Proof of the Condensation of Steam in Outside Cylinders, by comparison of the indicated consumption of steam with the measured consumption of water. Many experiments were made by the writer on this point; one was made with No. 42, passenger engine, on the Caledonian Railway, during a trip of 105 miles, from Glasgow to Carlisle, with an average train of 63 carriages, done in three hours 22 minutes, five stoppages included. Indicator diagrams were taken from the cylinder at intervals of one or two miles, and the notch of the expansion gear observed for each diagram, and the points of the line where each change of notch was made. The several points of cutting off, expansion, and compression were accurately ascertained by means of the slow diagrams; from which were calculated the exact quantities and pressures of sensible steam actually consumed in each interval of the trip, and the water equivalents for the several quantities of steam present in the cylinder; which, multiplied by the number of strokes of the two cylinders in each interval, gives the total quantity of water efficiently used as steam. The following final results were thus obtained:

The examination of the indicator diagrams in the manner employed before, by comparing the initial and final water equivalents of the steam during expansion, shows that at least 13 per cent. of this loss of 16 per cent. was due to condensation, and it is probable that no appreciable proportion was due to priming; indeed, the least loss was observed to take place with the least degree of expansion, and when the consumption of steam from the boiler is going on at the greatest rate, as we find on referring to the per centages of admission; which is the reverse of the effect that would be observed if priming were a material cause.

Experiments made by the writer with other outside cylinder engines, or imperfectly protected cylinders, corroborate the above deduction obtained from the performance of No. 42; and they are still further corroborated by his experiments on inside well protected cylinders, which show that in ordinary good condition there is no sensible excess of water of any importance, actually consumed from the boiler, above what is estimated. from the indicated steam passed through the cylinder. These results are also confirmed by the results of the trials of Mr. D. Gooch, with the Great Britain and similar engines.

The increased back pressure of exhaust affords additional evidence of the presence of water in the cylinder. The back exhaust pressure is the consequence of the want of facilities for the timely discharge of the exhaust steam from the cylinder; and the impediments to its discharge are much

increased by the presence of water amongst the steam, whether due to condensation or to priming. The presence of water is immediately made apparent by the increase in the back exhaust pressure, shown by the indicator diagram, as the writer has on many occasions had an opportunity of observing. The effects of priming from foulness of the water in the boiler are shown in fig. 3: A and B are indicator diagrams taken from the well protected cylinders of the Orion, in which very little, if any, condensation could be detected. The diagram A was taken before, and the diagram B after the boiler was blown off and supplied with clear water, both being taken at the same speed, and showing 7 lbs. back pressure caused by priming in the former case.

The diagrams C and D, fig.3, show that the total quantity of water from condensation is considerably greater, with the greater degrees of expansion, where a smaller quantity of steam is admitted, and consequently the loss is more seriously felt. These diagrams were taken from the outside cylinder goods engine No. 127, working at the same speed up and down an incline on the Caledonian Railway; the diagram C cutting off at twothirds the stroke, and the diagram D at one-sixth of the stroke. The latter, D, though it had the advantage of a much earlier exhaust, and only one-fourth of the quantity of steam to discharge, was affected with 10 lbs. more back pressure than the former, C, when working in full gear. This great back pressure was maintained over a continued run of twenty miles, when of course the cylinders had got into their working heat for that degree of expansion; and the inference is, that the steam was loaded with water of condensation, (proved also by the expansion curve,) which was with difficulty expelled, and which only became proportionably less when the degree of expansion was diminished; and, consequently, the mass of steam increased that was to be cooled within the same superficies of cylinder.

That the total mass of the steam has much to do with the condensation is proved by the diagrams E and F, fig. 3, taken under the same degree of expansion, and at the same speed, but with 75 and 20 lbs. steam respectively admitted to the cylinder. In the latter diagram, F, the back exhaust pressure is 7 lbs. greater than in the former diagram, E, although the total quantity of steam to be discharged was so much less. In the latter case, indeed, there was found to be an excess of 18 per cent. of the whole water used over the indicated steam expended, which was most probably altogether by condensation, as the rate of consumption was so moderate as to preclude any likelihood of priming.

Now here is a case where, in the same class of engines, the back exhaust pressure increases as the quantity of steam to be discharged becomes less, notwithstanding that the facility for exhaust increases at the same time. This is clearly a case of water in the cylinder, the quantity of which increases with the degree of expansion; and the water is as clearly a precipitation of steam by condensation. Also, though a full admission of steam at higher pressures may reduce the proportion of condensation, yet whenever expansive working is attempted by cutting off earlier, the heavy back pressure and the course of the expansion line alike show that no pressure of steam, however high during the admission, can mitigate the evils of condensation in exposed cylinders.

Evidence from the Proportions of the Valve Gear.-The greater the lap of the valve, the greater also is the inside lead, for the exhaust exposed cylinder, we should expect, would therefore require longer laps than well protected ones, seeing that wet steam exhausts with difficulty. Accordingly, it has been found that in Sharp's inside cylinder engines, on the Edinburgh and Glasgow Railway, which have only a 3-inch lap-probably the shortest lap in present practice for a 15-inch cylinder-the exhaust is as perfect as in the Caledonian passenger engines with 1-inch lap for the same cylinder. Further, in inside cylinders with clean boilers, it is practically a matter of indifference what amount of wear or slugger may have taken place in the valve gear, so far as concerns the exhaust: in outsides, on the contrary, it is a very important object to maintain the gearing in the highest order, so as to keep up the inside lead, as the wear of the gearing directly reduces the lead, and thereby increases the back pressure. The Caledonian is perhaps the first line in this country on which the special advantage of long lap for outside cylinders was experienced.

The formidable degree of condensation which accompanies high expansion in partially protected cylinders, accounts for the opinion held by men of experience of the inutility, for economical objects, of cutting off the steam earlier than at half stroke, for the proved advantage of expansive working in inside cylinders is neutralized in outsides by the condensation. Mr. Buddicom, of the Rouen Railway, led the way in the re-introduction of outside cylinders in this country: and to this day, he and some of his followers have adhered to the fixed gab motion.

Conditions on which the expansive working of steam in locomotives may be carried out with efficiency and success.-The first condition is to perfectly protect the cylinders, and to maintain them at a temperature at least as high as that of the steam admitted to them. Simple non-conducting envelopes are not sufficient; external supplies of heat must be employed, and the application of a steam jacket to the cylinder would be advantageous, when other sources of heat are not readily available. The writer tried an experiment with the Orion, Edinburgh and Glasgow Railway, which has its cylinders suspended in the smoke box, like the Great Britain's, in which, by the use of partitions, the hot air from the tubes was directed entirely round the cylinders, previously to its emerging by the chimney: but he could not detect the slightest change in the performance of the engine, probably because the hot air was really very little hotter than the steam, and the closer contact made no difference. For cylinders already well protected, more thorough modifications would be required to make a sensible improvement. The steam should also be surcharged, previously to entering the cylinder, by passing over an extensive heating surface, deriving its heat from the atmosphere of the smoke box, or, if necessary, from a hotter source.

Mr. W. C. Hare, of Stonehouse, Devon, to try the value of surcharging the steam, experimented on a small engine, with cylinder 3×8 inch stroke, and a boiler of 9 feet heating surface. He found that when the steam was passed over a surcharging surface of 5 feet in a coil of copper tube, and heated to 400° before entering the cylinder, the consumption of water from the boiler was three gallons per hour; and when the

communication with the surcharging pipe was cut off, and the steam led directly to the cylinder, the water used amounted to six gallons, or twice the other, while doing the same work, and involved a great increase of fuel consumed. To effect the reduction here noted, it appears that a surcharging surface equal to fully one-half of the heating surface has been necessary; and it is probable that for locomotives a considerable allowance must be made to produce a very decided change. The results of this experiment show that very much has yet to be done before the capabilities of the locomotive are fully developed.

As steam has been found so very sensitive to exposure on the one hand, and to surcharging on the other, it would probably be of advantage to lead the hot smoke round the barrel of the boiler and the fire-box, or the barrel only, previously to its discharge by the chimney.

The second condition of successful expansive working in locomotives is the combination of a sufficiently high boiler pressure of steam, with suitable proportions of cylinder and driving wheel, to admit of highly expansive working consistent with the required duty of the engine. It is probable that 150 lbs. per inch is about the highest pressure at which it is advisable to work a locomotive, consistent with the fair working and durability of its parts. The maximum pressure being settled, and it being assumed that the same pressure is to be maintained in the cylinder during admission, the degree of expansion to be adopted determines the capacity of the cylinder to develope the necessary average power. Long strokes are not advisable on the score of stability, at least for outside cylinders, and large diameters should rather be adopted; for the same reason, large wheels are preferable.

Thirdly, in the details of the mechanism, the cylinder should be arranged to have the shortest practicable steam-ways; as, for short admissions, a long steam-way deducts very much from the efficiency of the steam. Such an arrangement would be greatly promoted by the introduction of balanced valves, or such as have provision for preventing the steam pressure on the back of the valves; as, by being balanced, they could with facility be made large enough to embrace the whole length of the cylinder. The loads which ordinary valves are forced to carry on their backs are enormous; and though there is certainly no momentum in these loads to contend with, yet the friction of surfaces due to the loads is very great, even at the most moderate computation.

Discussion. Mr. Stephenson (the Chairman) observed that he felt much obliged to the author of the paper for explaining in such a clear and practical manner the action of the slide valve and the link motion; and the paper was particularly valuable for the actual numerical results that were given so completely of the variations in practical working, showing the improvements effected and the defects avoided.

Mr. M'Connell agreed that the link motion was the most advantageous and useful of any valve motions known for locomotive engines. He thought a hot air chamber should be contrived, passing round the cylinders, to be kept at a temperature sufficient to maintain the steam perfectly dry.

Mr. Clark said, that in the Great Western engines, Mr. Gooch had

carried the steam pipe straight down in front of the tubes, instead of curving it on one side as usual; and the pipe being of 3-inch copper, it absorbed the heat from the tubes rapidly, and surcharged or dried the

steam.

Mr. Stephenson observed, that with regard to the question of surcharging steam, he remembered being told by Mr. Trevithick of an experiment which he made in Cornwall in 1830. He had to repair an old engine there, which had no steam jacket to the cylinder, as most of the other engines had, to keep up the pressure of the steam; and he built a brick casing round the cylinder, leaving an air space all round, and applied a small fire to keep this air heated. About one bushel of coals in twenty-four hours was consumed in heating the cylinder, and he found at great increase was effected in the duty performed by the engine, with the same consumption of fuel under the boiler as before. He then removed the fire from the cylinder, in order to find the relative efficiency of the coal when consumed under the boiler or under the cylinder, and he found that it took five bushels of coals applied to the boiler to produce the same effect as the one bushel of coals applied to the cylinder. Mr. Stephenson said, he had been so much impressed with the results of this experiment, that in the Planet, one of the early locomotives made in 1832, he had the cylinders carefully inclosed inside the smoke box, instead of being outside, and there was found to be a considerable increase of power effected by the plan. That was the first locomotive constructed with heated cylinders, and it appeared the principle ought never to have been deserted; but it was singular how temporary prejudices sometimes caused a good thing to be departed from. Those inside cylinders were abandoned because the crank axles were found liable to break; but then, after that objection was subsequently removed by improved manufacture, the prejudice against the inside cylinders still remained; however, they appeared now to be going back to them. The construction of locomotives was still perhaps much influenced by these local prejudices arising from individual circumstances; and he was confident that this Institution would conduce greatly to the removal of them, by the mutual interchange of ideas and experience that was promoted by it; and nothing could assist more in forwarding such a desirable object than the reading of such papers as the present one by Mr. Clark. He quite agreed with the opinion stated in the paper on the great drawback to the application of expansion in locomotive engines caused by the condensation, from the cylinders not being heated; he considered some additional heat was required to be supplied during expansion to prevent condensation taking place, as it appeared the quantity of heat in steam was not sufficient to maintain the whole in the form of steam during expansion, but a portion returned to the form of water, as shown in the able investigation of the expansion of steam given in Lardner's "Treatise on Heat."

Mr. Cowper described some experiments that had been made by Mr. Siemens and himself, which he thought showed that condensation did not take place during expansion. They took a cylindrical tin vessel closed at the top, about 12 inches high and 2 inches diameter, the metal of which was very thin, and coated thickly with felt outside to prevent any loss of heat. A small steam pipe was connected at the top, but the