# Viewing page 8 of 195

```10

Thursday Dec 15th 1842

Mr Welsh, the engineer of the Raritan Canal; called on me this morning and gave [[strikethrough]] an account of a series of not very extensive but very interesting experiments, at which he had been assisting, relative to the velocity, range, and elevation of a ball, shot from Capt Stocktons large gun of wrought iron. This gun throws a ball of 212 lbs and is found to hit the mark with great precision, at the  distance of 2730 feet. Mr Welsh has devised a very simple method of getting the velocity approximately by means of the deflection from a straight line. Thus, he measures the [[image]] fall of the ball by means of a series of screens placed at regular distances, through which the ball passes, and from these falls or deflections aa' bb' cc' &c he gets the times.

The object of making this memorandum is to record my own idea of getting the velocity the ball at different distances, by means of a current of galvanism. Mr Welsh had thought that some method might be devised for using electricity, forgetting the velocity of the ball, but as he said, his knowledge of the agent was not sufficient to enable him to devise any means. The idea first occurred to me of using the revolving mirror as a means of determining the [[strikethrough]] ?[[/strikethrough]] velocity, but the electrical method which almost immediately suggested itself appears preferable.

[[image]] The general arrangement will be something like this. A number of screens being erected at equal distance, a circuit ^[[of]] wire is made from each [[strikethrough]] screen [[/strikethrough]]  to a galvanometer placed tangent or nearly so to a large graduated  wheel, which is made to revolve once in a second or oftener. Suppose the wheel revolves once in a second and the needle of the galvanometer be made to carry a bar

[[end page]]
[[start page]]

[[vertical line]] To test the truth of the indications of the needles, two sets of wires may be attached to the same.
[[end vertical line]]

N 13 Oct 1843 I have improved the plan of this apparatus, and published an account of it in the proceedings of the Phil Society.

on its end so as to make a dot on the revolving cylinder at the moment the current is stopped by breaking the circuit. In this way the time may be noted to the 1000th part of a second. All the registers may be made possibly by one galvanometer, by using a current of less intensity than the one connected with the screens to deflect the needle constantly against a fixed pin.
The connected current may also be used, if the register be made with a number of galvanometers. The ^[[same]] current may be continued around each. If it should be found necessary to use several galvanometers , then the inertia of each needle and the resistance of the point may not be the same. In this case, it would be necessary to test the accuracy of the register, and this could be affected by causing another wheel graduated like the first to revolve with two pins on it at intervals along its circumference. These could be made to interrupt two circuits [[strikethrough]] ? [[/strikethrough]] by striking against two wires in mercury, and as the distance of this pin apart are known, they would give the elapsed time which should also be given of the same duration by the other revolving wheel. The difference of the result given by the index wheel, and the other, would probably be constant, or at least it could be made approximately so. If one of the needles were a little heavier than the other, then the time noted would be a little later, comparatively, and the interval would be shorter than the truth if the heavier needle were the first to move. If the heavier were placed behind the other on the register, then the interval would be lengthened.

The wheel may be made to revolve by means of a weight with thread over a pulley or cylindrical [[part ?]] of the axis, and the same axis furnished with two pairs of centrifugal balls to regulate the motion. If the wheel was a foot in diameter and revolved one in a second, then the 1/100th of [[strikethrough]] [[?]] [[/strikethrough]]  second would indicate by about .4 of an inch, and the 1/1000th part of a second by .04 of an inch.

[[end page]]```

Transcription Notes:
"The general arrangement [[will]] be.."