of the original bearing and the new. The shaft was cleaned and measured as was the new bearing, prior to installing the new bearing. There was no obvious wiping with the old bearing.
Our instructions for installing and looking for a contact pattern were to blue, install lower bearings half, set shaft on bearing and then remove, without turning the shaft. The shaft is 4.5" dia. The bearing is a shell insert style, not spin casting.


Photos (1)
A feedpump I picture as a horizontal machine. But in my limited experience, the horizontal machines I have seen have axial distribution grooves spaced 180 degrees apart, vertical machines sometimes closer like yours. Is it horizontal or vertical machine?

The old bearing shows a pattern of heavy wear down the center that I associate with heavy load. With light load you should more get uniform wear. With heavy load, the bearing flexes but least flex occurs at the stiffest area where the rib supports the bearing from behind in the center and this is where the wear occurs during heavy load. It's tough to tell whether that wear pattern continues around the whole bearing, but if it does that suggests a heavy rotating load. Otherwise if in one area then heavy static load.

Just to think through all the possibilities: High temperature can come from high load or misalignment or clearance too tight or perhaps oil system malfunction. Plastigage should give you an idea about the clearance (I'm guessing it should be at least 6 mils although OEM should be consulted if possible). Alignment of bearing to shaft is checked somewhat by blue check. You can also do a feeler gage check at the corners of the bearing. Also check oil supply temperature and drain flow from the bearing. Anything unusual on vibration? From the removed bearing, it didn't look like misalignment to me (that should show up as an assymetric pattern), but did seem to look like a possible loading problem. But that is just a rough guess... definitely best to keep an open mind to all the possibilites.

Back to the original question: our standard practice would include blue check and a plastigage check of shaft/bearing and a plastigage check of bearing/housing.
Here is some more info on the pump in regards o some question. Horizontal machine, 5100 rpm, non-lubricated coupling with 10" spool. All clearances at .006 (.005-.007 manufacture rec.) Aligned to manufactures offset at cold condition. We don't have the equipment to perform continuous monitoring for alignment changes nor can we get isolation quick enough to perform a hot alignment check. No vibration indications of any problems.
Hello again.
What oil are you using? At this surface speed, approx 6000ft/min, you should probably be using a fairly "light" turbine type oil, with a viscosity of around 220 cSt @ 40˚C.
I had a largeish (12 ton rotor) steam turbine driving a blower for the FCCU that initially exhibited high bearing temperatures. We couldn't shut it down for operational reasons so we "changed out" the oil, on the run, to a slightly thinner oil, and the temperatures came down immediately.
As far as I know it's still running!
Oil is a synthetic 46 weight. Manufacture calls for a mineral 32. I don't know why we switched to the synthetic. We have four identical pumps using the same oil and bearing temps are 160-190. The 190 is our concern. As our cooling water starts warming up in July our projected temperature will be over 205 degrees. The manufactures rec. limit is 200.
Just to followup some more on my comment about wear down the middle.

The only time I ever saw it before like that is on a 2500hp sleeve bearing motor after a rotor problem caused 0.6 ips housing vibration for several minutes (shaft movement not measured).

There was fatigue damage on the bearing but also a heavy wear cirfumferntial wear pattern right down the axial center of the bearing (on both bottom and top halves, slightly heavier on bottom half).

If you look where the wear pattern is, it is exactly where the bearing is supported on the outside by a housing rib. On your bearing I think the rib is on the bearing but the result is the same. That is the basis of my theory: When the rotor pushes against the bearing, the outsides can flex but the center is stiff due to support from the rib. I don't think normal loading will cause that type of flexing (unless it is a very flexible bearing?) and I haven't seen it anywhere else.

Out of curiosity - has anyone else seen this type pattern? Would you agree it indicates heavy loading or are there other explanations?


One way to do a blue check:
Bearing is installed at the opposite end.
Shaft is set on blocks this end.
Bearing is removed. Coat bottom half of bearing completely in blue.
Carefully roll the bottom half bearing back under being careful not to scuff.
Remove the blocking.
Roll a few turns.
Put shaft back on blocks to carefully remove (without scuffing) to check the bearing.
You expect to see a fairly uniform strip of contact accross the bottom of the bearing with 80% contact accross that patch.

Another way to do the same thing is a dry roll scuff check. Scotch brite the bearing nice and shiny. Then roll and the lack of lubrication will cause a scuffing of the shiny bearing surface which gives similar pattern as the blue check.

I think the dry roll scuff check is a little easier to do because you don't have to worry about disturbing the blue pattern while installing/removing the bearing (easy to do). But you may not see the pattern t quite as distinctly.

If I said anything that doesn't sound right, I hope someone will correct me.
Generally the blueing is to check the bearing and fit properly out of the machine for bearing fit purposed and correction.

An in situ check would be lift check and plasti gage.

So I don't think you want to put blueing on a shaft in situ and turn it with weight of rotor on the bearing with no lubricant.
Sam - we do it in-place. Have always done it that way, and I think many others do as well. I don't know any problems caused by it. Note that the machine weight goes onto the dry bearings whether you roll or not. The roll does scuff the surface but I have always ASSUMED it was not destructive. The one possible problem I can imagine is if you got carried away with many many rotationgs you could put a wear pattern on the bearing that might tend to create a smaller radius (shaft radius) pocket in the bearing and disturb the delicate balance of shaft radius slightly below bearing radius which creates the proper preload. But from one or two turns? I'd be interested to hear more comments from Sam or others if this is good or bad or common practice.

If you don't get a good blue check, then the cause is either misalignment of the shaft within bearing as it sits when rolled... or bearing needs to be worked on a little. Investigate misalignment by checking level and do the feeler gage at 4 corners. As far as the bearing, you can try a light scotchbright, but most likely to correct a poor contact pattern you need to give it to someone knowledgeable in bearing repair, especially if you are going to do some scraping.

Having 25% contact area reduces the load carrying capacity of the bearing by 75% and will cause it to run hot. I agree with Pete that you need to have an experienced technician scrape the bearing in. You need to end up with at least 80% contact. On a high speed machine, you also need to make sure that the contact area doesn't get too wide because that can also cause heating. If you don't have anyone on site with bearing scraping experience you should check with your local large motor repair shops.I worked in the motor repair business for 30 years before moving to Trane. We always verifed the bearing to shaft fit using prussian blue.
You can have your standard and blue and check pattern scraping in true.

You can set the rotor into the bearing using plasti-gage (which I perfer). This will give you all cleanances in mils and tell you where to scrape. Blue for pattern; plasti-gage for clearance. I don't see where blue will give you clearance; only pattern - am I missing something? Some will use solder (rosin core) for checking cleanance on some machines. I could be missing something.

As Joe said, verification. Make the bearing round, parallel (taper free) and within tolerance specification.

Me personally, I would not like to set a large rotor on a bearing having only blue and turn the rotor on a 'dry' bearing. I would rather use a standard, scrape-in, then verify.
Cheeezzz! All this just to come back with what I said back on the 12th!
We've been bluing/scraping LARGE rotor weight bearings, and I'm talking 20 ton+ for decades, in situ. There is no harm at all to the bearing by rotating the rotor slowly, as that is where it starts with a great big bang every time you hit the button.
Scraping is not for the amateur. You need to know what you are doing or you will do more harm than good.
By the way, us old engineers also used to "fit" steam turbine casings by blueing and scraping! It's an art and unfortunately, along with apprentiships and good engineering practices, it's dying.
Originally posted by Sam Pickens:
Blue for pattern; plasti-gage for clearance.

I agree.
I don't see where blue will give you clearance; only pattern - am I missing something?

Who said that blue gives you clearance? No-one that I saw. I said that we do both plastigage and blue check (15 March 2007 08:38 AM) which is logical to me since they are checking different things.
When you say the bearing had wear (Pictures are often difficult to see things.), did you measure any wear; were the clearances off? Was there any babbit distress, cracks, missing, bonding issues? Babbit does not fail at 200 degrees if normally loaded, and the oil viscosity could contribute to the higher than desired temperatures.

Many applications today have precision bearings, and scraping is not allowed. Bearing geometry is often more complicated than being round. Many bearings are elliptical with horizontal clearance (greater) different from the horizontal clearance, there can be more than two pads (or less), bearing halves can be offset, pressure dams may be present (They have to have the right height and position.), pads may have grooves or other relief areas, etc. . Fit in the housing is important.
Thanks Pete. I'm not always the best at reading as close as I should all the time... too much hurry nowadays.

I don't think any of us are far apart and practice may be very close - but, isn't anyone using a standard? Blue the standard, transfer to the bearing and scrape some more off? I've done the scraping in the past or years ago but not in a position to actually do that type of hands-on stuff anymore due to labor restrictions, union grievences, not being of sound mind, cte....... Confused
Morning Bill,

You're on!!! I'm simple on what I'm assuming to be a discussion on 'simple'.

There are so many you just don't mess with. Our Elliott comperssors have tilting pads and operate ~51,300 RPM so I by-pass the electrical and run the aux oil pump just to turn the rotor for alignment. I think few do it and I really don't see the OEM tech reps doing it but it's a practice with me.
I agree Sam. As far as I know, there are not any written standards that cover performance of blue checks. It's good to look at what different people do and why.

Your approach for the Elliot compressor sounds reasonable to me. We start the thrust bearing oil lift pump whenever we do manual rotations of our vertical reactor coolant pump motors. In that case, the whole purpose of the lift pump is to help provide an oil film when shaft is rotating slowly during manual rotation, startup, and possibly shutdown. I have never worked with anything like an Elliot compressor.
First of all I would like to thank everyone for their comments and suggestion. Here's my story for asking so many questions.

We recently had increased temperature levels on this pump. We had some other issues with the plant at the same time and so we took one of our turbines off line to perform some work. Supposed to be 7 days. This bearing temperature issue was one of the tasks to be repaired. Our maintenance department does not have a lot of experience with sleeve bearings to start the explanation. I'm the reliability engineer for the site doing vibration, thermography, oil analysis, and alignment support (w/Maintenance). My recommendations were to perform a hot alignment check, verify adequate oil flow and check bearing contact pattern. Previous to working for this company I worked for one of the premiere apparatus repair facilities in the Midwest. For close to ten years I had the opportunity to perform thousands of alignments on equipment from 10 hp motors and pumps to large steam turbine generators. Along the way several motor installations and repairs including a majority of sleeve bearing application.
Our procedures for sleeve bearing installation were as follows:
1: Clean everything, shaft, bearings, seal surface and make sure everything was smooth, no nicks or burrs.
2: pour a small amount of oil on the top of the shaft, set the lower bearing half on and roll into place (off coarse the shaft would be lifted for enough clearance).
3: Bolt the upper bearing half together and then install the top bearing cap, end bell, or whatever to complete the bearing installation. 4: Rotate the shaft by hand to gain enough inertia to allow the rotor to rotate on its own a few turns.
5: Disassemble the bearing and inspect for the contact pattern on the lower shaft. Use the "scraping" method to gain contact pattern.
6: Reassemble and repeat this process until 80% contact was achieved.

Ok this all sound good and all but with out all your feedback and suggestion I cant convince our Maintenance Manager to perform the scraping (due the vendor telling him it wasn't "necessary because nobody scrapes bearings anymore") if for nothing at all but to rule it out as a possible source of the heating. After talking to the vendor rep, actually two of them neither one of them understood what scraping actually was.
Oh and by the way we install a new bearing, aligned, verifies oil flow and never did the contact pattern per vendor recommendation. Needless to say we started up and still have the hign bearing temperatures but it did change -2.5 degrees.

We have a bearing temperature problem. We have four identical pumps. Two of them run abour 20-25 degrees lower and one 10 deg. lower. I believe it is a contact issue (not equal contact or enough contact). The temp increases to 186 within 10 minutes and takes almost an hour or more to add another 3-4 deg. My experience tells me the quick temp increase is due to contact surface and the rest is system ralated changes. You will always have the sudden increase at start up due to friction but should peek out lower if contact is even (more surface area means more even oil film).
The bearing should not have much thermal lag to reach temperature. Alignment can gradually increase temperature (or decrease). Bearing load (factors such as alignment affect this) can cause a temperature increase. Are the oil supply temperatures the same?

Did the old bearing have any damage? This bearing didn't do much if anything - What tells you the old bearing was bad? You are talking bearing metal temperature and not oil temperature? Does the oil cooler keep the oil supply the same as the other machines? Are all oil viscosities the same? Are the bearings the same type (meaning geometry of the bearing)?

The bearing does have to be aligned with the journal. Are assembly clearances the same?

I cant convince our Maintenance Manager to perform the scraping (due the vendor telling him it wasn't "necessary because nobody scrapes bearings anymore") if for nothing at all but to rule it out as a possible source of the heating.

It is really difficult to un-scrape if the temperatures go up after scraping.

The oil supply temps are with-in one or two deg. for each pump. The old bearing had no damage but a noticeable uneven ware pattern (about 40% contact). Nothing about the old bearing was bad. Yes bearing temps and not oil temps both are continuously monitored. Oil temps are controlled near 100 deg. with same oil with all four systems. All bearing are the same. I am aware that scraping can not be reversed but if a good contact pattern is achieved I have never saw temps increase with this procedure only decrease.


The bearing inserts can only be installed one way so yes the pocket is the same on all of the bearings. The only other heat source is from the water in the pump and there is seal cooling water between there and the bearings. This temperature is also the same on all four pumps.
Are inboard and outboard bearings the same?

Having a pocket on the bottom decreases load carrying capacity and increases temperature. When this is done, there should be a reason. Also, other bearing designs may work and reduce temperatures.

Do the other pumps' bearings look exactly the same as this pump?

Also, it is difficult to say there is a problem here. Bearing metal temps are acceptable, and there was no bearing damage.
Yes the in and outboard bearings are the same. There may not be a problem with the bearing other than temperature. The manufacture says to not operate over 200 deg. if we continue with temps at 190 in February when cooling water is 40 deg, it is inevitable that we will bypass 200 degrees in August when cooling water is 70 deg. There lies the problem.
200 degrees should not not a problem, nor would 220 likely cause a failure. The bearing is probably good for somewhat higher temperatures,as well. We like to have our bearing design below 190 or even lower providing a margin, but design is not always what you get.

So, if you flip that bearing around to the other side, does the pocket change sides with regard to rotation and the bottom? We would ask for the bearing design to not allow installation of a top in a bottom or inboard in an outboard, unless the same design is used.

Are the bearings in the other pumps just like this one? Are the pumps similar?
We had mentioned there don't seem to be a lot of standards (other than OEM info – sometimes hard to come by). For sleeve bearings in electric motors, I did find some fairly specific guidance in IEEE 1068-1996 ("IEEE Recommended Practice for the Repair and Rewinding of Motors for the Petroleum and Chemical Industry"):


IEEE 1068-1996:
"Reassembly of bearings. Split sleeve bearings that are either new replacements or have been rebuilt should be fitted to journals by bluing and scraping as in the following:
1) Using a bearing scraping tool (typically a triangular Þle with the teeth ground off), scrape any side reliefs and lands to the clearances and contours recommended by the motor manufacturer. Apply a small amount of nondrying bluing compound to the shaft journal, spreading it out to form a uniform coating 1 - 2 in wide over the full length of the bottom of the journal. Lift the shaft slightly, roll the lower bearing half into place, then lower the shaft onto it, ensuring that the normal rotor weight is applied to the bearing. Turn shaft 1/2 to 1 revolution. Lift the shaft again, and roll the lower bearing half out. A pattern of very light blue and dark blue areas will be seen on the bearing surface. These correspond to high and low portions of the bearing surface, respectively. Scrape the high spots to make the light/dark pattern uniform; the fitting process should be repeated with bluing as required until at least 80% contact has been achieved. When this is complete, leave the lower bearing half in place with the rotor weight resting on it."

2) Lay two or three pieces of lead wire or other calibrated deformable gage material on the journal, perpendicular to the shaft centerline; their ends within 1/2 in of the horizontal split line. Make sure the upper and lower bearing halves are clean. If shims are used between the halves, make sure they are clean also. Place the upper bearing half over the journal on the wires or gage. Install the upper bearing cap or housing and tighten its bolts to speciÞcation. Then unbolt and remove the upper bearing housing and upper bearing half carefully. Measure the thickness of the lead wire or the clearance of the deformed material as instructed on the package. If the clearance is within limits, remove the wires or gage and proceed with reassembly. If it is not within limits, both bearing halves must be rebabbitted if too loose, or top half of bearing must be scraped if too tight."
Journal Bearings, Scrapping, heck I thought that went out with tasting the oil for babbitt. Mic out the shaft, I use a ball end micrometer for diameter all over and surface plate and indicator on a set of matched V blocks for flat. Anything over two tenths or more on the larger journal out of spec I send it out to be ground. If you send it out to be ground specify that that it must be polished. If you have the opportunity, have the bearing TIN coated or other on the shaft. The messy blue stuff, save that to check the fit under the bearing. Plastigage the final fit for conformation. That should be the only steps taken. Some other tricks are cross drilling and rounding the oil holes but that is another story. Modern journals extremely precise so if you do have a problem it is likely the shaft or the journal seat, do not scrape the bearing. Do not mess with the surface of the babbitt with scotchbrite, sand paper or any other abrasives. Bad contact under the bearing equals bad transfer of heat and thus high temperature. Did you check the relief valve?

Add Reply