The life of Isambard Kingdom Brunel, Civil Engineer Part 60

Dee 1366 Firefly 1364 Firebrand, as altered 1295 Phito 1215 Monarch 1323 Magnet 1310 Meteor 1490 Carron 1287 Average 1381

Great Western 127 Archimedes 129

This list shows that the result in the 'Great Western,' with which s.h.i.+p I have made the comparison, is in itself a favourable one, and that compared with many others the 'Archimedes' would stand much better.

This apparent superiority of the screw over the paddle as regards the resistance offered to it by the water may at first appear startling, but there is a great mistake committed in a.s.suming that the action of the screw is a very oblique action, tending rather to drive the water laterally with a rotatory motion than to push it steadily backwards.

Having witnessed and carefully observed the degree and the nature of the disturbance in the water caused by the screw, and comparing this with the violent displacement of the water by the action of paddle-boards, even under the most favourable circ.u.mstances, I no longer feel surprised.

The ma.s.s of water pushed backwards by the action of the screw appears to be very large, spreading from the screw probably in the form of an inverted cone, but there is little or no appearance of any rotatory motion, and the surface of the water is not put into rapid motion as in the case of the paddlewheel, which may be observed to impart a considerable velocity to the water, probably for a small depth only, but over a very large s.p.a.ce.

As regards the oblique action also, a great mistake appears to have been generally made, and very naturally made, by most persons when first considering the working of the screw. It is generally a.s.sumed that the inclined plane formed by the thread of the screw strikes the particles of water at that angle and with the velocity of the revolution of the screw, but it is forgotten that the screw is moving forward with the s.h.i.+p, and therefore that the angle at which the water is struck by the plane is diminished by all that much that the s.h.i.+p with the screw advances--indeed, it is evident that if the s.h.i.+p advanced the whole amount of the pitch of the screw, the screw, oblique as it appears, and rapidly as it revolves, would not strike the water at all, but simply glide through.

The angle at which any given part of the screw does in fact strike the water is only equal to the difference between the angle to which that part of the screw is formed and the angle or direction in which it moves by the compound motion of the revolution of the screw and of the forward motion of the s.h.i.+p and screw; and, contrary to one's hastily imbibed notions of the action of the screw, this angle at which the plane of the screw is driven against the particles of water, is in such a screw as that of the 'Archimedes' very nearly equal over the greater portion of the surface, diminis.h.i.+ng to nothing at the centre; and the motion imparted to the water, although perpendicular to the plane of the screw in point of direction, is small in extent or velocity, being also nearly the same over the whole surface of the screw, except close to the centre, where it is infinitely small.

In the 'Archimedes' screw, which appears to the eye so oblique, and the centre part of which would appear to act flat against the water, only causing it to revolve, the outer circ.u.mference being 18 feet and the slip 1 foot 8 inches, the angle at which this outer edge acts upon the water is only one in 11.

The total amount of motion imparted to the water at right angles to the plane of the screw by one entire revolution even at the outer edge is not quite equal to the slip, being only 167 foot. The rotatory motion is still less, the total distance to which any particle of water is displaced laterally, or at right angles to the axis of the s.h.i.+p, by one entire revolution of the screw being at the outer edge only 069 foot, and the maximum distance being in any screw only equal to half the slip, and occurring at that part of the screw where the circ.u.mference is equal to the advance of the s.h.i.+p due to one revolution. This maximum of lateral motion is 09 foot, and takes place at 099 foot, or about 1 foot from the centre. In this mode of considering the direction at which the particles of water are acted upon by the plate of the screw I have taken no notice of the effect of the friction upon the surface of the screw, which, causing to be carried with it a film of water, will modify more or less according to the degree of smoothness of the surface the effect of the screw upon the water; and towards the centre this friction, however smooth the surface may be made, will gradually become equal to, and at last greater than, the propelling effect of that part of the screw; but this defect applies only to a very small portion of the whole area of the screw, and the absence of any very violent impulse to the water in a direction approaching to a right angle with the axis of the vessel, and which has always been a.s.sumed as an unavoidable evil in the screw, will account for the absence I have observed upon of any apparent rotatory motion.

I would not pretend, however, to advance these circ.u.mstances which I have observed, or these reasonings, as arguments whereon to found an opinion of the action of the screw, the facts as proved by the experiments are what I rely upon; but it is satisfactory to be able to account for the results by circ.u.mstances actually observed, and the reasons which suggest themselves.

The effect of a propelling surface in the form of a screw, and moving at a certain velocity, as compared with an equal surface moving at the same velocity but applied in the shape of paddle-boards, having been ascertained, it remains to determine the comparative power required to give motion to that surface.

The difficulty of determining this with any degree of accuracy from any experiments which we could make on board the 'Archimedes' was very great, but considering such results as I could obtain in conjunction with experiments which I have since made in our own works, and with the results upon steamboats recorded by others, and of those of experiments made by Colonel Beaufoy on the resistance of bodies in water, I think we may arrive at approximate conclusions sufficiently accurate for our purpose, and which may safely be relied upon.

In the case of the 'Archimedes' the engines were certainly not effective well-working engines, the proportions of the gearing or wheel-work between the engine and the screw was bad--such that the engine could not attain its proper speed--the friction of the gearing (which, whether it be a source of resistance necessarily attending the use of the screw or not, I shall consider afterwards) was very great, and the surface of the screw itself, which I had an opportunity of examining out of water, was so rough as necessarily to create very much more friction than would be caused by a tolerably smooth metallic surface. With all these sources of resistance, and under these unfavourable circ.u.mstances, the power calculated for the effective pressure on the piston and without deduction for friction or other causes, which, for the sake of distinction hereafter, I shall call the gross power, was about 145 horses, the speed of the vessel being about 8? knots per hour, as actually measured by the land, and full 9 knots as measured with great care by heaving the common log, the mids.h.i.+p section being, as before stated, 122 feet, and the lines of the vessel not so good as those of fast boats; comparing this with the gross power of the 'Great Western'

engines when propelling that vessel at the same velocity, with the advantage of better lines and the other advantages arising from greater dimensions, there does not appear any such discrepancy as to indicate any loss of power by the use of the screw in the 'Archimedes'; on the contrary, the power expended in the 'Great Western' is actually as great as that in the 'Archimedes,' as compared with their relative mids.h.i.+p sections--and if any great allowance is to be made for the circ.u.mstances which I have referred to of larger dimensions and better lines, there would appear to be actually less power expended in proportion to the dimensions and form of the 'Archimedes' than in the 'Great Western.'

The results obtained with the 'Great Western,' which as regards speed are similar to those of the 'Archimedes,' are necessarily taken from experiments made when she was rather deep, and the speed thereby reduced to 79 knots; but I have compared these with results reduced by calculations from experiments at higher speeds, and I find them agree satisfactorily--indeed, at the draft and consequent immersion of paddles when in this state, I consider the 'Great Western' as very nearly at her best as regards economy of power and effect produced. I should observe that the particular experiments from which the following calculations are deduced were made with the 'Great Western' in smooth water in the Severn. I have added also some calculations deduced from data given by Tredgold as to the performance of the 'Ruby,' a good boat with immense surface of paddle-board.

The comparison stands thus:

+--------------------------------------+---------+-----------+-----+ GREAT WESTERN ARCHIMEDES RUBY +--------------------------------------+---------+-----------+-----+ ACTUAL DIMENSIONS: Mids.h.i.+p section 520 122 63 Area of board immersed 230 --- 64 Area of a disc of diameter of screw --- 26 -- RELATIVE DIMENSIONS AND POWER: Area of propelling surface, mids.h.i.+p section being = 1 0442 0213 1016 Gross power expended for one square foot of mids.h.i.+p section 1023 1026 0976 +--------------------------------------+---------+-----------+-----+-

The speed being the same, viz., 79 knots, the power expended is as nearly as possible the same in the three, and equal to one horse-power gross to one foot of mids.h.i.+p section; while the relative propelling surface in the 'Archimedes' is equal to only half that of the 'Great Western,' and one-fifth that of the 'Ruby.' This _gross_ horse-power, it will be observed, is _about_ equal to one-half a nominal horse-power.

I have made several comparisons with recorded observations made on board the 'Great Western' at different times, and with experiments made in other vessels, and I find the same result; in estimating the powers used more particularly in some comparisons with the 'Great Western,' I have taken the mean pressure as ascertained on both sides of the piston, while in the 'Archimedes' I only obtained that on the top of the piston, which appears generally to be the best, and consequently the estimate is made unfavourably to the 'Archimedes.'

Such general results are all that I could obtain from the experiments on board the 'Archimedes,' but since that time I have made some experiments upon the friction of a plate of metal in water, and have compared these results with the experiments of Colonel Beaufoy, and the conclusion I have come to is that the power absorbed by friction in a well-made screw, apart from all question of the means adopted for working it, would not be such as to interfere with its beneficial application.

The resistance created by the screw itself arises princ.i.p.ally from two sources--the resistance to the cutting edge and the tail-edge, and the friction of the surface in contact with the water. The amount of the first may of course be reduced to an unlimited extent by having a fine edge, and practically such edge ought to be much finer than that of the screw of the 'Archimedes.'

The friction upon the surface will of course materially depend upon the smoothness of that surface, and in the 'Archimedes' it was very rough, the iron being corroded at many places, with exfoliations and small holes--the corrosion arising apparently from the galvanic effect produced by the iron and the s.h.i.+p's copper.

The great number of revolutions required in the screw as compared with those of the paddlewheel, leads a person to a.s.sume, without much consideration, that a very high velocity is given to the cutting edges and to the surface of the screw, and consequently that great friction must be produced--this velocity is not, however, nearly so great as it at first appears.

In the present screw of the 'Archimedes' the velocity of the extreme point, following its oblique or spiral course, is only about three times that of the vessel, while the average velocity of either of these knife edges or of the surface is not twice that of the vessel.

Now without determining what the actual amount of these resistances may be, we can at once satisfy ourselves that it cannot be very considerable, by comparing it (which we have the means of doing) with the resistance caused by the cut.w.a.ter and any given portion of the s.h.i.+p's bottom. The resistance of a knife-edge will be about as the square of the velocity, and if we a.s.sume the surface friction to increase in the ratio determined by Colonel Beaufoy--namely, at the 175th power, or as the 4th root of the 7th power of the velocity--then the resistance of the knife-edge will be equal to the resistance of a similar edge of about five and a-half times the length of the diameter of the screw moving at the same rate as the vessel, and the surface friction will be equal to that of a piece of the s.h.i.+p's bottom about six and five-eighth times the area of the screw--or, in the case of the 'Archimedes,' the additional power absorbed by the friction of the screw would be about equal to that absorbed by the friction of little more than twice the s.p.a.ce of the dead wood which had been cut out to receive the screw--while the knife-edges would be about equivalent to three knife-edges immersed in the water, of the same depth as the s.h.i.+p's stem.

The actual amount of power absorbed in driving the 'Archimedes' screw was probably about twenty horse-power gross, or from ten to twelve nominal horse-power; but I have no doubt that a screw of similar diameter and in good condition would not absorb half that power: and this amount may be still further, and very much reduced, by increasing the relative size of the screw to that of the s.h.i.+p, and thereby reducing the slip, and proportionately reducing the number of revolutions required.

The great extent to which this is capable of being carried will at once be seen when I state that if the s.h.i.+p's progress were made to be 7 feet instead of 6 feet to each revolution of the screw, which a very slight increase of diameter and pitch of screw would effect, the power absorbed in driving the screw would be diminished in the ratio of the

6^{2}^{4} v6^{7} to 7^{2}^{4} v7^{7}--that is, as 6-15/4 to 7-15/4, or about as 3 to 2.

I must repeat here the observation I have previously made, and remind you that these calculations are not introduced as _proving_, but merely as _explaining_, that which appears to me proved by the general results of the experiments on the 'Archimedes'--namely, that the effect produced was, considering all the circ.u.mstances, fully proportionate to the power expended, while the experiments and calculations which I have since made also satisfy me that these results may be very much improved upon.

As regards the first of the two heads under which I stated that I proposed to consider the subject, namely, the mere efficiency of the screw as a propeller, I think but one conclusion can be drawn from the results of the experiments quoted, and that is, that as compared with the ordinary paddlewheel of sea-going steamers, the screw is, both as regards the effect produced, and the proportionate power required to obtain that effect, an efficient propeller.

I limit the comparison to the ordinary paddlewheels of sea-going steamers, first, because those are the circ.u.mstances which _we_ have alone to consider; and, secondly, because it is _possible_, by increasing the diameter and breadth of the paddles, which, for the attainment of an adequate object is practicable to any extent in a mere river boat, to render the action of the common paddle all but perfect, and probably more effective than any other propeller.

In considering the advantages and disadvantages likely to attend the use of the screw propeller, I will, commencing with the latter, consider such objections as have been advanced by others, as well as those which may have occurred to myself.

The only objections, however, which I think worth consideration are:--

First. The necessity of a peculiar form of vessel.

Secondly. The situation of the screw under water, and consequently to a certain extent unseen and inaccessible, and the liability to injury from its position from grounding or in other ways.

Thirdly. The probability of its being lifted out of water when the s.h.i.+p pitched deep.

Fourthly. The difficulty of getting up the required number of revolutions, and the great defects of the mode employed in the 'Archimedes,' and the shaking caused by the machinery.

As regards the form of vessel, undoubtedly a shallow boat, intended for shallow waters, would be very unfit for the application of the screw, which would probably require a greater depth of water than the whole draft of the vessel; but I see no defect or difficulty of this description in the vessel now under consideration, nor can I antic.i.p.ate any in any vessel this Company is likely to be interested in; a clean run is the most essential condition, and I should suppose no s.h.i.+p was ever built in which this principle of form was carried to a greater extent than in our new iron s.h.i.+p. Her present form I believe to be excellent for the screw, and with a very slight dropping of the keel towards the stern, which can easily be done now without any expense, a.s.sisted by the different trim, which, as I shall presently show, will be effected by the use of the screw, the required draft of water will be attained.

It may, perhaps, be as well to mention here, that the diameter of the screw, if in the same ratio to the mids.h.i.+p section as in the 'Archimedes,' would be only 12 feet 3 inches, my friend Captain Claxton having made a mistake upon this point in calculating it at 16 feet, and that if increased only to 14 feet 4 inches, the diminution referred to in a former part of my Report of one-third in the power lost in working the screw would be effected; considering the speed we wish to attain, probably 15 feet 6 inches would be a good diameter. Upon the whole I think the vessel is as well fitted for a screw as she is for paddles, and much better adapted for either than the 'Archimedes;' but if originally intended for a screw, possibly some trifling modification in the form and construction, princ.i.p.ally of the keel near the stern, might have been introduced which would have rendered the whole a more perfect job than she would now be if altered--but the absence of this would in no way lessen the efficiency of the screw, and I cannot think that any alteration we might now be obliged to make would exceed in cost the sum of 200_l_.

Secondly, the inaccessibility of the screw and liability to damage; this appears to me the objection most plausible, but I cannot say that I attach much weight to it, particularly in the case of a vessel intended for long voyages and across the ocean. During the whole pa.s.sage in deep water I consider the screw far less exposed to injury than a paddlewheel, and that the chances of injury are so remote that even if it were quite inaccessible it would still be altogether safer than paddles, which are so much exposed; but it is by no means inaccessible, the screw may be rendered stationary at any time or during any weather, when it would be barely safe to stop the engines with common paddles, and when it would be very difficult to do anything to the paddles even if the engines were stopped, while the whole of the screw, bearings, &c., may easily be examined and felt from above, and, if necessary, men sent down with common diving jackets and hoods to replace bearings, or attach tackle to move the screw, or clear away any obstacle entangled in it. When in port I still think the chances of injury very remote; an inspection of our model will satisfy you that from the form and size of her mids.h.i.+p section the vessel cannot lay in any position in which the screw would touch the ground, while at that time the whole screw may be very easily examined and replaced without any necessity for going into dock.

Thirdly, the probability of its being lifted out of water when the s.h.i.+p pitches deep.

This appears at first to be a very natural and an unavoidable objection, but the result of observations proves that the motion of vessels, of steamers at least, is not such as to cause the apprehended difficulty.

Among the observations made on board the 'Great Western' steam-s.h.i.+p by Mr. Berkeley Claxton, under my direction, were measurements of the angles of rolling and pitching, and from these it was evident that the vessel never pitches to so great an angle as that to which she rises; such a result might indeed have been antic.i.p.ated by considering the form of the vessel forward and aft, and the circ.u.mstance that a steamer is almost invariably meeting or pa.s.sing the seas, or, if overtaken by them, is still going at a good rate, which reduces the relative speed of the sea; consequently, although the vessel may be frequently thrown up very violently forward, yet the stern, which has no displacement under water, settles down quietly and heavily upon the surface; or, considering it in another way, the variation of displacement at the stern is very rapid, falling off almost to nothing at a few feet below the water-line, and spreading out to a great extent at a few feet above, whilst forward the difference of displacement is comparatively small, the centre of motion, therefore, is thrown very far aft, and while the bows, which are also opposed to the first shock, are thrown alternately high out of water or plunged deeply into it, the stern floats nearly steady, the vessel resting on its broad counter nearly as the centre of motion: whatever may be the explanation such is the operation, not only as measured by instruments, but more particularly as observed since, practically.

In the 'Great Western' the whole cut.w.a.ter and, it is said, a considerable length of keel, is frequently seen out of water from the bowsprit, while astern it is very doubtful whether more than half the stern part was ever seen; marks have been made by my direction on the rudder to observe this; as yet the 9-foot mark is the lowest seen, and this occurring rarely, and for very short intervals.

In the 'Archimedes,' during a voyage performed in her by Mr. Guppy from Bristol to Liverpool, and during which they were exposed on more than one occasion to violent pitching, the screw (which can be watched from the deck) never was uncovered; and Mr. Smith and others on board the 'Archimedes,' whose whole conduct was such as to inspire unusual confidence in all information obtained from them, a.s.sured me that such was always the case.

In the voyage to Oporto and back, in which I sent Mr. Berkeley Claxton, he made the same observation; these facts, in conjunction with previous and subsequent observations on board the 'Great Western,' convince me that nothing is to be feared on this head; but even if the screw were occasionally to be partly exposed, I know of no evil consequences likely to ensue, as I shall clearly point out when referring to the _advantages_ of this propeller over the common paddle.

Fourthly, the difficulty of getting up the required number of revolutions, and the great defects of the mode employed in the 'Archimedes.'

Upon this point certainly the 'Archimedes' offers but a miserable example, and the result is almost enough to prejudice the mind of any person against the whole scheme; the proportions of the gearing, as I have before stated, are so bad that the engines appear, even to the eye, to labour ineffectually to get up their speed. The required speed of the screw is not nearly attained, while the noise and tremor caused by the machinery is such as to render the vessel uninhabitable, and perfectly unfit for pa.s.sengers, I should almost say for a crew. I never attached much importance to these circ.u.mstances, because I felt convinced that such a mere mechanical difficulty would by some means be overcome, if, as I confess I did not then at all antic.i.p.ate, the screw itself should prove efficient.

The most simple and effectual means of overcoming all objections on these heads always appeared to me to be by the use of straps instead of gearing; and all my experience, and I have seen a great deal of the working of machinery by straps and ropes in the numerous works executed by my father, led me to the conclusion that there existed no difficulty whatever in sending the necessary power through a rope or hemp strap, but I was hardly prepared to find the result so entirely satisfactory as it has proved to be.

In an experiment made in your works at the yard, I have sent through two small whale lines, a power equal to about one-thirtieth of that which would be required in the strap if used in the new s.h.i.+p, and this without any slip or straining of the rope which would be injurious in practice, and without any peculiar means of ensuring adhesion to the drums; so that we have ascertained beyond doubt that sixty such whale lines upon a drum of only 4 feet 3 inches diameter is adequate to our wants, but if we suppose seventy lines of superior manufacture to that used in the experiments with a perfect mode of tightening and working upon a drum of 6 feet diameter, all of which can easily be had, it will ensure the perfect and easy working of a mode of obtaining the required number of revolutions of the screw without noise or tremor. The strap in question would be only about 3 feet or 3 feet 3 inches broad, easily replaced piecemeal, and even, if necessary, without stopping the engines.