An Improvement in Voice Keys
Journal of Experimental Psychology
The voice keys I designed earlier for use in the timing of association reactions were simplified and improved from the Romer key, and have always given some trouble in the adjustment of the contact. The difficulty has arisen from the method of leading in the current to the swinging lever. Sufficiently good conduction cannot be obtained through the bearing of the lever, and I therefore employed a fine wire connection directly from the lever to a binding post. Such a wire connection can be made
... ction can be made to work, but if the wire becomes bent (as it easily may, and frequently does, through removal of a cloth cover, or by accidental touches), it changes the pressure of the lever against the diaphragm, breaking the circuit permanently or decreasing the sensitivity of the key. With improvements in the chronoscope, the voice key has remained the only weak point of my Association Timing Apparatus. The final model of the voice key, illustrated in the cut, removes the last difficulty, offering a key in which the contact is operated by gravity alone, and in which the maximum of sensitivity with maximum of reliability is therefore obtained. The key consists, as in earlier models, of a brass ring (S-S) supported by a brass block (P) on a base (2?) of hard rubber or bakelite. Within the ring is supported a diaphragm (D) of thin aluminum, carrying at its center a gold contact plate mounted on a mica disc (F). The contact-lever (L), with its gold-pointed head (H), is pivoted in a bearing (G) attached to the top of the ring (S): thus differing from the early models, in which the lever was pivoted at the bottom. From the top of the lever extends a bent wire (JV) dipping into a mercury-cup (C).