The area to be surveyed is divided into a convenient number of districts, so arranged that, if possible, one instrument may measure the flow into the district. The district is formed by closing gates on the boundary. Care must be exercised to provide sufficient water to meet a sudden demand caused by a fire in the district. After the district is formed and the normal flow measured, the district is reduced by cutting off one or more blocks at a time, doing this work at night when the demand is at a minimum. The record of the pitometer will show how much reduction in flow is caused by each section of main shut off. Those sections which indicate a suspiciously high flow are further investigated, by pitometer or aquaphones, and the cause of the high flow ascertained. Usually it is found that one or more leaks exist on the main or service under suspicion. This method is one that must be carried out by men skilled in the use of these instruments if it is to be successful.

Hose and meter surveys. To determine the flow in one or more blocks a meter, usually of the 2-inch disc or current type, may be connected by hose lines to two hydrants, and the mains to be tested supplied by opening the two hydrants and closing valves, so that all the flow must pass through the meter. By mounting the meter on a light auto truck additional facility of measurement is afforded. This system requires many more measurements to be taken than the pitometer system, and is not so generally used, but the principle is virtually the same.

Aquaphone surveys. The aquaphone is one of the most valuable. of the devices available to aid in locating hidden leaks. It cannot tell whether leakage does or does not exist on a given long section of main or what is the rate of flow through the mains, but it can reduce greatly the cost of locating leaks after it is known that one or more leaks are on a certain section of main, on services or on other connections. By combining the tests of flow in mains with aquaphone tests, all leaks that are of sufficient size to justify their stoppage may be located and eliminated. Small diameter rods driven. down to the main aid greatly in locating leaks and avoid disturbances of pavement. In general, it may be considered good practice to test hydrants once a year, house services once a year, more or less depending upon whether the inspectors visit the premises for other purposes, and the mains about every two years or more.

Special devices of great sensitiveness have been manufactured to be used in place of the aquaphones in locating leaks through the sound made by escaping water. These devices are on the general principle of the stethoscope and in very quiet localities may be preferred to the aquaphone; the principle is, however, the same.

FINANCING AND MANAGEMENT

CHAPTER XX

RATES OF WATER CHARGES

Average rates of water charges and comparisons by groups are of limited value from the standpoint of precise information. In the opinion of some they may be misleading if used without an appreciation of the conditions for which they are representative. Nevertheless information obtained from the average rates of charges is of interest and seems to serve an educational purpose for those not interested in fitting rate schedules to the individuality of a particular water works problem.

Under date of February 16, 1925, questionnaires were mailed to water departments and companies in all cities of the United States and Canada having a population of 15,000 or more, requesting information in reference to water rates, water main installations, water treatment, payment for fire protection, payment of taxes, etc. Answers were received from 294 cities in the United States and 7 in Canada, having a total population of approximately 42,000,000. Included in this list are such cities as New York, Philadelphia, Chicago, Detroit, Minneapolis, St. Paul, Montreal, Toronto, San Francisco, Louisville, and New Orleans, together with smaller cities such as Bloomington and Evanston, Illinois; Kokomo, LaFayette, LaPorte and Muncie, Indiana; Shreveport, La.; Covington, Ky.; Boulder, Colorado; Auburn, N. Y.; Bay City, Ann Arbor and Adrian, Michigan, etc.

Basis of rate comparison

In the following table are four columns representing the charges which would be made on an average for each of the four groups below listed for water service as follows:

Column A-Charge per 1000 gallons for first 7500 gallons used in one month by domestic consumers, smaller stores, etc.

Column B-Charge for 75,000 gallons of water used in one month by apartment houses, stores, and smaller industries

Column C-Charge for first 1,000,000 gallons used in one month by factories, hotels, larger stores, etc.