Chapter XXVIII: Erecting (3)

"QUALITIES OF LUBRICANTS.--Good lubricants should have the following qualities: (1) Sufficient body to keep the surfaces free from contact under maximum pressure. (2) The greatest possible fluidity consistent with the foregoing condition. (3) The lowest possible coefficient of friction. (4) The greatest capacity for storing and carrying away heat. (5) A high temperature of decomposition. (6) Power to resist oxidation; or in other words, the influence of the atmosphere upon them. (7) Freedom from corrosive action on the metals upon which they are used. It will thus be seen that many conditions have to be carefully taken into consideration; and further, it may be stated that an oil which may be good for heavy bearings may not be desirable for use on light spindles, and for delicate machinery like clocks and watches, where very little power is required to be transmitted beyond that of overcoming their own inertia; and also that oil which is good for small machinery running at quick speeds is very often useless for heavy pressures and large shafting. For very heavy bearings tallow and other solid lubricants are used, such as mixtures of sulphur and tallow, asbestos, soapstone with asbestos, graphite, caustic soda, beeswax, and other similar mixtures, which find favor among locomotive engineers and those in charge of heavy machinery. The pressure that can be borne by a good lubricant for a useful length of time depends upon the nature of the bearings as well as upon the lubricant itself. The velocity of the rubbing action also must be taken into consideration. The maximum of pressure that solid lubricants will bear without destruction is unknown. For steel surfaces, lubricated with best sperm oil moving slowly, 1,200 lbs. pressure per square inch of bearing surface has been found permissible. Under the pivots of swinging bridges several thousand pounds per square inch have been found to work, and for iron journals 800 lbs. per square inch should not be exceeded.

"Lubricants in the market vary much in cost as well as in quality, and very often it is found that the varying prices bear little or no relation to the value of the article purchased. Probably the best test of value is one with which I was familiar some years ago. It consisted of a small engine very much overworked, which stopped and refused to move or go at the proper speed if the shafting had not been lubricated with good oil.

"TESTING BY DESTRUCTION.--The instrument here illustrated, in Figs. from 2495 to 2501, to which I call attention, consists of a bed-plate, having upon it a piece of shafting upon which friction is created by means of two brass steps, the speed at which it is driven being about 300 revolutions per minute. The friction is brought to bear by levers and weights somewhat after the manner of a friction brake as shown in Figs. 2495 and 2500. In the top step is a thermometer for indicating any increase of temperature caused by the friction. A small index indicates the number of revolutions that the shaft makes for any given temperature which the friction causes the thermometer to indicate. The machines used for testing oil have the friction shaft where the oil is destroyed three inches in diameter. Those for tallow are of larger dimensions. It will be seen that on ascertaining the number of revolutions which may be obtained without generating heat, or with the lowest possible increase of heat, that the value of the oil can be obtained. That oil which allows the greatest heat to accumulate with the fewest revolutions must be a bad lubricant. This tabular method of keeping an account of experiments has been found useful. The machine is stopped when the thermometer indicates 200 degrees, as it is considered that an oil has not much lubricating power left if it permits that heat.

------------+--------+--------------+--------------+------------------
Name of oil.| Price. |Revolutions to| Temperature |No. of revolutions
| |200 degrees F.|of atmosphere.| to each degree.
------------+--------+--------------+--------------+------------------
| | | |
------------+--------+--------------+--------------+------------------

"When testing with this machine a definite quantity of oil should be placed on the friction roller, the top step being removed for that purpose; the quantity should be about five drops. A glass tube or small tin measure should be used, as drops vary in size according to the temperature of the oil, and also differ with the specific gravity. The inventor of this machine is Mr. Heinrich Stapfer. I believe he may be considered the inventor of the first instrument for testing oils by destroying them by friction under the actual conditions in which oils are used as lubricants. In using this machine I found that, although it was supposed to test lubricants in the way in which they are used in manufactories, a slight difference existed, which prevented accurate results.

"BEHAVIOR OF THIN OILS.--The first machines were made with the brass steps lipped or recessed, to prevent the oil running away, (see Fig. 2496), which, when thus tested, gave results very much different to those which are accepted by those who are familiar with the use of lubricants. For instance, some thin mineral oils were found to be quite as valuable as, and in some cases superior to, sperm; and this was caused by the lips on the sides, which prevented the oil from running off the bearing when an increased fluidity was caused by friction, and by any slight elevation of temperature. This is a very important quality in lubricating oils, probably next to the capacity to resist oxidation, the most important to be criticised by those who wish to value a lubricant. Although this experiment points out to us that it may be advisable to make the journals of heavy bearings similar to these, if we wish to obtain the best results from cheap thin oils, yet, as oil should be criticised and prepared to be used on bearings with parallel necks, such as are used in works, it was considered proper to alter the tester to that shape to make the conditions similar. This illustration (see Fig. 2497) permits the oil when tested to run away from the bearing if its increased fluidity gives it a tendency to do so. It is this severe test which has enabled sperm oil to rise superior to all rivals, because it has these two apparently opposite attributes--body or thickness, which keep it on its bearing, combined with sufficient fluidity for lubricating purposes. Permit me further to illustrate what I mean in another manner. Suppose we take an oil, good as a lubricant in all other respects, and place it on a bearing, and that 40 per cent. works quickly away because of its extra fluidity when subjected to an increase of frictional temperature, and then compare it with another oil under similar conditions which only wastes, say, 5 per cent. This latter will be 35 per cent. superior as an oil having body, and even if slightly inferior as a lubricant, it may be the most valuable, because strong in this one great quality of remaining at its duty when placed in position. Still another illustration will inform us that in the one case we obtain, say, 60 gallons of lubricating material out of every 100 purchased, and in the other we obtain 95 gallons.

"THE BEST METHODS OF USING THIN OILS.--This will show us that oils which are deficient in body, but which are good in other respects, may be used with good results if doled out in small quantities, as required, by automatic oil-cups like the Lieuvain needle lubricator, Fig. 2498, or any other means. Journals which cannot be fed by means of automatic oil-cups in positions difficult of access should be fed with oil which has a good body. If time permitted, much might be said of the proper shape for bearings of machinery--a subject which would lead to valuable results if discussed by the members of this Society, many of whom must have great experience of those designs which have produced the best results, as well as of those mixtures of metals which are the most durable for light high speed and heavy slow shafting. If any member will take up this subject, or if several members will read short notes, giving their actual experience of different sorts of footsteps, pedestals, and spindles, as well as of the use of different sorts of oil-cups and lubricators, it will be highly advantageous knowledge, which must be of great value to all who use machinery.

Fig. 2501.

Fig. 2502.]

"FLUIDITY OF OILS.--Continuing my remarks on the thinness or fluidity of oils, I wish to call attention to an ingenious arrangement for testing the fluidity when subject to a slight increase of temperature, and also for detecting any tendency which they may have for combining with the oxygen of the atmosphere; this latter quality being advantageous in oils which are used to mix with paint, but which is a great evil when used for lubricating purposes. A piece of plate-glass placed at an angle is made warm to 200° Fahr. A drop of oil when placed on the upper end of this glass will flow down a few inches and thus indicate its fluidity when subjected to increase of temperature. Fig. 2499 shows a ready method I have designed for testing oil in this way. It consists of a tin box in which is fixed the glass, through which can be seen a thermometer. A graduated scale at the side of the box enables the track of the oil to be measured. The box has a door at the back which enables a copper vessel full of boiling water to be introduced; the box is lined with felt to prevent rapid radiation, and when the door is closed it will be seen that several experiments may be conducted before the apparatus becomes too cool for use. I think this a cleaner way than using a lamp for the purpose. The copper may also be used by itself for indicating the behavior of oil on copper when slightly warm in making it discolored or otherwise. As I have before stated, there are many oils which are good lubricants, but which become too thin when exposed to slight heat, and I do not hesitate to reiterate the statement, as I wish to have some influence on the future designs of bearing in this district. A correspondent writing to _Engineer_ from Queensland says that for six months in the year oil runs off the machinery like water and seems to have no lubricating power; he says that the thermometer registers in the summer 140° in the sun, and 110° in the shade. Great difficulty seems to have been experienced by him in keeping oil on the bearings; his experiments on locomotives show that it costs for lubricating a locomotive there about a halfpenny to three farthings a mile, according to the mixture used.

"INFLUENCE OF THE ATMOSPHERE ON OILS.--There are some oils which are excellent lubricants for the first few hours of use, but which have a low capacity for resisting the influence of the oxygen of the atmosphere upon them. The warm glass test may be used for indicating this weakness. If after the test for fluidity the oil be permitted to remain on the glass any exhibition of a resinous or varnish quality may be observed. Another test for this resinous or gummy quality is one which has been suggested to me by Mr. F. R. Wheeldon, of Bilston. He has made many experiments. He found that by permitting oil to remain on a Stapfer friction tester after one test which had been recorded, he tested again on the following day, without adding any fresh oil. This is a severe test, as the thermometer was made to indicate 200° Fahr. each time.

"LONGEVITY OF LUBRICANTS.--Supposing an oil to possess all the qualities which we think a good lubricant should have--that it has fluidity in season, and that it does not combine with the atmosphere and become varnish, that it does not become like water in summer and like mutton suet in winter, and is in most respects satisfactory. We then want to know its powers of endurance, its capacity to resist wear and tear--in other words, its longevity. A good test for longevity or durability of oil when subject to either heavy or light frictional pressure is one suggested by Mr. W. H. Hatcher, a very careful investigator, and chief of the Laboratory of Price's Patent Candle Company, who are extensive oil manufacturers. It consists in taking away the bottom step of the Stapfer tester and placing a small dish containing oil underneath the friction roller (as in Fig. 2500). This oil is carefully weighed before and after several hours' frictional wear and tear. The drawing (Fig. 2501) shows the application of this mode, which I have designed, for testing solid lubricants, such as lard and sulphur and other railway and steamship mixtures. It will be seen that the material is kept to its duty by the weighted lever, and its progress of diminution can be tested in its place by the scale-beam arrangement. When it is used with the pressure on the top step it is advisable to drive it at about 2,000 revolutions per minute; otherwise much time will be occupied in destroying a weighable quantity of oil. The large Stapfer tester (Fig. 2502) was designed a few months ago for this purpose for the Government railways of New South Wales, and it is also used by the Manchester, Sheffield, and Lincolnshire, the Lancashire and Yorkshire, and other railways. I have not been able to get any results of these tests in time for our subject on this occasion, but hope to do so at some future time. The frictional roller is 6 inches in diameter, the pressure amounts to 1 cwt. on each step. As it takes a considerable time to wear away half a pound of solid lubricant, it may be advisable to measure by minutes instead of using the speed index. The speed should be at least 1,500 revolutions per minute. The Stapfer tester should be used in a room of equal temperature, and should not be subject to draughts of cold air, as it will be obvious these will interfere with the indications of the thermometer. A recent alteration in the Stapfer tester permits the quantity of oil used for testing to be measured with greater accuracy than before. A small oil-hole is made in the top step (see Fig. 2502 at _a_ and at _c_) in which is placed a glass tube. This only holds a few drops, and can be filled by simply dropping the oil in, holding the finger at the bottom to prevent it running away, and then place it in the hole. If a small needle lubricator be weighed and then filled with oil of a definite weight, and placed in this hole (see Fig. 2502 at _b_), oil may be tested for longevity and for its anti-frictional properties for a longer period than with the small tube. If oil be placed in this at the same time that oil is placed in the lubricators in the works and the oil tester be driven from the same shafting, permitting it to stop and start when the engine stops and starts, the effect of a week's work upon the weight of the oil may be seen; notice should be taken of the difference of the temperature between the thermometer on the instrument and the temperature of the atmosphere of the workshop.

"TESTING FOR SALTS AND ACIDS, ETC.--It will be obvious that however good as a lubricant an oil is, and however valuable its properties may be when examined, if it possesses any corrosive quality which will be injurious to the metals upon which it is placed, it will soon become detrimental to the machinery, and may also cease to be valuable as a lubricant. Mr. William Thomson, analytical chemist, of Manchester, read a paper on this subject at the British Association at Glasgow, and he stated the results of elaborate experiments conducted by him to discover the influence of various oils of commerce upon bright strips of copper. He permitted the copper strips to remain entirely covered by oil. He also conducted similar tests with half of the strip below the surface of the oil, and the other half exposed to the atmosphere, in order to see what influence the oil had, when the surface line touching the metal would, of course, be acted on by the atmosphere. After noticing the effect upon the brightness or dulness of the copper, he carefully tested the oils in order to detect the quantity of metal which had been dissolved. Mr. Thomson found the following oils dissolved the largest proportions of copper, leaving the surfaces of the copper slips bright--rape, linseed, sperm, raw cod-liver, Newfoundland cod, and common seal oils; and that the following dissolved much smaller proportions of copper, also leaving the slips bright--seal, whale, cod, shark, and East Indian fish oils; and that mineral oils seem to have no dissolving power on the copper, the only effect being a slight discoloration on the copper slip of a greyish color.

"SWISS WATCHMAKERS' TEST FOR FLUIDITY AND CAPACITY TO RESIST COLD.--It seems, according to the _Watchmaker and Jeweller_ (a monthly trade journal), that the plan I have described, and what may be called the warm glass test, seems to be looked upon with favor for testing oil in Switzerland. The degree of heat used for testing the fluidity of oil is 200° Fahr., and if this causes the oil to become a varnish two or three days after the test the oil is considered unfit for use. Another test is one to which I have not alluded, and that is, capacity to resist low temperatures. Oils are tried for their capability to withstand low temperature in the following manner: Fifteen parts of Glauber salts are put into a small glass vessel, a small bottle of oil to be tested is immersed into this; this done, a mixture of five parts of muriatic acid and five parts of cold water is placed over the salt. By means of a thermometer the temperature is indicated, and when it shows a very low temperature, the behavior of the oil, subject to this freezing mixture, may be observed and noted. Mr. Thomson, however, considers that this mixture is not so good or so cheap as ice alone, or a mixture of ice and common salt.

"BLOTTING-PAPER TEST.--It seems it is considered that the blotting-paper test for fluidity is more reliable, according to the writer of the article, than the inclined plane experiment. In order to use this test we must saturate the strip of blotting-paper with oil, and watch whether the drops fall off in pearls or have an inclination to spread out. The latter is a certain sign, the writer says, of a viscid oil. Although this may be considered viscid oil, and may not be valuable for watches, it may, however, be a good oil for heavier machinery."

The amount of friction between a journal and its bearing varies with the kind of metal of which the journal and bearing are composed; on the area of surface in contact in proportion to the load or pressure sustained by the bearing surfaces; on the nature or degree of the lubrication afforded; on the diameter of the journal in proportion to its length; on the manner in which the journal fits or beds to its bearing, and on the kind of motion, as whether the same be continuous, intermittent, rotatory, or reciprocating.

Referring to the friction as influenced by the nature of the metals in contact: the friction varies with the hardness of the metal; thus, with hard cast iron, there will, under equal conditions, be less friction than with soft cast iron. The friction is greater when the surfaces in contact are both of the same metal than when they are of different metals. Mr. Rankine summarizes General Morin's experiments on the friction of various bodies not lubricated as follows:--

GENERAL MORIN'S EXPERIMENTS ON FRICTION.

--------------------------------------+----------------+------------
| Angle of |Friction in
Surfaces. | repose. |terms of the
| | weight.
--------------------------------------+----------------+------------
| degrees. |
Wood on wood, dry |14 to 26-1/2|.25 to .5
" " soaped |11-1/2 " 2 |.2 " .04
Metals on oak, dry |26-1/2 " 31 |.5 " .6
" " wet |13-1/2 " 14-1/2|.24 " .26
" " soapy |11-1/2 |.2
" elm, dry |11-1/2 " 14 |.2 " .25
Hemp on oak, dry |28 |.53
" " wet |18-1/2 |.33
Leather on oak |15 " 19-1/2|.27 " .38
" metals, dry |29-1/2 |.56
" " wet |20 |.36
" " greasy |13 |.23
" " oily | 8-1/2 |.15
Metals on metals, dry | 8-1/2 " 11-1/2|.15 " .2
" " wet |16-1/2 |.3
Smooth metal surfaces occasionally | 4 " 4-1/2|.07 " .08
greased | |
" " " continuously | 3 |.05
greased | |
" " " best results | 1-3/4 " 2 |.03 " .036
Bronze on lignum-vitæ, constantly wet | 3(?) |.05(?)
--------------------------------------+----------------+------------

"The 'angle of repose' given in the first column is the angle which a flat surface will make with the horizon when a weight placed upon it just ceases to move by gravity. The column of 'friction in terms of the weight' means the proportion of the weight which must be employed to draw the body by a string in order to overcome its friction, and the proportionate weight is sometimes called the _coefficient of friction_."[34]

[34] From Bourne's "Handbook of the Steam Engine."

In the following table are given some of the results obtained from Morin's experiments with unguents interposed.

------------------------------+-----------------+---------------------
Nature of surfaces in contact.| Coefficient of |
| friction during | Kind of
| motion. | unguent.
------------------------------+-----------------+---------------------
Brass upon brass | .058 | Olive oil.
Cast iron upon brass | .078 | "
" " " cast iron | .314 | Water.
Steel upon cast iron | .079 | Olive oil.
" " brass | .056 | Tallow or olive oil.
Wrought iron upon brass | .103 | Tallow.
" " " cast iron | .066 | Olive oil.
" " " wrought iron| .136 | "
------------------------------+-----------------+---------------------

Morin's experiments demonstrated that friction is always proportional to the pressure and independent of the area pressed in contact, providing that the pressure is not so great as to cause the surfaces to abrade in the manner or to the degree commonly known as cutting, which occurs when the area of bearing surface in proportion to the pressure is so small as to press out the lubricating material.

Now, between the degree of abrasion that is sufficient to cause a bearing to heat and the minimum, possibly lies a wide range that is very difficult of classification, and that influences the friction of the bearing and journal. Under any given dimensions of journal area and any given pressure of the same to its bearing, the abrasion, and, therefore, the friction, will be less in proportion as the fit of the journal to its bearing extends over its whole area and with an equal pressure of contact. Under these conditions, and with a bearing area ample for the given pressure, the surfaces of a journal and bearing have a smooth, glossy appearance, with a surface as glossy as plate-glass.

This degree of perfection, however, is only occasionally reached in practice, because of imperfections in the fitting and lubrication.

Now, between this condition of glossy smoothness and the degree of abrasion known to practical men as _cutting_ lies, as already stated, a wide range of degrees of abrasion, and each of these has its own coefficient of friction. This may be readily proved by freely lubricating the bearings of a number of journals working under the usual conditions of practice and smearing the oil just as it passes through the bearings upon a sheet of white note paper, when it will be found to contain fine particles of metal, the number and size of particles in a given quantity of the oil decreasing as the surfaces of the bearings are glossy, and increasing as those surfaces appear dull.

The order of value to resist wear is generally considered in practice to be as follows:--

1st in value, hardened steel running on hardened steel.

2nd (and by some considered equal to the first when the pressure per square inch of area is light), cast iron either upon cast iron, hardened wrought iron, or hardened steel.

3rd, under light duty cast iron upon wrought iron or steel not hardened.

4th, wrought iron upon hard composition or brass.

5th, wrought iron upon some anti-friction metal, as Babbitt metal.

Cast iron appears to be an exception to the general rule, that the harder the metal the greater the resistance to wear, because cast iron is softer in its texture and easier to cut with steel tools than steel or wrought iron, but in some situations it is far more durable than hardened steel; thus when surrounded by steam it will wear better than will any other metal. Thus, for instance, experience has demonstrated that piston-rings of cast iron will wear smoother, better, and equally as long as those of steel, and longer than those of either wrought iron or brass, whether the cylinder in which it works be composed of brass, steel, wrought iron, or cast iron--the latter being the more noteworthy, since two surfaces of the same metal do not, as a rule, wear or work well together. So also slide-valves of brass are not found to wear so long or so smoothly as those of cast iron, let the metal of which the seating is composed be whatever it may; while, on the other hand, a cast-iron slide-valve will wear longer of itself, and cause less wear to its seat, if the latter is of cast iron, than if of steel, wrought iron, or brass. The duty in each of these cases is light; the pressure on the cast iron, in the first instance cited, probably never exceeding a pressure of ten pounds per inch, while in the latter case two hundred pounds per square inch of area is probably the extreme limit under which slide-valves work; and what the result under much heavier pressures would be is entirely problematical.

Cast iron in bearings or boxes is found to work exceedingly smoothly and well under light duty, provided the lubrication is perfect and the surfaces can be kept practically free from grit and dust. The reason of this is that cast iron forms a hard surface skin when rubbed under a light pressure, and so long as the pressure is not sufficient to abrade this hard skin, it will wear bright and very smooth, becoming so hard that a sharp file or a scraper made as hard as fire and water will make it will scarcely cut the skin referred to. Thus in making cast-iron and wrought-iron surface plates or planometers, we may rub two such plates of cast iron together under moderate pressure for an indefinite length of time, and the tops of the scraper marks will become bright and smooth, but will not wear off; while if we rub one of cast iron and one of wrought iron, or two of wrought iron, well together, the wrought-iron surfaces will abrade so that the protruding scraper marks will entirely disappear, while the slight amount of lubrication placed between such surfaces to prevent them from cutting will become, in consequence of the presence of the wrought iron, thick and of a dark blue color, and will cling to the surfaces, so that after a time it becomes difficult to move the one surface upon the other. If, however, the surfaces are pressed together sufficiently to abrade the hard skin from the cast iron, a rapid cutting immediately takes place, which is very difficult to remove.

To obtain the best results from cast-iron bearings the bedding of the journal to the bearing must be full and perfect, and the surfaces bright and smooth, in which case it will wear better than hardened steel, unless it be very heavily loaded.

Again, a cast-iron surface will hold the lubricating oil better than either steel, wrought iron, or brass of any kind. Indeed, if a cast-iron surface be made very true and smooth so that it is polished and no marks are visible upon its surface, it will take _much patient_ rubbing and cleaning with a _dry clean rag_ to remove the oil entirely, whereas other metals will clean comparatively easy. In testing this matter upon surface plates the author has found that the only safe method, and by far the quickest, of removing the oil from cast iron is an application of alcohol or spirits of turpentine, because the oil will enter and to some extent soak into the pores of cast iron and gradually work out again as it is continuously wiped, so that if apparently quite clean (after having been oiled and wiped) a short period of rest will cause oil to again be present to some extent upon the surface.

As a general rule motion in a continuous direction causes more wear under equal conditions than does a reciprocating one, because when a revolving surface commences to abrade, the particles of metal being cut are forced into and add themselves, in a great measure, to the particles performing the cutting, increasing its size and the strain of contact of the surfaces, causing them to cut deeper and deeper until at least an entire revolution has been made, when the severed particles of metal release themselves, and are for the most part forced into the grooves made by the cutting.

In reciprocating surfaces, when any part commences to cut, the edge of the protruding cutting part is abraded by the return stroke; which fact is clearly demonstrated in either fitting or grinding in the plugs of cocks, in which operation it is found absolutely necessary to revolve the plugs back and forth, to prevent the cutting which inevitably and invariably takes place if the plug is revolved in a continuous direction. Furthermore, when a surface revolves in a continuous direction, any grit that may lodge in a speck, hollow spot, or soft place in the metal, will cut a groove and not easily work its way out, as is demonstrated in polishing work in a lathe; for be the polishing material as fine as it may, it will not polish so smoothly unless kept in rapid motion back and forth. Grain emery used upon a side face, such as the radial face of a cylinder cover, will lodge in any small hollow spots in the metal and cut grooves, unless the polishing stick be moved rapidly back and forth between the centre and the outer diameter. If a revolving surface abrades so much as to seize and come to a standstill, it will be found very difficult to force it forward, while it will be comparatively easy to move it backward, which will not only release the particles of metal already severed from the main body, and permit them to lodge in the grooves due to the cutting, but will also dislodge the projecting particles which are performing the cutting, so that a few reciprocating movements and ample lubrication will, in most cases, stop the cutting and wash out the particles already cut from the surfaces of the metal.

In determining the metals to be used for a journal and bearing it is preferable to use the softer metal, or that which will wear the most, in the position in which it can be the most easily and cheaply replaced, which is usually in the bearing rather than in the journal; and since two metals of a different kind run better together than two of the same kind, the bearing is usually of a different kind of metal from that composing the journal. It may be stated, however, that under _light duty_ cast iron will wear upon cast iron better than wrought iron or brass upon cast iron (for reasons which have already been stated), especially if the bearing area be broad and the lubrication ample and perfect.

To facilitate the removal of the bearings, brasses or boxes are provided, but in the case of small journals, as, say, of about 3 inches and less in diameter, the duty being light, the lubrication ample and equally distributed, and the journals an easy working fit when new, it is found that solid cast-iron boxes will last for a great length of time without sensible wear.

In some cases cast-iron boxes are cast with a receptacle for some soft metal, such as the various compound metals known under the general name of Babbitt metal.

Babbitt metal is composed of tin, antimony, and copper, mixed in varying proportions. A good mixture for general use where the duty is light is composed of 50 parts tin, 5 parts antimony, and 1 part copper. A harder composition, sometimes termed white metal, is composed of tin 96 parts, copper 4 parts, and antimony 8 parts. This mixture is especially suitable for journal boxes or bearings. It is mixed as follows: Twelve parts of copper are first melted, and then 36 parts of tin are added; 24 parts of antimony are put in, and then 36 parts of tin, the temperature being lowered as soon as the copper is melted in order not to oxidize the tin and antimony; the surface of the bath being protected from contact with the air. The alloy thus made is subsequently remelted in the proportion of 50 parts of alloy to 100 tin.

For brass bearings or boxes a mixture of 64 parts copper, 8 parts tin, and 1 part zinc is found to answer well; but for bearings not requiring so hard a metal, the quantity of zinc is increased, and that of the tin diminished.

Bearings or boxes that are to be babbitted are usually cast as in Fig. 2503, there being a rib at A, B, and C, forming a cavity at D, into which the melted metal is poured. The ribs (in new boxes) are sometimes bored out, or for rougher work may be chipped and filed out to fit the shaft, and hold it in line; and to prevent the ribs A, B, &c., from bearing and cutting the shaft, a piece of pasteboard is laid on ribs A and B, thus confining the journal bearing to the babbitt. The best method is to pour the bearing and then rivet the babbitt well into the cavity D, which is made wider at the bottom, to prevent the babbitt from coming loose, and then bore out the bearing in the usual manner.

The principal advantage of a babbitted bearing is the ease with which it can be renewed, and the fact that the metal will soon bed itself to the journal. This is of great advantage in the case of solid bearings in the framing of fast-running machines, and in situations where it would be awkward or difficult to take brasses or bushes out to fit them, or align them to the shaft, which in many cases would also require to be taken out to remove the brasses. On the other hand, any particles of grit that may find ingress to babbitted boxes are apt to become bedded into the babbitt metal and cut or grind away the journal.

Since the babbitt metal in a bearing is apt to close across the bore when cooling after being poured, a mandrel of slightly larger diameter than the diameter of the journal should be used in place of the working journal to run the bearing on. Some effect the same purpose by wrapping writing paper around the journal; but it is wrong to use the journal, for the following reasons: To get a good, sound, well-fitting babbitt metal box, the metal should be poured as cool as possible, for if made red hot it contracts so much in cooling that it does not fit well in the box or frame of the machine. On the other hand unless the metal be well hot it is apt to cool and set too soon and be unsound. To remedy this the journal, or whatever represents it, must be heated. The heating is very apt to bend it. It is obvious then that instead of the journal a temporary bar of iron of slightly larger diameter than the working journal should be used, heating it to a good black hot heat, so that the babbitt metal may be poured less hot than would otherwise be permissible, and the contraction of the babbitt in the box reduced to a minimum. A little powdered resin sprinkled in the box will help the babbitt to flow easily and make a sound casting.

The temporary spindle, or journal, should also be oiled, and as soon as the metal has well set, the temporary journal should be revolved to free it. Babbitt bearings cast in two halves should be fitted to the journal as already described for brasses, which will well repay the cost and trouble.

To prevent the metal from running out of the bearing, its ends are closed by means of either clay or putty closely packed against the bearing ends and the shaft, and in pouring in the melted metal it is best to pour it on the top of the shaft, and let it run down its sides into the cavity of the bearing. This heats the shaft equally, and prevents it bending from unequal expansion, as it would do if it met the heated metal on its lower half only, it being obvious that if the shaft bends the bore of the bearing will not be cast in line; hence, the shaft will bear at the end only, and will require to wear the babbitt down to a bearing.

Babbitting is sometimes employed to refit parts that have worn loose, or to bush the bores of work. Suppose, for example, that in a case of emergency a pulley of a certain diameter is required, and that the only one at hand has too large a bore, then we may take a mandrel or arbor of the diameter of the shaft the pulley is required for, and drive on it two thin washers and turn them to fit the bore of the pulley, and cut a recess in each to enable the metal to be poured through. We may then put the arbor and washers in the pulley (the washers serving to hold the arbor true), and fill in the bore with babbitt metal, leaving the pulley set-screw in place and set to just touch the arbor, so as to cast the thread in the babbitt bushing, and thus save drilling and tapping.

PROPORTIONS OF JOURNALS.--It follows from what has been already said that under a given amount of duty the friction will be less and the durability greater in proportion as the bearing area of a journal is increased. But it is an important consideration whether such area shall be obtained in the diameter or in the length of the journal, or, in other words, what shall be the proportions between the diameter and length of a journal. It is found in practice that a journal wears better in proportion as its length exceeds its diameter, providing that the stress is not sufficient to cause sensible flexure, because in that case the pressure is reduced at that part of the journal where the most flexure occurs, and increased where the journal is most rigid. As a result, the abrasion increasing with the pressure becomes locally excessive, the glossy smoothness is lost and increased friction ensues.

If, however, the length of a journal is limited by the exigencies of its location or the design of the machine, the diameter of journal must be increased if necessary in order to obtain sufficient bearing area to withstand the stress without causing undue abrasion.

Referring to the bearing area in proportion to the load, Prof. R. H. Thurston writes, in an article in the _Railroad Gazette_ of January 18th, 1878, as follows:--

"A pressure of 800 pounds to the square inch can rarely be attained on wrought iron at even low speeds, while 1,200 pounds is not infrequently adopted on the steel crank-pins of steamboat engines. I have known of several thousand pounds pressure per inch being reached on the slow-working and rarely moved pivots of swing bridges. In my own practice, I never, if I can avoid it, use higher pressures than 600 and 1,000 on iron and on steel, and, for general practice, make the pressure less as the speed is greater."

W. Sellers and Co. state that under a pressure of 50 lbs. per square inch, and with oil well distributed over the surface of the box, the metal of the journal will not touch that of the bearing box bore.

In practice bearings are made with a length varying from that equal to the diameter of the journal to about four times that diameter, and but few cases occur in which these limits are exceeded in either direction. It is to be observed, however, that diminishing the length is apt to increase the abrasion unless the duty is very light indeed, while increasing it increases the durability while not affecting the friction, unless the shaft bends.

There are special cases in which within certain limits the proportions of journals are nearly uniform in practice; thus the length of engine crank-pin bearings rarely exceeds once and a half times the diameter, while the main shaft bearings are often similarly limited in width from the exigencies of designing the engine so that the eccentric shall come in line with the slide-valve spindle. In the case of crank-pins the pin cannot be held sufficiently rigidly to prevent spring or flexure; hence it is desirable to limit its length so that its pressure shall be as short a leverage as possible to the crank. The solid bearings in the framing of machines usually admit of room enough to make their lengths three or four times the diameter. Again, in the case of line shafting, boxes having a length equal to three or four times the diameter may be employed, providing that the alignment be made correct, or that the boxes are self-adjusting. But in all cases the longer the bearings the greater the necessity for correct alignment, so that the axis of the bearing bore may be in line with the axis of the shaft, the error manifestly increasing with the length of the bearing.

PLACING TWO CRANKS ON A SHAFT SO THAT THEIR CENTRE LINES SHALL STAND AT A RIGHT ANGLE.--It is obvious that the keyways in both the crank and the shaft must be cut accurately in their proper positions, because it is a tedious operation to file out the sides of the keyways when the cranks are placed upon the shaft. To mark the keyways in the absence of any tools or appliances specially designed for the purpose we proceed as follows: Placing the shaft upon a marking-off table, we plug up the centres upon which the shaft has been turned by driving a piece of lead in them, leaving the surface level with those of the shaft; and then from the perimeter of the shaft we carefully mark, upon the lead plugs, the centres of the shaft. From this centre we describe a circle whose diameter will be equal to the required widths of the keyway, and then taking a square we place its stock upon the face of the marking-table, and bringing the edge of the blade even with the edge of the circle, we mark a perpendicular line upwards from the circle to the perimeter of the shaft, and then draw a similar line on the other side of the circle, as shown in Fig. 2504, in which A represents the shaft and B the circle, C the perpendicular line struck on one side of the circle, and D the square placed upon the marking-table E, in position to mark the line on the other side of the circle, F and G being wedges to keep the shaft A from moving its position upon the table. We next mark with a scribing-block or surface gauge the depth of the keyway as denoted by the line H, and the marking at that end of the shaft is completed. Passing to the other end of the shaft we find the centre of the shaft, and describe around it a circle equal in diameter to the required width of keyway, and from the edges of the circle to the perimeter of the shaft draw two lines with a scribing-block, as shown in Fig. 2505, A representing the shaft, B the circle, C D the breadth of the keyway, E the marking-off table, F and G the wedges, and H the depth of the keyway, which must, in this case, be marked with a square resting on the table.

If, however, the shaft is too heavy or large to be placed on a marking-off table, we may proceed as follows: Strike as before the circle B, Fig. 2504, equal in diameter to the required width of keyway, and adjust a straight-edge held firmly against the end face of the shaft, so that its upper edge is coincident with the perimeter of this circle, while the straight is horizontally level-tested by a spirit-level. Draw a line along the shaft face, using the straight-edge as a guide. This will give us the line C in Fig. 2505. By a similar process the line D, Fig. 2505, may be drawn. At the other end of the shaft similar lines, but standing vertical, may be marked, which will give the positions of the keyways.

We have now marked off on the end faces of the shaft a keyway at each end, one standing at a right angle to the other; but it must be borne in mind that we have paid no attention as to which crank shall lead; that is to say, suppose in Fig. 2506 A and B represent cranks placed upon the shaft C, and running in the direction indicated by arrow D, it is evident that the crank B leads in the direction in which the engine is to run, and hence the keyway E stands in advance of the keyway F; and therefore, as shown in the figure, the right-hand crank leads. To have made the left-hand crank lead, when the engine runs in the direction of the arrow D, we should, supposing the keyway F to be already cut, have to cut the keyway E on the directly opposite side of the shaft; or, what is the same thing, supposing the keyway E to be already cut, the keyway F would require to be cut on the diametrally opposite side of the shaft. It is obvious that if the engine ran in the direction of the arrow G, the left-hand crank would lead, supposing in each case the cylinders to stand at H. Here it may be necessary to explain the manner of determining which is the right-hand and which the left-hand crank. Suppose then that the figure represents a locomotive crank, the cylinders being at H, then as the engineer stands in the cab, facing his engine, A will be the left-hand and B the right-hand crank. It is usual in locomotives to make the left-hand crank lead when the engine is running forward, the practical difference being, that if the workman were by mistake to make the right-hand crank lead, the engine would run forward when the reversing lever was placed to run backward, and _vice versâ_. It makes no difference whether the shaft can be turned end for end or not: if the right or left crank is required to lead when the crank is required to revolve in a given direction the keyways in the shaft must be marked off in the relative positions on the shaft necessary to obtain that result.

The keyways may be carried along the circumference of the shaft by a square applied to its end face, or if that face is not flat by the ordinary keyway marking tool.

To mark off the keyways in the cranks, we place a centre-piece in the bore of the crank, as shown in Fig. 2507, in which A represents a crank having a centre-piece of sheet iron B placed in the bore. On the face of this centre-piece we mark the centre of the hole into which it fits, and from that centre we describe the circle C, which must be of exactly same diameter as the crank-pin if it is in its place, or otherwise of the crank-pin hole. We then draw the lines D and E, using as a guide a straight-edge placed one end upon the crank-pin journal, or even with the edge of the crank-pin hole, as the case may be, and the other end (of the same edge of the straight-edge) exactly even with the circumference of the circle C. From D and E we find the centre of the circle F, which must be central between D and E, and whose diameter must be exactly equal to the required width of keyway; and we then mark the circle G, describing it from the centre of the hole, and therefore of the circle C. By drawing the lines H and I, which must be even with the circumference of the circles F and G, using a straight-edge as a guide, we shall obtain the correct position for the keyway K, and the whole of the keyways may be cut, care being taken to cut them quite true with the lines, and of an exact equal width.

To put the cranks on the shaft, first provide a temporary key, a close fit on the sides, but clear top and bottom, so that it will bind just easily on the sides of the keyways in both the shaft and the crank. The shaft must be placed and wedged with its keyway downwards, so that in putting the crank on, the pin end may hang downwards, which will render it more easy both to put on, handle, and adjust. As soon as the shaft has entered the crank, say a quarter of an inch, we must insert the temporary key (which may have its end edges well tapered off to assist the operation of entering it) sufficiently far into the keyway of the shaft that it will not fall out, and we may then proceed to put the crank on the shaft to the necessary distance, keeping the temporary key sufficiently far in the keyway to enable it to act as a guide--that is to say, up to at least half the length of the keyway.

To put on the second crank, we first place the shaft so that the crank already on stands exactly horizontal, setting it by placing a spirit-level, as shown in Fig. 2508, in which A represents either the crank-pin journal or the crank-pin hole in the crank, and B a circle struck on the end face of the shaft and from its centre, the diameter of the circle B being exactly the same as that of A. If then we so adjust the position of the crank that a spirit-level applied to the exact circumferences of the circles A and B stands level, the crank will stand level, and we have only to put the second crank on with its centre-line standing perpendicular, and the two cranks will be at a right angle one to the other. We now proceed to put on the second crank, pursuing the same method employed in putting on the first one, save that the temporary key need not be inserted so far into the keyway, because, if the keyways have been cut the least out of true, it will make a great difference at the crank-pin, because of the increased distance of the latter from the centre of the crank-shaft. As soon as the second crank is placed to its position on the shaft we must ascertain if it stands vertical, which we may do by applying the spirit-level as shown in Fig. 2509, bringing its edges exactly fair with the edges of the circles A and B, and moving the crank until the bubble of the level stands true, and taking out the temporary key if it is necessary to adjust the crank at all.

If, however, the crank is to be forced on by hydraulic pressure, this latter adjustment should be made when the crank is just sufficiently far on the crank shaft to enable it to bind enough to well support its own weight, to facilitate which the end of the shaft is sometimes slightly tapered for a very short distance--a practice which is sometimes rendered unnecessary by reason of there being attachments fitted to the hydraulic presses which of themselves adjust the position of the cranks, and insure their being at a right angle one to the other.

After the cranks are on their places the keys may be fitted, care being taken that, if the crank last put on had to be moved to adjust it, the sides of the keyways be filed even, otherwise driving the key will tend to move the crank.