A NEW REFLECTING GALVANOMETER.

Fig. 1 shows an elevation of the instrument and a horizontal section of

the bobbins. Two pairs of bobbins, cc, cc, are so arranged that the axes

of each pair are parallel and in the same vertical plane. Each pair is

supported by a vertical brass plate, and the two plates make an angle of

about 106° with each other, so that the planes containing the axes of the

bobbins make an angle of about 74°. Two horseshoe magnets, m m, made of

1/25 inch steel wire, are connected by a very light piece of aluminum and

placed at such a distance from each other that, on being suspended, the

two branches of each of the magnets shall freely enter the respective

bores of the two bobbins fixed upon the same plate, and, when the whole

system is in equilibrium and the bobbins free from current, the two

branches of each of the magnets shall nearly coincide with the axes of

such bores. The magnets are not plane, but are curved so as to form

portions of a vertical cylinder whose axis coincides with the direction

of the suspension wire, and to which the axes of the bobbins are tangent

at their center, approximately to the points where the poles of the

magnets are situated.

[Illustration: FIG. 1. GRAY'S GALVANOMETER.]

The needles have been given this form so that their extremities shall not

touch the sides of the bore during considerable deflections.

In the instrument which the inventors, Messrs. T. & A. Gray, used in

their experiments upon the resistance of glass, the needles were arranged

so that their poles of contrary name were opposite.

[Illustration: FIG. 2.]

The system of needles is suspended from the extremity of a screw, p,

which passes into a nut, n, movable between two stationary pieces. On

revolving the nut, we cause the screw to rise or lower, along with the

entire suspended part, without twisting the thread.

The four bobbins are grouped for tension, and have a total resistance of

30,220 ohms. They contain 16,000 feet of No. 50 copper wire, forming

62,939 revolutions, nearly equally divided between the four bobbins. When

a current is passing through the bobbins, the poles of one of the

horseshoe magnets are attracted toward the interior of the corresponding

bobbins, while those of the other are repelled toward the exterior by the

two other bobbins. We thus have a couple which tends to cause the system

to revolve around the suspension axis. A mirror, which is fixed upon a

vertical piece of aluminum, a, gives, in the usual manner, a reflected

image upon a scale, thus allowing the deflections to be read. A

compensating magnet, M, is supported by a vertical column fixed to the

case, above the needles. This magnet may be placed in the different

azimuths by means of a tangential screw, t. The extremities of the bobbin

wires are connected with three terminals, T, T¹, T², and the

instrument may, by a proper arrangement, became differential. These

terminals, as well as the communicating wires, are insulated with

ebonite.

Thus arranged, the instrument is capable of making a deflection of one

division of 1/50 inch upon a scale placed at a distance of a little more

than a yard, with the current produced by one daniell of 10 ohms. This is

a degree of sensitiveness that cannot be obtained with any of the astatic

instruments known up to the present. By regulating the needles properly,

a greater degree of sensitiveness may be attained, but then the duration

of the needles' oscillation becomes too great. The sensitiveness of the

instrument is sufficiently great to allow it to be used in many cases,

even with a moderate duration of oscillation.

In their experiments upon the resistance of glass, the inventors employed

an instrument that was not arranged for giving great sensitiveness, and

one with which resistances of from 10^{4} to 10^{5} megohms could be

measured by the use of a pile of 120 daniells.

The instrument can be given another form. The four bobbins may be

arranged symmetrically in the same plane, and the two horseshoe magnets

be supported by an S-shaped aluminum bar. The latter traverses the plate

that supports the bobbins, in such a way that one of the magnets enters

one of the bobbins that correspond to it on one side of the plate, and

the other on the other side, as shown in Fig. 2. The bobbins are so

connected that, when they are traversed by a current, both magnets are at

the same time attracted toward the interior or repelled toward the

exterior of the bobbins. Such a form of the instrument has the advantage

of being more easily constructed, while the regulation of the magnets

with respect to the bore of the bobbins is easier.

The chief advantage of the instrument results from the fact that, owing

to the arrangement of the magnets and bobbins, a large portion of the

wires of the latter is situated very near the poles of the magnets, and

in a position very favorable for electro-magnetic action. The instrument

presents no difficulties as regards construction, and costs no more than

an ordinary one.

We might even arrange a single horseshoe magnet, or an S-shaped one,

horizontally, and employ but a single pair of bobbins, and thus have a

non-astatic apparatus based upon the same principle. But in astatic

instruments it is better to place the magnets in such a way that the two

branches shall be in the same vertical plane.

Were the line that joins the two poles vertical, the system would be

perfectly astatic in a uniform field, since each magnet in particular

would then be perfectly astatic. A pair of horseshoe magnets may thus be

regulated in such a way as to form a perfectly astatic system in a

uniform field and to preserve an almost invariable zero, this being

something that it is very difficult to obtain with the ordinary

arrangement of needles, especially when a compensating magnet is used;

for, in such a case, one of the needles becomes more or less magnetized,

while the other becomes demagnetized, according to the position of the

compensating magnet.--_La Lumiere Electrique_.

* * * * *

Scientific American Supplement, No. 492, June 6, 1885 · 第 40 / 59 题