Page:Handbook of Meteorology.djvu/255

 established at that time. The pressure of the pen on the sheet is bound to vary slightly from day to day, and this in itself is likely to cause the rate of the clock to change. The pen should touch the paper positively but lightly.

The pen requires frequent cleaning; whenever the nib appears clogged it should be wiped with a bit of soft rag. Half a dozen times a year it should be removed and made thoroughly clean, scraping off the sediment that washing will not remove. A clean pen and good ink will leave a record as clear and clean as though the lines had been made with a drawing pen.

The wiring plan for the triple register is set forth in detail in Circular D, Instrument Division, and the details need not be rehearsed here. Wherever wires pass through woodwork, porcelain tube insulators are required by insurance regulations. If the wiring is situated where contact with electric light wires is possible, heavily insulated wires, such as are prescribed by local regulations, should be used. All permanent wire joints and splices should be snugly twisted a length of 2 inches—or, better, soldered—and wrapped with tape. All outside wires should be held by insulators.

Batteries.—In operating a triple register, 10 cells of battery are required; for the wind-direction register, 4 cells; for the anemometer, 3 cells; for the sunshine and rainfall recorder, 3 cells in common. For a 2-magnet register, recording wind-velocity and sunshine, 3 cells in common are sufficient. Unless the electromotive force is strong, however, the sunshine recorder may fail to register when the anemometer contact is on the bridge. This, however, need not lead to error; the bridge contact is always a long offset on the anemograph.

The requirement of battery cells for magnet registers is steadiness rather than strength. Dry cells run down in electro-motive force so quickly that, if the wind is on the bridge for more than a few minutes, the sunshine recorder fails to register unless operated by a separate battery. The ordinary “wet cells” are not much better.

When storage batteries suitable for the work are not available, the cells of the Edison primary type are the best. Such a cell properly charged gives a feeble but steady current on a closed circuit for nearly 400 hours. It will operate a 2-magnet register for more than a year.