What are white metal bearings

The task of bearings is to support or guide parts that can move relative to one another and to absorb and transmit the forces that occur in the process. Bearings can be divided into plain bearings and roller bearings according to the nature of their movement conditions:

bearingsroller bearing
The sliding movement takes place between the bearing and the stored part.Rolling elements cause rolling.
Plain bearings are mainly used where high speeds and loads are required with a long service life (endurance runners such as turbines, pumps, etc.) or where the bearing has to absorb strong shocks and vibrations (punching, pressing, hammering, etc.). Inferior plain bearings are also popular where a low price is decisive and the storage should meet low requirements (household appliances, agricultural machinery, lifting equipment, etc.).Rolling bearings are used for maintenance-free and operationally reliable bearings with normal requirements. They are predestined in the application area for low speeds and medium-high loads.
advantagesdisadvantageadvantagesdisadvantage
Due to the large, dampening bearing and lubricating surface, plain bearings are insensitive to shocks and vibrations, run quietly and are not very sensitive to dirt: they hardly need any seals. Plain bearings allow unlimited high speeds and, with liquid lubrication, achieve an almost unlimited service life. The split design enables easy installation and removal of the axles and shafts on bearings. Adjustable or multi-surface plain bearings result in high running accuracy.The always dry starting friction results in an unfavorably high starting torque, combined with high wear. The high consumption of lubricant requires constant monitoring of the bearings.When installed correctly, rolling bearings guarantee an almost smooth run. The starting torque is only slightly greater than the operating torque. The consumption of lubricant is very low, which is why roller bearings are undemanding in terms of care and maintenance. Rolling bearings do not require a running-in period and, thanks to extensive bearing standards, they are easy to replace.Rolling bearings are very sensitive to shocks and vibrations, especially when they are at a standstill and at low speeds. The service life and the speed are limited. The sensitivity to contamination requires a high expenditure on bearing seals, which brings additional wear and tear and bearing losses (efficiency) with it.

In order to enable the guided shaft / axle (journal) to move in the bearing shell (bush), the bearing bush is made slightly larger than the bearing journal. This creates a cavity between the two bearing elements - the so-called bearing play - which is filled with liquid or gaseous lubricants.
The position of the pin in the bushing is by no means defined within the bearing play and depends on the load condition of the bearing and the operating speed "n".

The critical phase of a plain bearing is the run-up and run-out: When the journal starts to turn, it touches the bearing shell for a short moment until the lubricating film has built up. The mechanical friction caused by the contact between the bearing shell and the bearing journal that is already moving causes high bearing wear and is also responsible for the unfavorable starting torque.

The following statements can be formulated from the findings described:

  • A plain bearing is only subject to mechanical wear when it starts up and when it coasts to a stop.
  • Overcoming the starting friction requires a lot of energy.
  • The maximum load capacity of a given bearing depends on the speed and the viscosity of the lubricant.

The emergency lubrication property (sliding behavior of the material combinations without lubricant) can be influenced with a suitable material pairing between the bearing shell and the bearing journal: Due to the strength criteria, the axles / shafts are usually made of an iron alloy (steel). The bearing shells, on the other hand, consist of bronze alloys (CuSn), copper-lead alloys (CuPb), tin alloys (SnPb or SnSb), cast iron with a high graphite content, carbon graphite, sintered metal, plastics (PTFE, POM, etc.), depending on the application. ), Rubber or even wood.

In railway vehicle construction, the plain bearing is mainly used in steam locomotives and old electric locomotives. At the time when these locomotives were being constructed, rolling bearings could not be manufactured with the required precision and metallurgically could not cope with the stresses occurring in the railway industry. At that time, two types of plain bearings were mainly used:

  • the white metal bearing (SnSb)
  • the bronze bearing (CuSn)

Although the plain bearing would actually be predestined for railway construction in terms of its shock resistance and resilience, it falls victim to a "side effect" of railway operations: the mechanics of the railway vehicles are not subject to permanent stress but are constantly accelerated, braked and finally stopped again. This operating mode subjects plain bearings to high mechanical wear. In post-war designs, the plain bearing was gradually replaced by roller bearings.

White metal is a colloquial term for a tin-antimony alloy (SnSb). This soft material is very easy to process and, in combination with steel, has excellent emergency lubrication properties when accelerating the plain bearing from standstill or when the lubricant film breaks (interruption of the lubricant supply). White metal bearings were often used as shaft bearings in locomotive transmissions and for wheel axle bearings on locomotives and wagons.

In the case of the connecting rod drive, however, it could not always prove itself. Driving rods on steam locomotives and old electric locomotives are mounted with a relatively large bearing clearance, which is necessary for the unconstrained compensation of the spring movement of the locomotive wheels. This large bearing play, however, results in a very selective load on the bearing. Due to the resulting high surface pressure, this causes deformation (knocking out) of the soft bearing bush after a relatively short time. As a result of this process, the liner diameter and thus the play increase steadily, which continuously accelerates bearing wear and ultimately destroys the bearing.

Small bearing clearanceLarge bearing clearance
If the bearing play is negligibly small, the bearing force is distributed over the entire projection area of ​​the bearing. The surface pressure (force per unit area) is therefore considerably smaller with the same bearing load compared to bearings with a large bearing clearance.Theoretically, the bearing force is transmitted from the journal to the bearing shell on an infinitely narrow contact line. In practice, the elasticity of the material and the lubricating film (not shown) are responsible for distributing the force over a narrow band. The bearing force is thus distributed over a very small area: the so-called surface pressure is correspondingly large.

Bronze bearings are mainly used in railway construction where white metal bearings no longer meet the requirements due to the surface pressure (bearing pressure: force per unit area) and a constructive enlargement of the bearing is out of the question for reasons of space or weight.

In contrast to white metal bearings, where no special demands are made on the steel quality of the bearing journals, the journals of bronze bearings are often hardened or at least tempered. This measure is taken to prevent smearing (rubbing) of the pin surface on the bronze bearing shell with a correspondingly high surface pressure. In addition, the bearing journal generally gains strength as a result.

Due to the toughness of the alloys, bronze bearing bushes are considerably less sensitive to knocking out than those made of white metal.

In the case of the rod drive, an elegant solution was found for dealing with the problem of high surface pressure with large bearing clearance, which has already been described for white metal bearings: the bronze bushing is not pressed into the bearing bore in the solid body, as shown in Figure 1, but also slides in it with a corresponding clearance (Figure 2). As a result, the specified total bearing play is distributed over two bearing pairings and thus halved for each friction surface pairing. Due to the now smaller difference in diameter between the pin and the bores, the contact surface is enlarged. As a result, the surface pressure decreases and the permissible load capacity of the bearing increases to the same extent.

Figure 1: Large bearing clearance with a small contact surface and correspondingly high surface pressure.Figure 2: Large bearing play distributed over two bearing combinations: The contact surface is correspondingly larger, which results in a pleasing decrease in surface pressure.

The form of double bearing shown in Figure 2 with the aid of a "floating" bearing bush should only be used under the condition of the required increased bearing clearance. Apart from the larger number of wearing parts, both sides of the bronze bushing also suffer from shrinkage due to mechanical compression, which greatly increases the bearing play during the first few uses. The deformation shrinkage is approx. 0.05 mm per pair of bearings, depending on the application and load: With a "floating" bearing bush, the total bearing play increases by approx. 0.1 mm. This compression of the material must be taken into account in the manufacturing tolerances!

Left: Fully assembled undercarriage with leaf spring and axle box. Middle: After the axis has been axially extended, it encloses the plain bearing. Right: The dismantled axle box releases the plain bearing. (Photo: Ch. Kramer)

This report describes the manufacture and revision of the white metal axle bearings of a LcK long timber car of the Rhaetian Railway RhB. With the exception of the fact that the forge has been replaced by two more modern ovens with controlled temperature, a hundred-year-old manufacturing technology is used.
The type of bearing used is special in that the bearing journal is not completely enclosed by a round bearing bush. The sliding surface only forms a half-shell that encompasses only about 120 °, which lies on the pin and bears the weight of the vehicle.
The bearings are lubricated by means of two lube rags which - similar to the wick of a candle - suck up oil from the axle box and wipe it off the exposed sliding surface of the bearing journal.


Left: White metal half-shell bearing liner with lubricating rag. Right: bearing pin. (Photo: Ch. Kramer)

Since the bearing half-shells described are not new products, the old white metal layer must first be removed from the base body. This is best done with a Bunsen torch or an oxy-fuel welding system.

The white metal melts at a temperature of approx. 400 ° C. The main body made of brass or bronze liquefies at approx. 900 ° C. At the moment when the white metal drips off the base body in a liquid state, there is still sufficient thermal reserve to protect the base body from destruction. Nonetheless, be careful, especially with an acetylene burner: the high temperature supply means that the melting point is quickly reached, even with non-ferrous metals. Furthermore, stress cracks can also form if the localized heating is too high.

With acetylene-operated burners, it is advantageous to work with excess oxygen and a greater distance between the flame and the workpiece. Otherwise slag formation must be expected. The slag prevents the white metal from dripping off and makes its recycling process more difficult.

The melted white metal is collected and returned to the supplier / manufacturer for reprocessing. Under no circumstances may already used material be used for the pouring of new bearings, as the trapped impurities (burned old oil and dirt) result in poor pourability and possibly destroy the sliding surface of the journal with the effect of an emery paste.

After melting, the base body must be freed from the remaining impurities in order to later enable a coating with tin as a binding agent between the white metal and the bronze: This cleaning process is done by sandblasting.

During sandblasting, sand particles are thrown onto the workpiece surface with the aid of compressed air. As with sanding, this technique removes the top layer of the solid. The sand also fulfills its effect in small inaccessible corners of the workpiece. After sandblasting, the bearing shell body is free of any contamination. The workpiece is now ready for tinning.

A tin layer serves as a binding agent between the bronze and the white metal.
CastoTin 1 - a soldering paste made from pure tin - is enriched with a 19.5 percent zinc chloride solution until it can be evenly spread over the surface to be tinned with a brush.

The workpiece is now heated as evenly as possible with the Bunsen burner. If the supply of heat is sufficient, the tin melts and bonds with the surface of the bronze body.
The flame should not be aimed directly at the tin layer, but rather dosed at the back of the bearing. This prevents the flux from burning.

Since the tinning process is not carried out in an oven with a specified temperature setting, the correct temperature dosage must be visually estimated due to the discoloration of the tin: Once the required temperature has been reached, the tin turns dark and the flux previously bound in the paste floats in the form of dark brown Droplets on the clearly visible liquid tin layer on top. Once this state has been reached, the temperature supply with the Bunsen burner must be stopped immediately!

Now the superfluous flux on the still hot workpiece is rubbed off with a cotton cloth. A leather glove provides sufficient protection from the high temperature of approx. 250 ° C. (The melting temperature of pure tin is 231.97 ° C.) When the flux is wiped off, the tin, which is still liquid, is also thoroughly distributed on the non-ferrous metal surface.

The complexity of the mold naturally depends on the geometry of the workpiece. In the present case, the very simple design of a plain bearing for car axles allows for a very rudimentary construction: The shape used consists only of a so-called "core" designed as an L-profile and two U-shaped sheets of metal. The latter form the framework for the axial white metal plain bearing coating.

The bearing body is attached to the core with wire. This is followed by the assembly of the two sheets, which are also fixed with wire.
The wire must be tightened by twisting it so that the individual parts can no longer easily be moved against each other with normal muscle strength. The subsequently applied sealing compound would be destroyed if the individual parts were moved!

The fully assembled casting mold is sealed with Lenit. When doing this, particular care must be taken to ensure that the lenit is not pressed into the cavity of the mold when grouting. Otherwise the lenit would leave a hole in the bearing metal in the appropriate places.

A thin layer of lenite is less prone to cracking when it subsequently dries out in a 400 ° C oven. For this reason, the Lenit should only be applied very sparingly.
The drying time in the hot oven should not exceed 25 - 35 minutes, since at these temperatures the tin, which is already liquid again, reacts undesirably in the air.

After removing the mold from the furnace, the temperature of the bronze body is checked again with a tin stick. If the tin rod begins to melt when you touch the workpiece, the mold is ready for the white metal to be poured on.

The Rhätische Bahn AG uses a white metal alloy for plain bearings on wheel axles for medium loads and medium speed ranges: Sn80 (2.3770). This alloy consists of 80% tin (Sn), 13% antimony (Sb) and 7% copper (Cu). The melting temperature of Sn80 is approx. 380 ° C. However, this bearing metal is cast at 470 ° C. At this temperature the viscosity of the white metal is comparable to that of water. Thanks to the approx. Six times higher density, the alloy finds its way into the open even through the smallest cracks in the casting mold. Accordingly, the loss due to leaks in the mold can never be completely eliminated. However, it is reduced to a minimum if the following points are observed in the casting technique:

  • As already mentioned, the lenit layer should be kept thin to avoid cracking when it dries out.
  • Right from the start, the mold is cooled from below with water during the casting process. The hydrostatic pressure in the liquid metal is always kept low with the batch casting and the continuous solidification in the lower areas, which has a positive effect on the tightness of the mold.

The mold is filled with the white metal in several batches (filling processes). After each batch, the liquid white metal is stirred with a solid object: This makes it easier for impurities and air bubbles to float to the surface. This simple technique enables the material to be cast largely free of voids.

After the white metal has solidified, the wires that hold the bearing body, the casting core and the sheets together are cut. With strong hammer blows on the side of the base body, the bearing is freed from the core.After cleaning the finished raw casting with the aid of a circular brush, the bearing half-shell is ready for machining.

After casting, the bearing shells are machined to the exact geometry. The finished dimensions of the plain bearing are to be individually adapted to the bearing journals. The bore diameter of the half-shell must under no circumstances fall below the actual value of the bearing journal. It is recommended to pre-turn the bore to the actual dimension of the journal. The subsequent scraping results in a bore diameter of approx. 0.1 mm plus.
In practice, the axial play is chosen to be much larger and should be approx. 0.5 to 1 mm.

If only a few bearing shells have to be produced, boring out with a boring bar in the vice of a horizontal milling machine is ideal for machining. If a B-axis is available, both faces and the bore diameter can be finished. However, the advantage of quick and easy clamping cannot compensate for some disadvantages:

  • Without a B-axis, the cast parts have to be manufactured in two work steps. When reclamping the raw bronze castings, errors occur in the parallelism of the two end faces.
  • Without an optical tool measuring system, the diameter of the bore cannot be measured precisely, since the bearing shells only cover about 120 ° of the full circle.
  • The transition radius between the bore and the face can only be machined on a conventional milling machine with shaped steel and a boring bar. The exact adjustment of the cutting edge geometry as well as the infeed of the mass is more complex with a rotating tool than with turning.

All inaccuracies during machining have to be corrected later by hand by scraping. The time required for this should soon justify the manufacture of a special clamping device, even with smaller quantities, which allows a higher manufacturing accuracy of the bearing half-shells. For these reasons, turning out the bearing half-shells with a special device on a lathe is presented here.

In order to be able to machine the bearing half-shells on the lathe, a device is required that meets the following criteria:

The device must ...

  • ... can be clamped in the three-jaw chuck of the lathe.
  • ... be adaptable to different storage sizes of the same storage type.
  • ... allow the boring of several diameters with different circle centers.
  • ... enable fine adjustment of the bearing half-shells.
  • ... also work if only one bearing half shell is to be clamped.
  • ... allow the processing of the shells on both ends without the half-shells having to be removed from the anchors.
  • ... achieve a positional and form tolerance deviation of max.0.1 mm when reclamping the half-shells.
  • ... can be realigned or converted at any time.

The device for bearing machining consists of two components: the driver, which is clamped in the three-jaw chuck, and the clamping frame, in which the bearing shells are clamped. The bearing shells are aligned and fixed in the device with the help of screws that serve as stops.

In order to be able to face and bevel both faces of the bearing shells, the workpieces must be turned. Thanks to the removable clamping frame, this can be done within the required tolerances. The accuracy of the device and not that of the workpiece is decisive for the repeatability during reclamping.

According to the CAD sketch, bearing journals in the diameter range 80 mm to 85 mm can be machined.

If the underlay is less strong, the shells are shifted outwards by 4 mm each. If the turning diameter of the tool is now increased by 4 mm, a lubrication inlet can be turned on. During operation, this ensures that the lubricant is evenly distributed over the entire width of the bearing, also supplies the radii at the end of the bearings with oil and thus replaces every lubrication groove. Furthermore, this measure reduces the support angle between the bearing shell and the journal to "only" 120 °: this practically prevents the bearings from jamming later.

The machining itself is very unproblematic: The soft bearing metal makes no demands on the tool. Only the imbalance caused by the cast body limits the maximum speed of the lathe spindle to approx. 300 revolutions per minute. With a bore diameter of 80 mm, this results in a cutting speed of approx. 80 m / min. A conventional HSS tool steel is the right choice in view of these technological data. A carbide insert for aluminum would also work very well.

The machining feed rate depends on the stability of the tool: Due to the long clamping length of 160 mm, the largest possible clamping shank diameter should be selected. However, the tool holder of the lathe often sets the limits in this regard!
If the tool stability is satisfactory, it would be possible to work with relatively large infeed depths and feeds.
However, since the interrupted cut creates a kind of hammer blow, which impairs the accuracy of the device, large "roughing orgies" were dispensed with here. In the present case one worked with 0.1 mm feed per revolution and about 5 mm infeed depth (in the radius).

When it comes to the final assembly of plain bearings and the associated fitting work, opinions differ over the choice of assembly technology, even among experts. The person in need of information usually receives recommendations from the craziest insertion technique with special templates up to and including the complete waiver of any additional manual adjustment after the boring has been spindled. Of course, this report also only documents another individual view for solving this "problem"; - namely that of the author. However, this cannot be denied at least justifying the choice of his preferred manufacturing technique for this type of bearing application. And - who knows - maybe this report also contains an idea approach for the solution of your individual storage problem ...

Although - as already mentioned - opinions differ widely regarding the choice of the right fitting process, one statement can be made with certainty:

The optimal fitting process varies with the requirements for the corresponding plain bearing.

A turbine bearing, which is in constant use at a constant speed and is permanently supplied with lubricant by an oil pump under pressure, cannot be compared with a railroad bearing. This is subject to constant start-up friction when it is continuously stopped and driven down and is therefore subject to great wear and tear. In addition to wick lubrication, in the best case it also has centrifugal oil or rolling ring lubrication and must be able to allow the bearing journal in the plain bearing shell to tilt slightly (swinging) when the suspension springs. The bearing must not be affected by large fluctuations in outside temperature, or by dirt particles in the lubricant, long downtimes or condensation.

The type of fitting process is determined by the expected wear and tear.

The vehicle tilts to one side due to centrifugal forces when cornering or on uneven tracks: the axle and vehicle chassis are no longer parallel. The difference in inclination must be compensated for by the wheel suspension (axle box guide and / or wheel bearing).
In the case of wagons, the storage boxes are usually guided through the steering knuckle with a relatively large amount of play. Only the torsionally rigid leaf springs or any friction dampers provide limited resistance. The axle box can thus compensate for the wobbling movement of the axle almost freely without misunderstanding the bearing shell. For this reason, the type of wear described in trailers only occurs in rare cases and is much less pronounced.
However, in order to avoid the formation of cracks in the main frame of locomotives due to the strong alternating loads caused by the rod drive, the storage boxes of locomotives are pretensioned in the vertical guides. Instead of a hammer-blow-like alternating load, this creates a swelling load that is gentler on the material. However, this is done at the expense of the ability of the bearing bush to follow the wobbling movements of the axis without restraint.


Under these difficult conditions, it is understandable that a bearing pairing that is common in mechanical engineering (e.g. in the tolerance fields H7-g6) does not make sense for axle bearings in railway construction, or would even have a destructive effect. In order to counteract these signs of wear, a bearing shell with a deliberately inserted bearing clearance should be prepared for future wear development when it is installed: A slight cut-out at the axial ends of the bearing reduces the pressure points during oscillation and thus the shearing off of the lubricating film. The bearing journal thus remains true to shape for longer thanks to the lower dry friction.

Apart from free-form milling, the machining of rotationally asymmetrical geometries is not possible on a computer-controlled processing machine. For this reason, appropriate manual work with the scraping tool after pre-turning is almost inevitable. The following procedure is recommended:

  • The white metal bearing is turned on the lathe to the same diameter as the bearing journal.
  • Subsequent scraping achieves the bearing clearance recommended for new bearings of approx. 0.1 mm in diameter over the entire length of the plain bearing.
  • In the case of locomotive axle bearing shells, the axial ends of the bearing in the bearing base are then scraped over a second time, which results in an additional camber of approx. 0.1 mm in the soft white metal.

The flat, spindled surface becomes slightly uneven as a result of the scraping. The depressions take on a function as lubrication pockets in later operation and, in the event of an unintentional excessive punctual bearing load, allow the excess white metal to be displaced more easily to a limited extent. The elevations, on the other hand, serve as points of support for the bearing shell on the bearing journal when the system is stationary and, similar to polygonal or sliding surface axial bearings, during operation as a gusset point (gusset area = pressure increase due to narrowing of the cross section).

When scraping plain bearings, the following quality features must be observed:

  • The number of points of contact between the bearing shell and the bearing journal.
  • The regular distribution of the points of contact over the entire contact surface.

The scraping technique should be chosen in such a way that as many points of contact as possible are created when the bearing shell and bearing journal are then touched with touching lacquer on the bearing journal. Although the bearing shell never touches the bearing journal during operation, but is separated by a film of lubricant, the evenly distributed colored dots guarantee even surface pressure when the plain bearing is optimally used.

It is not just the number and regularity of the support points that have to be taken into account. It is also important to correctly cut the depressions in the white metal bearing shell.

Even a perfectly manufactured bearing pairing will never work if it is not adequately lubricated. In contrast to classic mechanical engineering, where the bearing is either immersed directly in the lubricant, splashed by it or supplied by a line, in old railway bearings - similar to a candle - a wick usually transports the oil to its destination. Wick lubrication is also very useful in railway construction, as it has neither pollution problems nor minimal oil levels. As long as there is a residual amount of lubricant, it can be sucked up and conveyed by the wick.
There are different versions of wick lubrication: A very simple and reliable variant is "wedging" the lubrication pad between the oil pan and the bearing journal. The cushion lies directly in the oil, is soaked in it and thus conveys a sufficient quantity of it from below to the plain bearing surface.

The design described in this article is much more delicate, as the two cushions are not soaked directly in the oil. Long wicks are immersed in the lubricant and pull it up to the two side-mounted lubrication pads. A sufficient quantity of the oil can only be passed on to the bearing journals after the cushions are saturated.
The structure of the lubrication pads is relatively simple: A metal sheet serves as the base body, to which the fastening eyelet and a pressure spring are riveted.

The sheet is covered with a woven mesh made of undyed pure sheep's wool. Pure wool must be used so that if the bearing gets warm, no synthetic materials melt and get between the bearing pairing as an additional friction factor. Allegedly, the wicking effect should also be better with pure wool.

When attaching the wicks, make sure that they are attached as far as possible to the lower edge of the lubrication pads. Otherwise, the oil would primarily be distributed on the back of the sheet metal incorporated in the poster and not reach the contact surface with the bearing journal. Likewise, the wicks must not be "strangled" by sewing them in, as this would interrupt the flow of oil in the wick.

Before assembly, the lubrication pads must be immersed in warmed plain bearing oil for several hours. This saturates the wool with oil. Dry wool is unable to function as a wick.

The oil-soaked lubricating wicks tend to stick to the bearing journal. When the axle begins to rotate, it winds the adhesive wicks onto the bearing journal. This can tear away the wicks or even the entire lubrication pad. Most of the time, however, the wicks are jammed between the bearing bush and journal. This creates additional friction, which heats the wicks until they burn. After a short time, both of these cause a hot runner and thus a defect in the bearing.
The winding is prevented by a simple, loose looping of the opposite lubricating wicks during assembly. Here, too, the loop must never be pulled so tight that the oil flow is tied off.