LXXXI .
A Description of Two Methods , by which the Irregularity of the Motion of a Clock , arising from the Influence of Heat and Cold upon the Rod of the Pendulum , may be prevented ; by John Ellicott , F. R. S. THE first of these methods consists in a particular construction of the pendulum itself , which occurred to me several years ago .
In the beginning of the year 1738 , I put into the hands of Mr. Machine , then one of the secretaries , a description and a drawing of such a pendulum , in order to their being laid before this honourable Society : but Mr. Machine , soon after , acquainting me , that a gentleman , of whose skill and judgment in mechanical contrivances I had always entertained the highest opinion , made some objections , I was advised to defer communicating my invention to this Society , till I should have examined into the weight of those objections , and , by a fair and impartial trial , should be fully assured , that the contrivance would answer the end proposed .
And having now at length obtained that satisfaction , I beg leave to give a short narrative of some of the most remarkable observations I have made during this inquiry , which I hope will not prove unacceptable to this honourable Society .
Read June 4 , 1752 .
About the year 1732 , an experiment , which I made , in order to satisfy some gentlemen , that the rod of a pendulum was liable to be considerably influenced by moderate degrees of heat and cold , led me to consider , that , as metals differ from each other in their density , it was highly probable they might likewise differ from each other in their expansion ; and that this difference of the expansions of two metals might be so applied , as in a great measure to remove those irregularities in the motion of a clock , which arise from the effect of heat and coid upon the length of a pendulum .
With this view , not long afterwards I contrived the pendulum now described by Fig. 1 .
In which ab represents a bar of brass , made quite fast at the upper part by pins , and held contiguous , at several equal distances , by the screws 1 , 2 , 3 , &amp;c. to the rod of the pendulum , which is a bar of iron ; and so far as the brass bar reaches , is filed of the same size and shape , and consequently does not appear in the figure ; but a little below the end of the brass bar , the iron is left broader , as at d d , for the convenience of fixing the work to it , and is made of a sufficient length to pass quite thro ' the ball of the pendulum to c. The holes , 1 , 2 , &amp;c. in the brass , thro ' which the shanks of the screws pass into the iron rod of the pendulum , are filed as in the drawing , of a length sufficient to suffer the brass to contract and dilate freely by heat and cold under the heads of the screws .
eeee represents the ball of the pendulum : f f , two strong pieces of steel , or levers , whose inner centres , or pivots , turn in two holes drilled in the broad part of the pendulum-rod , and their outer ones in a strong bridge , or cock , screwed upon the same part of the rod , but omitted in the draught ; because , when put on , it covers this mechanism .
g g , are two screws entering at the edge , Philos. Trans. Vol. XLVII .
TAB. XX .
a 2 Fig. 1 .
p. 480 .
Fig. p. 48 1 Fig. 2 .
p. 481 .
ca e e db~ ff gg dbk e ig h d i e e hg a m ~h g dbeeb no Fig. 3 ii a p. 486 .
ab k ~ 481 .
a a ba 3 2 3 1 2 f 1 kf k c g Fig. 4 .
p. 488 .
f k and reaching into the cavity near the centre of the ball .
The ends of these screws next the centre are turned into the form represented in the drawing , which , pressing with the weight of the ball against the longer arms of the levers , cause the shorter arms to press against the end of the brass bar at 6 .
Things being in this situation , let us suppose , that the rod of the pendulum , and the brass annexed to it , grow longer by heat ; and that the brass lengthens more than the iron of the same length : then the brass , by its excess of dilatation , will press the short ends of the levers downwards at 6 , and at the same time necessarily lift up the ball , which rests upon the long ends of the same levers at f f , to any proportion necessary : And provided the ends of the screws do press upon the levers at a proper distance from the centres , the said ball will be always kept at the point of suspension , notwithstanding any alteration the rod of the pendulum may be liable to from heat or cold .
What this distance ought to be , may very nearly be determined , if the difference of the expansion between the brass and iron bars is known ; for the proportion the shorter arms of the levers ought to bear to the longer ones will always be , as the excess of the expansion of the brass is to the whole expansion of the iron , as may be thus easily demonstrated .
Fig. 2 .
Let the line a b , drawn perpendicular to the line e f , represent a bar of iron ; the line cda bar of brass , the pricked line b g , the expansion of the brass bar by the same degree of heat : let the line gi be drawn parallel to the line e f , then will ih represent the difference of the expansion of the two metals : thro ' the points h , g , draw a right line cutting the line e f , as in k ; this line may be supposed to represent one of the levers turning upon its centre at g , h the point where the brass bar acts upon the shorter end of the lever , and k the point where the screw acts upon the longer end of the lever , which being the place where it intersects the line e f , it is evident the ball of the pendulum will be as much raised by the lever , as it would have been depressed by the expansion of the iron ; but the triangle ihg is similar to the triangle bg k ; and therefore , as i h , the excess of the expansion of the brass , is to b g , the whole expansion of the iron , so will h g , the shorter arm of the lever , be to g k , the longer arm of the lever .
Q. e. d. At Fig. 1 .
is placed a strong double spring , whose ends pressing against the under edge of the ball , hinder it from bending the brass bar by its forcible action thereon at the point b , which , when the ball is of a considerable weight , it might otherwise be very liable to do .
The description here given is exactly agreeable to the original contrivance ; and the only alteration I have since made in it , consists in placing the screws gg within the ball of the pendulum , as represented in Fig. 4 .
But as the success of this contrivance depended entirely upon the supposition , that metals were expanded differently by the same degree of heat , before I attempted to put it in execution , I thought proper to inquire what experiments had already been made upon this subject , when Mr. John Eames , a late very worthy member of this Society , put into my hands Mr. Graham 's account of his quicksilver pendulum ( as it is now commonly called ) published in the Philosophical Transactions , No 392 , which account I found to be introduced by the following paragraph : Whereas several , who have been curious in measuring of time , have taken notice , that the vibrations of a pendulum are slower in summer than in winter ; and have very justly supposed this alteration has proceeded from a change of length in the pendulum itself , by the influences of heat and cold upon it , in the different seasons of the year ; with a view therefore of correcting , in some degree , this defect of the pendulum , I made several trials , about the year 1715 , to discover whether there was any considerable difference of expansion between brass , steel , iron , copper , silver , &amp;c. when exposed to the same degrees of heat , as nearly as I could determine ; conceiving it would not be very difficult , by making use of two sorts of metals differing considerably in their degrees of expansion and contraction , to remedy , in great measure , the irregularities , to which common pendulums are subject .
But although it is easily discoverable , that all these metals suffer a sensible alteration of their dimension by heat and cold ; yet I found their differences in quantity , from one another , were so small , as gave me no hopes of succeeding this way , and made me leave off prosecuting this affair any farther at that time .
' ' The reading this paragraph proved at that time sufficient to make me lay aside all thoughts of succeeding in a contrivance founded upon principles , which a gentleman of so great abilities , and known accuracy in making experiments , had , after trial , judged to be insufficient .
And it was not till about the latter end of the year 1734 , that I again resumed them on the following occasion .
A gentleman desirous to make some experiments concerning the expansion~of metals , employed me to make him an instrument like one invented by Mr. Muschenbroek for that purpose , which he calls a pyrometer .
Upon looking over Mr. Muschenbroek 's experiments , I not only found the difference between the expansion of some of the metals much greater than I expected , but , as I thought ( if they were to be depended upon ) sufficient to answer my former purpose .
This led me to consider the structure of the instrument , which Mr. Muschenbroek made use of in his trials , and upon examination , I thought it liable to some objections , which I imagined would make the result of experiments made by his instrument very uncertain .
I therefore endeavoured to contrive one of a different construction , that might be more to be depended upon .
Such an instrument I some time afterwards completed , and had the pleasure to find it so far met with the approbation of several very worthy members of this Society , that , at their particular desire , I drew up a description of it , which was read , and the instrument itself shown to the Society on the 8 of April 1736 * : and though it was not in every respect * This appears from the minutes of the Royal Society ; tho ' the description and manner of that instrument was , by some mistake , placed six months later in the Philos. Trans. Vol. XXXIX .
N. 443 .
for October 1736 .
The other dates , mentioned in this paper , relating to that instrument , are likewise justified by authentic vouchers , which have been produced by Mr. Ellicott .
spect so accurate as I could wish , I am fully persuaded , that such experiments , as are carefully made with it , may be depended upon , as very near the truth .
Having made a great variety of experiments with this instrument upon bars of different metals , as nearly of the same dimensions as possible , I found , upon a medium , their several expansions by the same degree of heat to be as follows : Gold Silver Brass Copper Iron Steel Lead 73 103 95 89 60 56 149 And as I found so great a difference between the expansion of brass and iron , I immediately determined to make a pendulum after the manner above described , composed of those two metals , and likewise ordered a clock to be made , with the utmost care and exactness ; and , as I then apprehended , with some considerable improvements , with which I intended to make the experiments .
These were both finished in the beginning of the year 1738 ; and having no reason to doubt of success , I showed the pendulum to the late Mr. Machine , and gave him a drawing and description of it , in order to its being communicated to the Royal Society ; but , as I have already observed , objections were made to it , of which the only one , that appeared to have any weight , was , that it had been found by experiment , that two bars of different metals , screwed together , so as to be in contact with each other , would not expand regularly and smoothly , but by jerks .
In order to examine into the force of this objection , I directed two bars of equal dimensions to be made , one of brass , the other of iron , of about two feet in length , fastened together after the same manner , as the two rods of the pendulum , which I intended to place so , that , by acting very near the centre of an index of a considerable length , even the smallest alteration in the bars would be made sensible , and by the motion of the index I should be able to form a judgment , whether the rods moved regularly and freely , or not : but before this was put into execution , I contrived , by fastening the two bars to the back plate of a clock , not only to make them answer the end above proposed , but , at the same time , to lengthen or shorten a pendulum of a common construction , in such a manner , as sufficiently to correct the irregularities aris~ing from the influence of cold or heat upon it .
The manner of applying them is described by Fig. 3 .
In which , aaaa represents the back plate of the clock , bb b , a triangular piece of brass , screwed by two screws , thro ' the slits c c , to the plate , yet so that it may be drawn backwards or forwards by means of the screw at d. of is a brass bar , about two feet in length , made fast at the bottom , by a screw and two pins at f , to an iron one of equal dimensions , to which it is likewise screwed by the screws 1 , 2 , 3 , ~c .
after the same manner as the rod of the pendulum already described .
The iron bar is fastened at the upper end to the triangular piece of brass , nearly under that part of the brass bar marked e. gh is a strong brass or iron lever , moveable upon a centre at g , and is supported by the upper end of the brass bar ; ii is the cock , on which , in a common clock , the pendulum is hung ; k k , part of the rod of the pendulum , whose spring passing thro ' a fine slit in the cock i i , is fastened to a stud riveted into the lever at l. The slit in the cock must be made so close , as to prevent the spring from having any lateral motion in it .
From this description it is evident , that , if the brass bar expands more than the iron one , it will raise up the lever , and , consequently , the pendulum , which is fastened to it ; and , as the length of the pendulum is only from the centre of oscillation to the under part of the slit , thro ' which the spring passes , the pendulum will be thereby shortened ; and , by making the point of the brass bar to act upon a proper part of the lever ( to which it is capable of being adjusted by means of the screw d ) the pendulum may be shortened to whatever degree shall be necessary .
To prevent the pendulum from bending the bars , which it would be liable to do , if the ball of the pendulum was of any considerable weight , the end of the lever , farthest from its centre of motion , is hooked to the end of a chain , which is wound about and fastened to a small pulley at m. Upon the same arbor , to which this pulley is fixed , is fastened another pulley , of a much larger diameter , to which is hung , by a silk line , the weight or counterpoise n. By means of this counterpoise , any part of the weight of the pendulum , that shall be desired , may be taken off from pressing against the brass bar .
And if , upon the end of the arbor , to which the pullies are fixed , an index be placed , so as to point to a graduated circle , the least motion of the lever will not only be easily perceived , but also whether that motion is uniform and regular , or not .
And upon having , some time after , made a clock with this contrivance added to it , I had the pleasure to find the index not only to move very sensibly , but very regularly , and never , that I could perceive , by jerks .
And I doubt not , but , when the point of bearing of the brass bar upon the lever is once well adjusted , it will be found to lengthen or shorten the pendulum to as great a degree of exactness , as any other method whatsoever .
But , as I have not as yet thought of any other method of adjusting it , except from actual trial in different seasons of the year , I must prefer the pendulum to this method , which , from the great ease , as well as exactness , with which it is capable of being adjusted , will , I think , appear to have much the advantage over any other contrivance yet made use of for this purpose .
The method I take for adjusting the longer arms of the levers of the pendulum to the shorter ones is described in Fig. 4 .
To a strong post , fixed to the wall , is fastened a small shelf , supported by two brackets a b. In the middle of this shelf is fastened a wire , by the screw ~ ; to the end of which the pendulum is to be hung .
Below this shelf , at the distance of about 40 inches , is placed the index c d , turning freely upon a center : The length of the index is 50 inches .
At the distance of half an inch , upon a part of the index produced beyond the centre , is placed a steel pin ; and in the back of the pendulum , as near the centre of oscillation as may be , is drilled an hole to receive this pin ; when the pendulum is hung upon the wire against the post , and the wire is screwed higher or lower by the screw e , till the pin resting against the upper part of the hole ( which is filed into a proper shape for that purpose ) keeps the index nearly in an horizontal position .
Below the bottom of the pendulum is placed a second index index f g , exactly like the former , except that it is kept in an horizontal position , by the screw k bearing against the end of the iron rod .
When the experiment is to be made , the pendulum is first put into a box , and gradually heated by a large fire , to a considerable degree , being often turned , that every part may be equally exposed to the fire .
And having continued shut up in the box for some time after it is removed from the fire , that the two rods may be heated as uniformly to the same degree as possible , the pendulum is hung upon the wire , and the two indexes made to stand nearly in an horizontal position .
The two graduated plates hi are then slid upon a wire , till the divisions in each marked o are pointed to by the indexes .
As the pendulum cools , the lower index will be seen gradually to descend ; but if the ends of the two screws , in the ball of the pendulum , act upon proper parts of the levers , the upper index will continue in the same place .
If the ends of the screws are either too far off , or too near the centres of the levers , the index will either rise or descend ; and , by comparing the number of divisions it has varied , with those which the lower index has varied , a near estimate may be made , how much the screws require to be altered ; and , in a very few trials , they may easily be adjusted to a very great exactness .
In order to make an actual trial , how far this contrivance of the pendulum will answer the end proposed , it is necessary , that the clock , to which the pendulum is fitted , be made with great exactness , and entirely to be depended on : For otherwise the experiments will be very uncertain , as I found in the clock I first made use of .
I have already observed , that , in order to render this clock as perfect as possible , I made it , in several respects , different from the common ones , in hopes of removing some imperfections I apprehended they were liable to .
But as , in this attempt , I fell into an error , which it was a considerable time before I discovered , my making the trial was thereby greatly retarded .
And in order to prevent others from falling into the like mistake , I shall beg leave to give some short account of it .
In a common clock the pendulum is usually hung by a spring to a cock on the back plate of the clock , whilst the wheel and pallets , by which the pendulum is kept in motion , are placed in the middle of the frame ; and the pendulum is moved by a piece of steel ( called the crutch ) riveted to one end of the arbor , to which the pallets are fastened .
This disposition of the pieces I apprehended liable to some considerable objections : To remedy which , I contrived to fix the pallets to the upper part of the pendulum itself , above the centre of motion ; and , in order to make the pendulum vibrate as freely as possible , it was made to turn upon two steel points , and was hung in the middle of the frame , exactly under the swing-wheel , and so as to vibrate in the same plane with it .
By this means I was in hopes , that it would have moved with much greater freedom and regularity , than when hung after the common method ; and , upon trial , it was found to move with so great freedom , that a pendulum of above 20 pounds weight , when hung in its place without the clockwork , and made to vibrate thro ' an arch of two degrees , was found to make above 1200 vibrations , before it had lost half a degree , and was observed to have a sensible motion above 20 hours afterwards ; and the clock , when first put together , was kept going , for several days , by a weight of only eleven ounces , hung to the end of a single line .
But it was not long , before I discovered , that this great freedom made it liable to be considerably affected by the least motion .
A remarbable instance of this I communicated to this Society , which was published in the Philosophical Transactions , No 453 .
But the greatest objection to this method was , the points being subject to wear ; and I found , that the least alteration in them would occasion the clock to vary much more , than ( without having made the trial ) I could have imagined .
To remedy this inconvenience , I made the pendulum to move upon edges , like those , on which the beam of a pair of scales turns ( a method I had good reason to believe had been made use of with success ) ; but I found these likewise liable to wear , tho ' not in so short a time as the points ; so that , after much time spent in making several experiments , in order to remedy this inconvenience , I found myself obliged to lay this method wholly aside , and to hang the pendulum upon a spring , as usual .
In making this alteration , I observed one circumstance , which I think deserves to be taken notice of .
Before I made any alteration in the work , I took particular notice to what height the pendulum required to be raised , before the pallets would escape from the wheel .
I next observed the number of degrees of each vibration of the pendulum , when moved by the clockwork ; and then , the clockwork being removed , the pendulum was made to describe an arch of two degrees ; and particular notice was likewise taken , in what space of time it had lost half a degree each vibration .
Having then made the necessary alterations for hanging the pendulum by a spring , and particular care being taken that the pallets should scape off from the wheel exactly at the same angle as before , the pendulum being hung by its spring , and made to vibrate thro ' an arch of two degrees , it was observed to lose half a degree in about half the time it did when turning upon edges .
But , upon being set a-going by the clock-work , the pendulum was found to describe an arch of near two degrees more than before : For , when it turned upon the edges , it described an arch of only three degrees ; whereas , now it was hung by the spring , it vibrated near five degrees ; which was very different from what I expected .
This alteration being made , I soon found , that the clock went very regular ; and , after a sufficient trial , was fully satisfied the pendulum would answer my expectations .
But , fearing left I might be thought prejudiced in favour of my own invention , I engaged the Rev. Mr. Professor Bliss to make trial of it ; and .
accordingly , in the beginning of the year 1750 , I sent to him , at Oxford , a clock for that purpose ; and , in January last , I received from him a letter , giving his opinion of it , of which the following ( so far as relates to the clock ) is an exact copy .
SIR , I have now had thorough trial of the clock ; and am perfectly satisfied , that your pendulum takes off the effect of heat and cold as well as either the gridiron pendulum ( as it is commonly called ) or the quicksilver pendulum ; and this upon sufficient trial for near two years .
It has this advantage of both the fore-mention 'd ones , that it may , by lengthening or shortening the levers , be easily adjusted to the exact proportion of the difference of the iron and brass , which neither of those kinds is capable of , without very great trouble and difficulty .
I was indeed prejudiced against the method of doing it by levers , as I had heard the late Mr. Graham say , that he had tried levers in different ways , that he found they did not work regularly and freely , but by jerks .
However , in your method , I am satisfied , by the fullest experience , that they succeed as well as either of the other sorts , or perhaps any other kind , that may be invented hereafter .
' ' Before I conclude this paper , I shall beg leave to acquaint this honourable Society , that , in the year 1748 , I made a model of a contrivance to be added to a pocket-watch , founded upon the same principles , and intended to answer the like purpose , as the pendulum above described .
And , at a meeting of a council of this Society , on February 15 last , I produced a watch ( which I had made for a gentleman ) with this contrivance added to it , and likewise the model , by which was shown to the gentlemen then present what effect a small degree of heat would have upon it .
But , as I have not yet had sufficient trial of this watch , I shall defer giving a particular description of this contrivance , till I am fully satisfied to what degree of exactness it can be made to answer the end proposed , I am , June 4 , 1752 .
Gentlemen , Your most obedient humble servant , J. Ellicott .

