National Metal Finishing Resource Center (NMFRC)

Historical Articles

September, 1954 issue of Plating

Some Considerations of Importance in Ball Burnishing

Presented at the Forty-First Annual Convention of the American Electroplaters’ Society,
July 12, 1954.

Arthur S. Kohler, Technical Director, Frederick Gumm Chemical Company, Inc.

While ball burnishing in tumbling barrels has been practiced for many years, little material has been published that considers the basic problems involved in obtaining a brilliant metallic finish in barrel burnishing operations. Some operators secure fine results, others are less successful, and many results are found to vary between batches, even in the same barrel and with what is supposed to be the same stock of metal parts.

The purpose of this paper is to discuss various factors which are of material importance in determining the final quality of the work. It is hoped that some of the points discussed here will give the operator a better understanding of the problems involved and produce a general improvement in the results obtained.

In barrel tumbling, there are numerous tumbling media and types of compounds used; but in general, there are only three basic actions that can take place, as far as the tumbled metal parts are concerned, namely: (1) the metal may be worn away gradually by a grinding action of abrasive, (2) with a brittle metal (such as hardened steel) sharp edge burrs may be broken off, or (3) the metal surface crystals can be deformed or caused to flow due to pressure against the surface exerted by contact with the rolling medium or other work pieces in the barrel. While there- is a general similarity between deburring and burnishing, there is also a sharp distinction between the two operations, since deburring employs grinding action, metal flowing, and sometimes chipping, whereas the’ only action in ball burnishing is metal flow.

Before the problems involved in the burnishing barrel are considered, a brief discussion of the properties of metal will be quite helpful. Metals vary considerably, and each metal may be said to have its own personality. Metals vary in color, melting point, hardness, density, corrosion resistance, and malleability. Furthermore, when fabricated parts are dealt with differences are found in size, shape, thickness, weight, and surface conditions in general.

All metals are composed of crystals, of either large or small size, and in many there may be two or more types of crystals present in the same alloy. In a rough way, metals may be divided into two classes: those whose crystals are easily deformed, for example, copper or lead; and those whose crystals are set in a fairly rigid state, such as hardened steel. When the gentle pressure of a hard surface is brought to bear on a piece of metal, the metal bends slightly. When the pressure is released, the metal returns to its original shape. The reason for this is that the metal has a certain amount of elasticity. As long as the pressure applied to the metal is less than a certain amount (which differs for each metal) the same action occurs. Now if the pressure applied to the surface exceeds the amount mentioned above, it is found, upon release of the pressure, that the metal no longer returns to its original shape but has become deformed permanently. This ability of a metal crystal to deform under pressure without having the metal break is called malleability. The degree of malleability varies enormously between metals; such metals as copper, soft brass, and pure aluminum are quite malleable. Hardened steel has, however, practically no malleability and will rupture rather than bend when the elastic limit is exceeded.

When the metal is deformed due to physical changes in the metal crystals themselves there is a tendency for the crystals to interlock more rigidly and become less malleable. The metal is now said to be work hardened. As the amount of deformation increases the harder the metal becomes, the more difficult it is to deform it further. If one wishes to deform the metal further, it is necessary to heat it to a high temperature. The various internal strains are relieved and the metal returns to approximately its original malleability. The metal is now said to be in the annealed condition and can be worked again.

In addition to the physical properties mentioned above, one needs to consider the state of the surface itself since this is where the burnishing action takes place Initial surface conditions of a metal vary enormously and range from a rough matte to a mirror like polish. A matte surface is dull because it consists of numerous hills and valleys, or grooves and ridges. The problem in ball burnishing is to push down these hills and produce a smooth brilliant surface. In a sense, what a road-roller does on a large scale, the burnishing balls do on a microscopically small scale. The burnishing medium is applied against the surface under pressure until the surface crystals are deformed to the desired extent. Since surface roughness varies enormously in a microscopic way, it is reasonable to suppose that if two pieces of the same metal with different amounts of roughness are both burnished, the one with the lesser roughness will be easier to smooth.

In order to flatten the hills, it is necessary to apply a force on the burnishing ball which is sufficiently high enough to exceed the elastic limit of the metal crystals (Fig 1). It would be perfectly possible to allow a single small burnishing ball to roll over a surface for an indefinite time and have no noticeable change produced in the apparent surface roughness. However, when sufficient pressure is applied, for example, by pressing against the ball with a flat board, and rolling it over the surface, it will be found that the path of the ball shows up as a bright line on an otherwise matte surface. Here the elastic limit of the surface crystals has been exceeded, and they have been deformed permanently; some of the hills have been pushed down, and the surface made more even. While these hills have been pushed down, the surface crystals of the parts have been work-hardened and as a result further burnishing is much more difficult. When the burnishing is continued for a sufficiently long time, either the surface becomes smooth and polished in appearance, or it becomes so work-hardened that further burnishing is impossible. At this point, if the surface is not sufficiently burnished, one has a choice of applying more pressure; for example, by using a larger and a heavier load, or the parts could conceivably be annealed and reburnished. Neither of these methods is practical. Here is where the previous treatment of the work is of importance. This is somewhat like the rule for being very tall—the surest way is to pick one’s self tall ancestors.

In the consideration of two surfaces in cross section (Fig. 2), it is seen that A is coarse and B is fine. After these surfaces have been burnished for a given time, A appears like C, and B appears like D. The burnishing barrel has done the same amount of pressing and pounding on both pieces, but A had much further to go and had work-hardened to the point where further burnishing is a waste of time. In most cases, if a surface cannot be burnished within a few hours, the surface has not been prepared properly. Also, barrel operations are similar to wheel finishing. Thus, when a rough casting is to be wheel finished to a high polish, one does not start with a fine polishing rouge and expect a smooth uniform brilliance in a short time. On the contrary, one progresses from a coarse cutting wheel to finer wheels and compositions and finally to a coloring operation. In barrel operations, the deburring barrel does the rough cutting and brings the surface to the proper degree of smoothness so that burnishing produces the smooth, brilliant luster desired.

Point 1—the surface of the metal should be in a proper state of smoothness before burnishing.
In connection with surface roughness, the question of the malleability of the metal is also of importance. It is possible to finish two metals, for example, brass and hardened steel to the same degree of surface roughness. However, when both pieces are put in the same barrel and burnished for the same length of time, the brass develops a brilliant smooth finish whereas the hardened steel is still matte and lifeless in appearance. The reason for this difference is that the brass is fairly malleable and the surface crystals are able to flow readily under the conditions of pressure developed in the barrel, whereas the steel has a rigid structure which is only very slightly malleable. In the case of hardened steel, the surface finish has to be in the neighborhood of a few microinches before any appreciable burnishing can be obtained. With mild steel, a rougher surface is permissible and so on as the metal increases in malleability.

Point 2—The harder or less malleable the metal the finer the surface should be before burnishing.

A frequent source of disappointment in barrel burnishing is lack of depth of color. For example, two pieces of brass are deburred in different ways and brought to the same degree of light matte finish. Both are barrel burnished in the same barrel for the same length of time, yet one comes from the barrel with a brilliant mirror-like appearance, and the other, while just as smooth and shiny, has a flat lifeless appearance. The second piece lacks ”depth of color.” If one looks at the reflection of his eye in the two pieces, he will find a clear distinct reflection in one case, and in the ”flat” piece, he will see only a hazy indistinct reflection (Fig. 3). If a blackened cylinder is rested against the surface of the ”flat” piece (Fig. 4) and one again looks at the reflection of his eye, it will be found that he now sees a clear distinct reflection similar to Fig. 3A.

Why is there such a difference in the reflection when the cylinder is used from what is seen when no shield is employed? The reason is that the surface with poor depth of color is covered with numerous fine holes whose surfaces are not parallel to the surface of the plate. As a result, side light is picked up by these pits and part of it is reflected to the eye. This mixture of reflections produces a fogged appearance. Fig. 5 will help to make this point clear.

The eye looks perpendicularly at the surface and sees its reflected image. However, stray light from the side hits the tiny surface pits and is scattered in all directions. Part of the scattered light reaches the eye and the general effect is a fogging of the image. When the black cylinder is inserted between the eye and the plate the side light is cut off and the only light being returned to the eye is the clear reflected image.

Actually, the surfaces of the two pieces A and B were as shown-in Fig. 2, B burnished to the surface like D and A burnished to a surface like C to produce a flat finish. While probably 90 to 95 per cent of the surface area of C is perfectly smooth and mirror-like, the 5 or 10 per cent pitted area produces the fogged reflection and lacks ”depth of color.”

Point 3—To obtain good ”depth of color” the surface should have a fine matte surface and be free of pits before the burnishing operation.

Up to this point, we have assumed that the surface of the metal is perfectly clean and free of all impurities. In practice, this is never the case since all fabricated metal has an oxide coating (however thin it may be), and in addition, may be coated with oil and shop dirt. Most oxides of metals are relatively hard and often are somewhat abrasive in character. Furthermore, these oxides usually are not metallic in appearance but have various colors. These oxides generally are not removed by barrel burnishing but adhere to the surface and give the work an off-shade color. Oil and dirt tend to stick to the surface or be pounded into microscopic pores and again cause discoloration. In practically all cases, a two-step cleaning cycle should precede ball burnishing. First, the parts should be soak-cleaned or tumble cleaned with alkaline cleaners to remove all dirt and oil. Then, after rinsing, the parts should be given a light descaling or pickling operation. Usually, such a treatment removes surface oxides and gives the surface a very clean light matte finish. For burnishing, such a finish is to be preferred to a smoother surface, since the tiny hills are of microscopic size and are~pressed down easily by the burnishing balls, and fresh-clean bright metal rapidly is exposed to view.

Point 4—Remove all oil, dirt, and scale from the surface before burnishing.

In the case of work which has been barrel deburred, there is a good likelihood that there will be a certain amount of abrasive impregnated in the surface of the work. If even a small amount of abrasive finds its way into the burnishing barrel, there will be trouble, and the work will be dark in color. As was said before, ball burnishing is concerned only with the flow of surface crystals. As soon as there is abrasive in the barrel there will be some grinding action produced. This may roughen the balls slightly, and they will then act like fine files and will further grind the work. Such grinding action produces some fine metal particles which appear blackish in color. Some of these particles are pushed into the surface of the work and tend to give the black color sometimes obtained in burnishing.

Point 5—Remove all abrasive from previous cutting down operations, since any abrasive carry-over to the burnishing barrel will cause dulling and discoloration.

Let us now consider the medium. For most burnishing, highly-polished, case-hardened balls are preferred, although at times diagonals, cones, and pins are used for special jobs. In the light of the previous discussion, certain points will be readily apparent. In order to accomplish any burnishing, it is necessary to apply a pressure against the surface greater than the elastic limit of the metal crystals at the surface, and the greater the pressure the more the surface will be deformed, and the hills will be leveled with the valleys. The pressure on the surface is due to the weight of the load above, pressing the steel balls into the surface of the work. Consider a flat surface in a tumbling barrel. The total pressure per unit area of work depends on the weight of shot above the balls. Fig. 6 shows two examples, in case A, large burnishing balls are used and in B, small ones. Both A and B are the same length but where there are 4 points of contact on A per unit length, on B, there are 12. Or per unit area, there would be- 16 points of contact and 144, respectively. If the total weight above is the same in both cases, each point of contact in case A will carry 144/16 or 9 times as much pressure as in case B. As a result, the burnishing action in A is much greater and the time required much less than for B. In practice, it is found that the larger size medium is much more effective than the small sizes. However, another problem enters at this point. Unfortunately, the larger sized shot, while more rapid in action, tends to damage the work much more than small shot.

There are two principal reasons for the damaging of work by large medium. The larger medium with high pressure at points of contact acts not only on the surface crystals, but also, due to excessive pressure, works on subsurface crystals and distorts the surface to a greater depth. This type of heavy action produces ”ball pattern” or ”orange-peel” effects. The second drawback to large sized shot is the lack of fluidity of the mass when large size medium is used. In a rough qualitative way, this matter of fluidity can be shown by considering three containers loaded with steel balls. In one, let us say, we have 1/8-inch balls, in the second, 1/4-inch balls, and in the last, 3/8-inch balls. If a stick is pushed into each pail in succession, it will be found that the stick can be thrust into the 1/8-inch shot to quite a depth with little effort, with the 1/4-inch considerably more effort is required to move the stick even part way into the load, and in the case of the 3/8-inch shot it will be found that the stick can be pushed only a little way even with considerable effort. In other words, the large sized shot constitutes a mass, which is relatively rigid in character, whereas the small shot acts more like a fluid. In the tumbling barrel, Fig. 7, we find that the main mass of material is moving as a solid mass (to a large extent) and rotates with the barrel until it reaches the point where it starts to slide -cross-wise of the barrel and down hill. Parts which project from the mass, ”A” (Fig. 7), moving with the barrel into the landslide zone are struck by the down flowing work and medium. If the shot is large, the parts projecting into the landslide are held in a rigid position and these parts receive the full impact of the down-sliding shot and work. Furthermore, the energy with which large balls strike the work will be much greater with each collision than would be the case with smaller shot. In the case of the smaller shot, due to the more fluid character, the projecting part ”A” (Fig. 7), will be able to cushion the shocks and shift its position so as to protect itself.

The hardness of the work being burnished also should be considered here, since in practice harder metals can be burnished better with larger-shot than would be the case with soft metals. The writer finds that for most soft metals a 5/32 inch ball is quite satisfactory. Where diagonals and cones are required, it is preferable to use the smaller sizes whenever possible, and usually a better job-is obtained if a mixture of balls and diagonals (or cones) is used rather than the diagonals only. Diagonals, due to their sharp edges, are inclined to cause a choppy, orange-peel type of surface.

Point 6—Use as large a shot as possible without damaging the work.

Before leaving the subject of burnishing shot, it is well to consider the condition of the shot itself. Burnishing balls should be sparkling clean and look like drops of quick-silver at all times. If the shot becomes rusty, rough, or ;dirty, a clean brilliant color cannot be obtained on the work. If the shot is merely dirty, it can be cleaned by tumbling it several times-with a good grade of burnishing compound and thoroughly rinsing the barrel each time. Sometimes the addition of a little alkaline cleaner to the burnishing compound will be found helpful. If the shot is moderately rough or rusted, then the best thing is to recondition the surface entirely. Polishing compounds are available which will bring back a sparkling surface to the shot in a 12-24 hour tumbling time. After the polishing, the barrel must be rinsed thoroughly and the balls tumbled with several charges of burnishing compound before being put back into service. Dirty shot rubs some of the dirt onto the surface of the work and spoils its brilliance. Dull shot has sufficient roughness so that it grinds small metal particles from the work and then these particles are pounded into the surface of the work to lower the luster and darken the color.

Point 7—Keep the burnishing shot clean and brilliant for best results.

The matter of size of barrel loads frequently is overlooked even by relatively experienced operators. If one piece of work were to be burnished at a time, a fine looking job could be turned out in most cases. However, if a half barrel load of work were to be tumbled, alone without any burnishing shot being present, there is a good likelihood that the pieces would be nicked badly from collision between parts. The nicking is more pronounced with larger sizes and heavier pieces than would be the case with small, relatively light work. Obviously, between the uneconomical condition where only one piece is tumbled and the damaging condition where no shot is used and pieces collide and nick, there must be some point where a reasonable load of parts can be tumbled and an acceptable finish obtained. Just where this point is, must be determined by trial in the equipment available. However, a ratio of 3 volumes of shot to 1 of work is a fairly average value for work ranging from 1 to 2 inches in major length and for larger parts higher ratios will be found best in practice.

Point 8—Avoid too low a ratio of shot to work.

Let us now consider the matter of burnishing soap or compound. While soap was used exclusively years ago, today most ball burnishing is done with a ”built” soap or compound. It is recognized that most industrial water contains varying amounts of calcium, magnesium, and iron salts. These salts react with the soap to form insoluble soaps which are useless for burnishing and which are sticky in character, adhering to the work and spoiling the color and brilliance of the finished material. Burnishing compounds usually contain appreciable amounts of water softening ingredients which prevent the curd formation. In addition, most compounds contain buffers which maintain the burnishing solution at the proper pH. It is found in practice that compounds preferred for steel may not be entirely suited to brass or aluminum because brass may tarnish or aluminum may pit when tumbled with a compound designed for steel. On the other hand, carbon steels and cast iron may pit when tumbled with compounds designed for aluminum. Zinc-base die castings present another problem, since generally a different type of compound is best for these alloys. In addition, compounds may contain detergents, lubricants, and other special chemicals. As a result of differences in the chemical activity of different metals, it may be found desirable to use several different compounds in burnishing operations.

Point 9—Choose the proper compound for the job for maximum brilliance and luster.

While on the subject of burnishing compounds, attention should be called to a matter that is often overlooked and yet is a very common cause of trouble in barrel burnishing. As a rule, burnishing compounds are used at a concentration of about 1 oz/gal of water. As far as chemical solutions in the plating room are concerned, this is a very dilute solution. Cleaners may operate at 6-12 oz/gal; plating solutions from 6-50 oz/gal, and acid dips from 6-120 oz/gal. Now if chemicals are dragged into the burnishing barrel from plating solutions or acid dips—and this is a very easy thing to do when the work has cup-shaped recesses or where flat pieces tend to form closely packed stacks— these acids and salts may react with the burnishing compound or soap and produce discoloration or pitting of the work and shot. Obviously, with concentrated solutions such as described above, only a small amount need be dragged into the burnishing barrel to spoil completely the burnishing solution.

Point 10—Be sure to rinse all work thoroughly in good clean water. If necessary, use two rinse tanks before putting the work in the tumbling barrel.

While the above ten points do not cover all of the problems in ball burnishing, the author feels that close adherence to the above rules will overcome the vast majority of the difficulties encountered in practice. It has been the writer’s purpose to try to explain some of the underlying principles involved so that the operator can apply these ideas to his own problem in an intelligent rather than a hit-or-miss manner.

ACKNOWLEDGMENT
The writer wishes to thank Mr. R. E. Moore for his aid in preparing the drawings and photographs.