Page images
PDF
EPUB

which, when raised, cuts off the supply of gas. Whenever the pressure becomes too great, the bell rises and reduces the flow of gas, while too small a pressure makes the bell descend and admit more gas. Beyond the regulator a T is inserted, and connected with a gauge I, which gives the pressure. In its simplest form this is a bent tube containing water, which should be of considerable size, if accuracy is required. Sometimes a floating bell is used, which rises and falls as the pressure varies, and moves a long index over a graduated scale. The gas next passes to the burner B, first traversing one or two stopcocks, II, to regulate the quantity consumed. A variety of burners should be procured, which may be used in turn and compared.

A slide G is placed on the photometer bar, carrying a Bunsen photometer disk (Experiment 67). When this is placed between two lights, if the brightest is in front, the small circle looks light on a dark ground, if the brightest is behind, it appears dark.

A clock D, marking seconds, is also needed in this room, the best form striking a bell at the beginning of each minute, also five seconds before it, and having what is called a centre seconds' hand. It is often convenient to have a separate gas-pipe, meter and burner at A, the candle-end of the photometer bar, and to have the latter arranged so that it can be swung horizontally to it. In fact, in almost every new research some change of arrangement will be found desirable, and the apparatus should therefore not be fixed, but arranged so that the connections can be easily altered.

Experiment. To measure the candle power of burning gas. The law in the State of Massachusetts requires that the gas fur

D

B

Fig. 51.

nished, when burnt at the rate of five feet per hour, under the most favorable circumstances, shall give a light at least equal to

twelve sperm candles (6 to the pound), when consuming 120 grains per hour. To make the experiment the gas and candle are burnt at opposite ends of the photometer bar, their relative intensities compared, and the consumption of each measured. The amount of wax burnt is measured by the candle-balance A. This consists of a sort of steelyard, with a light weight or rider K, moving over its longer arm, which is divided so as to give grains. The centre of gravity of the beam is at such a distance from the point of suspension that the sensibility shall not be very great, and an index is attached, which moves over a scale, each of whose divisions corresponds to a change in weight of one grain.

Attach the candle to the shorter end of the beam and light it; set the rider at zero, and place weights in the scale-pan, until that end of the balance is somewhat the heaviest. Now as the candle. burns it becomes lighter, and soon begins to rise, its diminution grain by grain being shown by the index moving over the scale. Light the gas also at the other end of the beam, and place the photometer-disk on the bar ready for use. Precisely at the beginning of a minute, as given by the clock, read the gas-meter, recording the feet and thousandths, also the position of the index of the candle-balance. The observations commonly extend over five minutes, and in this time 10 grains of wax should be consumed ; set the rider therefore at 10, and if the candle is burning at the standard rate, the position of the index at the end of that time will be the same as at the beginning, if not, the difference shows. the correction to be applied.

Next measure the intensity of the two lights by the photometer disk. As stated above, this possesses the property of appearing light on a dark background, or the contrary, according as the brightest light is in front of, or behind it. By moving it backwards and forwards therefore, a point will be found where the spot will disappear almost completely, owing to the equality of the two lights. Read its position by the graduation and record, then move it, and set again several times. At the end of the minute read the meter, and then take some more readings of the disk. Try also setting the disk so that the spot shall be first slightly brighter, and then equally darker than the adjacent paper. This is called taking limits, and the mean gives the true reading. Pro

ceed in this way for five minutes, reading the meter at the end of each minute, and taking two or three intermediate settings of the disk. At the end of the time read the index of the candle-balance also.

From these data the candle power may be computed as follows. The consumption of the candle is obtained by subtracting the first reading of the index from the last, and adding the difference to 10. This gives the consumption in 5 minutes, and multiplying it by twelve gives C, the number of grains per hour. It is, however, safer to extend the reading of the candle-balance to a longer time, as fifteen or twenty minutes, to diminish the errors. The consumption of gas per minute is obtained by subtracting each reading of the meter from that which follows it, and multiplying by 60 gives G, the rate per hour. Call X the ratio of the two lights, as given by the mean of the readings of the scale attached to the bar, and apply the following corrections. First, for rate of candle, it is assumed that the light is proportional to the consumption. C Hence the corrected candle power L' : L = C : 120, L' = L 10 120

Again it is assumed that the light of the gas is proportional to its G

consumption, or to and dividing by this fraction gives what

5'

would be the candle power if just 5 feet were burned, or the true

5

candle power I′′ = L'; hence L′′ = L·

C 5 120 G

This example serves to show how the photometer is ordinarily used, but it may be applied to a great variety of investigations. For instance, different burners may be compared, or a single burner under varying consumption. The amount of light cut off by plain and ground glass at various angles may be measured, and the effect of changes in moisture, in temperature, or in barometric pressure studied.

70. LAW OF REFLECTION.

Apparatus. In Fig. 52 a circle divided into degrees is attached to a stand, and carries two arms with verniers, or simple pointers, C and D. The first is attached to a centre plate, which carries a vertical mirror placed at right angles to BC. This mirror is silvered on its front surface, or may be made of blackened glass,

and a vertical line is ruled on it, which is brought to coincide exactly with the centre of the circle. A vertical rod or needle is attached to D, whose reflection in B is to be observed at different angles of incidence. A is a piece of brass with a small hole pierced in it to look through.

[ocr errors]

Fig. 52.

8

A

Experiment. Bring C in line with B and A, and turn it so that on looking through the latter the reflection of the hole may be brought to the centre of the mir ror and bisected by the line marked on it. The reading of the index C gives the zero, or starting point. This observation should be repeated two or three times, dividing the degrees into tenths by the eye. Turn Ca few degrees and bring D into such a position that the reflection of its needle shall coincide with the line on C. Now the difference in reading of D and C will equal the angle of incidence, and the difference between the reading of C and the zero equals the angle of reflection. By the law of reflection these two angles should be equal. Repeat this observation with different parts of the graduated circle, at intervals of about fifteen or twenty degrees. Small deviations from the law serve well to exemplify the different kinds of errors of observations. Thus if the needle is not exactly on the line connecting its index with C, a constant error will be introduced. If the mirror is not exactly over the centre of the circle, the difference will vary in different parts of the circle, causing a periodic error. If the differences between the angles of incidence and reflection are sometimes positive and sometimes negative, they are probably due to accidental errors, such as errors in graduation, in reading, unequal fitting of the parts, etc. Finally, if a single observation gives a large error, it is probably due to a mistake, or totally erroneous reading.

71. ANGLES OF CRYSTALS.

Apparatus. The instrument most commonly employed to measure the angles of crystals is Wollaston's reflecting goniometer, represented in Fig. 53. A is a vertical circle divided into degrees, and turned by a milled head B through any given angle, which is measured by a vernier C. A second milled head D is attached to

a rod passing through the axis of this circle with friction. The crystal is fastened by wax to a small brass plate E, bent at right angles and resting in the split end of the pin F. By this it may be turned horizontally, and the joint G gives a vertical motion. The whole is mounted on a stand, which should be placed on a table opposite the window, and ten or twelve feet from it. A spring stop is attached to the stand, and a pin placed in the graduated circle, so that when the latter is turned forward it cannot pass the 0°, or 180° mark, but may be brought by the milled head exactly to this point. Several crystals to be measured should be provided, some, as quartz, galena, alum or salt, well formed and polished, and therefore easily measured, and others of greater difficulty. The best material with which to attach them to E is a little beeswax.

C

Fig. 53.

B

Experiment. The crystal must be fastened to the stand in such a way that the edge to be measured shall lie exactly in the axis of the instrument prolonged, and the main difficulty in the experiment is to make this adjustment with accuracy. Attach the crystal to the plate E by a little piece of wax, and adjust the edge as nearly as possible by the eye, turning it horizontally by the pin F, and vertically around the joint G. It is thus brought parallel to the axis, and may be made to coincide with it by sliding the plate in the pin F. Select now two parallel lines, one of which may be a bar of the window, and the other the further edge of the table, or a line.ruled on paper, and set the axis of the instrument parallel to them. On bringing the eye near the crystal an image of the window will be seen reflected in one of its faces, and by turning either milled head the image of the bar may be brought to coincide with the second line. If they are not parallel it shows that the face is not parallel to the axis of the instrument, and the crystal must be moved. Do the same with the other face to be measured, and when both images are parallel to the line on the table, both faces, and consequently their intersection, are parallel to the axis. This adjustment is most readily made by placing one face as nearly as possible perpendicular to the pin F, when the image in this face may be rendered parallel by turning G, that in the other by turning F

« PreviousContinue »