The Clock Jobber's Handybook

By Paul N. Hasluck

Brought to you by:

Tick Tock Productions  

Anitque Clock Repair Courses on DVD

Next page  Previous page     

Home Page

The

CLOCK JOBBER'S HANDYBOOK.

PENDULUMS THE CONTROLLERS.


In planning a clock the pendulum claims first attention.  Though apparently a simple adjunct to a clock, the pendulum is in reality the most important part of its construction, for the value of the clock as a reliable timekeeping machine depends upon its free and regular movement. The function of the pendulum is to control the velocity of the going train with uniformity, and at a fixed rate, and it must be uninfluenced by the train except in receiving a sufficient amount of impulse to keep up its vibrations.  Before commencing to make a pendulum for a new clock, or to supply the place of a lost one, it is very desirable to know something of the laws and properties of pendulums. 

The simplest form of pendulum may be described as consisting of a weight suspended by some flexible substance and free to swing when moved on one side and then released.  The power which operates upon the pendulum is gravity, and the velocity it attains is proportional to the height fallen, notwithstanding the fact that the curve which the weight describes offers a resistance tending to neutralise in some degree the gravitating force. The effective force of gravity in producing the motion of the pendulum depends upon the position of the weight in relation to the vertical. The greater the distance the pendulum is moved from the vertical, the greater is the impelling force of gravity. From this, two important facts may be learnt. One, that a pendulum of a given length moves quicker in proportion to the distance it swings, therefore it will move through a large arc in the same time as a short one, and vice versa. In other words, when the extent of vibration is very little, gravity exercises but little force ; but, as the vibration increases in amount, the force of gravity becomes proportionately greater, causing the pendulum to move through a large arc in the same time as through a short one.

 
In one instance it moves through a large space quickly, in the other through a small space slowly, the time occupied being the same in both cases. Strictly speaking, this is not true of a pendulum moving in a circular arc, but it is so with a pendulum moving in what is known as a cycloidal curve.  A cycloid is a curve of the shape traced out by a point in the rim of a circle rolling upon a straight plane. This cycloidal curve corresponds for a certain distance from the vertical with the circle.  The vibrations of pendulums are generally of small extent, and any pendulum suspended by a spring never moves exactly in a circle. For these reasons it has been found sufficiently correct for all ordinary purposes to reckon that in pendulums of the same length unequal arcs are equal timed. This peculiar property of the pendulum is called its isochronism, and the difference between the time of vibration of a simple pendulum influenced only by gravity, swinging in a circular arc, and one of the same length moving in a cycloidal curve is known as the circular error.  

Another important fact is, that theoretically the vibrations of a pendulum are not altered by the weight or material of the bob, unless it is so light as to suffer from the resistance of the air. Consequently a pendulum of a given length may have a bob of any material either light or heavy, and it will vibrate in the same time. In practice, it is found that from various causes weight, and, therefore, material, does make some difference in the time of vibration of a pendulum. There is another cause which disturbs the uniform rate of vibrations in a pendulum which must be just noticed, that is, the varying density of the atmosphere. The effect of this is known as the barometric error, and to reduce it as much as possible, the " bob " must be made as small as it can be for its weight, and also of such a shape as will pass through the air with the least resistance and without any tendency to "wobble." In pendulums swinging 2.5 derees each side of zero, the barometric error is stated to be exactly compensated by the circular error. 

The above reasoning shows that the velocity which the pendulum attains, or its time of vibration, is proportional to the height fallen. The circumference of a circle may be considered to be 3.1416 times its diameter, and it is proved that the time of vibration of a simple pendulum will be 3.1416 x the time required for a body to fall vertically a distance equal to half the length of the pendulum.  It being well known that the times of falling from different heights are proportionate to the square roots of the distances fallen, it follows that the time of vibration of a pendulum varies as the square root of its length. Perhaps this will be better understood by stating that a pendulum i ft. long would vibrate four times during one swing of a pendulum 2 ft. long, and nine times during one swing of a pendulum 3 ft. long.  This reasoning applies properly to what is termed a simple pendulum, that is, one in which the rod is supposed to be without weight, the entire weight of the pendulum being at one point at the extremity. Such a pendulum cannot actually be made, and therefore the application of the rule has to be considered in relation to pendulums as they are usually met with. Pendulums commonly in use have the rods made either of wood or metal, sufficiently large and strong to support the heavy bob at the bottom. 

The principle upon which compensated pendulums are constructed may be briefly stated as a proper application of the expansion of metals. The most simple arrangement is that in which the "bob" expands upwards in such proportion to the lengthening of the pendulum rod that the centre of oscillation is always kept the same distance from the point of suspension. The cheapest and most simple form of compensated pendulum for vibrating seconds is made of yellow deal. It should be well-seasoned and straight, not sappy, nor of strong grain full of turpentine. The rod should be about 46 in. long, and f in. in diameter, and either well varnished with good spirit varnish or painted and gilt. The bob should be of lead, about 14 in. high, resting on the regulating nut at the bottom. The mounts at the top and bottom of the rod, as also that which receives the crutch-pin, are made of brass.

A clock or any other timekeeper cannot be easily regulated to keep mean time, because the mechanical adjustment of the regulator is not sufficiently fine to allow of it. As an example, suppose we have a pendulum 40 in. in length vibrating some 3,600 times per hour, by altering the length only 1/1000 part of its length, about one twenty-fifth part of an inch, it will cause a variation of one minute per day. These figures are only approximate, but quite near enough for the argument; e.g., for convenience in calculating taking 40 in. for length of pendulum. The exact length of one to vibrate 3,600 times an hour in London is 39.1393 in., whilst at the equator a pendulum 39.017 beats seconds. A clock going within seven minutes per week of mean time would be considered very badly regulated, and yet the alteration of 4/100 of an inch in the length of a seconds pendulum, or 1/100 of an inch in a half- seconds pendulum, will cause seven minutes a week difference in the rate.  

Coming to the mechanical adjustment, we find that the pendulum bob is raised and lowered by a nut on a screw, having perhaps some 50 threads per inch, so that one turn of the nut will make, say, 3.5 minutes per week difference in the rate of the clock. We can, however, divide the nut into, say, one hundred parts at its periphery, and then each division will represent a gain or loss of 18 seconds per month of 30 days, or about 31 minutes a year — not a very close rate after all. However, in practice the final adjustment is made by sliding a small weight on the rod. By a consideration of the above calculation, it will be easy to understand how minute must be the alteration in the regulation of a clock to cause it to gain or lose only, say, half a minute in 24 hours.

The gridiron pendulum of Harrison's is now almost entirely disused on account of the expense and trouble of making it, and also of its appearance. It consists of four pairs of brass and steel rods, and the steel rod which supports the bob. The mercurial pendulum, though very simple in construction, is as near perfection as can be desired, the only objection being its great expense. There are two forms in use; in one the mercury is contained in a straight glass vessel standing in a stirrup at the bottom of the rod, and in the other the mercury is in a cast-iron jar, into which the end of the rod dips. The great feature of the mercurial pendulum is the ease and accuracy with which the compensation can be tested and adjusted by simply taking away or adding mercury, as may be found necessary.  

Whichever form of pendulum is selected, whether plain or compensated, it is of the greatest importance that its suspension should be well made, and quite free from any looseness when the pendulum is set in motion. When the pendulum is long and the bob heavy, it is always desirable to suspend it from the back of the case, and not from a cock attached to the movement itself. On page 63, parts of such a suspension are illustrated.

It is of importance that the underneath of the "chops" which clip the spring should be quite square, and not rounded as they often are, because the spring will be liable to impart a twist to the pendulum at every vibration, if not perfectly free to bend in the correct manner. The bend of the pendulum spring should be exactly opposite the centre of the pallet-arbor pivot, in order that the up and down friction of the crutch may be as little as possible. 

The method of making a pendulum spring for an English clock is to soften a piece of wide watch main-spring, and then " draw it down " between two files — that is, pinch the spring in the vice by its lower end, and then tightly grip it between two files and draw them along its whole length. This is rather a troublesome and unsatisfactory plan, and it is much better to buy prepared pendulum spring, which can be obtained at a very moderate price, of the spring makers.

The accompanying illustrations show a useful form of spring suspensions having double springs, which greatly control any tendency of the pendulum to wobble.

 

The lighter the pendulum bob, the thinner the spring should be. The suspension springs are very often left too thick; or much too long and narrow. Generally the suspension spring should be as thin as possible, provided it is not so slight as to bend abruptly close to the chops or unsafe to support the pendulum weight. 

When the pendulum rod is of sufficient size to admit a pin, it is better to use that form of crutch in preference to the fork. If the rod is made of wood, it will be necessary to make two brass plates and carefully fit them into the recesses in the rod, at the proper place. The mortice through the wood should be made a little larger than the holes in the brass, so that the crutch-pin may rub against the metal only and not touch the wood. Care must be taken that there is only the necessary freedom at the crutch. If it binds, the clock will be sure to stop, and if the freedom is excessive, there will be a great loss of power, and probably the same result. The brass plates are secured in place by screws at top and bottom, which pass loosely through the front brass and screw into the back one. The best material for the bob is lead, on account of its specific gravity being greater than other material that can be employed. A lentil-shaped bob offers less resistance to the air, whilst it moves truly in its plane of vibration ; unfortunately, should it wobble, the resistance becomes irregular and incalculable, and therefore a pendulum should be hung carefully to ensure regular vibrations.

 _____________________________________________________________________________

  Learn clock repair with these DVD courses!  Course manuals are included. 

Watch, study and learn antique clock repair through DVD course instruction using actual live repairs!! 

 

Clock Repair 1 & 2    Advanced Clock Repair    PRO advanced clock repair  

Clockmaker Watchmaker Lathe Basics  Clockmaker Watchmaker Lathe Projects   Clock Case Repair & Restoration   Wooden Works Movement Repair

Sitemap

 

© Copyright 2001-2009 by Tick Tock Productions   © Copyright 2001-2009 by John Tope  All rights reserved.

 

 

Back to clock information page.

 

Hasluck, Paul N.  The Clock Jobber’s Handybook.  London: Crosby Lockwood and Son, 1889.

This and the following pages are excerpts from the book.