A voltmeter in which the current passing through its conductor heats
such conductor, causing it to expand. Its expansion is caused to move an
index needle. By calibration the movements of the needle are made to
correspond to the potential differences producing the actuating currents
through it. The magnetic action of the current plays no part in its
operation. It is the invention of Capt. Cardew, R. E.

The construction of the instrument in one of its most recent forms is
shown in the cut. On each side of the drum-like case of the instrument
are the binding screws. These connect with the blocks m and n. To these
the fine wire conductor is connected and is carried down and up over the
two pulleys seen at the lowest extremity, its centre being attached to
c. From c a wire is carried to the drum p, shown on an enlarged scale on
the left of the cut. A second wire from the same drum or pulley connects
to the spring S. The winding of the two wires is shown in the separate
figure of c, where it is seen that they are screwed fast to the
periphery of the little drum, and are virtually continuations of each
other. By the screw A the tension of the spring S is adjusted.

On the shaft of the little drum p is a pinion, which works into the
teeth of the cog-wheel r. The shaft of r is extended through the dial
of the instrument, and carries an index. The dial is marked off for
volts; g g and h h are standards for carrying the pulleys.

The action of the instrument is as follows. The current passing through
the wire heats it. This current by Ohm’s law is proportional to the
electro-motive force between the terminals. As it is heated it expands
and as it cools contracts, definite expanding and contracting
corresponding to definite potential differences. As the wire expands and
contracts the block or pin c moves back and forth, thus turning the drum
p and cogwheel r one way or permitting it to turn the other way under
the pull of the spring S.

In this construction for a given expansion of the wire the piece c only
moves one half as much. The advantage of using a wire twice as long as
would be required for the same degree of movement were the full
expansion utilized is that a very thin wire can be employed. Such a wire
heats and cools more readily, and hence the instrument reaches its
reading more quickly or is more deadbeat, if we borrow a phraseology
properly applicable only to instruments with oscillating indexes.

In the most recent instruments about thirteen feet of wire .0025 inch in
diameter, and made of platinum-silver alloy is used.

If the potential difference to be measured lies between 30 and 120 volts
the wire as described suffices. But to extend the range of the
instrument a resistance in series is required. If such resistance is
double that of the instrument wire, and remains double whether the
latter is hot or cold the readings on the scale will correspond to
exactly twice the number of volts. This is brought about in some
instruments by the introduction in series of a duplicate wire, precisely
similar to the other wire, and like it, carried around pulleys and kept
stretched by a spring.

[Transcriber's note: If the series resistance is twice that of the
voltmeter, the indicated voltage will be ONE THIRD of the total
voltage.]

Thus whatever ratio of resistance exists between the two wires cold, it
is always the same at any temperature, as they both increase in
temperature at exactly the same rate. Tubes are provided to enclose the
stretched wires and pulleys, which tubes are blackened.

The voltmeter is unaffected by magnetic fields, and, as its
self-induction is very slight, it is much used for alternating currents.
The tubes containing the wire may be three feet long.

Its disadvantages are thus summarized by Ayrton. It absorbs a good deal
of energy; it cannot be constructed for small potential differences, as
the wire cannot be made thicker, as it would make it more sluggish;
there is vagueness in the readings near the zero point and sometimes
inaccuracy in the upper part of the scale.