[ 777 ] XXIII .
On the Determination of the Constants of the Cup Anemometer by Experiments vnth a Whirling Machine .
By T. R , Robinson , D.D. , F.R.S. , \amp ; c. Received March .
14\#151 ; Read April 4 , 1878 .
[ Plates 66-70 .
] In a communication to the Royal Irish Academy ( Proceedings of the Royal Irish Academy , December , 1875 ) , in which I examined the experiments made by M. Dohrandt to determine the constants of the cnp anemometer by means of Robins ' whirling machine , I expressed a hope that I might have an opportunity of making similar experiments free from some influence which seemed to me objectionable .
This opportunity was soon afforded me by a liberal grant from the Donation Fund of the Royal Society .
I lost no time in having the necessary apparatus constructed by Mr. Howard Grubb , of Dublin , and I nowT propose to give the results which I have obtained with it .
( 1 .
) I was at first embarrassed by the difficulty of finding a suitable locality for the machine .
In the St. Petersburg experiments the anemometers passed in their rotation at unequal distances from the walls of the building , and were too near its roof , which seemed likely to introduce extra disturbances into the air vortex which is formed by the whirl of the machine , and which , as will be seen , is a notable element of uncertainty in these investigations .
But this was removed by Mr. H. Grubb placing at my disposal the central dome of his works at Rathmines , then recently constructed for mounting the great equatorial which he is making for the Vienna Observatory .
Its basis is a regular dodecagon , 42 feet least diameter , and 42 feet from its floor to the summit of the dome .
Round it , at 10 feet from the ground , runs a gallery 2 feet wide , with a light iron railing , from which several doors give access to workshops in which polishing and other light work is executed .
On the ground , open arches communicate with shops , in which large lathes and other heavy tools are established .
This seemed all that I could wish , but , unfortunately , I could have the use of it only till the end of June , wdien it was wanted for the erection of the equatorial .
There were , however , two inconveniences to be feared .
One was that the open arches might admit irregular currents of air ; the other was that when heavy turning or planing was going on , tremors were produced in the ground which might be expected to make the friction of the apparatus vary .
I supposed , however , that this disturbance would equally affect the motion of the anemometers and the measures of their frictions ; but 5 g 2 this influence proved to be of far greater importance than I bad anticipated .
On the other hand , the vicinity of skilled workmen , ready to make any necessary repairs or any alteration suggested by my experiments , was of great importance , and not less so the help which Mr. Grubb and his assistant , Mr. Vereker , gave me during the whole of this work .
( 2 .
) The vertical axle of the machine was placed in the centre of the polygon , supported by a wooden pyramidal framing .
It is an iron tube 3^ inches diameter and 10*4 feet long .
It turns in a collar above and a perforated step below , supported by a transverse piece of the frame .
On its top is fixed an iron box , 10 inches square and 12 inches high , open at two opposite sides .
The third side supports , on a bar 40 inches long , a counterpoise disc of lead , 40 lb. weight , set edgewise .
The fourth carries a horizontal arm of sheet steel rivetted together , of the section shown in Plate 67 , fig. 1 ; at the box 5^ inches by 3 inches ; at its outer extremity 2 inches by fths of an inch .
In this turns the axle of the anemometer between three friction wheels at each end .
These were at first of bronze , 1 inch diameter , but on May 22 the outer three were changed for others of hard steel l-^- inch .
On the top of the box is an upright , bearing a steel wire attached on one side to the counterpoise , on the other to the arm to prevent its flexure .
The box also contains a brake apparatus ( described at paragraph 21 ) and provision against centrifugal friction .
The vertical axis carries near its bottom a driving pulley 24 inches in diameter , and both it and the horizontal axis are connected by circuit interruptors with a chronograph .
The centre of the horizontal axis is 14 feet from the floor , and the plane of the anemometer 's arms is 9 '07 71 feet from the centre of rotation .
Hence it follows that THE CONSTANTS OF THE CUP ANEMOMETER .
the cups of the instruments are 12 feet from the walls , 10 feet from the railing of the gallery , and , in the lowest case , 12 feet from the floor .
The annexed sketch will show the disposition of the apparatus ( except the driving machinery ) .
A , A is the inner wall of the building ; D , Da portion of the dome ; G , G the gallery ; Y , Y the vanes of the apparatus for measuring the air vortex subsequently described in paragraph 13 .
( 3 .
) The plan of having the plane of the anemometer 's rotation vertical instead of horizontal , is liable to three objections , which will be noticed hereafter , but I was induced to adopt it on two accounts .
Had the brake apparatus been at the outer end of the horizontal arm , the centrifugal friction would have been twice as great , and the action of the brake less certain ; and secondly , the cylindric zone of disturbed air in which the cups move is little thicker than their diameter , while in the other case it would be nearly five feet across , and as the velocity of the air vortex decreases outwards , the uncertainty of its influence would be greatly augmented .
( 4 .
) From the above we find that calling N the number of revolutions of the vertical axis in S seconds , the velocity of the anemometer 's centre V=~xL-1 ( 1*58979 ) in miles per hour .
The anemometers used had arms 2 feet and 1 foot long respectively .
A Calling A the turns in S , v~- XL-1 ( 0*93288 ) is the velocity of the centres of the cups in miles per hour ; and half this for the shorter arms .
( 5 .
) The revolutions of the vertical shaft and the anemometer are recorded on an electric chronograph a little modified from Krille 's construction ; which was placed in one of the rooms opening into the gallery .
It has three tracing points , of which the left records N by transverse marks on the helical line which it traces on the barrel of the chronograph as it is moved uniformly by a screw .
This line is drawn on peculiar paper enamelled with some white composition , which is coated over with black .
The point cutting through the black exposes a white line beautiful }7 distinct on the dark ground.'* The right hand point records A similarly .
The middle point was connected with Mr. H. Grubb 's regulator , and recorded S by interruptions of the helix , so that the three elements of the experiment were closely contiguous .
The motion of the barrel was regulated by a governor clock , such as Mr. Grubb applies to his equa-torials , and its accuracy is very great , as is shown by the line of seconds ; so great , indeed , that before long I dispensed with the seconds record , each revolution of the barrel being exactly a minute .
There was attached , of course , a commutator , which brings each point separately into action .
( 6 .
) The driving apparatus is carried by a vertical frame resting on the ground and attached at top to the gallery ; it is an application of Huyghen 's maintaining power , and shown in fig. 2 .
A B is one of the top bars of this frame ; it carries the driving pulley D and the free pulleys I , C , G , H , F , and K. The endless rope ( ^ inch diameter ) passes from .
D round these .
I and K are set obliquely to the plane of the * This paper is supplied of excellent quality by Messrs. De La Rue .
others , so that the rope after leaving them and crossing , goes round the driving pulley on the vertical shaft .
On the axle of D is fixed a large drum , E , which gives motion to it , being driven by a band from a Porter 's wheel .
These pulleys are 15 inches diameter , and their spindles -fths of an inch .
To G and H weights of 60 lb. are hung to maintain the requisite tension on the driving pulleys .
If an additional weight , D , be hung to G , and the drum turned so as to raise it , it is obvious that when G descends the rope C I will be drawn up , F K drawn down , and H up ; therefore the vertical shaft will be turned with a uniform speed by a force half D ( neglecting friction ) as long as G remains suspended .
It works freely .
When the anemometer was removed from the end of the horizontal arm , Y=52 miles with D=20 lb. I made , however , a great mistake here , from motives of economy , in using manual labour as the motor of this apparatus , instead of adopting the suggestion of Mr. Grubb of constructing an additional piece , by which the power of his steam engine could have been applied to drive it with any required speed .
It would have been far less expensive than the labour proved to be ; and I could have carried my experiments to a far higher velocity .
When D exceeded 30 lb. , two men were required at the wheel ; and in one instance , where it was 105 lb. , four men were required , and with difficulty could go beyond three minutes of the exertion .
( 7 .
) Opposite the driving frame is a platform which gives easy access to the anemometer for measuring its friction , and cleaning or oiling the various bearings .
It is a trapezium 10 feet long , the longer end of which is 8 feet long and is hinged to the platform of the gallery ; the shorter end , which is 2 feet long , reaches towards the anemometer .
When in use it is supported by a strut resting on the ground ; at other times it hangs vertically , and its influence to disturb the air vortex was found to be insensible .
( 8 .
) Five anemometers were experimented on .
No. I. is of the Kew type ; its cups 9 inches diameter , its arms 24 inches long ; but instead of being of thin plates moving edgewise , it was necessary to make them of steel tube 0*5 inch external diameter , as stays could not be applied to oppose the centrifugal force , and even these were scarcely strong enough.* Its weight = 110*87 oz. In No. II .
the cups are 4 inches diameter , and the arms 24 inches ; the weight =72*5 oz. No. III .
has cups 9 inches and arms 12 inches ; weight =68*25 oz. No. IV .
has cups 4 inches and arms 12 inches ; weight = 40*5 oz. No. V. was tried as a matter of curiosity ; its cups were semicylinders with mouths 9 inches square , the planes which closed their ends set paraUel to the * The equation ( C ) given when treating of centrifugal friction enables us to compute the outward throw due to this cause .
One of the 9-inch cups weighs 12 oz. , and its arm 15*7 oz. At the highest Y which I obtained with them \#151 ; 36*29 miles this throw =124*8 oz. acting at their C G by a leverage = 17*2 Now we found that 80 oz. acting on the arm by leverage =19*5 inches deflected the centres of the cups outwards 0*275 inch .
This would make the real Y in this case 0*091 mile greater than the estimated one .
I did riot , however , think it necessary to take this into account , as there are so many other and more important cases of uncertainty .
THE CONSTANTS O THE CUP ANEMOMETER . .
plane of their rotation ; the arms ( to the centres of the square mouths ) 24 inches , and the weight =129 oz. The other parts ( beside the anemometers ) which move with the horizontal shaft weigh 178*82 oz. ( 9 .
) The method followed in experimenting was this .
The machine was put in motion with as small a D as would make the anemometer revolve ; when this had attained as uniform a motion as could be judged of by the eye , the electric circuit was completed by the commutator , and the action was continued generally for four minutes .
During this time a person watched the chronograph to guard against its failing to record , which sometimes happened from oxidation of the contact maker , or from the points becoming blunt.* I made the observations for vortex motion , to be soon described , and Mr. Yereker kept watch over the labourers .
When this observation was finished , a larger D was applied , to give increased velocity , and so up to 76 lb. , which I only passed in one instance .
Even this was very severe on the men ; the more so as the temperature of the place was often as high as 70 ' .
( 10 .
) When this series was completed a weight was hung on the brake , and a new one taken .
On account of the increased friction the anemometer would not move with the first D of the preceding series , and I adopted the plan of keeping the Y 's nearly equal .
I laid down on a card the average length of the mark made by the point which records N in each experiment of the first series ; and then set N alone to act ; I altered D till its trace was nearly as long as the corresponding one on the card .
Then the A point was made to record ; and thus another series was completed .
This was continued till the load on the brake was as large as could be used with safety to the levers in the case of No. 1 ; and with the others such as would admit of a sufficient value of v. The chronograph sheets , which are carefully preserved , were tabulated independently by two persons .
( 11 .
) Before proceeding to discuss these observations , it is desirable to consider the conditions which determine the amount of an anemometer 's motion .
This obviously depends on the impelling force of the wind and the various resistances which oppose it .
These resistances can only arise from the action of the anemometer itself on the air , and the friction of its parts ; and therefore Y , the velocity of the wind , is a function of v and F. The nature of this function cannot be determined d priori in the present state of hydrodynamics ; but a general conception of its form is easily obtained .
Suppose the wind makes an angle 6 with the mouth of a cup ( or the arm which carries it ) which is revolving with such angular velocity that the centres of the cups move with the velocity v , I have shown in the paper already referred to that the velocity R with which the wind is incident on the cup = -v/ Y3+'y3=F2Yt ; sin 6 , the negative sign belonging to the first semicircle .
The pressure of this to turn the cup=R3X\#171 ; 1 , cq * These points should have been diamonds , as in the chronograph at the Armagh Observatory , but they could not be immediately procured .
being the pressure of an unit of wind on the cup estimated in a direction normal to the arm .
But as the arm carries on the other side of the anemometer 's axis another equal cup equidistant and in a reversed position , the action of the wind on its convex surface will oppose the motion with the pressure B The actual force therefore =cq ( V2-\-v2\#151 ; 2 V-vsin 0)\#151 ; a\(V2-\-v2+ %Vvsin 0 ) = ( oq\#151 ; a\ ) ( V2+ \#151 ; 2Vvsin 0(cq + This force is opposed first by the moment of friction at the centres of the cups , secondly by the resistances depending on v2 , of which the chief are the resistance to the arms and the reaction of the cups on the air which forms an air vortex in the plane of the anemometer .
When these opposing forces balance each other through a revolution , the angular velocity of the instrument must be constant ( except for small periodical fluctuations , the effect of which is much lessened by the moment of inertia of the arms and cups ; still more so if four cups are employed ) .
If we could find the mean values in a revolution of cq ; a\ ; cq sin 0 ; a\ sin 0 , we could express this state of permanent motion by the equation aV2\#151 ; 2/ 3 W \#151 ; yv2\#151 ; F=o ( I. ) , in which F can be obtained by measurement , cq and a\ must be functions of the angles and M* ' which the resultants It and It1 make with the arm that carries the two cups , and which are given by the , .
V sin 0+v equation sm xSr= \#151 ; \#151 ; --\#151 ; - .
R or R ( 12 .
) But even in the case where the cups are at rest and v=o , we do not see our way to a determination of these functions .
In the case of the concave surface , one would naturally suppose that when 0=180 ' , the wind being parallel to the mouth of the cup can exert no pressure on it ; but so far is this from being the case that a single cup will only be in equilibrio 30 ' beyond this position , notwithstanding the pressure on the convex surface .
I cannot say how it behaves at 0=o , for then the equilibrium is unstable and the cup gets into rotation .
Yet more ; the centre of the wind 's pressure on the concave varies with 0 ; before 90 ' it is within the centre , after it outside ; and its place depends on the deflection of the air stream-lines , the law of which is unknown .
Equally uncertain is a\ ; but it is evidently a different function .
It acts through the entire circumference ; the surface which the convex exposes to the wind varies according to a different law , and the deflection of the stream-lines on it is of an entirely different character .
I thought it possible that eddies might introduce terms depending on the first powers of Y and v , but it will hereafter be shown that this is not sensibly the case ; though the expression for V implies that v should lessen a by diminishing the arc of 0 through which It is effective .
But as cq must be small at the beginning and end of the semicircle , it is possible that this influence is not important .
These considerations , I think , justify me in believing the equation ( I. ) to be a close approximation to the general conditions of anemometer motion .
If we had a series of observations in which V ' , v , and F were accurately known , we could determine by minimum squares the three coefficients a , ( 3 , and y. If on applying these to the successive values of v and F , we reproduce the values of V7 , the correctness of ( I. ) is established ; if not , the march of the errors may enable us to THE CONSTANTS OF THE CUP ANEMOMETER .
trace whether they be casual or depend on extra powers of v ; and I entertained such a hope when I entered on the experiments , but it has been but very imperfectly realised .
( 13 .
) In the first place , we have instead of real wind the transport of the anemometer with the velocity V through air which is not quiescent , but moving in the same direction with a velocity W. Therefore we must use instead of V , Y'=Y\#151 ; W. Here are two elements of uncertainty .
It is not certain that a body moving through a fluid even if this be quiescent , is equally resisted with one at rest sustaining the impulse of a current fairly uniform , much less so if the fluid be in a state of disturbance ; and secondly , though V and v are given with sufficient accuracy by the chronograph* it is otherwise with W. We cannot measure it in the actual track of the anemometer , and must reduce our measures to that track 's centre on some hypothesis ; while we may be sure that it varies in every part of the circumference described by the cups .
But even in the line of its measurement it will be found very irregular and disturbed by powerful eddies ; and besides these vorticose motions in the direction of V , there is another Z of very irregular character in a direction normal to this , so that the air moves in spirals not in circles , and instead of Y\#151 ; W , we should use V/ ( Y-W)2+Z3 , but I see no possible mode of estimating the effect of Z correctly , on account of its intermittent character .
Woltman 's fly , which was used by M. Dohrandt to measure W , seemed unknown to our opticians ; but at a latter period I was informed by a scientific friend that it was called here an air meter , and he lent me one , which I found useful .
But I could scarcely have used such a one for habitual measurement without chronographic registry , and I wished for something that would show the changes of the vortex more evidently .
Fig. 3 ( Plate 7 0 ) shows the method I adopted .
A slip of deal , J- inch square and 23 feet long , was suspended by\gt ; a fine thread from the summit of the dome ; to prevent bending , it was braced by other threads fastened 4 feet above it ; to it were suspended two of those thin caoutchouc balloons which are sold for playthings for children .
They were about 8 inches diameter and a little higher , and hung 4 feet below the rod , their centres on a level with that of the anemometer , and 14 inches outside it .
Threads connecting them with the ends of the rod prevented their being drawn in among the cups .
From this it will be seen * A second of time measures on the chronograph 0-208 .
Now the highest value which I obtained for N was one revolution in a second , for v 1*67 ; so that even in this last case x^th of a unit of A=0'012 inch , a quantity quite visible .
It must , however , be remarked that my V belongs only to the cup whose arm is perr2 pendicular to the'horizon , whose 0=90 ' .
For any other it =V(1 +^ Xcos~\lt ; 9 ) .
lor Nos. I. and II .
this is 1 +0-243 X cos2 6 .
At its maximum the addition is / T of the whole ; in its mean value through the semicircle half this .
But this increase of Y is counteracted by its obliquity to the plane of the anemoV meter motion , which lessens its power to tnrn the cups The tangent of this = j^cosd ; it \#151 ; 12 26 at maximum .
If we knew the forms of a and cl we could compute the effect of this obliquity ; but as it is , we can only say that both these disturbances are greatest when cl and cl are least , so that probably their influence may be neglected .
that the free part of the thread is 23 feet long , and its torsion force is insensible in these experiments ; after being twisted during several series when left to itself it resumed its original position so slowly that ten minutes were required for about a third of a revolution .
The lightness of the balloons and their large surface make them obedient to very slight currents , and extremely sensitive to any irregularity in its motion .
The process of observing W is this : When the experiment has lasted for about a minute , the time b of two revolutions of one of the balloons was noted by a stop-watch ( by Robert , of Paris ) whose beats are 0'2S .
From the radius of the circle described by them , we find that W in miles per hour=L x- , ( 1-94335 ) b X reduction to cross .
This reduction is made on the assumption that the velocity in any part of the vortex is inversely as the square of the distance from the centre .
This agrees as well with experiment as under the circumstances could be expected .
Thus on one occasion , at the normal distance 6=83s , when the distance was increased to 30 inches it became 104'6S .
The inverse square gives 106'2s .
Hence the factor for reduction =L-1(0'10586 ) .
These balloons were very perishable ; drops of oil were projected occasionally from the bearings at high speeds , which perforated them like shot , and by the commencement of June we had exhausted the stock in the neighbouring shops , and most of those in Dublin .
They were then replaced ( June 7th ) by vanes of tissue paper 1 foot square , stiffened at their edges by rods of deal 0T25 inch thick , and suspended from the rod by slips of deal 0*5 inch broad , 0'062 inch thick , and 4 feet long .
These moved edgewise and were kept steady by lateral slight trusses of the same material and scantling .
As the vortex reaches far above the rod these suspensions will not sensibly retard the vanes .
Their centres are 20*25 inches from the cross , but they give 6 0*003 smaller than the balloons .
Their reduction to the cross = L-1(0*14810 ) .
( 14 .
) This apparatus shows that the vortex current is very irregular : the balloons sometimes dart rapidly forwards , then move more slowly , and this not only when the anemometer is passing them ; sometimes they are violently projected upwards and outwards , showing that in addition to the general translation of the air , it is affected by both horizontal and vertical eddies of considerable power and possibly magnitude .
The vanes can only show irregularities in the circular motion , but fully confirm its lawlessness .
Under such circumstances we can have no exact knowledge of the W which exists at the cups ; the balloons give its mean value through the circumference , but it is probably greatest immediately after the cups have passed , and decreases by the eddies and the friction of the walls till they again meet it .
Any fixed meter would yield the same result , and the only proper plan would be one capable of giving the value of W at the cups for every instant .
A pressure gauge carried with the cross and connected with a chronograph promises best ; but in air so agitated the relation between the velocity and pressure would be very doubtful .
( 15 .
) The horizontal arm has little effect in producing the vortex ; in the case already THE CONSTANTS OF THE CUP ANEMOMETER .
mentioned where it was whirled for several minutes without the anemometer , scarcely any motion of the balloons was observed though V=52 .
Nor does the anemometer 's own rotation seem to have more effect , for \#151 ; does not change with a given value of V though v be reduced by brake friction to half its original amount ; it decreases , though slowly , with V. Out of many comparisons , those given in the following table may suffice :\#151 ; Table 1 .
D = 75 Brake .
V V w D Mean V. Mean W. V w No. of Observations .
0 29-81 12-79 75 29-53 2-213 13-34 8 6 29-49 12-66 65 27-15 2-128 12-75 8 12 29-81 12-20 57 25-40 2-016 12-60 8 18 29-81 12-71 46 21-66 1-825 11-87 7 24 29-97 1318 39 19-71 1-689 11-67 7 30 29-17 12-64 28 15-19 1-370 11-08 7 36 29*25 12-96 23 12-57 1-296 9-67 5 20 10-40 1-039 10-01 4 18 8-705 0-923 9-43 4 The three first columns belong to those single observations of the second set of No. I. , in which the driving weight was 75 , and allowing for the irregularities which I have described , may be considered to show that ^ is constant .
The remaining columns relate to all in the first set of No. I. , where the driving power is the same without regard to brake friction ; here the decrease of the ratio is obvious , though it might not have been evident from single observations .
For instance , the first of the eight ratios belonging to D=75 is 11*89 , the fifth is 14*30 .
( 16 .
) I obtained some additional information about the vortex by the air meter already mentioned .
It is a small windmill of eight vanes ( five would have been better ) , 2*8 inches diameter , connected by a stop and wheelwork with a series of indices registering its revolutions to a million .
It is very sensitive , though less so than the balloons or vanes .
A revolution is stated to be a foot ; but there was in its case a table of corrections which seemed to me so strange that I did not use them .
Attaching it to the horizontal arm without the anemometer , and whirling for 8m with V=17 , I found a division =0*867 foot .
On placing the meter on the platform with the necessary precautions , with its axis of rotation in the plane of the arm and in the direction of W at 14 inches from the cross , I got in 2m 212 divisions with D=35 .
This on the above scale is 1*53 foot per second , the balloons at the same time giving 1*57 feet .
It is , however , needless to turn the divisions into feet , as they serve for comparison .
When the meter was placed to measure the radial current it gave 70\#151 ; a third of the other\#151 ; but with this unexpected peculiarity : that during the two minutes it sometimes moved very rapidly and sometimes almost stopped .
In another trial with the same D at 28 inches distance the tangential velocity =162 , the radial 60 ; with D=22 , distance=20 inches , tangential 122 , radial=82 with the same irregularity .
It follows from this that the vortex motion is not , as I had supposed , in a regular spiral , but in a track of very complicated character , which it is not practicable to define .
At the same time I tried to verify with this instrument the law of inverse squares , which cannot easily be done with the balloons , as it was very troublesome to shift them .
A scale was fixed on the platform , so that the meter could be exactly placed at any distance from the cross .
Four distances were taken for each value of D , and the machine was whirled for 5m before beginning to measure , and kept in motion till the four were finished .
The irregularities are considerable , partly owing to the small size of the vanes , for even in real wind the current is far from uniform , narrow veins of great intensity being contiguous to others comparatively feeble .
The results are given below with their reduction to the cross by the factor ( 108 + dj* ( 108)2 Table II .
D=25 , Y = 1555 .
D = 30 , Y=17-82 .
D = 35 , V=19-12 .
Dist .
Meter .
Reduced .
Do .
-M .
Meter .
Reduced .
Do \#151 ; M. Meter .
Reduced .
Do .
-M .
12-25 233-0 288-9 + 11 274-0 339-7 -34-6 291-0 360*7 -1-4 20-25 -38-5 i 194-5 274-3 -13-5 -57-5 216-5 307-3 + 2*2 -40-5 250-5 353-2 -9-9 28-25 -15-0 179-5 285-7 -21 -24-5 192-0 305-6 + 0*5 -1-5 243-0 386-8 +24-7 36-25 -10-0 169-5 302-4 + 14-6 -42-5 149-5 .
267-9 -37-2 -48-0 195-0 347-9 \#151 ; 14*2 287-8 o-oo 305-1 \#151 ; 17-3 362-1 -0-2 The irregularities of the differences show that the errors of the formula are merely casual .
( 17 .
) Another question occurred to me : Does W go on increasing for any considerable length of time % To ascertain this a set were recorded during ten successive minutes as soon as the anemometer got into full motion ; the s are separated by the time of one revolution ; the circumstances were very favourable , and the results are a good specimen .
THE CONSTANTS OF THE CUP ANEMOMETER .
Table Til .
No. N. A. b. V. V. w. I. 28 45 78-8 18T5 6-43 1-421 II .
28 46 18-15 6-57 III .
2817 45-5 71-4 18T5 6-50 1-566 IV .
V. 28-17 45-25 66-4 18-26 6-46 1-689 28 45 18T5 6-43 VI .
28 45 66-4 18-15 6-43 1-689 VII .
28 44-5 18-15 6-36 VIII .
28 45 68-2 18-15 6-43 1-643 IX .
27-5 44-8 66-4 17-82 6-40 1-689 X. 27-5 44 17-82 6-28 W becomes sensibly constant after four minutes , and this time was allowed in subsequent experiments .
( 18 .
) On the whole , the estimation of W is a weak , perhaps the weakest , part of this investigation ; and what I have observed respecting this air vortex makes me doubt whether the conclusions deduced from such experiments can be fully applicable to real wind .
Still , it may be hoped that they will give useful information .
( 19 .
) Friction is an important agent in anemometric motion , and one in which I found more difficulty in the measuring than I had anticipated .
It was fourfold : ( 1 ' ) that due to the instrument 's weight ; ( 2 ' ) that caused by the application of the brake ; ( 3 ' ) that caused by the lateral pressure of the axle on the upper bearing , caused partly by the action of the wind , partly by the force which , as in the gyroscope , is required to change the plane of the anemometer 's motion ; and ( 4 ' ) that due to centrifugal force .
There are three degrees of this resistance ; the highest is that which occurs when a machine is started from rest ; the intermediate is that which acts during continuous motion , which is the one to be considered here , and the least which prevails when the parts that rub are put into a state of vibration by tapping or jarring , which may be sometimes only half the first .
Friction occasionally varies from the effect of temperature or chemical change on the oil used to lubricate the bearings , or else of dust finding its way into them ( as sometimes happened ) .
( 20 .
) Nos. 1 and 2 were measured thus .
A cast-iron disc , weighing 103T7 oz. , was fixed on the shaft in place of the anemometer ; on its circumference a groove was cut whose bottom was l'O foot diameter , in which a fine silk thread was wound ; to this , weights were appended till when the disc was barely moved they continued the motion .
This weight , divided by four , is considered to represent the moment of friction at the centre of the cups after a correction for the increase of friction due to its own pressure .
1 .
The normal friction f is larger than in ordinary instruments on account of the long shaft , its horizontal position , and the weight of the brake apparatus :* with the disc in place the moving parts weigh 282 oz. The measurements at Rathmines gave values from 100 to 120 grains ; but I adopt one which I obtained at Armagh by a better process : that used by Atwood in determining the constants of his celebrated machine .
If a weight W be hung to the disc , W \#151 ; f accelerates a mass W+M ; M being the moment of inertia of the moving parts reduced to the circumference of the disc , and we have by the law of uniformly accelerated motion ~ S being the space through which W descends in the time T. We get S most easily by taking a given number of the disc 's revolutions .
M was got by suspending the disc and its appendages in the sector tower of the Observatory with a bifilar suspension whose lines were 62'5 inches long and 3*01 inches apart .
A graduation fixed on the disc enabled me to note the arcs of vibration ; 300 complete vibrations were taken for each trial , and by the usual formula the square of the distance of the This x W centre of gyration was found .
------ \#151 ; 7 , \#151 ; gave by ten sets M= 25,773 *6 grains .
Hence we get f far more consistently than in the former method.f I obtained 72 observations with W from -|th of an oz. to 1^ oz. , and divided them into three groups to see iff varied with the velocity .
The first 24 gave114*18 ; v=l*33 ' .
The second gave / =110*41 ; t'=2*13/ .
The third/ =108'20 ; ^=2*87 .
There is a decrease , but I think it is mere error of observation , for the effect of an error in noting the time acts inversely as T3 ; | so I take the simple mean= 110*93 .
To obtain from this the f for each of the anemometers , 2 lb. were hung on the axle , and ten observations gave * In its present position with its axes vertical the normal friction is only 22 grains .
f As the rotation of the disc is liable to be disturbed at its commencement by any casual irregularity of the friction , it is better to reject this part and take two spaces immediately consecutive .
In this case it is easily shown that \#151 ; " Q \#151 ; \gt ; or ( if S'=S)=^-f-\#151 ; \#166 ; Both S and t are given by the J yW + M tt'(\#163 ; +0 tt'(t+t ' ) chronograph , of which , however , I could not then avail myself .
This method has over that which I used at Rathmines the advantage of giving the friction during actual motion over a considerable space .
If when the descent of " W has given a rapid angular motion to the disc it be detached from it , the disc will continue to revolve till stopped by the friction .
This is equivalent to supposing W=0 in the equation ; and if / be so small that it makes twenty or thirty revolutions , it is ascertained with great precision .
It is much to be regretted that this process did not occur to me at Rathmines , for it would probably have very much improved my results .
Had I then possessed my present knowledge , I would have made an arrangement to give the mean friction during each , which could have been effected by a modification of the means ( hereafter described ) used to measure a alpha .
J Some observations made with the anemometer 's axis vertical show that f does not change with v , at least within the limits of 29'6 and 7T miles .
THE CONSTANTS OF THE CUP ANEMOMETER .
increase of f\#151 ; 9'7 grains for the added pressure = 0-294 grains for every ounce , or 4'71 for every pound .
Hence for the different anemometers\#151 ; No. I. / =113T9 .
No. II .
" 101-27 .
No. III .
" 201-32 .
No. IV .
" 184-98 .
No. Y. " 118-53 .
( 21 .
) 2 .
Brake friction.\#151 ; This is produced by means shown in Plate 70 , fig. 4 .
A is the box mentioned ( 2 ) , to the bottom of which is screwed the strong brass frame b. In this turns the lever D , with its arbor E. The longer arm of this lever is connected with the piece F , slotted to give room for the shaft and one of its three friction wheels , as shown in the figure , and attached below to the wire W , which passes down through the vertical axis and its perforated step , and carries a scale-dish below .
The shorter arm of D is connected by the link G to the right angled lever HH ' , whose arbor is at L. If now a weight be placed in the dish it draws down F , and thereby D , and G raises H , makes H7 press the piece M , which turns with the arbor N , and presses the rubber It which it carries against the friction-disc C , six inches in diameter .
A duplicate of this mechanism is placed opposite ( but not shown , to avoid confusion ) , and being also actuated by the wire W , presses a similar rubber against the corresponding part of C. K is one of the counterpoises to balance the scale-dish , which , however , was much too heavy , as I had no idea what weight might be required , and it produced unnecessary friction .
The rubbers are of stout cloth , 1"25 by 3*1 6'5 0*8 inch , and show no signs of wear .
The leverage of the brake is \#151 ; X p_ ; =8'03 .
By this elegant arrangement ( for which I am indebted to Mr. Grubb , F.B.S. ) the friction , while it can be varied at pleasure , causes neither vertical or lateral pressure on the supports of the shaft .
When , therefore , the brake is made to act , it merely adds to f another friction , which can be similarly measured .
The measures , however , of this friction were much more discordant than those of f : two differing in one instance 234 grains .
This irregularity is probably caused by the tremors of the ground ( already referred to ) acting on the brake-weight ; when thrown upwards it would relax its pressure momentarily , and permit the measuring disc to revolve with a less pull than what really represented the friction .
When I measured this by Atwood 's process at Armagh , the results were very regular , but far larger than those obtained at Bath mines , except on one occasion when Mr. H. Grubb took a set very early in the morning before there was work going on .
This agreed well with them .
As , however , all the anemometer experiments were affected by this vibration , I deemed it best to use the Bathmines frictions .
But I found an unexpected fact : that the coefficient of this friction is not constant , but decreases with the pressure ; this is probably owing to the elastic nature of the rubbers , which becoming condensed by the pressure tend to act like a hard body .
On laying down the curve of this coefficient , it looked so like an equilateral hyperbola with coordinates parallel to its asymptotes , that I tried the Armagh values of it by the equation Y=B+u~irX ' \#174 ; being the load on the brake in ounces \#151 ; 200 grains , the F force required to bring the rubbers into contact ; each value of \#151 ; gives an equation of condition , combining which we get values of the constants .
Substituting these in F each , we get values of \#151 ; whose errors enable us to approximate still more by the ctxi cLub equation \#151 ; ^r^2- Thus we obtain 315*52 ; 814 ; w=6'662 .
F These give when B=0 , -= 125*28 ; when B= oo , 77*81 ; and afford values of F differing in most cases from the observed ones far less than the probable errors of the latter .
I have therefore used these values in reducing the observations .
The following table gives these : their differences from the observed ones , the number of observations , F and \#151 ; for each value of B. JD Table IV .
No. Ou Brake .
Computed Friction .
o \#151 ; c. No. of Observations .
Friction B 1 oz. 3 284-75 + 2-3 7 11303 2 6 574-3 + 13T 6 105-85 3 9 841-7 -18-2 4 96-42 4 12 1098-1 -14-0 7 93-51 5 15 1347-6 + 11-2 5 9345 6 18 1593-5 \#151 ; 15-S 5 .
89-95 7 24 2026-9 + 13T 4 86-66 8 30 2556-0 - 8-5 5 86-24 9 36 3031T + 7-5 5 85-50 3 .
A lateral friction is produced by the wind pressing the anemometer 's axis against its outer bearing .
Omitting the consideration of the arms , this pressure will be the sum of the mean pressures on the cups during a revolution , and the same reasoning as in the case of I. shows that it = eV'2\#151 ; 2/ cV/ v-f-ey2 .
If , therefore , the constants of I. be determined by the observations , the effect of this friction will merely be to diminish a and j3 and to increase y ; e and k are larger than the other , and k much less than e , so that probably 2kV'v\#151 ; ev2 is small in comparison of the first term , and the pressure P is simply = eY 2 .
This is not merely confirmed by these experiments , but a good measure of it is obtained .
In these the force which turns the vertical axle is \#151 ; the frictional resistance of the driving apparatus : this when the motion has become uniform , acting at the end of the horizontal arm=airs resistance=P at the point of THE CONSTANTS OF THE CUP ANEMOMETER .
bearing .
Now if we examine Plate 67 , fig. 2 , it is obvious that the spindle of the pulley G is pressed by the weight W+D ; that of C by the same \#151 ; friction of G , and so on to I. hrom this the tensions of the cords GC , Cl , and of U , the cord which leaves I for the driving pulley can be determined .
The same thing can be done for IT , FK , and HF .
Now tension of U\#151 ; tension of U'\#151 ; friction of vertical axle is the force which drives the axle ; and if this be worked out , supposing / equal in each pulley , and so small that powers above its square may be neglected , we have U \#151 ; \#151 ; ( 2W-|-D)xy !
When D=12 the axle sometimes moves , but seldom ; I think 11 is the limit , and as 2W=120 we have f 1=131 , anc^ therefore r 5-5 n 1*025 P P= \lt ; ( 2W+ ) X \gt ; \#187 ; 9-Q78 ' an\lt ; ^ we can ^ *s constant .
This proves to be the case as is seen in the following table , which gives the results with Nos. I. and II .
Table Y. No. I. D P t ' .
p-\#163 ; y/ 2 .
No. of Observations .
22-3 0-583 0-004278 -0-017 3 28-5 0-903 0-004464 + 0-015 4 33-8 1-174 0-004418 + 0-004 5 44 1-700 0-004531 + 0-048 6 55 2-267 0-004412 + 0-006 7 64 2-730 0-004479 + 0-047 7 74-3 3-260 0-004380 -0-018 7 105 4-843 0-004239 -0-186 1 No. II .
D P P-t'V2 .
No. of Observations .
25-75 0-760 0-001332 -0-051 4 28-8 0-917 0-001370 -0-002 5 35-8 1-277 0-001383 -0-006 6 40 1-507 0-001332 -0-062 6 48 1-906 0-001464 + 0-014 7 56 2-318 0-001441 + 0-078 1 The agreement is good ; for the 9-inch cups e=0'004400 ; for the 4-inch\#151 ; - 0'001387 ; and I think it likely that they would answer well even for real wind .
( 22 .
) The observations for determining a ( vide par .
27 ) show that the effect of may be neglected , as in them v=0 .
I take three from No. I. D V ' P P-eV'2 .
20 9-49 0-464 0-067 29 14-36 0-993 0-047 52 21-50 2112 0-079 It will be observed that for the smaller cups e is relatively larger than in the proportion of their areas .
This is , in great measure , owing to the arms , 0*5 inch diameter and 24 inches long , bearing so much greater proportion to them .
From the preceding it follows that/ , =eY'2X 4'71 .
As to what I call gyroscopic friction , for want of a better name , the pressure which produces it \#151 ; ( A\#151 ; C)(co/ \lt ; u/ // ) ; A and C being the moments of inertia round the principal axes , and a\gt ; / co/ // the angular motions round them .
C was determined , but it would be rather awkward to get A. As , however , the angular velocities are as Y and v , the effect of this friction will be merely to lessen / 3 .
( 23 .
) 4 .
In the rotation of an anemometer at the end of a revolving arm , the centrifugal force produces an outward pressure which must be resisted by some stop , and produces there a friction This pressure is given by the formula ( C ) \#166 ; p"'___ M'xG \lt ; 7 x R2 X Y2 ( not Y'2 ) , where M ' is the weight of the parts which revolve round the is of the anemometer , G the distance of the centre of gravity of M ' from the vertical axis , and R the length of the horizontal arm .
This pressure was in some of these experiments considerable ; in No. 133 it was 52 lb. ; on another occasion it broke a steel arm of No. 111 , fths of an inch thick .
As the resulting friction varies nearly as Y'2 , it might , like/ " , have been included in a ; but as I hoped that the results of these experiments might be available for real wind ( where f " has no place ) , I thought best to measure it and add it to f I determined it thus : A strong upright was fixed in the gallery , and secured by a strut to the platform ; to this was fixed a pulley 1'73 inch diameter , as nearly as the eye could judge in a line with the seam of the horizontal arm , which was my nearest guide to the direction of the axis of the shaft .
Over this passed a fine iron wire attached to the centre of the shaft , to which weights were suspended which pulled it in the direction of its length .
The friction thus produced was measured by weights placed in the cups , and the normal subtracted from this .
The tensions were corrected for the friction of the pulley , which when this line crossed it at right angles was -2^th of the load .
The only other mode of measuring this friction was the setting the arm vertical with the 12-inch disc below , placing weights on this and proceeding as in the case of / This , however , would have required the total dismounting of the apparatus and constructing a. proper stand , for which time could not be spared .
With P'"=10lb .
, F=180 grains ; with P'"=21 lb. , F=280 , each a mean of three trials .
These give " 7 in the first instance=7'032 , in the second\#151 ; 8*361 .
It seems THE CONSTANTS OF THE CUP ANEMOMETER .
from this that here the coefficient of friction increases with the pressure .
I see no mechanical reason for this , but must accept the result , for the difference is clearly marked .
The outward pressure was resisted by a single roller bearing on one side of a plate fixed to the shaft ; this must have tended to make the latter press more on its bearings , and as it acted on the plate 1 '6 inch from its centre , the friction was needlessly great .
Therefore on May 18th it was replaced by two opposite ones , larger , carried by the shaft , and bearing on the back plate of the frame O'5 inch from the centre of the shaft .
This reduced the friction by 0*6 of its first amount .
The measures were now repeated .
With P=56 lb. , F=277"5 grains ; with P=70 lb. , F=392 , each also a mean of three .
Here , however , f must not be subtracted ; for with the two rollers the pull raises the shaft sensibly from the outer bearing .
These give for the first ^77 , 5'216 , for the second 5*895 .
The simplest mode of representing these is the formula F=P ' , / a\gt ; +P/ , 'ay .
These give for the first set ( which were used till No. 79 ) jc=5'8232 ; y=0'1272 , and for the second $=2-5010 ; y=0'0509 .
All the constants of the equation ( C ) except G have been already given ; I give it here , and with it the factors for P for each instrument .
No. I. G=5'4605 P " =Y2xL~1 ( 8-56260 ) .
No. IT .
5-0232 L-1 ( 8-46338 ) .
No. III .
5-0482 L"1 ( 8-53793 ) .
No. IV .
5-0099 L"1 ( 8-40158 ) .
No. Y. 5-5965 L"1 ( 8*60018 ) .
( 24 .
) I suppose this variation of the friction is somehow connected with the rollers , but both Coulomb 's experiments and railway experience indicate that rolling friction is simply as the pressure .
( 25 .
) On the whole , I consider that these friction measures , though perhaps not so uncertain as those of " W , are yet sufficiently so to increase materially the difficulty of drawing accurate conclusions from these experiments .
I expected that increasing m by additional friction would give a wider range to the coefficients of the equations of condition , and this is so .
But experience shows that the advantage thus gained is neutralised by uncertainty when the value of v is small , for then the latter is greatly affected by the irregularities of friction to which I have already alluded , this pio-bably arises from the momentum of the apparatus not being sufficient to overcome any casual increase of friction .
But it is also possible that as the friction when a body is started from rest is greater than when it is in motion , the passage from oiie state to the other may be gradual through a certain small range of v. ( 26 .
) Lastly , I shall describe the means which I employed to determrne directly the coefficient a of equation ( I. ) and the results which they gave .
The outer end of a strong clock-spring was attached to an arm of the anemometer set horizontal , the inner end to the horizontal arm of the machine : a circle divided to 100 parts was 5 i 2 fixed on the shaft and read by an index on the arm .
I hoped the tension of this spring would be so nearly as the angle of its torsion that mere reading of the circle would give it ; but this was not quite the case , and I had to form a scale for it .
The instrument was turned through a quadrant , and held while weights were hung on the outer edge of the cup till they balanced the spring .
As the friction caused some uncertainty , the cup was a little raised and allowed to descend gently , and its point of rest read ; it was then depressed and allowed to rise , the mean of the two readings being considered the true point of equilibrium .
This was repeated for the second quadrant , and so on for three complete revolutions .
The numbers so obtained were reduced to the centres of the cups , and corrected for the angular deviation from the horizontal position .
The same spring was used for No. II .
and No. III .
with separate scales ; but for No. I. it was necessary to combine two springs .
As the equilibrium points were scarcely ever at the quadrants , they were reduced to them by interpolation , and tables formed with first and second differences , which easily gave T the tension in grains corresponding to a given 6* ( 27 .
) Now things being thus arranged , if the whirling machine be put in action the anemometer will turn and tend the spring till its elastic force balances the pressure aV/ 2 ; when this has occurred I place on the brake a weight of 6 lb. , which produces a friction far surpassing the tension ; the chronograph is then made to record N , generally for three minutes .
The machine is then stopped , the brake holds the anemometer immovable , and 6 is read off on the circle to a tenth of a division .
But the tensions thus obtained are too large : the anemometers motion is accelerated up to the point where aY/ 3=T+F ; there it has acquired a velocity which carries it on beyond this .
The force which brings it back is now T\#151 ; F , and it will rest when aV'2= T\#151 ; F.'f Even the a thus obtained must be a trifle too large , for V ' ought to include , as a component , the radial vortex-motion Z. I subjoin the measures which I have taken in the following tables .
Of the headings , D , S , N , b , Y , W , and f " ' have been explained ; C is the reading of the circle , T the corresponding tension in grains , 6 the angle made by the highest of the cups with the horizon .
* It seemed unnecessary to give these tables .
t This supposes that the moment due to the final velocity is greater than 2F , which is a minor limit of a. I have used it as giving results nearest to those obtained by minimum squares .
If the moment does not exceed F , _T + F a y'2 THE CONSTANTS OF THE CUP ANEMOMETER .
Table VI .
No. I. / = 113-3 ; two springs ; June 20 , Bar .
=2977 ; Therm. =73*5 * No. D. S. N. b. V. W. f " .
C. T. 6 .
a. I. 17 180 31-25 0 6-75 0 5-4 16-5 564 59'4 9*749 II .
18 99 39-25 0 8-48 0 10-2 21-8 638-6 785 7*145 III .
19 99 42"5 214 9 18 0-61 12-5 38-5 1035-5 48-6 12-359 IV .
20 99 46-75 214 10 10 0-61 14-6 40 1068-6 540 10*422 V. 21 9\gt ; 50'6 198-6 10-93 0-66 17-13 42 1155-4 61-2 9-917 VI .
22 99 55-7 161-6 12-03 0-81 20-9 66-5 1616-7 55-8 11-793 VII .
23 99 58-5 162-4 12-64 0-80 23-2 66-5 1616 7 55-8 10-394 VI IT .
25 99 66-5 128-8 14-37 1-01 286 88-7 2171-6 53 2 12*148 IX .
27 99 68 116-6 14-69 112 31-8 101-2 2650-9 85-9 14*868 X. 29 99 72-5 100 15-66 1-30 36-4 102-2 26947 82 1 12-327 xr .
32 99 79 886 17-03 1-47 43-55 115-1 3236-1 54-4 12*697 XII .
35 99 82-5 87-6 17-82 1-485 48-1 129-5 3712 3 75-8 13-276 XIII .
38 99 89-3 85-2 19-29 1-53 57-15 145 4041-6 72 12302 XIV .
41 99 93-5 86-1 20-21 1-51 64-15 169 4543-0 68-4 12-461 XV .
44 \gt ; 9 97 84-8 20-96 1-37 69-2 190 5156-2 54 12*978 XVI .
48 99 100-5 72-3 21-71 1-80 74-2 213 6012-6 46-8 14-669 XVII .
52 99 108-8 45-1 23-50 200 88-79 215 6090-9 54 12-714 1 The mean of the 17 = 1T896 .
They differ more than might be expected , considering that V and T are pretty certain .
The discrepancies evidently are connected with the fluctuations of W and F , and are a sort of measure of their uncertainty .
It is also evident that they follow no law which might indicate the presence of any power of V ' in the equation ( I. ) , except the squares ; the values of 9 also show the effect of the disturbing influences .
If , as is probable , the mean value of 0 is that at which a is a maximum , it is here 62''3 , and the a of No. V. should be the largest , the other should decrease as their 0 's recede from that value ; but nothing of the sort is observable .
Table VII .
No. II .
f= 101'3 ; June 15 , Bar , = 29'63 ; Therm. =62 .
No. D. I. 14 II .
15 III .
18 IV .
21 V. 24 VI .
bis .
27 VII .
bis .
30 VIII .
33 IX .
36 X. 39 XI .
45 N. b. V. W. f " .
C. T. e. a. 47 204-6 10-15 0-64 116 20-7 245 3 72'-5 1-456 55-2 214-3 11-48 0-61 15-0 30 358-7 72 0 2-045 72 152-4 15-555 0-85 28-15 50 612-3 90-0 2 229 83-2 115 16-88 1-13 33-4 53-5 658-2 77-4 2-104 95-8 91-4 20-15 1-42 48-9 96 962-0 75-6 2-310 106 86 8 22-90 1-70 64-7 r 121-5 1363-3 77-4 2-657 f 128 1428-7 79-2 2-803 114-75 81 24-79 1-61 77-2 f 129 1144-5 1460-8 1596-4 75-6 70 2 2-379 2-631 121-5 80 1 26-25 1-63 87-8 146 1607-2 75 6 2-333 138-5 64-6 i 29-92 2-01 118-5 195 2093-7 72-0 2-400 147 57-8 !
31-76 2 25 140-3 196 2103-7 75-6 2132 158-5 63-6 !
34-34 2-05 164-5 269 3003-0 68-4 1 2-634 The mean of the 13 = 2'316 .
Here also there is no appearance of the equation * The coefficients a , / 3 , and 7 depend on the air 's density .
This must he allowed for in reducing the equations .
containing any power of V but the square .
If we reduce the mean a of No. I. and No. III .
in the ratio of the areas -g-f and the air 's density , we get 2*477 , so nearly that actually found that we may fairly assume the a to be proportional to the areas of the cups .
Table VIII .
No. III .
/ =201*3 ; June 19 , Bar .
29*83 ; Therm. 67 .
No. D. S. N. 6 .
V. W. C. T. e. a. I. 18 180 52-5 120-4 11-34 1-08 17-4 67-25 1326-6 62'-l 10-496 II .
21 586 118-4 12-675 1-10 22-0 92 1918-2 61-2 12 625 III .
24 68 101 14-69 1-29 29-9 107-5 2293-9 630 11-466 IV .
27 ) ) 744 82-2 16-07 1 58 36-2 142 3118-1 64-2 13-385 V. 80 81-3 93-2 17-56 1-40 43-8 158 3477-5 61-2 12-063 VI .
33 \#187 ; 84-5 68-8 18-255 1-89 47-6 167 3650-2 61-2 12-393 VII .
36 ) ) 91-6 63-4 19-70 2-05 56-2 188 4178-2 45-2 12-571 VIII .
39 )\gt ; 99 58-6 21-39 2-22 67-4 201 4509-4 89-4 11-521 IX .
42 j ; 103-7 59-6 22-40 218 74-7 221 5074 5 75-6 11-714 X. 45 yy 107-5 64-2 23-06 2 03 79-7 248 5482-3 82-8 11-843 XI .
48 \#187 ; 115 51-4 24-84 2-53 94-4 341 6828-0 57-6 13-098 The mean of the eleven \#151 ; 12*107 ; that of No. I. reduced to the same air density = 12*383 .
Making reasonable allowance for errors , I think we may infer that the a is not changed by reducing the arms of the anemometer to half their length .
( 28 .
) I tried similar measures with No. IV .
, but had very indifferent success .
The main cause of this was the bad quality of a supply of oil which had just been sent in for lubricating the machinery , and which probably prevented the excursion of the anemometer beyond the point of balanced tension .
From the high temperature , also , it was necessary to have all the windows open .
Eleven were taken , of which from the second to the seventh , C changed only from 23 to 25*5 , though V varied from 18*1 to 28*5 .
C then passed at once to 94 , and continued increasing .
The extreme hypothesis would make a= ^/ 2 - ; it is precarious , but I give those that are not palpably wrong for what they are worth .
Table IX .
No. IV .
y=185*0 ; June 21 , Bar .
29*66 ; Therm. 75*0 .
No. D. S. I. II .
VIII .
IX .
X. XI .
15 17 33 34 39 42 180 V y9 9\gt ; yy yy N. 6 .
V. W. I C. T. 66 361 14-26 0-36 20-2 20 364-8 720 84 153 18-15 0-85 33*6 23 426-9 82-8 141 117-4 30-46 1-11 105-3 94 1967-7 68-4 150 708 32-41 1 84 121-5 96 2017-5 75-6 159 91-6 34-35 1-42 129-4 121 2594-8 89-4 159 72-3 34-35 1-80 129-4 97 2042-8 79-2 2-940 2-147 2-614 2480 2 677 2-168 THE CONSTANTS OF THE CUP ANEMOMETER .
The mean of the six=2*502 .
The a of No. II .
reduced to the present density =2-288 .
The irregularity of W is notable , as also that XI .
with a larger D has only the same V as X. As it was at this period necessary to prepare for the erection of the Vienna equatorial , I could not repeat these experiments , but the results obtained with Nos. I. and III .
leave no room for doubting that a similar agreement would be found here with No. II .
( 29 .
) No. V. was not tried this way .
The less exact method of gradually increasing the brake friction til ] the anemometer stopped , gave by two observations a= 10-400 on June 2 , Bar .
30"06 , Therm. 71 .
( 30 .
) After these preliminary details I proceed to state the experimental results , beginning with those of No. 1 .
The first 79 of these were obtained while the centrifugal pressure was opposed by a single friction roller , and the second set ( 80 to 123 ) were taken after two were applied .
Most of the headings of the tables have been already explained ; m = the V ratio of the wind 's velocity to that of the anemometer ; p is the factor to reduce the air 's density when a was measured to that at each observation .
Table X. No. I. f= 113*2 ; May 8 , Bar .
30*37 ; Therm. 52 .
No. D. S. 1ST .
A. b. Y. V. W. i Log. p. m. 1 15 240 33 39 290 5-35 1-39 0-45 6-5 0-06616 3-519 2 16 180 32-6 34-8 242 7-04 1-66 0*54 11 99 3-935 3 18 180 41-3 52 173-3 8-92 2-475 0-74 17-2 99 3-301 4 20 180 48-71 65 147-3 10*52 3-09 0-89 22-9 \#187 ; 3-115 5 24 180 60-4 85-17 122-3 13-05 405 1-095 39-2 99 2-948 6 28 180 69 103-7 104 14-91 4-94 1-26 47-1 99 2-7655 7 33 240 1033 160-2 96-2 16*74 5-72 1-36 77-35 99 2-689 8 39 240 119 191-5 81-8 19-28 6-84 1-59 101-6 99 2-587 9 46 240 1339 216-2 74 21-695 7-72 1-765 138-3 99 2-582 10 52 180 106-33 181-5 67-5 22-97 8-64 1-93 158-6 99 2-435 11 58 180 113 198 63-5 24-41 9-425 2-05 1 184-7 \gt ; 7 2-372 12 66 180 123-5 216-9 58 26-68 10-32 2-25 ; 229-95 .
99 2-366 13 75 240 178-5 310 53-6 28-92 11-07 2-43 296-7 99 2-393 B=3 oz. ; \#151 ; 398*0 .
No. t\gt ; .
S. N. A. b. Y. V. W. Log. p. m. 14 18 240 42-5 19-5 159-8 6-89 0-70 0-815 10-5 0'06616 9-872 15 20 240 56-58 51 67 101-6 9-17 1-845 1-28 18-9 99 4-272 16 20| 240 59-9 64-12 110-8 9 71 2-18 1-18 21-4 99 3-955 17 23i 240 71-67 82 95-6 11-61 2-93 1-36 32-0 99 3-502 18 25 240 84-8 114-5 .
91 13-74 4-09 1 43 45-9 J* 3-011 19 28 240 93-5 1314 98-8 15-15 4-69 1-32 57-8 99 2-948 20 33-2 180 80-3 111-8 84-4 17-35 5-62 1-515 79-9 99 2-8135 21 39 180 91 142-5 81-8 19-66 6-78 1-64 107-5 99 2-656 22 46 180 100-9 166-8 74 21-80 7-93 1-83 139-9 99 2-519 23 52 180 108 180 67-5 23-33 8-57 1-95 165-7 91 2 496 24 58 240 1515 263-9 63-5 25-09 9-42 1-97 219-5 99 2-454 25 66 240 169-7 297 58 27-49 10-60 2-14 258-8 99 2-391 26 75 240 184 332-3 53-6 29-81 11-51 2-20 323-4 99 2-400 B=6 oz ; /+B'= 687-6 ; May 30 , Bar .
= 30-20 ; Therm. 61 .
No. D. S. N. A. b. Y. V. W. Log. p. m. 27 17-5 240 56-45 24-85 130 9-15 0-89 1-005 19-4 0-03873 9179 28 18 yy 60-9 35-5 115-4 9-87 1-27 1-13 222 yy 6-894 29 20 y\gt ; 69 6 57-5 110 11-28 2-05 1-185 30 yy 4-916 30 25 yy 88-1 98-4 95-4 14-27 3-51 1-37 52-3 yy 3-674 31 27 yy 108 127 86-4 17-50 4-53 1-51 70-8 yy 3-527 32 38 yy 120-1 173-3 75 2 1947 6-19 1-73 95-5 yy 2-8665 33 45 300 168 253 73-4 21-78 7-23 1-78 139-5 yy 2-768 34 52 240 148 228 9 75 23-98 8-17 174 178 yy 2-722 35 58 yy 158-5 254-5 64 2 25-68 9-09 2-03 214-4 yy 2-603 36 65 yy 168-5 272 61-8 27-30 9-71 211 276-8 yy 2-594 37 75 sy 182-5 302 584 29 525 10-78 2-235 316-8 y 9 2-531 B=9 oz ; y\gt ; +B/ =955*0 ; May 15 , Bar .
= 30*16 ; Therm. 58 .
No. D. S. N. A. b. V. V. W. Log. p. m. 38 19-5 180 49-25 9-5 133 10-64 0-45 0-98 26-3 0-04636 21-579 39 24 240 7475 399 102-6 12-11 1-42 1-27 35 yy 7-688 40 28 v 92-9 85-4 90-8 1505 305 1-44 56-9 yy 4-465 41 31 yy 99 25 111 85-2 16-08 3-96 1-53 66 yy 3-6715 42 42 yy 126 175-3 70-6 20-42 6-26 1-85 118-8 yy 2-968 43 46 yy 136 192 3 77-8 22-04 6-865 1-68 143-9 yy 2-966 44 55 yy 148 227-5 65-4 23-98 8-12 1-995 178 yy 2-707 45 59 180 124-5 195-5 67-2 26-89 9-31 1-941 243-5 yy 2-682* 46 65 240 166-0 265-5 60-8 26-90 9-48 2-15 273-9 yy 2-611 47 75 yy 182-5 298-5 57-2 29-57 10-66 2-28 315-9 yy 2-561 B=12 oz. ; /+B'= 1211*3 .
No. D. S. N. A. b. V. V. W. Log. p. m. 48 25 240 82 37'6 102-6 13-29 1-34 1-27 46-9 0-05971 8-950 49 30 yy 98-4 86-2 95-6 15-95 3-08 1-365 65-9 yy 4-741 50 32 180 76-9 70-7 86 6 16 61 3-365 1-51 72-6 yy 4-488 51 40 240 120-5 140-7 74-8 19-525 4-84 1-74 105-6 yy 3-670 52 46 yy 131-7 166-2 68-4 21-34 5-93 1 91 132 yy 3-376 53 54 yy 147 203 5 68-8 23-82 7-265 1-92 174-9 yy 3014 54 58 yy 155-1 216 65-8 25 13 7-71 1-98 202 yy 3 003 55 65 yy 166-5 249-8 66 26-98 8-92 1-98 243-5 yy 2-804 56 75 \#187 ; 183-3 2814 62-8 29-70 1005 2-08 3198 99 2-750 * At the end of this it was noted that one of the cups was disfigured by something striking it ; 45 must therefore be rejected : it is given here to show how little v was affected by a considerable deformation of the hemisphere .
I am not certain but the accident may have occurred earlier , the cup was immediately restored to shape , but the refixing it put it a little out of balance , which was not corrected till the 16th .
Some dirt also had got to the outer friction wheels , so that Nos. 46 and 47 and 73 to 79 were not of equal weight with the others .
THE CONSTANTS OF THE CUP ANEMOMETER .
B=15 oz. ; /+B'= 1460*9 ; May 12 , Bar .
= 30T2 ; Therm. 53 .
No. D. S. N. A. b. V. V. W. Log. p. m. 57 28 240 92-8 34-2 93 15 04 1-23 1-40 56-7 0-05971 11-095 58 31 55 98-4 57-9 85-2 15-95 2-07 1-53 65-3 6-975 59 39 55 119-4 124 5 76-8 19 35 4-445 1-70 103-3 3 971 60 45 55 133-75 160 69 21*67 5-71 1-89 137-8 3-463 61 52 5\gt ; 148-25 193 70-6 24-02 6-96 1-85 178-8 3-185 62 56 55 152-8 207-4 64 24-76 7-405 2-04 194-1 3 068 63 64 55 165-45 234 60-6 26-805 8-35 2-53 241-3 2-951 64 75 5\gt ; 181-2 267-5 60-2 29-36 9-55 2-17 309-4 yy 2-8475 * B : = 18 oz. ; j-\~B ' =1707*9 ' May 13 , Bar .
= 30'34 ; Therm. 55 .
No. D. s. N. A. b. Y. 1 w. / " ' .
Log. p. m. 65 28 180 72-7 33-3 97 14-00 1-585 1-345 48 0-05725 7-9825 66 31 240 100 56-2 91-9 16-20 201 1-42 67-8 yy 7-367 67 39 55 120 105-5 81-4 19-44 3-77 1-60 104-5 yy 4-737 68 45 55 131-5 140-75 68-8 21-31 5 025 1 92 131 yy 3-8575 69 54 55 147-75 187 67-1 2426 6.68 1-94 183-8 yy 3*343 70 58 55 156-25 202-75 64-6 25-32 7-24 2 02 206-2 yy 3*219 71 65 55 166-75 226 60 27-02 8.07 2-17 246-5 yy 3-079 72 75 55 183 265-75 59 -6 29-65 9.49 2-19 318 yy 2-895 B=32 oz. ; / +B'=2827-9 ; May 15 , Bar .
= 30T6 ; Therm. 58 .
No. D. S. N. A. b. V. V. W. / ' " .
Log. p. m. 73 32 240 102-4 36 96 16-59 1-285 1-359 71 0-04636 11-853 74 41 yy 122-4 80 72-2 19-83 2-85 1-81 109-8 yy 6-311 75 56 yy 136 117 66-4 22-03 4-18 1-96 143-8 yy 4-804 76 l yy 158-8 161 62-8 25-73 5-85 2-09 215-2 yy 4-037 77 60 yy 160 180 62-2 25-92 6-43 2-10 220 yy 3-708 78 66 y y 169 200 63-2 27-38 714 2-07 254-4 yy 3-545 79 75 yy 185 244 61-4 29-975 8-71 2-125 328 yy 3197 A change was now made in the friction rollers , in consequence of which it seemed desirable to take another series of No. 1 .
B=0 ; / = 113*2 ; May 24 , Bar .
29-702 ; Therm. 60 .
No. 1 D. S. N. A. b. Y. V. W. / ' " .
Log. p. m. i 80 15 240 37 38 294-4 5-99 1-36 0-45 5-8 0-03423 4-088 81 18 61 81-9 148-2 9-88 2-92 0-88 9-9 yy 3-3805 82 22 79-5 116-5 107-2 12-88 4-16 1-22 16-8 y* 2-806 83 28 98 151 83-8 15-88 5-50 1-56 27-4 yy 2-605 84 35 112 185 83-8 18-15 6-60 1-56 37 yy 2-512 85 45 133-5 225 69-8 21-63 8-03 1-87 57-4 yy 2-461 86 56 153 265 59-2 24-79 9-46 2-20 81-4 yy 2-387 87 65 169-2 294-8 55-6 27-41 10-52 2-345 106-9 yy 2-383 88 75 184 324 56 29-81 11-84 2-33 134-5 yy 2-322 89 85 200 354 54-8 32-41 12-64 2-38 171 y ' 2-375 90 105 yy 224 399-6 53-4 36-29 14-27 2-44 238 0-05728 2-373 B=6 oz. ; / +B'=687*6 ; May 25 , Bar .
29*90 ; Therm. 56T .
No. D. S. 91 17 240 92 22 \gt ; \#187 ; 93 28 ff 94 33 99 95 43 if 96 54 9 9 97 63 )9 98 74 if 55-9 80 100 110 132-8 153-5 16 6 182 A. 22-7 71-75 114-25 137-6 187 234 264 294 173-6 86-4 72-2 63-4 56*6 Y. V. W. 9-06 0-81 0-75 7-8 12-96 2-56 1-17 17 16-20 4-08 1-485 27-9 17-82 4-91 1-51 35-4 21*52 6'68 1-81 56-7 24-87 8-35 2-06 82-1 26-90 9-425 2-30 101-5 29-49 10-50 2-33 131-0 Log. p. 0-05625 m. 10-251 4-602 3-609 3-3205 2-952 2-731 2-609 2-588 B=12 oz. ; / +B'= 1211*3 .
No. D. S. N. A. b. Y. V. W. Log. p. m. 99 23 240 76 26 174 12-31 0-93 0-75 15-4 0-05625 12-460 100 29 99 97-7 63 92-4 15-83 2-25 1-41 27-2 if 6-411 101 34 a 109 92 83 17-66 3-28 1-57 34-7 4-899 102 44 a 136 151 73-8 19-44 5-39 1'77 43-9 a 3'278 103 55 a 153 204 62-2 24-79 7-28 2-09 81-4 a 3-117 104 64 a 171 236 58-4 27-71 8-425 2-235 HOT a 3-023 105 74 a 184 272 53-4 29-81 9-71 2-44 134-6 \#187 ; 1 2-8185 B=18 oz. ; / +B'=1707-9 ; May 26 , Bar .
29*84 ; Therm. 58*2 .
No. D. S. N. A. b. Y. V. W. Log. p. m. 106 29 240 94 35 85-6 15-23 1-25 1-52 24-8 0-04113 10-970 107 33 108 52 85-6 17-50 1-61 1-52 34-4 a 9-941 108 44 ) ) 136 116 73 22-04 4-14 1-79 60-3 a 4-890 109 55 9 ) 154 163 63-2 24-95 5-82 2-06 82-8 a 3-933 110 64 99 168 204 60-4 27-22 7-28 2-16 105 if 3-442 111 74 99 184 242 55-6 29-81 8-64 2-32 134-5 fi 3-182 B=24 oz. ; / 4-B'=2140T .
No. D. S. N. A. b. Y. V. W. Log. p. m. 112 113 114 115 116 34 44 55 64 74 240 if if 99 99 Ill 136 156 169 185 42 84-2 140 179 220 91-6 72 63-4 61 57-6 17-985 22*04 25-28 27-38 29-975 1-50 3-01 5- 00 6- 39 7*85 1- 425 1-81 2- 06 2-14 2-27 36-2 60-3 85-8 91-5 136-6 0-05625 a 99 99 99 11-090 6*728 4-646 3-950 3-529 B=30 oz. ; / +B'=2669*3 ; May 26 , Bar .
29*84 ; Therm. 58*2 .
No. D. S. N. A. b. Y. V. W. Log. p. m. 117 118 119 120 44 55 64 74 240 99 99 99 131 152 165-5 180 40 84 129-2 175 73-2 62 62-8 56-6 21-23 24-63 26-815 29-165 1-43 3- 00 4- 61 6-25 1- 78 2'10 2- 075 2-30 54-8 79-9 100-7 126-9 0-04113 99 99 99 13-617 7-5105 5-634 4-299 THE CONSTANTS OF THE CUP ANEMOMETER .
B=36 oz. ; /+B'=3144'4 .
No. D. S. N. A. b. V. V. W. Log. p. m. 121 55 240 148 70-4 67-4 23-98 2-51 1-92 79-5 0*04113 8-774 122 64 99 165 112 63 26-73 4-00 2-17 99-9 6*155 123 75 99 180-5 152 57-8 29-25 5-43 2-26 128-4 99 4-974 Table XI .
No. II .
./ = 101*3 ; June 6 , Bar .
29'695 ; Therm. 62 .
No. D. S. N. A. b. Y. V. W. Log. p. m. 124 121 420 113-4 55-4 241-6 10-50 1-13 0-54 8-2 0-00095 8-813 125 15 240 92 71 184 14-91 2-535 0-82 18-1 9 ) 5*518 126 18 99 113-0 99-5 156-2 18-31 3-55 0-97 31-3 99 4-910 127 22 99 133 131 115-2 22-52 4-75 1-31 47-7 99 4-415 128 25 99 152 156 102-2 24-628 5-57 1-48 61-7 99 4-157 129 28 99 164 166-0 104-4 26-57 5-93 1-45 72-3 99 4-266 130 35 99 195-2 204 82 31-27 7-28 1-84 112-0 99 4-090 131 40 99 217 237 74 35-16 8-46 2-045 155-4 99 3-914 132 48 99 240-8 276-7 68-4 39-02 9-84 2-24 210-1 99 3-739 133 56 180 196 232-2 67-2 42-34 11-05 2-25 268*7 99 3*627 B=1 oz. ; 167*0 ; June 7 , Bar .
29'895 ; Therm. 63 .
No. D. S. N. - i b. Y. V. W. Log. p. m. 134 18^ 180 106 80 j 138-6 17-175 2-86 0-94 25-2 9-99883 5-685 135 23 99 135*5 125i 111-2 21-955 4-47 1-04 44-8 99 4-678 136 26 99 152-6 148i 102-2 24-725 5-30 1-27 59-9 99 4-432 137 29 99 158 157 94-4 25-60 5-605 1-38 65-65 99 4-322 138 36 99 185f i92i ; 75 30-10 6-83 1-73 100-8 99 4-154 139 41 99 156 168-5 65 34-03 8-02 2-00 141-4 99 3-992 140 48 120 115-5 129-5 64-2 37-43 9-24 2-02 236-7 99 3'833 B==2 oz. ; / +B'=280-4 .
No. D. S. N. A. b. Y. V. W. .
Log. p. m. 141 23 240 138 110| 104 22-36 395 1-25 49-75 9*9,9883 5 339 142 26 161-3 146-3 81*2 26-13 5-22 1-60 69-2 99 4-697 143 29 166 155 87 26-90 5-53 1-495 74-75 99 4-554 144 36 205 214*8 72-6 33-215 7-67 1-795 132-05 99 4-097 145 41 223-2 241 62-6 36-16 8-60 2-08 168-6 9 ) 3-9615 146 48 180 186 213-5 56-4 40-18 10-16 2-30 233 1 99 3-727 B^^3 oz. ; j-{-B \#151 ; 398 '0 .
No. D. S. N. A. i b. Y. V. W. / " ' .
Log. p. m. 147 148 149 150 151 26 29 36 41 48 246 99 99 99 180 164-5 175 206-2 225-5 187 130* 149 200 233-5 203-7 82 74 70-4 62-6 51-6 26-65 28-35 33-41 36-54 40-40 4-44 5 32 7- 14 8- 34 9- 70 1-59 1-76 1- 85 208 2- 52 730 85-7 184-3 173-7 232-8 9-99883 99 99 99 99 5-6315 5-000 4-420 4136 39065 B=4 oz. ; / '+B/ =486*2 ; June 8 , Bar .
29*745 ; Therm. 59*5 .
No. D. S. N. A. b. V. V. W. f " .
Log. p. m. 152 29 240 175-9 132 81-6 28-50 4-71 1-59 86-9 999156 5-709 153 36 if 203 187-3 60-2 32-88 6-69 2-16 128-4 " 4-595 154 41 ft 223 214 58 3613 7-64 2-24 167-8 4-436 155 48 it 245'9 253 55-8 39-835 9-03 2-33 224-9 tt 4-042 B \#151 ; 5 oz. ; / +B/ =582*5 .
No. D. s. N. A. b. V. V. W. Log. p. | m. 156 36 240 202-4 163-4 61-4 32-79 5-85 212 127-3 9-99156 5-246 157 41 ft 221 199-8 63-6 35-80 7-14 2-045 1637 ft 4-732 158 48 180 182 181-5 58-2 39-32 8-64 2 235 2150 ft 4-292 No. III .
was originally constructed with arms of steel plate |-th of an inch thick , as I thought that with a length of 12 inches there would not be much danger of flexure .
Sixteen observations were made with it in this state ; but during the seventeenth , with D = 75 , the screw which attached one of the arms to the centre piece snapped off , though 0*32 inch thick .
The arms were now made of steel tube similar to No. I. On June 8 an observation , 164 , was made with D = 35 , which gave exactly the Y of 163 ; and a v only O'071 less .
At the time I thought this difference was merely casual , and did not repeat these sixteen observations ; but it is more probable that the difference is real , in part at least , and must be kept in mind when discussing the observations .
These are given in Table XII .
, No. III .
Table XII .
No. Ill f= 201*3 ; May 30 , Bar .
29*94 ; Therm. 67*9 .
No. D. S. N. A. b. Y , V. W. Log. p. m. 159 14-5 240 35| 60 283-2 5-79 1-07 0-48 5-9 9-99927 4'964 160 18 ft 64-9 179-5 111-4 10-515 3-20 1-23 20-4 tt 2-896 161 22 ft 83 .
246 85-4 13 45 4-39 1-59 34-6 tt 2-699 162 28 tt 101 312 697 16-36 557 1-95 54-5 tt 2-589 163 35 ft 120 392 66-2 19-44 7-00 2-05 81-6 tt 2-4855 164 35 tt 120 388 66-4 19-44 6-93 1-89 81-9 tt 2-534 165 45 ft 145 7 482-8 56 23-61 862 2-42 130-7 tt 2-461 166 56 180 124 424 56 26-79 10-09 2-42 184 tt 2-415 167 75 240 195-2 676 61 31-63 12-07 2-23 290 8 tt 2-4365 B=3 oz. ; /+B/ =770*8 .
No. D. S. N. A. b. 168 17-5 240 65-3 44 103-4 169 22 ft 84 154-9 93 170 28 tt 102 249-6 81-2 171 37 tt 123-5 351-5 76 172 47 ) t 150-8 472-4 77-2 173 57 t ) 165-6 535 59-6 174 66 it 182-6 608-3 66-2 Y. V. W. f " .
Log. p. 10-58 .
0-77 1-34 20-7 9 99927 13-61 2-765 1-46 35-7 It 16-53 4-445 1-67 56 it 20 01 6-27 1-79 88 tt 24-43 8-43 1-76 143-4 5* 26-83 9-55 2-28 184-9 ft 29-59 10-86 2-05 246 tt m. 11-963 4-394 3-324 2905 2-689 2571 2-519 : THE CONSTANTS OF THE CUP ANEMOMETER .
B\#151 ; 6 oz. ; / +B'\#151 ; 1349'9 ; June 3 , Bar , 29'56 ; Therm. 61'5 .
No. D. S. N. A. b. Y. V. w. Log. p. m. 175 176 177 178 179 180 181 22 28 36 46 56 66 75 2.40 99 \gt ; \gt ; 99 \gt ; 5 99 99 83-5 102-2 124 146 164 183 198-5 66-3 165 279 393-6 481-4 570-7 631-3 118 91-2 84-8 65-2 66 59-4 58-2 13-53 1656 20-09 23-66 26-57 29 65 32-16 1-18 2-945 4-98 7-03 8 59 10-20 11-27 1-28 1-66 1- 78 2 39 2- 29 2 55 2 60 351 56-25 88-8 134-7 180 245-4 304 1 0 01060 99 99 99 99 10-245 5 060 3-676 3 028 2-826 2-6545 2-623 B\#151 ; 9 oz. ; y+B'^1884'7 .
No. D. 1 S. N. A. !
b. Y. V. W. f " .
Log. p. m. 182 28 240 100-5 45 I 99-2 16-28 0-82 1-52 54-7 0 01060 18-376 183 36 J 9 .
123-5 168-5 90-2 20-01 3-01 1-68 88 6 096 184 47 99 148 337-5 72 23-98 6-02 2-10 136-6 3-633 185 57 9 9 167 436 66-4 27-06 7-78 2-27 189-3 99 3185 186 66 99 183 517-7 !
62-8 29-65 9-24 2-41 245-4 99 2-948 187 75 99 197 585-2 56-2 31-92 10-445 2-69 297-5 99 2-798 B=12 oz. ; /+B'=2397*5 .
No. D. S. N. A. b. Y. V. W. f " .
Log. p. 111 .
188 36 240 126 142-3 94-4 20-42 2-54 1-65 92-4 0-01060 7-397 189 47 99 150-3 310-4 74 24-35 5-54 2-04 1421 99 4 0265 190 57 9 } 166-5 369 69-4 26-98 6-59 2-18 187-8 99 3 765 191 66 99 177 431 63 28-68 7-69 2-40 225-5 9\gt ; 3-416 192 76 99 198 530 57 32-08 9-46 2-65 302 1 99 3111 B=15 oz. ; /+B/ =2896*5 .
No. D. S. N. A. b. Y. V. W. f " .
Log. p. m. 193 47 240 146 195 76-4 23-66 3-48 1-98 131-5 0 01060 6-228 194 57 164 332-9 68 2 26-57 5-94 2-22 180 99 4100 195 66 179 408 61-3 29 00 7-28 2-45 232 99 3-646 196 76 99 196 496 .
60-8 31-76 8-85 2 49 294 99 3 306 B=18 oz. ; /+B'=3390-5 .
No. D. S. N. A. b. V. I V. W. f " .
Log. p. m. 197 57 240 164-4 274-3 67-6 26-64 1-90 2-24 181-3 0 01060 4-984 198 66 99 181-3 384 62-2 29-375 6-85 2-43 239-5 99 3 931 199 76 180 147 339-5 56-6 31-76 8-08 2-67 293 99 3-599 The results with No. IV .
are less reliable than any of the others ; the shorter arms and smaller cups having less power to overcome the casual irregularities ol friction .
Table XIII .
No. IV .
j6=185*0 ; June 21 , Bar .
29*66 ; Therm. 75.* No. D. S. N. A. b. V. V. W. f " .
Log. p. m. 200 14 240 827 104 184 13-40 1-86 0-71 24-3 9-96688 6-836 201 16 99 106-4 181-8 143 17-22 3-245 0-91 42-8 9 ) 5-025 202 20 79 121-8 242 3 152 19-73 4-325 0-86 58-4 99 4-364 203 18 \gt ; \gt ; 136-8 292-6 167-6 22-16 5-22 0-78 76-9 99 4 095 204* 22 99 142-8 314 111 23-14 5-60 1-17 851 9-97611 3-916 205 24 99 149-3 339 111 24-19 6-05 1-17 95-2 9-96688 3-831 206 21 99 155-5 359-8 106-1 ; 25-19 6-42 1-22 104-6 9-97149 3-733 207* 26 99 166-6 392-7 96-2 26-99 701 1-35 125-8 9-97611 3-581 208* 28 99 175-6 425-5 82-9 i 28-45 7-59 1-58 143-4 9-97149 3-539 209 28 99 181-9 439 86 29 47 7-835 1*51 157-5 9 96688 3-568 210 30 99 194-2 485-8 78 81-465 8-67 1-61 187-8 99 3-437 B=I*5 oz. ; / '-f-B'=432*0 .
June 22 , Bar .
29*95 ; Therm. 73 .
No. D. S. N. A. 4 b. Y. V. W. / ' " .
Log. p. m. 211 16 240 120 212-8 178 19-485 3-80 0-73 56-6 9-97611 4-938 212 18 99 134-9 274 167-6 21-86 4-89 0-78 74-5 99 4-290 213 20 99 151 333 153-3 24-47 594 0-85 97-7 99 3-9775 214 22 99 162-2 370-9 93-3 26-28 6-62 1-395 117-5 99 3-761 215 24 99 178 424 82-4 28-84 7-57 1-58 148-7 99 3-603 216 25-5 99 184 444-2 82 29-81 7-89 1-59 163-4 99 3-576 217 27-5 99 196 495 78-8 31-76 8-835 1-65 192-8 99 3-408 218 29-5 \#187 ; \gt ; 199 506-7 69-6 32-24 904 1-87 201-7 99 3-359 B= = 3 oz. ; ; / +B '=754*5 .
No. D. S. N. A. b. Y. V. W. / ' " .
Log. p. m. 219 18 240 125-7 99-3 133 20-365 1-77 0-98 63 9-97611 10-939 220 20 99 147-6 217 120 23-915 3-87 1-08 92-5 99 5-8945 221 22 99 160 274 97-2 25-92 4-89 1-34 113-5 97 5-027 222 24 99 172 328 92-6 27-87 5-85 1-405 135-9 99 4-520 223 26 99 187 379-6 63-2 30-30 6-775 2-06 168-9 99 4-169 224 29 99 205-7 456 79-2 33-83 8-14 1-64 227 97 3*893 B : = 4*5 oz. ; B'= 1052*0 ; June 23 , Bar .
29*68 ; Therm. 72 .
No. D. s. N. A. b. Y. V. W. / ' " .
Log. p. 1 m. 225 23 240 176 .
238 80 28-52 4-25 1-63 144-3 9*97782 6-330 226 25 99 188 304 72-6 30-46 5-43 1-795 168-8 99 5-283 227 27 99 199 342'8 67-8 32-24 612 1-92 200-7 99 4'956 228 29 \gt ; 9 208-5 394 64-2 33-785 7-03 203 226 77 4*516 * Those marked * were observed on June 22 .
There is a great confusion in the D 's .
This may have arisen from irregular friction of the vertical axes .
See the remarks preceding Table IX .
THE CONSTANTS OF THE CUP ANEMOMETER .
B\#151 ; 6 oz. ; y'-f-B'zr : 1333*6 ; June 22 .
No. D. S. N. A. b. Y. V. W. Log. p. m. 229 23 240 170-2 108 77 27-58 1-60 1-69 132-7 9-97611 16-140 230 26-5 192 224 71 31-11 400 1-83 182 7-323 231 29'5 9 ) 202 267 69-8 32-89 4-77 1-86 209 yy yy 6*511 Table XIV .
No. V.\#151 ; Cylinder Cups .
/ = 118*5 ; May 31 , Bar .
30*165 ; Therm. 67*9 ; Standard for p , Bar .
30*05 ; Therm. 71 .
No. D. S. N. A. b. Y. V. W. / " ' .
232 15 240 40 38 196-2 6-48 1-36 0-77 4-3 233 22 )\gt ; 65-2 77-3 112-2 11-05 2-76 1-36 13-3 234 28 yy 82-3 102 73-6 13-335 3-64 2-05 19-6 235 35 yy 96-8 124-5 70-1 15-68 4-445 216 29-4 236 45 yy 114-3 151-5 72 18-52 5-41 2-10 53 1 237 56 yy 129-3 173 65 20-95 618 2-33 59 238 65 yy 141 190 57-4 22-85 6-78 2-64 73-3 239 75 yy 152 208 57 24-63 743 2-65 89-7 Log. p. 0-00946 m. 4-209 3-512 3-098 3-043 3-0155 2-979 2-9595 B=6 oz. ; y+B/ =692*8 ; June l , Bar .
30*17 ; Therm. 70 .
No. D. S. N. A. b. Y. V. W. Log. p. m. 240 25 240 79 44-9 87-8 12-80 1-60 1-72 17-7 0 00442 6911 241 37 yy 102 97 87 16-53 3-46 1-74 33-5 yy 4-271 242 46 yy 117-2 124-9 65 18-99 4-46 2-33 46 yy 3-737 243 57 yy 132 150 61-6 21-39 5-355 2-46 62-2 yy 3-535 244 66 yy 143 169 53-8 2317 603 2-81 76-1 yy 3-375 245 76 yy 155-5 189-2 50-1 25195 6-75 3-02 95-5 yy 3-283 B=12 oz. ; 1216*6 .
No. D. S. N. A. b. Y. V. W. Log. p. m. 246 35 240 98 41-5 72-8 15-88 1-48 2-08 30-3 0-00442 9-316 247 46 117-4 89-3 63-8 1902 319 2-40 46-2 yy 5-215 248 57 132 121 59-8 21-39 4-355 2-53 62-2 yy 4-330 249 66 * 144 144 56-8 23-33 514 2-64 77-5 yy 4-024 250 76 yy 157 164-5 51-6 25-46 5-87 2'93 98-3 yy 3-836 B=18 oz. ; / +B'= 1715*2 .
No. D. S. N. A. b. V. V. W. Log. p. m. 251 46 240 119 49 64-4 19-28 1-75 2-34 47*8 0-00442 9-685 252 56 } ) 132 98 59-6 21-39 3-23 2-54 52-2 yy 5-387 253 66 yy 144 125 53-6 2333 4-46 2-82 77-6 yy 4-596 254 \#166 ; " 76 yy 156-3 146-6 55-2 25-32 5-23 2-74 96-8 yy 4-315 B=24 oz. ; / +B'=2145'4 .
No. D. S. N. A. b. V. V. W. Log. m. 255 56 240 131 69 61 2123 2-46 2-46 60-9 0-00442 7-618 256 66 JJ 144 105 55-4 23 33 3-75 2-73 77-6 5-763 257 76 156-8 130-5 53 25-405 4-66 2-85 97-7 4-837 i On examining these tables it is seen that m , the ratio of the wind 's velocity to that of the anemometer , is different in each of the four instruments , and varies in each .
In No. I. the extreme values'are 21'58 and 2*32 ; in No. II .
8*81 and 3'63 ; in No. III .
18'00 and 2'41 ; in No. IV .
16'14 and 3*44.* It decreases with the increased size of the cups and length of the arms , and with increased v it increases with F. On examining its rate of decrease with an increasing v it is seen to be of an asymptotic character , such that it may be expected to remain finite even when v is quasi infinite .
All these conditions are satisfied by equation ( I. ) , the positive root y / \#166 ; ?
of which gives \#151 ; =\#151 ; j , + ?
\#166 ; : = X ( III .
) .
av2 pav3 V 2 If , as is probable , / 3 and y are proportional to a , the variation of m depends upon alone ; and its limiting value=\#163 ; c+ *J % .
It follows from this that the variation of m will be less in proportion as F is less , the cups larger , and the arms longer .
( 31 .
) We can now proceed to determine the constants , and as already indicated , by combining the equations given by the observations by minimum squares .
The present case , however , is not favourable for the employment of this method , which supposes that the error of each equation , besides being small , depends solely on errors of the constants , and that their coefficients are exact .
This is by no means true here .
V ' and v are both liable to errors which are variable , and F to variable ones which may be of considerable amount , and also to casual ones even larger .
Still , the result so obtained is probably better than what would be obtained by dividing the equations into three groups and proceeding by elimination .
( 32 .
) I put the equations into the form ( II .
) , a~2\#163 ; X^-yX^2=^ ; or , a-2/ 3\#163 ; -y\#163 ; 2=r ) ( IV .
) , both to avoid the large numbers that would occur in dealing with the original form ( I. ) and to diminish the influence of errors of V ' .
The measures of a , given in Tables VI .
, VII .
, and VIII .
, show that ( I. ) contains no term of V ' ; and in the first instance I tried one but found that it led to values of V/ so much astray that the presence of it is inadmissible .
( 33 .
) As it was possible that a , \amp ; c. , might be functions of I divided the equations * These numbers explain my assuming in my original paper that the limiting value of m is 3 .
The anemometer with which I experimented had 3-inch cups , 6-inch arms , and considerable friction .
THE CONSTANTS OF THE CUP ANEMOMETER .
of* No. III .
( with which I commenced , operations ) into three groups \ the first those in which v\lt ; 5 ; the second those where it is between 5 and 9 ; the third those above 9 .
The final equations are :\#151 ; J. aX13 \#151 ; 2/ 8X 2*501 \#151 ; yX0*591 = 71-641 a= 12-117 aX 2-501 \#151 ; 2 0-591\#151 ; yX0-160= 11-226 giving / 8= 24-436 aX 0-591\#151 ; 2 0*160\#151 ; yX0 047 = 2'217 y\#151 ; -59-672 2 .
aX13 \#151 ; 3'S31\#151 ; yX 1-170\#151 ; 49-175 a= 9-593 aX 3-831\#151 ; 2 1-170 \#151 ; yX0*368 = 13'451 P= 6-568 aX 1*170 \#151 ; 2/ 8X 0-368\#151 ; yX0-021= 3782 21*554 3 .
aX14 \#151 ; 2 5-077\#151 ; yX 1-864= 29\319 CL \#151 ; 11-975 aX 5-077 \#151 ; 2/ 3X 1-864\#151 ; yX0-692== 10*037 / 3= 14-287 aX P864 \#151 ; 2fix 0*692\#151 ; yX0-259= 3*485 y= \#151 ; 3-617 All .
aX 40 \#151 ; 2/ 8X 11'409\#151 ; yX 3-625 = 150135 a= 9*472 aX 11 '409 \#151 ; 2/ 8x 3-625\#151 ; yX 1-220= 34714 / 3= 8-469 aX 3*625 \#151 ; 2/ 8x 1-220\#151 ; yX 0*427= 9*484 7= 9-814 There is here no evidence of dependence on v ; but there is very great discordance .
The values of a are most consistent , and do not differ much from my measures of it , being ( except in one number ) a little less , as I had expected , ft is more aberrant ; but the range of y is extravagant : its value in 1 is unreal , for it cannot have a negative value greater than a. In the case of this set , one cause of error is obvious .
When v is small , the moment of the anemometer is not sufficient to master these casual irregularities of friction , of which I have already spoken ; but at higher speeds their effect is much less sensible .
Another is the smallness of the coefficients of ft and y compared to those of a. In 1 the coefficients of / 3 and y compared to those of a are 0*38 and 0"045 that of a ; in 2 are 0*59 and 0'09 ; in 3 are 072 and 0T3 .
We may therefore expect a to be better determined than j3 , and j3 than y. To make this clearer by an example .
If we determine the a and y of the first group , keeping the independent terms of the three equation as symbols , we have\#151 ; a=Hx 2*46\#151 ; H'X 26 01 + H , / X 57 59 , y=\#151 ; Hx55-79+H/ X 659-19\#151 ; H"x 1564-41 , from which it is evident that errors in H will affect y very largely in comparison of a. If we suppose the probable errors of the frictions in the observations to equal those given by my friction measures at Pathnames , we can compute those of a and y due to this cause of error .
In three of the thirteen observations the PE of F\#151 ; \#177 ; 5-4 ; in three , di35-4 ; in three , ^9"7 ; in two , d=17'8 ; in one , ^47-0 ; and in the last , T:3-2 .
The errors must have the same sign in each H. Hence I find PE of a=d=r30 ; PE of y=d~39'06 ; thirty times that of a. The errors of W must also be injurious ; the more so as they affect every term of the three equations , but it is impossible to measure them , and if known the calculation of their influence would be very complicated .
( 34 .
) We cannot test surely the accuracy of these constants by trying which of them gives the best value of V ' , for it is to be noted that they , and especially y , may be changed to a considerable extent without ceasing to satisfy , at least approximately , the original equations.* Supposing F unchanged , we have for one of them . . . .
/ 7Tb | 7Tb m2Aa\#151 ; 2A/ 3xm\#151 ; Ay=0 .
Combining this with a second one we obtain Aaf\#151 ; -\#151 ; j = A/ 3 ; A y=\#151 ; Aa Xmm'(V .
) ; and the constants so increased will satisfy these two 7Tb \#151 ; |\#151 ; 7Tb^ equations .
But it will be found that , except v is very small , \#151 ; -\#151 ; and do not differ much for any other pair ; and therefore if we take their means for an entire set we shall be able to find the / 3 and y belonging to a small change of a. Twenty pair from No. I. , and as many from No. III .
, give A/ 3=Aa X 3'293 ; Ay= \#151 ; Aa X ll'463.t ( 35 .
) As y seems the chief difficulty , if it could be found a 'priori the others would be much better determined ; but no means of doing this has occurred to me .
By causing No. I. to revolve in quiescent air , convexes foremost , it was found that the resistance =v2x28 ; but this coefficient is far greater than what it would be in a current of air .
When moving down that current , whose velocity is always more than 2v , the convexes experience hardly any pressure , when against it , the pressure is largely included in the expression of a ' ( V2+f2 ) .
Therefore y must be much less than 28 \#151 ; a. We may , however , consider a as tolerably determined by the process described in paragraph 27 .
( 36 .
) In this state of uncertainty I consulted Professor Stokes , and his reply was so instructive that ( with his permission ) I annex it in an Appendix .
In it he suggested that as the equation ( IV .
) has only two variables \#163 ; and rj , it could be plotted on a plane surface ; that such plotting might give valuable information , not only as to the existence of the errors which I suppose to affect the V , \amp ; c. , but as to the equation ( I. ) really representing the conditions of the anemometer * This is well shown in the Appendix .
t The same may be done without changing a by making Are / = \#151 ; 2A/ 3 X mean m. I tried this in rather an extreme case on the constants given in paragraph 38 , changing them so that \#151 ; aq2 .
I got a value for x-y identical with that obtained on this hypothesis from thirty-three of the equations in Table XII .
It may be remarked that this supposition gives a formula analogous to M. Dohrandt 's V , =va ; 1+ This , of course , gives good results for many of the observations , but is astray both for high and low values of v. Its maximum AV'= + 5'56 ; its minimum = 4-2 .37 ; while those of Table XVI .
are +2-68 and \#151 ; IT7 .
Its probable error is twice that of ( III .
) .
THE CONSTANTS OF THE CUP ANEMOMETER .
809 motion , and possibly some information as to the value of constants .
In the parabola of equation ( IV .
) its ordinate at the origin of \#163 ; =a ; its tangent there = \#151 ; 2/ 3 ; where the curve meets the axis of \#163 ; the \#163 ; =\ is the limiting value of and the tangent there =\#151 ; 2(/ 3-b\y ) .
Plate 68 shows the plotting of No. III .
, in which \#163 ; is on a scale twenty times that of rj to prevent crowding ; the dots are marked with numbers expressing the series to which they belong .
The irregularity of their distribution is notable : least so at the higher values of \#163 ; , where for a considerable distance from the limit their general direction is nearly a right line , which would be represented by A slight curvature downwards may , however , be traced , indicating a y of small magnitude .
Near the origin of \#163 ; the dots ramble so much that no curve can be drawn through them with any certainty ; but their predominant tendency is in favour of a downward curvature .
Four of them are so far above the right line as to explain fully the enormous negative value of y derived from the first set of equations .
It is evident that this right line r]=a\#151 ; 2/ 3'\#163 ; or its primitive \#151 ; must very nearly satisfy the observations ; and , at least for the higher speeds , would suffice to give V ' in terms of v and F. This supposes that the coefficient of that part of the resistance which depends on v2is =0 , a supposition by no means improbable .
( 37 .
) Assuming y=0 , the second and third of each of these equations\#151 ; 1 .
a\#151 ; 9*999 2 .
a=ll'308 3 .
a'=ll*550 All .
a= 9*989 / 3'=11*645 \#163 ; '=12*768 \#163 ; '=13-037 \#163 ; '=10*928 These are much more consistent than the results when y is also sought .
It is obvious that nine or ten of the aberrant dots can in no wise contribute to a correct determination of a and / 3 , and therefore that observations when \#163 ; is low and rj large had better be omitted in the minimum squares .
And lastly , it may be inferred from this plotting that even if we had accurate values of a , \#163 ; , and y , we cannot expect to get very correct values of V ' .
( 38 .
) These considerations induce me to assume my measure of a , or rather 0*9 of it to be the true one , and to substitute it in the second equation .
This reduction of it is arbitrary , but I have given reasons for thinking my measures a little too large , and the general tendency of No. III .
and No. I. is to give it even smaller .
If on comparing " V ' with that given by the constants thus obtained , it is found systematically erroneous , they may be corrected by ( V. ) , combined with dV__\#151 ; rrftda + 2md/ 3 + die , " -j- , v 2(am\#151 ; /3 ) I also reject the observations where v\lt ; 5as unlikely to give reliable results .
I find from v\#151 ; 5to v\#151 ; 9 ; a=10'896 ; \#163 ; =11*360 ; 176.'r * The first and third are deduced from values of a ' and ft ' by the formula V. ; I consider them better than the direct value .
v\#151 ; 5 to v=9 , a= 10*896 ; \#163 ; =11*360 ; y=5*176 All above 9 .
10*896 11*227 5*291 All above 5 .
10*896 11*830 2*437 All . . . .
10*896 13*104 \#151 ; 6*505 10*896 11*880 + 1*600 The values of -=sc=1*0903 ; -=y=0*1468 ; and 1*8356 .
a cl ( 39 .
) The observations with No. I. constitute two series : the first ( 1*79 ) taken while there was only one roller to sustain the centrifugal pressure ; the other ( 80'123 ) after two were applied .
Plate 66 shows the plotting of the two : those of the first being represented by dots , those of the second by small circles .
There is the same irregular distribution , and it is still more difficult to trace the curvature .
It is to be remarked that the system of small circles falls below that of dots , which implies that the friction for it must have been greater than I assumed ; but for both the average curve is nearly the right line .
In applying minimum squares , my previous experience made me omit those where v was very small .
I obtained\#151 ; First set 73 .
a=12'299 ; \#163 ; =17*488 ; y=-15*372 ; a'=ll*350 ; \#163 ; '=13*401 .
Second set 39 .
a= 8*721 ; \#163 ; =12*974 ; y= \#151 ; 14*397 ; a'= 7*753 ; \#163 ; '= 8'990 .
Total 112 .
a= 11*183 ; \#163 ; =16*954 ; y=-19*862 ; a = 9*880 ; \#163 ; '=11*538 .
The second , as might be expected from the plotting , is not to be relied on ; and in all of them the negative value of y is too great .
It will be seen in the subsequent values of AY ' that those for May 26 are too small , so as to confirm my suspicion that the F which I used was too small .
This may have been owing to the engine having been stopped during part of the time , or else doing lighter work .
I have already mentioned that a set of measures taken by Mr. H. Grubb while the engine was not acting gave larger results than those in Table IY .
For B=36 the excess was 406*4 ; the AY ' for 123 would correspond to an excess of 291 .
Under the circumstances this set by itself is useless ; but as the earlier part of it seems correct , it may be retained in the total .
But I give a double weight to the first .
Using -po measured a as before .
First set 73 ... ... ... ... ... ..a=10*706 ; \#163 ; = 12*328 ; y=0*524 " reduced by ( Y ) . .
" 12*269 ; 0*937 " omitting v\lt ; 3 . .
, , 12*178 ; 0*936 Total 112 ... ... ... ... ... ... .
" 13*193 ; -2*991 10*706 12*492 \#151 ; 0*148 Hence x=1*1648 ; y=\#151 ; 0*0138 ; 2=1*3414 .
In both cases \#163 ; seems tolerably certain ; but it is not so for y , especially in No. III .
THE CONSTANTS OF THE CUP ANEMOMETER .
The confusion of the summits of the curves ( where the curvature is most influenced by y ) may account for this .
( 40 .
) The near equality of / 3 and y for Nos. I. and III .
seems to show that in these two anemometers they are nearly proportional to a. There should be a little excess in the first on account of the greater resistance of the arms , which are nearly twice as long , and instead of being ( as they should be in working anemometers ) of thin plates with sharp edges , are of J-inch tube .
This must increase both ( 3 and in No. I. more than in No. III .
Still the difference can be but very small , and I think that probably the mean values of them will satisfy both sets , within the limit of error of observation .
This may be tried by computing V ' by equation ( III .
) with the x and 2 deduced from their mean .
We have on this assumption a ; =l'1282 ; z= 1'3400 .
AY ' the observed\#151 ; the calculated V ' is given in Table XY .
Table XY .
No. AV ' .
1 -018 2 + 0-93 3 + 0-77 4 + 1-28 5 + 1-45 6 + 1-34 7 + 1-25 8 + 1-04 9 + 112 10 + 0-25 11 -0-23 12 -0-27 13 + 0-55 14 + 0-08 15 -0-38 16 -0-29 17 + 011 18 + 0-04 19 + 0-37 20 + 0-47 21 + 023 22 -0-29 23 -0-34 24 -0-61 25 -0-94 26 -0-75 27 -0-64 28 -0-65 29 -0-49 30 + 0-01 31 + 1-30 32 -012 33 -0-02 No. AY ' .
+ 0-33 -0*32 -0-22 -0-42 + 0-06 -0-03 + 0-30 -0-25 -0-44 + 015 -0-72 -0-28 -0*90 -0-83 + 0-25 +0-26 + 0-36 + 0-52 + 0-16 + 0-15 + 0-50 -0T0 + 0-02 + 1-05 + 0-94 + 0-24 + 015 + 0-23 -0*07 -0-38 + 010 AV ' .
-1-26 + 0-23 + 0-65 + 011 + 013 + 0-07 + 0-03 -018 -1-79 + 1-44 -117 -0-34 -1-08 -1-25 -113 + 0-43 + 1-01 + 110 + 1-07 + 0-99 + 0-67 + 017 + 0-26 -0-46 + 0-34 + 0-39 -0-22 + 0-80 + 111 + 119 + 1-20 + 0+1 No. AV ' .
97 + 0-40 98 + 0-66 99 + 0-42 100 + 1-47 101 + 1-55 102 -0-45 103 -0T0 104 + 1-67 105 + 0-11 106 + 0-05 107 + 1-82 108 + 2-38 109 + 2-24 110 + 1-82 111 + 1-61 112 + 1-10 113 + 2-60 114 + 2-52 115 + 2-29 116 + 2-12 117 + 111 118 + 3-45 119 + 2-79 120 + 2-77 121 + 2-41 122 + 2-86 123 + 2-94 The end of each series of a set is marked by a single line , and that of each of the two sets by a double one .
It must be remembered that each set begins with a low value of v which increases in each term ; in each also there is a different brake friction .
The values of AY ' are not very close ; but this was to be expected from the uncertainty of W and F ; and as the aberrant ones are not regularly distributed they are evidently due to casual errors .
Thus in the first series of the first set , six have large positive values ; but this is not the case in the corresponding terms of the other series , in which they are small and mostly negative , except the last , in which are four large negative and two positive .
In the second set the three first series are of the same character as the previous one ; but there was evidently some disturbing influence at work in the last four .
I have already mentioned what I believe this to have been .
The brake friction of the fourth occurs in the last but one of the first set , and the last of No. III .
, without producing such anomaly , so it is not due to error in the measure of the friction .
The results for No. III .
are given along with those obtained by supposing y=0in Table XVI .
No. AY ' .
Do .
y = 0 .
No. AY ' .
Do .
y = 0 .
159 -0-43 -0-43 181 -0-64 -0-68 160 \#151 ; 0T7 \#151 ; 0T1 161 -0-27 \#151 ; 0T8 182 ' + 2-68 + 2-68 162 + 0-08 + 0-20 183 + 1-49 + 1-20 168 -0-08 + 0-09 184 -0-09 -o-oo 164 + 0-33 .
, 185 -0-32 -0-20 165 -0-06 + 0-28 186 -0-61 -0-45 166 -Oil + 0T4 187 -1-08 -0-69 167 + 0-27 + 0-58 188 + 0-67 + 0-90 168 -0-20 -0-29 189 -0-33 -0-07 169 + 0-23 -0-25 190 + 0-33 + 0-63 170 -0-25 -0-30 191 -0-27 -0-02 171 -0-25 -0-22 192 -0-27 + 0-06 172 -075 + 0-20 173 -0-72 -0-56 193 + 0-78 + 0-70 174 -0-59 -0-03 194 -0-35 -025 195 -0-40 -0-31 175 -0-31 -0-30 196 + 0-39 -0-36 176 -0-36 -0-33 \#151 ; 177 + 1-30 -0-99 197 -0-49 -0-45 178 -077 -0-66 198 -0-49 -0-41 179 -1-03 -0-63 199 -0-46 -0-35 180 \#151 ; IT 7 -0-98 I Here also V ' is sufficiently well represented ; but it is to be noted that y= 0 gives rather better results than the other .
The probable error = ffiO'384 , the maximum = +2'68 , the minimum = \#151 ; 0*98 .
For the other the probable error =\#177 ; 0'\amp ; 55 , the maximum also = -{-2'6 8 , the minimum = \#151 ; IT7 .
I did not think it necessary to THE CONSTANTS OF THE CUP ANEMOMETER .
813 compute the entire of N o. I. on this hypothesis , but ten on which I tried it showed similar advantage .
( 41 .
) I think these results warrant us to believe\#151 ; first , that if the constants were properly determined , the equation ( III .
) with or without would give V ' with sufficient accuracy for all practical purposes ; and secondly , that for 9-inch cups the same constants avail for arms of 24 and 12 inches .
It also deserves notice that the substitution of |r-inch tube for knife-edged arms in No. III .
, makes very little difference in the value of V ' , as is seen by comparing 164 with 163 .
( 42 .
) By using the differential formulae ( V. ) and ( VI .
) to correct the constants , the mean value of AV might be considerably reduced ; but I think this unnecessary , and I hope to get more accurate values of them by another process .
The plotting of No. II .
is shown in Plate 67 .
It is far more confused than the previous two , and shows that the disturbing agents were still more powerful .
It is not possible to detect the curvatures , and we can only infer that a right line can be drawn giving the average direction of the system , while the breadth of that system shows that we may expect the range of AV ' to be considerable .
This was to be expected , as the impelling force a ( V'2+v2 ) is only one-fifth of its value in I. , while the frictions are nearly the same , therefore any irregularity on the latter will be felt more powerfully ; the same result will also be produced by the lesser momentum of No. II .
, which has less power to equalise fluctuation of motion .
That such fluctuation exists is evident from examination of the chronograph sheets on which the time of each revolution of the anemometer is recorded .
For instance , in 125 the maximum time of one revolution during the four minutes of the experiment was 4*58s , giving 1 *8 7 ; the minimum time 2'39s , v\#151 ; 3'58 ; and there were many intermediate values .
For higher speeds the difference was not so great : in 132 the maximum time was l*06s , v=8'\6 ; minimum 0*72s , 11*97 .
With 9-inch cups these variations are scarcely perceptible in the high speeds , but notable when v is very small .
Thus in 182 maximum time is 6'80s , v=0'62 ; minimum 4*50s , v=0'95 .
In such cases it is evident that the mean v cannot give the mean V ' , as m is not constant .
If the arms of the anemometer had been of sharp plate , it is not improbable that the x and z of Nos. I. and III .
might have availed here also ; but as their section is 0*87 of the area of the 4-inch cups , y , and in still greater degree / 3 , bear a much greater proportion to a. The character of the plotting prepared me for bad results from minimum squares : 34 gave a=3T00 ; / 3=9*6607 ; y=\#151 ; 31*140 , a cannot be larger than that given by my measures ; and the y would give V ' imaginary for v above 8 .
The result was not improved by omitting three of the smallest i/ s. Assuming 0 , I had a. =1*681 ; / 3'=2*602 ; both evidently too small .
I then tried , as in previous cases , nine-tenths of my measured a , and had by six various combinations , a=2'084 ; / ?=4'4065 ; y= \#151 ; 7*626 .
Still , as formerly remarked , y is too large , and avoiding it entirely I got a'=2*084 ; fi'=3*489 .
Then taking twelve distributed over the entire set , I computed the values of AV/ for them , and from these found by ( VI .
) y~ 0 8146 ; x=l'7390 ; z=2*209 .
The results are given in Table XVII .
No. AV\ No. AV ' .
No. AV\ No. AV ' .
124 + 0-54 134 + 0-75 144 -0-04 152 + 0-63 125 + 1-24 135 + 0-37 145 -0-82 153 -0-66 126 + 1-40 136 + 1-17 146 -1-09 154 -0-62 127 + 1-94 137 + 1-00 155 -119 128 + 1-35 138 + 1-26 IP7 -0-54 129 + 2T8 139 + 1-33 \#177 ; 4/ 130 + 1-95 140 -0-22 148 -1 i A + 0-87 156 + 0-38 131 + 1-81 149 \#151 ; 0T5 157 -1-40 132 + 0-63 141 -F0-33 150 \#151 ; 0'59 158 \#151 ; 151 133 -0-27 142 + 1T0 151 \#151 ; 143 143 -I-0-04 1 \J Utr Here also the entire set is nearly as well represented with y=0 as by the entire equation .
The probable error of Table XVI .
=di0'630 .
The maximum value of AV ' in it = + 2T8 ; the minimum = \#151 ; 1*51 .
In the other case the probable error = \#177 ; 0'637 ; the maximum = + l'66 : the minimum =\#151 ; 218124 .
It was observed on a different day from the rest of that series , and its difference from its neighbours seems to imply an excess of friction in the latter .
The table might evidently be improved by farther approximation .
( 43 .
) The plotting of No. IV .
, Plate 69 , resembles the other except in one notable circumstance .
The dots of the first series ( 200 to 210 ) are lower than the rest , although their general direction is nearly parallel to that of the others , This indicates that the friction during their course was greater than on the following days .
This may have been caused by bad oil , as already mentioned , and a rough measure of f taken then gave it =250 grains instead of 185 , which I computed it to be from the disc measures .
The viscidity of the oil would increase but if , asis probable , it was also applied to the pivots of the brake , it would lessen the pressure of the rubbers and the brake friction .
Both these effects seem to have taken place to such an extent that I find it impossible to derive from the observations any constants which will represent the entire set .
It was scarcely to be hoped that minimum squares could give a good result .
I tried them , omitting the first 11 , and had a=3-682 ; \#163 ; =7'541 ; y=\#151 ; 18T72 .
These are inadmissible ; a is too far above my measures , and y far too negative .
Assuming y=0 the result is no better .
Taking , as in the others , a=0'9 , the measured one for II .
, the result is still worse .
It seemed however possible that though a and / 3 were both astray , their ratio might approximate to the truth .
This gives x=1*0240 ; ai2=z=l'049 .
I computed by these AV , intending to correct them by the formula ( VI .
) .
The second series was well represented ; in the first the V ' was too small ; in the rest too great , and it was not possible by any correction of the constants to represent the .
whole .
The constants just given for No. II .
failed utterly ; but to my surprise those used for Nos. I. and III .
were not THE CONSTANTS OF THE CUP ANEMOMETER , inferior to the preceding .
I have little doubt that with the true frictions they would give correct results .
Aberrant as both are , I give the results in-\#151 ; Table XVIII .
No. AV ' Do .
I. and III .
No. AV ' .
Do .
I. and III .
200 + 0-22 -0-02 217 + 0-92 + 0-47 201 + 2'46 + 145 218 + 0-95 + 0*55 202 + 2-38 + 172 203 + 3-24 + 2-46 219 -2-83 -2-99 204 + 3-20 + 2-28 220 -221 -2-66 205 + 3-29 + 2-29 221 -1-96 -2-58 206 + 3-56 + 2-49 222 -1-37 -2-00 207 +4'01 + 276 223 -1-38 -2-35 208 -1-4-16 + 2-67 224 \#151 ; 0-44 -1-64 209 ' + 4-40 + 3-04 \#151 ; 210 + 4-50 + 2-98 225 -2-57 -3-08 \#151 ; 226 -2-47 -248 211 + 0-47 -0-23 227 -1-01 -2-80 212 -070 -1-39 228 -2-08 -2-95 213 + 0-09 -0-80 214 + 042 -079 229 -3-02 -347 215 + 070 -0-51 230 -2-81 -2-96 216 + 1-00 -0-27 231 -2-42 -3-01 If , as I suppose , the viscid oil increased fand f ' " and lessened the brake frictions , there must be a certain part of the set where these opposite effects will balance each other ; and 'this seems to occur in the second series .
The increase of F wdiich would nullify AV ' for 210 in the second system =139 ; that for 231=\#151 ; 225 ; both quite possible .
( 44 .
) No. V. cylinder cups is in marked contrast to No. IV .
Its plotting ( Plate 70 ) is more regular than any of the others .
Minimum squares give a=9T52 ; / 3= 13*416 ; y= \#151 ; 3-481 .
The a is a little less than what I got by two doubtful measures , so I retain it , and have a ?
= 1*4659 ; 2=1*769 .
The results of these are given in\#151 ; Table XIX .
No. AV ' .
No. AV ' .
No. AV ' .
No. AV ' .
232 -0-81 240 -0-24 247 -0-49 254 -0-57 233 + 0-39 241 -0-27 248 -0-56 234 -049 242 -0-60 249 -0-51 235 -0-03 243 -0-42 250 -0-33 255 -0-63 236 + 042 244 -0-62 256 -1-06 237 + 0-26 245 -0-59 257 \#151 ; 1-03 238 + 048 251 + 070 239 + 044 252 -0-36 \#151 ; 246 -1-57 253 -146 These might be considerably improved by a small increase of a ; for instance , if I had taken it =0*9 X 10*4 , the error for 257 would have become \#151 ; 0*66 .
( 45 .
) I was surprised at finding the cylinders so much more consistent than the hemispheres .
I attribute it to two causes : 1 , the cylinders are 18 oz. heavier than the 9-inch hemispheres , and their greater momentum tends to neutralise any fluctuations of friction or of the vortex 's current ; 2 , any radial currents will probably have more power to increase the resistance by acting on the convex hemispheres than they can have on the planes which terminate the cylinders .
It must be observed that in this form of anemometer the lateral friction is greater than with the hemispheres .
D=75 gave Y'=22'5 , while with No. 1 it was 27'6 .
I shall now briefly recapitulate the conclusions which , in my opinion , may be drawn from these experiments .
( 46 .
) The equation ( III .
) represents the observations from V=5 to V/ =40 and through a wide range of friction .
The discordances follow no law which might indicate a relation between V ' and v different from that expressed by ( III .
) ; they are evidently such as would be produced by casual disturbances of W and F. ( 47 .
) If the equation ( I. ) contains any other functions of V ' and v than squares and products , their effect is so small that it is masked by these disturbances .
( 48 .
) The coefficient of the impelling forces a appears to be as the area of the cups , and to be the same for radii of 24 and 12 inches .
( 49 .
) There is reason to believe that ( 3 and y are proportional to a , or that x and 2 are constant in all anemometers whose arms are of sufficient length to prevent the wake of one cup from interfering with the cup which follows it .
For Nos. I. and III .
this is highly probable ; it is not disproved by No. IV .
, which is not worse represented by it than by any other system of constants .
No. II .
requires a larger x and 2 than the others , but I think I have assigned a sufficient cause for this .
If this opinion be well founded , the difference between the indications of different anemometers would F ... V ' depend only on the fraction - , and the limiting value of \#151 ; given by the constants of paragraph 39 would =2''286 instead of 3 , which I originally assigned to it .
More exact observations may show that x and 2 are not rigorously constant , and diminished section of the arms may alter this limit ; but probably the changes will not be of much practical importance .
( 50 .
) If these observations were repeated in a quiescent locality , if the frictions were measured at each experiment and by the means described in the note to paragraph 20 , more accurate results would certainly be obtained ; but there would remain the inevitable difficulty arising from the impossibility of rightly estimating the effect of W. Any measures of it which can be taken give only an average of the whole circumference , and one outside the anemometer 's track , while the experiments with the balloons and air-meter show that it varies notably in various parts of the circumference , and with the distance from the centre .
There is also an outward current of considerable magnitude , but very irregular .
The rotation of the anemometer itself , whether vertical or horizontal , produces a secondary vortex , which must THE CONSTANTS OF THE CUP ANEMOMETER .
modify the primary one ; to what degree we cannot say .
The effect of these variations of W is shown by the changes of v in individual revolutions , while V is perfectly uniform , ( 51 .
) For this reason , combined with the difficulty of obtaining satisfactory measures of the centrifugal friction , I have given up the idea of repeating these observations , and I propose instead to try the method described in the end of my paper .
I have established on the flat roof of the dwelling house here the anemometer No. I. , with its axis vertical and 22 feet due east of that of the Kew instrument already there .
Both of them are about 16 feet above the roof .
The chronograph and its battery of six Leclanche 's cells are placed in the wooden house which supports it , and are connected with it and the other instrument by insulated wire and contact makers .
From the altered position of the axis it is necessary to apply the brake weights by means of a right-angled lever ; the counterpoises already described have been replaced by springs to prevent the chance of accidental disturbance of their position .
The axle now rests on a toe of hard steel , 0*25 inch in diameter , turning in oil ; and to prevent it from cutting under the pressure of the axle and its appendages , the rollers formerly used to lighten the centrifugal friction have been altered by Mr. H. Grubb so as to relieve the toe of any required portion of the weight .
This arrangement works so well that the friction is little more than in the Kew one , though its weight is 1*6 times as great .
The arms are now of plate steel -|th of an inch thick , with a sharp edge , strengthened by wire stays , and with means of placing both the 9-inch and 4-inch cups at 24 , 12 , and 8 inches from the centre .
( 52 .
) The mode of experimenting which I intend to follow is this .
Considering the Kew as the standard S , and the other E , and arranging one of the chronograph points to register the v of S , the other of E , apply a brake friction to the latter which will diminish v. When this has lasted long enough to make it tolerably certain that each instrument has been acted on by the same amount of wind , apply a different brake friction to E and take another set .
We have hence three equations , but four unknown quantities , V , a , / 3 , and y ; and since a appears as a coefficient of \ - , it cannot be obtained separately .
But if , as I believe , it is given with tolerable precision by the measures given in Table VI .
, the others may be determined .
The formula ( VI .
) enables us to estimate for the Kew anemometer the error in " V caused by Aa=l .
When v= 1 it is \#151 ; 0*20 , 0*059 of the computed V ; for v=20 it is \#151 ; 3*30 , 0*072 of V. The highest v I ever observed was in a squall in March , 1849 , when .n during five minutes it was 41 *2 ; this would give A"Y = 8*14 , 0*0/ 2 of 112 80 .
It is possible that more exact measures of might be obtained by moving the measuring apparatus in a right line , and in an enclosed space of sufficient dimensions , but the requisite conditions for such an experiment will nob be easily realised .
( 53 .
) If these constants give for any other pair taken with a different wind A s \#151 ; VE they are right ; if not , the formula ( VI .
) will show us whether the erroi O tises from using a wrong a , or whether the constants vary with v. 5 M % ( 54 .
) It is not easy to tell beforehand what difficulties may beset this mode of investigation .
The most obvious one is the irregularity of wind which may be expected to vary from one anemometer to the other , and also at each during an experiment .
If it blow in the direction W.E. , or there is danger that the eddy caused by the windward one may reach the others ; if it be S.N. , different streams of the current may fall on each .
The extent of these disturbances may be studied by making the friction of E equal to that of S ; and it will also show what length of time is required to make the average Y the same in both .
The changes of Y during the experiment may , I think , be eliminated by sorting the two v into sets , of which the individuals are all in the same proportion , and comparing them separately .
This can easily be done by measuring the intervals on the chronograph sheets .
How far these precautions will avail can only be ascertained by trial , but I hope that it may be given to me to make this trial to its full and decisive extent .
Appendix .
The object of the experiments being to determine the relation between the velocity of actual wind supposed uniform ( the air also being at , or reduced to , a normal density ) , the velocity of the cups and the friction , I assume in the first instance as correct , the values of those two quantites given by the experiments with the whirling machine , and proceed to consider the relation .
Let V ' be the velocity with which the air passes the anemometer , that is , in the case of the actual experiments , the velocity of the centre of the anemometer itself corrected for the velocity of the wind produced by it ; let v be the velocity of the centre of the cups , F the moment of the total friction .
Then supposing the density of the air normal for a given anemometer , v will depend only on V ' and F , that is , there will be a functional relation between the three variables Y ' , F , leaving two of them independent .
In investigating experimentally the relation between two variables , it is often very useful to plot the results of experiment , as the general character of the relation sought , and the allowance to be made for errors of observation can thus be estimated .
The relation between three variables would be expressed graphically by a surface instead of a curve , and it is troublesome to model a surface .
If , however , we can find a relation between the variables which is satisfied , provided some other relation is satisfied , we can thereby reduce the number of independent variables from two to one , and employ ordinary plotting in investigating the relation between the variables .
In fact , the relation sought is reduced from one of the form Y'=\lt ; f\gt ; ( v , F ) to one of the form / ( Y ' , v , ( Y ' , v , F ) } , where \lt ; \#163 ; , i/ / denote unknown , and \gt ; j known functions .
THE CONSTANTS OF THE CUP ANEMOMETER .
In the present case , since by hypothesis the anemometer is in a permanent state , the moment of the friction is equal to the total impelling force of the air , the total pressure arising from the motion of the air , without distinction of impelling or retarding , but reckoning the latter as a negative impelling force .
Now , in cases of rapid motion like that of the air passing the cups of the anemometer , it is well ascertained that the resistance varies as the square of the velocity , all other circumstances being the same .
Hence , with a given anemometer , when only the scale of the velocities changes , i.e. , when V varies as v , the moment of the total impelling force may be expected to vary as the square of the velocity .
When the density changes it may be expected also to vary as the density .
Hence we may expect that when v varies as V ' , then F varies as pV'2 , or in other words that r=Pv^(|7 ) , ... ... ... ... ... ... ... ... ... ..(i ) where ( f\gt ; denotes some function the form of which is not at present under consideration .
Let \#151 ; -y^=r ) , and for each observation let the point whose coordinates are f , rj be laid down on paper .
If Fp_1were merely some arbitrary function of V ' and the points so laid down would be spread out over the paper , but if equation ( 1 ) be true they will lie in a definite curve .
The actual experiments were executed in series , in each of which only one independent variable was changed , so that if the experiments were infinitely numerous and infinitely exact the locus of the point whose coordinates are \#163 ; 77 would be a definite continuous curve .
And the test of the truth of ( 1 ) is that the curves belonging to the different series shall coincide , instead of being arranged in some order of sequence .
Plate 68 shows the result of plotting the observations taken with anemometer No. III .
On inspecting the figure it will be seen that the different series fit very well into one another .
Departures there are no doubt in the individual observations from a mean curve , but these appear to be casual , not methodical and depending upon the order of the series .
The result of the observations then is confirmatory of the fundamental supposition made hitherto , that when the friction is so arranged that the velocity of the air passing the anemometer bears a given ratio to the velocity of the cups , the moment of the total impelling force varies as the square of either velocity .
Assuming then the truth of equation ( 1 ) , * we have next to inquire what is the form of the function ( f\gt ; ?
* It formed no part of the object of the ' experiments to determine the relation of the impelling1 force to p , which merely comes in as a small correction for reducing observations made on different days to a common standard .
It is the dependence of F on V ' and v that is contemplated in the text , A complete hydrodynamical solution of the problem is altogether beyond our power .
On the other hand , the irregularities of the observations prevent us from going , by observation alone , more than a certain way towards the determination .
We must , therefore , endeavour to combine as best may be the indications of mechanical theory with the results of experiments .
In his paper " On the Cup Anemometer/ ' in the Transactions of the Royal Irish Academy , Dr. Robinson has shown that ( supposing the density constant , say =1 ) the relation between the moment of the impelling force and the moment of the friction is either accurately or approximately of the form 'F = aV'*\#151 ; 2/ 3V/ v\#151 ; yv2 , ... ... ... ... ... ... ... . .
( 2 ) which would give for the locus of the point whose coordinates are f , 77 the parobola r ) = a\#151 ; 2/ 3g\#151 ; 'y\#163 ; 2 ... ... ... ... ... ... ... ... ... ( 3 ) If now we turn to the plotting of the observations , we see that the best smooth curve to represent the observations , free from sinuosities which the observations do not warrant us in supposing real , is either accurately or approximately a straight line , 77=0/ \#151 ; 2/ T\#163 ; ; ... ... ... ... ... ... ... ... ... ..(4 ) in fact , so nearly does a straight line represent the observations that it is not easy to say to which side the concavity of the line , if curved it be , should lie .
On the whole there appears to be a slight indication of a gentle concavity towards the origin .
It may be remarked in passing that the formula ( 4 ) which the experiments show to be at least very approximately true , leads to a very simple expression for v in terms of V ' , namely\#151 ; v=dV'-^\gt ; where a and b are constants .
The figure shows that there cannot be much doubt as to the distance from the origin at which the curve intersects the axis of f , nor as to the direction of the curve at that point ; and generally that the curve is well determined in its right hand half , though it becomes more uncertain towards the left .
If be the value of f at the point of intersection , and \#151 ; t the tangent of the inclination at that point , the equation of the curve , assumed to be a parabola , will be r , =t{\-\#163 ; )-C ... ... ... ... .(5 ) or again , if we suppose known two points ( y\gt ; , h ) and ( , lying in the well-determined part of the curve , its equation will be r)=(q\#151 ; p)-i{h(q\#151 ; \#163 ; ;)+k("\#151 ; p)\#151 ; C'(\#163 ; \#151 ; p)(t ; \#151 ; q ) } ; .(6 ) and as ( X\#151 ; or ( \#163 ; \#151 ; p)(\#163 ; \#151 ; q ) will be small throughout the well-determined part of THE CONSTANTS OF THE CUP ANEMOMETER .
the curve , the constant C or O ' will admit of considerable latitude of variation without much affecting the satisfaction of the observations .
Conversely , if we attempt to determine the constant C or O ' from the observations , in addition to the two elements X , t , or h , k , the determination will be extremely precarious .
And if we arrange the formula ( 5 ) or ( 6 ) according to powers of so as to throw it into the form ( 3 ) , the precariousness of the determination of C or O ' will more or less affect all the three constants a , / 3 , y. Accordingly , if we take this formula , and attempt to determine the three constants a , ( 3 , y , from the observations , it may be that by different processes we shall arrive at results differing considerably , not only as regards y , but even , though to a less degree , as regards a and ( 3 .
It is not until we use the values of a , ( 3 , y , so obtained for the determination of two out of the three elements of the parabola which are well or fairly determined by the observations , that we perceive the accordance which underlies the apparent discrepancy .
If the simple formula ( 4 ) so nearly fits the observations , it is by no means merely as an empirical formula of interpolation presenting two arbitrary constants whereby an approximate accordance may be brought about , or in the way that a small arc of an arbitrary curve may be approximately represented by a straight line .
The observations were also plotted by taking for coordinates V/ : v and F : instead of : V and F : V'3 , and in this case the curvature of the curve was very decided .
Accordingly , though the observations may be satisfied by the first two terms of the formula ( 2 ) almost as well as by the three , that is by no means true of the last two , though in both cases alike we have two arbitrary constants at our disposal .
Explanation of the Plates .
In these plates the results of the experiments made with the five anemometers aie graphically represented .
In the plottings the abscissa is throughout V. In Nos. I. and III .
the ordinate represents ( / -pB )~^P^ ~\gt ; Nos. II .
, IV .
, and V. it represents the same divided by the ratio of the area of the mouth of the cups to that of the mouth of the cups in Nos. I. or III .
; that is , it represents the above expression multiplied by 81 -f-16 for Nos. II .
and IV .
, or by 7t-t-4 for No. V. In the plottings , 0*75 inch is taken as the unit .
The reference numbers represent the order of the series , 1 meaning no weight oil the brake , 2 the lowest weight , and so on .
In the experiments with anemometer No. I. , the full dots represent the first set , experiments Nos. 1 to 79 ; No. 45 being omitted for a reason alieady stated .
The small circles refer to the second set , experiments Nos. 80 to 12b .
In all the plottings , except that for No. IV .
, a parabola or straight line , or both a parabola and a straight line , are laid down for comparison .
The elements chosen for these are as follows :\#151 ; For No. I. , a= 11T83 , \#163 ; =16*954 , y=-19*862 , a ' = 9*880 , \#163 ; '=11*538 .
For No. II .
, a'=2*084 , \#163 ; '=3*489 .
For No. III .
, a=9*472 , \#163 ; =8*469 , y=9*814 , a'=9*989 , \#163 ; '=10*928 .
For No. V. , a=9*152 , \#163 ; =13*416 , y= \#151 ; 3*481 .
The ordinates of the straight line and parabola were of course reduced in the same ratios as those of the dots , giving for the reduced constants in their equations\#151 ; For No. II .
, a'= 10*550 , \#163 ; '=17*663 .
For No. V. , a=7*188 , \#163 ; =10*537 , y=\#151 ; 2*734 .
PLATE 67 .
Fig. 1 .
Section of the horizontal arm ( paragraph 2 ) .
Fig. 2 .
The driving apparatus ( paragraph 6 ) .
PLATE 70 .
Fig. 3 .
Apparatus for measuring the vortex current ( paragraph 13 ) .
Fig. 4 .
Brake apparatus ( paragraph 21 ) .
Robinson bPhil .
Trans. 1878 .
Pla Harrison 8c Sons , Litii , S !
Martins Lane .
W !
C. Mobinsbn JPhil .
Trans.1878 .
Plate , fit .
] .
Harrison ic Sons , Lith. .
S ?
Martins Lane3\ .
C. Robinson Phil. Trans.1878Plate , 68 .
Harrison fc Sons .
Lith .
S !
Martins Lane.'W .
C. _Robinson Phil. Trains .
187\amp ; Plate , 69 .
N ' IV .
\#163 ; ^ 4* .7 .* 4# j ._______________ i________________ i + S 6 0 __1 Harrison Sc Sons .
Litb .
, S !
Martins Lane .
W. C. Robinson JPhil .
Tran$:1878 .
70 .
Harrison \amp ; Sons .
Lith .
, S ?
Martins Lane/ W. C.

