Pump Bearing-Scraping help

Hello all,

I’ve been a long time reader and finally decided to take the plunge and post.

Some background on me, I began my career as a machinist (manual and basic CNC) and spent 5 years in the trade....barely scratched the surface?!?? I then had the opportunity to apprentice as a millwright, and have been doing that ever since (16 years give or take).

I’ve encountered an issue scraping a babbit bearing on a horizontal split pump. These pumps date back to the early 1950s, the impeller Diameter is 48” pumping roughly 26 million imperial gallons per day.

Shaft size at the bearings is 6.5” (+/-). The bearing housings are not cast as part of the pump housing, they are doweled and bolted in place. I’ve verified their location and all seems good. For thrust, the outboard bearing is designed with 3 collars, spread out axially roughly every 3” over the bearing area. The bearing has babbit cast to accommodate these collars, collars and groove are roughly .375” larger than the shaft diameter. Total thrust clearance according to the pump manual is .025”.

Post machining I verified the babbit dimensions and all was within spec, engineers on this are hoping to achieve somewhere in the neighbourhood of a .004” oil wedge.

Although the bearing housing is split horizontally (“standard”), the bearing itself is fit into the housing with the split in the vertical position. This is where I’ve encountered issues, rolling the bearing in and out while bluing yields many false bluing readings (smears).

My process-

-Lightly blue shaft
-use hydraulic jack to lift shaft
-roll in one bearing half (naturally it settles at the bottom half of the housing)
-drop second bearing half onto first half (ensure locating dowels for bearings halves are “engaged”)
-roll entire assembly into place so that bearing split is in vertical position
-release hyd jacks (lower shaft)
-rotate shaft repeatedly
-raise shaft using jacks
-rotate bearings in housing until split is horizontal
-remove first half
-remove second half

When looking at the blue transfer the bearing has smear marks along much of this surface area. I say smear marks because they aren’t high spots as would be expected.

My idea to solve this was was to remove the shaft/impeller from the housing (using crane) and chain falls to ensure assembly lifts slowly and level. Once removed, blue shaft, install bearings halves with split vertical. Use long hose clamps to keep bearing halves on shaft (apply just enough pressure to keep halves together). Reinstall shaft/impeller in housing, and release hose clamp. Rotate shaft for blue transfer, reinstall hose clamps, and pull shaft/impeller assembly.

Does this make any sense?

Have I completely missed something in regards to getting readings that can the trusted?

With a lot of work on ball mills and pumps I’ve been lucky enough to have been involved in bearing scraping since I started in the trade, however I’ve never had to scrape anything with the split in the vertical position that required so much rolling about the shaft.

Any help is greatly appreciated,

Thanks,
Peter

If you can post pictures. It might help the rest of us get a better idea of what your playing around with.

A split in the vertical plane on a horizontal shaft will cause issues...

The oil wedge will fail at the split..

All the dowels , that I’ve seen are designed to keep the split line in the horizontal position.

Time Saver lapping compound might help you out....

The mistake your making is rolling the shaft to blue up the bearings. I have scraped several Cincinnati centerless grinder spindles that use 3 shoes. I don't turn it I push it to and fro. When you turn the shaft it will roll up the bearing and give you a false reading. Also prepare the shoe (bearing) by relieving the center of the shoe(s) aprox. .002" or center 30% so it is not high in the middle and rocks and gives you a false reading. Also if you want to scrape easier instead of using a spoon scraper. buy a Dapra 1/2 moon flaker blade that has a radius bottom. I have some photo's on it I will post. If you use the hose clamp idea make a gage to keep the shoe square to the shaft. I would bevel the leading edge of the wedge at a 45 degree angle so the oil funnels into the wedge. I would be happy to consult with you on the phone. I have also taught babbit scraping classes at Kingsbury Babbit in Philadelphia and American Babbit in West Virginia and if you need the shoes rebabbited contact American. My number is 651 338 8141

Even if you use Timesavers you will need to lower the middle and bevel the wedge edge.

Industrial Babbitt Bearings Repair Manufacture Services | American Babbitt Bearing & Kingsbury Thrust and Journal Bearing System

Pic's L to R: 1/2 moon blade. For you I would say a Radius 60 or 90 mm, Grinder housing I scraped and a flat pie babbitt bearing I taught Amer. Babbitt to scrape.

I found another demo of ID scraping . It is Aluminum, but principals are the same.

That's a BIAX 1/2 moon flaker blade attached to a Russian brand blade holder we used in Austria

Here's an alternative:

An old pump expert once showed me what he called the "bright spot method" where the shell is manually buffed with something compliant but non-abrasive (he used scraps of clean fire hose but non-scratch ScotchBrite for non-stick cookware will be easier to find) to generate a matte or unidirectional texture. No blue or other marking media is used. The shaft slides against babbitt creating shiney spots in the matte finish indicating points of bearing. This is tedious and time consuming best used for picking up the final points

Streaks are a persistent problem printing babbit with blue. Long uni-directional strokes (rotations) promote marking media smears. Regardless of marking medium, instead of rotating the shaft several turns, I suggest you give the shaft an eighth turn back and forth a few times. .

Jobs like this justify a scraping mandrel, an accurately machined thick walled sleeve having the same dimensions as the journal (plus the clearance.) It's far quicker to print pieces one man can handle than crane in the impeller assembly every time. The shell has to be checked for alignment so one out of three or four passes must be checked installed with the rotating assembly in place. In between installed checks, the shell may be selectively scraped to correct alighment.

It's easy to over-scrape bearing shells so the shaft running height winds up below case concentricity. There's a lot of Kentucky windage in scraping and fitting Babbeted bearings, too much to write in a forum response; oil pockets and leads, orienting the split to bear resultant loads, controlling running concentricity, ramping fixed multi-segment thrust bearings, etc. For years, I've looked for a book on the topic. Sad to say, the unwritten lore is dying out with the passing of the experts who possessed it.

When properly fitted and maintained, babbitt bearings may last a generation. Poorly done, they may fail in a week or a month. The problem is, the difference is easily lost in incomprehensible subtleties. An inexpert eye can compare the well-fitted bearing to one doomed to early failure and see no obvious differences.

I'm not trying to buffalo anyone, just trying to make a point: pay attention to details

Hello all,

Thanks for all the various replies, it’s much appreciated, and on my end I’m sorry for the delayed response. I got shipped off to a shutdown for a few weeks, and then all of a sudden COVID-19 put a stop to everything and I’m only getting back to this project now.

Richard- I don’t understand your comment regarding not turning the shaft? Is this your process for scraping at all times? Or was that simply in regards to thrust bearings? I think without pictures I failed to accurately explain the configuration of this bearing set-up.


In regards to the split location, my understanding is that the splits is in the vertical plain for various reasons. Namely the orientation of the pump inlet and size/pressures associated with the pumping. The inlet piping is 36” and pressures are in the 100psi plus range, my brain tells me the manufacturer designed it as such to effectively have the loaded area of the bearing not be anywhere near the bearing split. Also this particular bearing is hydrodynamic (there is no pump), and relies on slinger rings and appropriately placed oil grooves and holes to achieve its oil wedge.

I’ve attached some images of a similar set up (different pump but same collars for thrust), hopefully this helps provide some clarity.

Thanks again for your patience with this,
Peter

Thank you for the pictures.

Forrest's suggestion of a scraping mandrel is probably your best bet.

I even learned something. I've never seen that kind of setup.

I haven't read through every post here so I apologize to all in advance for my ignorance. I just wanted to post my 2cents on the matter and I'll tell you first off I have never scraped a pump. My experience comes from lathe and grinder spindles. When scraping any round bearing the number one mistake I see is people scrape all blueing evenly. That may sound weird to hear but as a shaft drops down into a bearing half while scraping it should be obvious to note that the bearing, in this case babbitt, can only go deeper, if you make the bearing too wide near the parting lines you'll never get that bearing back. Most of your scraping should occur at the bottom of the bearing and progressively less as you work up the sides toward the parting line. The only exception to this is if you have the shaft chromed and ground and it is larger than the old shaft diameter, then you may need the widen the bearing to get it to drop down in, after that your focus should be on the very bottom and maybe 75% of the way up from there. Almost parabolic. As Richard and Forrest pointed out it is very easy to false blue babbit bearings, and like Rich said rotating the shaft to take a print can get you into trouble as the shaft climbs up the sidewalls like a gerbil in a wheel and blues everything. I dont usually use bluing for plain bearings, not that it can't work, but rather red brick dye, it's cheap and abundant, comes in powder form, mix it with oil to get a paste and apply it to the babbit. Wipe off the excess until it is just a haze and print the shaft to it dead clean with no blue. In your case the thrust rings would prevent you from sliding the shaft linearly to take a rub which is ideal. You will have to do as Forrest suggests and just barely wiggle the shaft rotationally, but be leary of the high spots showing up near the sides or parting line, it's good to have contact there but I wouldn't scrape those points unless its holding the shaft from touching the bottom, which in my experience never will. Babbitt is so soft it can conform easily in those areas. If you want to make sure you have the appropriate clearance all the way around when done you need to assemble the unit and torque the bolts and set up an indicator to measure in the vertical and horizontal directions and manually move the shaft around, if you need more vertically add shims if you need more horizontally CAREFULLY scrape the sides.
My great uncle taught me a trick once with babbitt. To do the final scraping, use no blue or dye whatsoever and simply scrub the babbitt with a piece of burlap, it will erase any hanging burrs and leave a matte finish so the high spots show up like mirrors after rubbing to the shaft.
I hope you found this helpful, there are a lot of gentlemen on here who know much more than I, i just wanted to share a bit on the matter as i work in a shop with lots of plain bearings and most of this information is not easily gained these days. And on that note, best of luck finding burlap in 2020.

CHRIS well said! The burlap trick ...thanks.. I still need to come meet you sometime seeing your what 3 hours away?

Richard King said:

CHRIS well said! The burlap trick ...thanks.. I still need to come meet you sometime seeing your what 3 hours away?

Click to expand...

4 ish, yeah. I need to make my way over there someday and also stop to visit with Rick at A&D, he does a lot of grinding for us.

ironsmith89 said:

4 ish, yeah. I need to make my way over there someday and also stop to visit with Rick at A&D, he does a lot of grinding for us.

Click to expand...

I'll meet you there....and we can go have lunch or stand 6' away..and wave ...Yeah Rick and I have been friends for years. I sold him a lot of "stuff" had him plane my straight-edges...etc... maybe in May if were lucky