Chapter III

THE MAGNETIC COMPASS AS A PART OF A SURVEYING INSTRUMENT
OR SEPARATELY--BROAD AND EDGE-BAR NEEDLES--MANUFACTURE
OF THE NEEDLE--MAGNETISATION--SUSPENSION--DIP AND ADJUSTMENT--
LIFTING--INCLINATION--DECLINATION--VARIATION--CORRECTION--
COMPASS-BOXES--DESCRIPTION OF COMPASSES--RING COMPASSES--TROUGH
COMPASSES--PRISMATIC COMPASSES--STAND--SURVEYING WITH COMPASS--POCKET
COMPASSES.

116.--=The Magnetic Needle=, which forms part of a great many surveying instruments, is made of the form adapted to the special purposes of the instrument in which it is placed. There are two prevailing forms commonly in use--one in which the needle is made pointed at one or both ends to read directly upon a divided circle fixed upon the instrument, and the other form in which it is made to carry and to direct a divided circle by its magnetic force.

The magnetism which gives directive force to the needle has been found by experiment to reside in every separate part of the magnet, that is, it is assumed to be a _molecular_ force. Therefore, it would not appear to be very important, within certain limits, of what form the magnetic needle is made, and this is found by experiment to be to a large extent true. The only important conditions appear to be that the needle shall be of such form that the inducing magnet, to be described, arts. 120-123, which is used for magnetising may be brought into contact upon every part of its surface, and that the molecular continuity of the parts should mutually support the general directive influence of the magnetism longitudinally in parallel lines.

117.--Magnetic needles are generally made in the form of flat bars, which are balanced upon a standing point falling into a cup which forms the centre. When the greatest section of the bar is placed horizontally it is termed a _broad needle_, as shown Fig. 26. This may be made of the lozenge form shown, or be parallel throughout. When the greatest section is placed vertically it is termed an _edge-bar_ needle, as shown Fig. 27. The north pointing end of the broad needle is commonly tempered dark blue, or has a deep cut across it, if the needle is left open. This is not necessary if it carries a ring. The edge-bar is generally used where it is required to read into a fixed circle of division, in which case its ends are brought to fine knife-edges.

118.--From the difficulty of reading a sharp point in bright metal against the black line of a divided circle, the author occasionally makes one point of the needle with a fine cut, sawn vertically for a short distance from its end, so as to form a kind of _split_ which is afterwards closed, so that it presents the appearance of a fine black line of the same character as the divisions into which it reads. With this, as shown Fig. 28, the reading is found to be much more easy. The point is also more readily adjusted by grinding, as the end of the needle being broad, less care is necessary to avoid reducing it so much that it may leave the interior of the circle short where it reads into the divisions. This form of needle is not adapted to mining instruments, which have often to be read in an oblique direction.

119.--_In the Manufacture of the Needle_ it should be made of the finest cutler's cast steel, or, better still, of steel containing 3 per cent. of tungsten. If not left in a parallel strip as it is drawn or rolled, it should be brought as nearly to its form as possible by forging at a low heat. The steel should not be over-heated for hardening. It should be hardened in cold water or oil, and be tempered afterwards down to a very pale straw-colour--in fact, the temper colour should only just appear. Long needles may have the temper sufficiently lowered at the centre to set them approximately straight during the tempering; but the temper should not be lowered even in the centre below a pale blue, _spring temper_. After tempering, the setting and working up to balance is best done by grinding, and for the final adjustment, by stoning with Water-of-Ayr stone.

120.--_Magnetisation of the Needle_ may be performed in many ways by means of a permanent magnet or an electro-magnet, or electrically by means of a solenoid. When the magnetism is induced from another magnet it is only important that the properly hardened needle should be regularly and equally magnetised over its surface by pressure upon it of the proper poles of the inducing magnet--that is, that the north pole of the magnet should induce magnetism in the southern half of the needle only; and the south pole in the northern half only.

121.--_Method of Magnetisation by Single-touch._--This method is more generally applied to touching up needles than magnetising them at first. The northern pole of a strong permanent magnet is stroked down the southern end of the needle from its centre to its end three times on one side of the needle. The needle is then turned round, and the northern end is stroked down in like manner with the southern pole. The needle is then turned over, and the process is repeated on the other side. This may be done a second time and the edges of the needle be stroked down also.

122.--_Method with both Poles._--In this process the needle is held down firmly with pegs on a board, and a strong horse-shoe magnet with rather close poles is laid on the bare needle without its cap, in a manner that both terminals press upon it. It is then drawn backwards and forwards from end to end of the needle several times, lifting the magnet finally from about the centre. The process is then repeated on the opposite side of the needle and its edges.

123.--_Method of Divided-touch_ is a somewhat quicker process, which does not entail removing the cap, the general plan of which is shown in the engraving below. The poles of the magnets, or one of them, is marked. Two good straight bar magnets are used. The needle is fixed down on a board and the poles of the two magnets are laid upon it at an angle of about 30°, applying one north or marked pole, and one south or unmarked pole. The magnets are then drawn apart in a horizontal direction along the needle, with constant pressure upon it, so as to reach the opposite ends of the needle simultaneously, and then again pressed back to the centre. After this operation is performed three or four times on one side of the needle, it is turned over and the process is repeated on the other side, being careful as before to use the same ends of the magnets upon the same ends of the needle. The operation may be repeated several times to be sure of saturation of the needle. It is better to lift the magnets off at the termination of the operation at the centre of the needle.

124.--It is found that the needle is magnetised a little more quickly if it is laid upon a strong magnetised bar during magnetising, or upon the ends of two bars, as shown in the engraving, Fig. 29, or on the two ends of a wide horse-shoe magnet.

125.--Needles are now more generally magnetised electrically by placing them in a solenoid or coil of stout insulated copper wire through which a strong direct current is passing from a dynamo or powerful battery. This method is employed in the best shops. The touch system above described is convenient for the profession for remagnetising a needle when weak, as a horse-shoe magnet at small cost may be kept for the purpose. It is generally used in small shops, as being at all times ready to hand, less expensive, and sufficient to ensure saturation if it is skilfully done.

126.--With every care in the manufacture of the needle there remains a little difference in the qualities of needles which are apparently otherwise identical. Little local differences in the quality of the steel, slight over-crystallisation from over-heating in hardening or unequal tempering, or unequal magnetising, are liable to form weak parts, or even what are termed _consequent points_. These are points in which the magnet possesses a reversal of its general longitudinal polarity. This can be made quite evident by experiment, as it is possible to make a needle not only with poles at each end, but with intermediate poles which are easily detected by sifting iron filings over it. The filings are found to adhere strongly at other local points than those near the ends, where a good magnet is alone strongly attractive.

127.--_Mounting of the Needle._--The needle for a surveying instrument has a female centre upon which it is suspended. The centre, termed technically _cap_, is generally formed of a hard precious stone, agate, chrysolite, ruby or sapphire, the latter being best, simply from the high polish it attains in grinding out with diamond dust. Rubies and sapphires are like minerals, except in the colour, which varies very much; the _off-colour_ stones, which are of small value for jewellery, are used for scientific purposes. The cap is mounted in a brass or aluminium cell made as light as possible for sufficient stability.

The needle is supported upon a hardened steel point, upon which it is perfectly balanced. The base of the point is tempered down to a low degree in order to admit a certain amount of bending to counteract the slight warping which generally occurs in the hardening.

128.--_Correction of Errors._--The needle, after it is mounted, although in balance may not have the steel placed symmetrically about its axis through slight curvature, unequal thickness about the cap, or otherwise, so that the magnetic direction is not perfectly linear between the points and the centre. If the points and centre are not magnetically linear, the correction for declination, which will be presently considered, cannot be made accurately. On this account it is better for the manufacturer to mount the needle on a slate bed with two sliding heads that may be brought up to the points of the needle. The heads have upon their upper surfaces lines drawn perfectly linear with the centre point of suspension of the needle, and a few lateral divisions to these lines for determining errors. On this bed the needle is placed upon the centre point to be examined how nearly its reading points are true with the axis. The error being recorded, the needle is demagnetised, and remagnetised end for end, and again examined. Corrections are then made by grinding or stoning from observations of bisections of the points cut in the separate readings, until the needle is made symmetrical and invariable, whichever end is magnetised for the north or south.

129.--_Lifting the Needle._--The needle of a surveying instrument should never be supported upon its centre except for the time it is in use for observation, as a fine steel point against a hard stone must, by any jar in conveyance from place to place, receive a certain amount of abrasion that will make it duller. For this reason a lift for the needle is always provided in scientific instruments. In the engraving, Fig. 30, an edge-bar needle is shown in section with its lift. The lift is made in the form of a bent lever, whose fulcrum is upon the bottom of the box. On the left-hand side of the broken line at _B_ the needle is shown lifted. On the right-hand side _A_ the needle is shown at its position for use, floating just slightly above the divided circle _D_. The pressure of the milled-head screw _C_ depresses the bent lever or lift on the bottom of the box and thereby raises the point under the centre of the needle. This point has a hollow cone formed upon it which fits over the standing-point to keep the lift in position. The cone fits externally into the cap to lift the needle vertically. The screw _C_ should always be clamped down when the needle is out of use. In place of the screw a wedge shaped sliding piece is sometimes fixed inside the compass-box, which is moved by a stud projecting through the outer case. Another plan of raising the lift is by a cam, or what is technically termed a _kidney-piece_, applied to the exterior part of the lift. Either of these plans answer, but the screw first described, being the gentler motion, jars the needle least. A screw is occasionally used longitudinally to the needle connected with a cam lift, the object in all cases being to lift the cap entirely clear of the standing-point.

130.--=The Inclination or Dip of the Needle= is the position a needle balanced level upon a free centre _before_ magnetisation takes in the vertical plane _after_ magnetisation. This inclination or dip varies in different parts of the globe, and at different times. At the present time at Greenwich (Jan., 1914) the angle is 66° 50′ from the horizontal. It is uniformly nearly _nil_ at the equator, and increases until over one of the magnetic poles, where it becomes vertical. There are two magnetic poles in the northern hemisphere active in directing the needle, one in Siberia, but the most active is about Melville Island; also two in the southern hemisphere, which are supposed to be nearly together, but the exact positions of which are not ascertained. As we require only the horizontal component in surveying and not the dip, it is necessary to balance the needle in opposition to the direction of the dip until it keeps in a horizontal position. This may be done by making the needle lighter on the dip side--that is, the northern in this hemisphere. But the plan adopted in all scientific instruments is to place a rider over the needle, as shown Fig. 30 under _B_. This clips the needle sufficiently to hold it firmly to its place, and yet is loose enough to be moved by the fingers to balance. The rider has to be shifted when the instrument is taken into a country where the dip is different from its position at home. When a needle is taken abroad without any rider, it may be balanced by means of a little sealing-wax placed upon its uptending end.

131.--To get at the needle for suppression of dip when it is placed in the compass-box, it is necessary to raise the spring ring, which is placed over the glass to keep it down, by inserting the point of a pocket-knife between the ring and the glass, moving the knife entirely round it and using a little twist upon it if necessary until the ring is free. This must be done gently or the glass will break. The needle is then adjusted to read correctly to the plane of the divided circle and is replaced in its box. The glass is then replaced and the spring ring is pressed down by passing the finger firmly round it until it is tight upon the glass. Sometimes a little extra pressure by a hard body is needed, but this must be done with care or the glass will be broken.

132.--=The Declination of the Needle=, that is, its variation in pointing in a true northernly and southernly direction, is necessary to be known and considered by the surveyor where the needle is used, both in relation to the locality and to the time, as this declination not only varies in different countries but also from year to year. For instance, this year (Jan., 1914) it points 15° 12′ West at Greenwich. The following chart, Fig. 31, gives the declination variation for 1914. The whole system of declination lines is now moving westward at the rate of about seven minutes per annum, but the rate varies slightly and from year to year. The declination lines, independently of correction, which will be presently considered, may not be exactly represented by the symmetrically curved lines shown in the figure. There are small local deflections from the theoretical curves here given, which are permanent and need local consideration when using the needle for obtaining very correct bearing. These have been ably considered by Professor Rücker and Dr. Thorp, but the subject is too complicated to be entered upon here, except for this note of observation.[2]

133.--For new countries, where the needle often becomes most important from the impossibility of tying up lines by direct observation through forests and other obstructions, reference must be had to magnetic charts which give systems of lines easily worked through by symmetry, even for unexplored countries. At present the declination is west in Europe and in Africa; east in Asia and the greater part of North and South America.

134.--=The Magnetic Variation of Declination in Time=, becomes important in reference to old plans in which the magnetic north of the period has been plotted for the true north very much to the pecuniary advantage of the legal profession when engaged upon actions with regard to disputed boundaries. The following table gives an idea of the variation in declination for Greenwich approximately for a few dates:--

Year 1580, Dec. 11° 36′ E. | Year 1860, Dec. 20° 40′ W.
" 1663, " 0°   | " 1870, " 20° 19′ W.
" 1700, " 8° 20′ W. | " 1880, " 18° 58′ W.
" 1818, " 25° 41′ W. | " 1890, " 17° 9′ W.
" 1850, " 19° 31′ W. | " 1900, " 16° 30′ W.

It will be seen by the above table that the needle pointed due north in 1663, that it attained its greatest western declination in 1818, and that it is now losing its westerly declination at the rate of about 7′ annually.

135.--=Annual Variation.=--The declination is subject also to a small annual variation which is greatest about spring time, diminishes towards the summer solstice, and increases again during the following nine months. It varies at different periods, and seldom exceeds 16′ of arc.

136.--=Declination Correction= to true north may be made for the compass by observation in this hemisphere of the pole star, which is practically due north in January at 6 p.m., February at 4 a.m., March at 2 a.m., April at midnight, May at 10 p.m., August at 4 a.m., September at 2 a.m., October at midnight, November at 10 p.m., December at 8 p.m. Most surveying instruments, except the transit theodolite, are not made convenient for this observation. More generally observations of the position of the sun may be made where a sun-glass is provided to the telescope of the theodolite, Fig. 19, _SG_, page 45, with the aid of a chronometer or a good watch. For this observation we may remember that the sun is true south at twelve o'clock on the 16th April, 15th June, 1st September, and 25th December. The following table may be useful for some intermediate times to show how much the chronometer (mean time) is faster or slower than the sun's southing approximately at noon:--

Jan. 1 subtract 4 min. | July 15 subtract 6 min.
" 16 " 10 " | " 30 " 6 "
" 31 " 14 " | Aug. 14 " 4 "
Feb. 15 " 14 " | Sept. 13 add 4 "
Mar. 2 " 12 " | " 28 " 9 "
" 17 " 8 " | Oct. 13 " 14 "
April 1 " 4 " | " 28 " 16 "
May 1 add 3 " | Nov. 12 " 16 "
" 16 " 4 " | " 27 " 12 "
" 31 " 3 " | Dec. 12 " 6 "
June 30 subtract 3 " | " 31 subtract 3 "

137.--As variation in time of southing is from fourteen minutes fast to sixteen slow, or a difference of thirty minutes, correction becomes important, as the sun passes over 7½° in this period. In these observations the diaphragm lines, webs, or points must bisect the sun's disc. This is done more exactly by taking the mean positions of the sun's eastern and western limbs or its semi-diameter, which is given for every day of the year in the _Nautical Almanac_.

138.--=The Compass-box.=--The needle, as it is generally mounted for the theodolite, mining-dial, and many other instruments, reads into a divided circle of 360°. The circle is raised up from the bottom of the compass-box to the height of the top of the needle, as shown in section Fig. 30, _D_, and is generally silver-plated. The bottom of the compass-box is sometimes divided with a _compass-rose_ giving the points N. E. S. W. The E. and W. in some cases are reversed from their natural directive positions from the centre of the box, so as to read the letter indicating the point nearest to the division instead of that opposite to it. In modern surveying instruments, however, no regard is paid to the points of the compass, north being 0°, east 90°, south 180°, west 270°.

139.--In the manufacture of the compass-box very great care should be taken that the metal is quite free from iron, and that no iron comes near it. On this point the maker cannot be too guarded. The author has in several instances found the compass-box of perfectly free metal; but a single foul screw made of commercial brass wire, being used to fix the ring or the rose plate, has by its influence entirely destroyed the value of the compass.

140.--In the construction of the compass-box the author has found the most certain method of getting the divisions correct with the centre is to make the division directly from the standing-point of the compass, and not to try to get this point correct to the divisions afterwards. The standing-point may be fixed directly to the box by screwing, or be attached to a brass plate before fixing. It is adjusted to the compass-box by bending until the needle turns freely, but at the same time nearly touches the circle. The needle is then removed and the circle is divided with the point as its centre. Where the divisions read to the point of the needle, or to a line upon it without a magnifier, the divisions of the circle may be made directly upon the lathe by a lever to the slide-rest if the lathe has a well-divided headstock. When the divisions are magnified and require great accuracy, or where a floating ring is used upon the needle, the circle should be divided upon the dividing engine, which will be described further on, the centre used being still the point or pivot on the bottom of the case, from which the divisions are to be made radially.

141.--=Preservation of the Magnetism in Needles.=--It is most important that the magnetism of the needle, particularly in mining-dials where so much depends upon it, should be preserved to near saturation in order to secure certain direction in opposition to the friction of the centre, necessarily always present. This is often much neglected from carelessness, or want of knowledge of the principles of magnetic action. In the first place we know that a bar of soft iron, possessing no evident magnetism, if it be placed in the magnetic meridian with proper dip, will after a time manifest strong magnetic properties. Thus, such a bar in London placed due north and south, with a dip of 67° to the north, becomes a weak magnet. From this we may also infer, and this experiment shows, that a needle placed in this position will not lose its magnetism. But what is most important to observe is that if the needle is placed in a _contrary direction_, as, for instance, with its northern end towards the south, it is in constant opposition to the influences of terrestrial magnetism, and will certainly become weaker. Therefore, although it is necessary to lift the needle when carrying the instrument, which must necessarily place its poles in all directions, it is not at all necessary that the needle should be lifted when the instrument is put by out of use. Indeed, magnetism is _materially preserved_ by releasing the lift to let the needle take its true bearing. This does not at all injure the standing-point, as there is no movement upon it to cause wear. Of course if the needle is at first magnetised beyond its permanent condition it will lose this surplus magnetism, but the residual magnetism in this position will remain nearly constant.

142.--A valuable precaution for a needle in constant wear is occasionally, say twice a year, or much oftener if it is used in a dusty mine, to take it out of its box and wipe out the cap with the point of a small sable brush. The standing-point may at the same time be sharpened if necessary by gently rubbing it all round with a slip of oiled Arkansas stone at its former pointing angle. The sharpness of a needle is easily ascertained by sliding the thumb-nail over the point at an angle of about 30° to it. If the point sticks and holds the nail, it is sharp; if it glides upon it, it is dull. The author has often had compasses of various kinds sent to him for remagnetisation whose only fault has been dulness of centre.

143.--=Ring Compasses.=--In modern theodolites, levels and prismatic compasses, the magnetic needle carries a light divided circle, which is now generally made of aluminium on account of the extreme lightness of this metal. A broad needle is used of about ¼ inch in width and 1/18 inch in thickness. There is considerable difficulty in mounting the circle to get it truly concentric and correct for bearing, therefore ring compasses are often found to be inaccurate. The author has followed two methods of construction, either of which answers fairly well:--The one is to leave a bar across the compass when cutting out the compass ring from a plate of aluminium. In this case, when the outer edge of the ring is chucked in the lathe to be turned, a centre hole is also made in the cross-bar which exactly fits over the cap of the needle, so that the adjustment for centre is practically secured, and attention is only necessary to get the adjustment correct for bearing--that is, the 0° at true magnetic north to the axis of the needle. Another method, which was suggested to the author by the late Mr. Thos. Cushing of the India Office, answers perfectly, and only entails a little extra trouble in setting for dividing. This is to permanently mount the ring on the needle without any means of after-adjustment, and to divide the circle from a point placed in the axis of the dividing engine, upon which the ruby centre is placed, being of course particular that the zero line 0° cuts the magnetic axis true north in the graduation.

144.--=Mariners' Compasses=, and an inexpensive class of prismatic compasses, are made with a paper disc in place of the ring above described answering the same purposes. The paper disc is generally made in two thicknesses with a thin sheet of talc placed between them. Mariners' compasses have frequently the divisions painted directly upon talc for transparency by lighting from beneath, also for general lightness combined with stiffness.

145.--The reading of mariners' compasses, and the compasses on levels where the needle carries a divided ring, is taken from a line drawn vertically up the inside of the box or a pointer. This _lead_ line in the mariners' compass gives the direction of the head of the vessel; a pointer in the level compass gives a direction in line with the axis of the telescope. In high-class theodolites, a microscope is used by the author reading to a spider's web in the diaphragm.

146.--=Trough Compass=, sometimes termed a _long compass_. Where an instrument possesses a double vertical axis and a divided circle, as the theodolite, the division of the circle may take the place of the divided ring of the compass and save the repetition of the graduation, at the same time the needle may often be made longer, as the bulk of the compass-box is proportionately less. In fact in all cases where the magnetic north only is required the trough compass is to be preferred. The ordinary construction of this compass is in the form of a narrow box, Fig. 32, _A_ representing a plan, and _B_ a parallel section taken through it horizontally. About 10° are graduated on each side of the meridian line, _aa_ being adjusting screws to bring the scale true with the needle.

147.--=Magnification of Reading.=--With the trough compass it is very common to have some form of microscope for reading the needle more exactly. This may be done by a Ramsden eye-piece being placed directly over the needle, as is common in some German instruments. A much more convenient plan for certain instruments is to read the needle longitudinally. This is generally done by means of a transparent scale being placed across the end of the needle which is divided upon glass or horn. This may read to either the near or distant point of the needle. A very good form of needle reading is found in some French instruments. This is shown Fig. 33, where the compass is shown entirely enclosed in a tube _C_ which protects it from dust. The needle _N_ has a vertical point fixed upon its end at _P_ which reads pretty closely to a scale of 10° divided upon glass at _G_ by the eye-piece _E_. It has a lifter _L_ pressed up by a milled-head screw _M_. Fig. 34 shows the graduated glass. This compass is attached beneath the limb of a theodolite, or in any other convenient position upon an instrument. The author has placed a compass constructed upon this principle in a telescope, in such a manner that the needle may be read with the eye-piece, so as to cut a line with a distant object coincident with the line cut by the principal telescope of the instrument at 0° of its graduation. This plan will be more fully explained with tacheometers,