Imágenes de páginas
PDF
EPUB

sufficiently long to prove that the faults in Nos. 1, 2 and 3 rendered them unfit for sea use at that time; but that confidence might have been safely placed in No. 4.

[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][ocr errors][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]

In the choice of a chronometer, the object of the mariner should be to select one which keeps a steady rate in those temperatures to which he will be most frequently exposed at sea; and the uniformity of the daily rate of No. 4, in temperatures ranging from 58° to 84,° would doubtless lead him to select this for a warm climate. Without such records to refer to, he has nothing to guide him in his choice.

The four chronometers here selected, appeared equally good to the eye, although three of them were quite unfit to take to sea at that time. Yet, a few months after the above results were obtained, No. 1 did keep, and Nos. 2 and 3 were made to keep rates nearly as steady as No. 4.

The way in which these improvements were brought about, we will endeavour to explain, beginning with No. 1. When new, chronometers have a tendency to gain on their rates, for periods varying from two or three, to six or eight months. This tendency to gain, appears to be caused by a change which gradually takes place in a new balance spring; the spring must be made hard to insure good performance, and the harder the spring the longer the chronometer appears to gain. Old chronometers, with new balance springs, gain as much as new chronometers. The effect of this tendency to gain is shown in No. 1. On the first day of the twenty, the daily rate was one-tenth of a second; on the last day, it was two seconds and four tenths. In this case, the chronometer continued to gain for upwards of four months, and the rate increased from one-tenth of a second to upwards of ten seconds a day. Five months subsequent to the period at which these results were obtained, the rate became steady, and the chronometer might then have been taken to sea, with every probability of good performance. A chronometer having this fault, might increase its gaining rate ten or fifteen seconds a day, in a voyage of three or four months; and the mariner not knowing the cause, would become dissatisfied, and possibly part with it at a great sacrifice, just at the time when good performance might be expected from it. It is quite a common thing for captains to bring their new chronometers to the Observatory after, instead of before, the first voyage, for the purpose of having them tested; in consequence of bad performance during the voyage. In such cases, we frequently find the fault to be this tendency to gain, from being new, and notwithstanding the great increase of rate which has taken place, it often happens that the tendency to gain has then nearly or quite ceased; and without anything being done to the chronometers, they are generally found to give satisfaction the next voyage.

We will now proceed to an explanation of the cause of the irregularity in the daily rate exhibited in No. 2. This is a case of imperfect compensation for change of temperature. The chronometer loses nine seconds a day more in 84° than it does in 55°. The shipmaster in possession of a timekeeper with this imperfection might be subjected to the most serious inconvenience, and possibly, after giving it a trial for a voyage or two, he would come to the conclusion that it was worthless, and part with it for a fault which could be removed for a few shillings. A chronometer maker who understands his business, on being shown the

tabulated results here given, would instantly know how to make the necessary alteration; the weights on the compensation balance, in this case, should be shifted a little towards the ends of the curved arms. In No. 4, it will be seen that no sensible alteration of rate took place on changing the temperature from below 60° to above 80°, and there is nothing whatever indicated in the performance of No. 2, to lead us to suppose that it might not be made to perform equally well. The defect is one of very common occurrence, as may be seen by our records; hundreds of similar cases have been detected during the past ten years at the Liverpool Observatory. It sometimes happens that chronometers are sold with both the faults shown in Nos. 1 and 2, and the increase of gaining rate during the first voyage is on some occasions so nearly equal to the loss from defective compensation, that the ship's longitude by chronometer is found to be sensibly accurate. On the next voyage, the tendency to gain having ceased, defective compensation acts alone, and the result is great disappointment. We have known captains in such cases to say that the performance of the chronometer was satisfactory during the first voyage, but that it had never gone well since.

In No. 3 the irregularities cannot be attributed to either of the faults exhibited in Nos. 1 and 2. The change of rate is not so gradual, continuous, and in the same direction as in No. 1; nor is it dependent on temperature, as in No. 2. The irregularity is quite as decided as in either of the preceding cases, but the cause is not so apparent; precisely similar irregularities are found to arise from a great variety of causes. The oil may be glutinous, or it may have escaped from one or more of the jewelled holes; a jewel may be loose or defective; the escapement may be imperfectly adjusted, or some other imperfection may exist, which the skilful workman will probably detect on taking the chronometer to pieces. Another and similar trial will prove whether he has detected and removed the fault. Sometimes the chronometer may pass through the hands of the most experienced chronometer maker several times, before he detects the fault, for it requires, on some occasions, no ordinary amount of skill; but the test, subsequent to the last examination by the maker, will show if the desired improvement is effected.

Supposing, therefore, that the mariner has access to an establishment in which experiments, similar to those exhibited in the preceding examples,

are made for him, he can himself, with the greatest ease, detect those faults which chronometers are so frequently found to have, and which are to him of so much importance. He has only to look down the column of daily rates, and to see that the temperature, shown in the side column, has been changed 20° or 30°. The difference between the results given from day to day will indicate at once if the chronometer has a steady rate. He must, however, bear in mind, that absolute perfection is impossible, variations of a few tenths of a second will take place. Chronometers which will bear a change of temperature of 20° or 30°, and in which the extreme difference between any two days for thirty or forty days in succession, does not amount to more than one second, or one second and a half, may be safely taken as first-rate marine timekeepers. If it were possible to make chronometers perfect in all other respects, the defect of the ordinary compensation balance would alone produce a change in the daily rate, as large as that above named. This may be seen in the following examples, in which we have selected three chronometers which were found to have nearly the same rates in 50° and 80°. By examining the rates about half way intermediate between 50° and 30°, it will be seen that each of them gained from one second to one second and a half more, than they did in either the high or low temperatures. The daily rates here given are all gaining, and the averages have been taken instead of the rate for each day, as in the preceding examples, the errors to be detected being very small.

[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][subsumed][subsumed]

These rates were obtained from chronometers having the ordinary compensation balance. This balance is in use in the merchant service almost without exception. It does not perfectly compensate for the change of elasticity in the balance spring. With it the chronometer can be made to go the same in any two temperatures, but it cannot be made to go precisely the same in any three temperatures. All chronometers with this description of balance go faster in some particular temperature than they do in any other temperature, and the chronometer maker has the power of fixing

this maximum gaining rate, either in a high, low, or medium temperature, at pleasure; by so adjusting the weights, on the compensation balance, as to give the chronometer the same rate in two temperatures at equal distances from that in which the maximum gaining rate is required. In example No. 6, the maximum gaining rate is fixed at 65°, as near as possible, and the loss at 50° is one second and seven tenths, and at 80° one second and six tenths. The temperature at which the maximum gaining rate is fixed, is a matter of the highest importance in chronometers with this description of balance; because we find, that the further this point is departed from, the greater will be the variation of rate for a given change of temperature. At the distance of 15°, as we have seen, the change of rate is about one second and a half, but at 30° distance it is found to be much more than twice that amount. Ships at sea are for the most part exposed to temperatures ranging from 50° to 80°, and captains should therefore take care that their chronometers have as nearly as possible the same rate in these two temperatures. Suppose a chronometer with the ordinary compensation balance to go mean time in 50° and 80°, then, as we have seen, it may be expected to gain about one second and a half in 65°; in 90° it would probably lose about two seconds, and in 40° it would probably lose from two to four seconds a day. In temperatures below 40° there is less certainty; the thickening of the oil and the consequent decrease in the arc of vibration, appear to be the cause of this. By winding up or letting down the main spring, and thereby increasing or diminishing its power, the maker can cause the chronometer balance to vibrate in a large or a small arc; and there is a certain length of the balance spring, at which the daily rate of the chronometer is found to be the same, whether the arc of vibration be large or small. Now it is very probable that some makers so adjust the balance spring as to cause the chronometer to have the same rate in the large as in the small arcs; and that others so adjust it as to make the chronometer gain more or less in the small arcs. In the former case, the loss from additional friction, caused by the thickening of the oil, together with that from the defect of the balance, might be excessive; while, in the latter case, it would be more or less compensated for by the gain consequent on the diminished are of vibration. Mariners who have chronometers compensated for high tem

This circumstance is either not known by chronometer makers, or it is not sufficiently attended to: the records of the Liverpool Observatory clearly prove this.

« AnteriorContinuar »