The sterilization of public water supplies has been the subject of careful investigation by many municipal and private corporations for the last two or three years, and more so in the past year, owing to the shortage of water and the tact that many supplies, including wells, reservoirs and small streams became badly polluted due to wash when there were rains, and at other times when there was not sufficient water to properly flush the streams or surface to prevent pollution from such wells through the sand strata. These investigations have invariably met with the obstacle which has been the stumbling block in all such enterprises, that is, first cost and operating cost of the purification system.

The Baltimore County Water and Electric Company have been operating a filter plant for a number of years, and being desirous of obtaining a cheaper method of sterilization, have from time to time been investigating the various methods of filtration, namely, slow sand, American or rapid sand, and the up-draught system of sand filtration here, as well as the European. This company has also experimented for several years with the application of alum and hpochlorite of lime applied directly in the reservoirs or storage basins themselves, with considerable success, as will he shown in this paper.

In addition to the above-mentioned careful investigations, the sterilization of water by ozone was considered about two years ago. The United Water Improvement Company controlling the James It. Bridge patents, submitted a sample 3,000-gallon ozone sterilization plant for test, which was placed alongside the American or rapid, and slow sand filters, handling raw water directly without sedimentation. Some of the results obtained from this sample plant have already been reported in the previous paper on the use of hypochlorite of lime, which has also been in use both in the American and English filters for some two years. The results from the use of hypochlorites are given from careful tests made by our chemist. Shepherd Powell. In addition to these tests, a rough determination oi the amount of nascent oxygen in a grain of chloride of lime has been made, and the approximate calculation of the volume of gas on a basis of I grain per gallon, to get an approximation for the volume of ozone, which was roughly .00096 cubic inch per gallon, hot this depends upon the concentration and will vary with the condition of operation and temperature. The specific gravity of ozone throughout the test has been taken at 1.05 as compared with air as 1.

With the data at hand, the ozone plant at Lindsay. which had just been started, was investigated. and it was decided to install a sterilization plant at the Herring run station of the Baltimore County Water and Electric Company, to have an ultimate capacity of 10,000,000 gallons in 24 hours with roughing filters to remove the suspended matter, to handle the raw water at double or triple the usual rates per unit area in rapid or American sand filters. Plans were gotten out and work started, hut considerable difficulty was nut with owing to quick-sands. The sterilization pit was Ho feet deep and within 25 feet of the river hank. Ihc walls were not completed until it was late in the fall. Work had to be abandoned during the winter other than to test the many different kinds of mixing heads or aspirators, including various nozzles, converging tubes, orifices, etc.

As it was important to get information as to the amount of air that could he mixed with a given volume of water under our low head, it was found that under low heads some aspirators would not give a continuous draught of ozone or air. and when the draught was heavy, the specific gravity of the mixture would become so light that there would he a blow-back, which would momentarily blow the ozone input back, and while it was possible with low heads to get about one-third ozone to water in volume, the lowest point or head that should he considered is about 12 feet. Though some changes which were made in the design of this plant and adopted at Ann Arbor, Mich., will allow the use of heads of from six to ten feet, same will not give a steady input of ozone, as it is intermittent. This, together with the use of a small blower, seems so far to be satisfactory.

* Paper read at the annual convention. American Water Works Association. New Orleans.

The area required for plant of size installed is only about one-tenth that required for American plant, and is not to be compared with the area required for slow sand filters. The sterilization plant proper constitutes raw water chamber, aspirator heads, ozone chambers, ozone generator as well as time contact basin or mixing cell. The latter compels the water to travel horizontally as well as vertically, so as to give water the benefit of as long contact as is possible with ozone.

The results of the many tests of the different cones will he given in some subsequent paper.

The ozone generator consists of boiler plate, shell and head, containing each 109 2-inch aluminum tubes, inside of which are micanite tubes or dialectrics, inside of which is another tube or cylinder which rests on a tray, which is connected to one terminal of a special 10,000 to 15,000 volt static transformer. The other terminal, which is a ground terminal, is connected to boiler plate shell; inner and outer tubes being so close and only separated by micanite tube, allows a continuous violet discharge to take place through the dialectrics, and the passage of air through this discharge produces the ozone, the production running from eighty to one hundred grammes per kilowatt, and the concentration of five or less, depending upon the condition, velocity and the temperature of the air discharged.

Sparks must be prohibited, as when a sparking condition takes place nitrous oxide is generated, and also high temperature should be provided against in the generators.

There is under consideration another type of ozone generator, which from test on a small experimental unit has given as high as 60 grammes of ozone, with concentration of 20, but it is too soon to say how this will work out in much larger sizes. This company expects, however, to have one of these generators in use soon for test, and more definite results will be obtained. The ozone generators may be either water, oil or air cooled. If it is air cooled, the outer shell has to be removed and the top and bottom heads supported by four rods, or a frame used, and air caused to circulate by any means between and around the outside of the fixed tubes.

The ozone generator in use here will he cooled by oil. to prevent, if possible, the sweating of the tubes when water is the cooling medium, as the oil will remain warm for some time after current is cut off. preventing moisture collecting and giving trouble at starting. The only trouble to he feared with oil cooled system is leakage, owing to the fact that it is a hard matter to get joints that will stay tight under oil pressure, although the pressure will not exceed from two to three pounds under any condition. Oil is circulated by means of small 1/2-horsepower motor belted to rotary pump, the oil being passed through copper coils in a tank or cylinder through which water is flowing continuously. Each generator has its thermometer for determining the temperatures, and a controlling panel on which will be mounted indicating wattmeter, potential transformer plug, high tension switch, voltage regulating head which will permit of handling 2,300 volt current on the primary side of the transformer in such a manner as will give voltage of the secondary side of the transformer range from 7,500 volts to 15.000, or to be varied to intermediate volts. In addition to this there is a voltmeter reading to 15,000 volts, power factor meter and frequency indicator. These panels are located over a concrete vault in which are located the transformers. Each 1,000,000-gallon unit requires 5 kilowatts in transformer capacity. All the switching is done at 2.000 volts on the primary side, and the high tension side is connected directly to the ozone generator through varnished cambric lead incased cable two-conductor, placed in bitutnitiized fiber tubes, encased in concrete.

Owing to the weather conditions and the fact that some of the appartus lor this plant, such as insulators, etc., were special, and the departure from the original design, continuous tests have only just begun on this plant.

Considerable information was obtained, on some of which further experiments will be carried out, to determine what would appear to be the slip of the hubbies of ozone or air, and the fact that they do not seem to fall at the same velocity with the water, or for some reason do not receive or hold the kinetic energy imparted to them, and the fact that the bubbles seem to rise at about .75 of a foot per second, this, together with the fact that the specific gravity of the air bubbles and water, when the amount of ozone or air reaches certain entrainment, is so low that it loses its kinetic energy and the blowoff is caused, previously mentioned, and is similar to the action of the air lift, the mixture looking like soap bubbles in the down pipes momentarily.

When the temperature of the atmosphere is at or below zero, the strength of the ozone seems to be greater, and the absorption by the water varies as the strength of the ozone given off by the water in the relief chamber over the time contact basin or pit is noticeably less to the smell, and a longer or shorter time is required in high temperatures or extremely cold weather before the ozone is noticeable. But the ozone in this relief chamber becomes noticeably thin when the temperature is high, but varies. Bunsen and Carius found that one-hundred volumes of water near freezing absorb four volumes of oxygen and two volumes of nitrogen, and the absorption is quickened by stirring or mixing. Tests are under way to ascertain more fully this relation to the absorption of ozone as water absorbs more air, and the more of it the lower the temperature ; also it absorbs more oxygen than nitrogen. It is to determine this relation when ozone is mixed with water, as no data are available at the present time, that experiments are being made. From this, it seems that these gases when absorbed are given up as pressure lessens, that is, given up in minute bubbles, similarly to water which is highly charged with air, and allowed to stand in a glass, which gradually becomes clear at atmospheric pressure. A cubic foot of water holds in solution about .029 cubic feet of air at about freezing point, and it is possible that the absorption or change increases or decreases the CO2 in the water when ozone is applied.

The generators require about eighteen watts per square foot surface in the electrodes at 60 cycles, but it is hoped to materially reduce the wattage with the new form of generators.

The experiments carried on by our chemist, Mr. Powell, are a part of this paper. The tests below, which have been carried on under Mr. Powell’s supervision, will undoubtedly prove very interesting:


The company’s filtration plant at Avalon, Md., which had already been mentioned, consists of slow sand and mechanical filters, and an experimental ozone apparatus. Since the construction of the rapid filters, about two years ago, hypochlorite of calcium has been used for sterilizing the water in connection with filtration at this plant. The hypochlorite has not been used continuously, as at times the raw water contains but a few hundred bacteria per cubic centimeter. The object in using the bleach has not been necessary because of the failure of either system of filtration. but to lessen the cost of operation and at the same time to deliver water which was practically sterile.

Experiments also were carried on in the laboratory for a better knowledge of the action of the bleach upon the raw water, and to determine the bacterial efficiency under abnormal conditions. The hypochlorite of lime used was received in 750pound sheet-iron drums. Only a small opening was cut in the drum so that it could be tightly sealed to prevent rapid deterioration of the chemical. Average analyses of the bleaching powder contained 34.2 per cent, of available chlorine. A 0.5 per cent, solution was used and was applied to the water through a graduated orifice. The storage solution tanks are of reinforced concrete and have a capacity of 41.5 gallons. The solution flows from the storage tanks to the orifice boxes which also arc made of reinforced concrete. The head of solution upon the orifice is kept constant by a valve controlled by a ball float. The tanks and boxes before being used were painted on the inside with elaterite paint, so there has been no trouble from leaking. The orifices were graduated at the plant and have been checked from time to time as to their accuracy.

In making up the solution, the required amount of bleaching powder is first dissolved in a small quantity of water to insure thorough breaking up of all lumps. This solution is then put in the storage tank and the required quantity of water run in.

Under ordinary conditions the capacity of each tank is sufficient for four or five hours’ supply. By making up the solution so often, the hypochlorite does not lose its strength to any appreciable degree, and it is not necessary to change the setting of the orifice unless there is a change in the character of the raw water.

The table given here shows the amount of bleach in grains per gallon applied to the water at different settings of the orifice, ft will be noted that the chemical is given in grains per gallon and not as parts per 1,000,000 of available chlorine or oxygen. The only object in doing this is that grains per gallon is more easily understood by the filter operators than the other terms mentioned.


Capacity of storage tanks, 41.55 gallons.

Charge of bleach, 2 pounds.

Orifice graduated in inches.

Rate of filtration, 1,025,000 gallons per 24 hours. Razo Ozone Raw Ozone Razo Ozone Razv Ozone

There was an apparent increase in the bacterial reductions from the bleach when used in connection with sulphate of alumina, over the reduction obtained when bleach only was used. Experiments were undertaken in the laboratory to determine if this condition existed. Tests were made upon the raw water with conditions as near those at the filtration plant as was possible to obtain. One set of samples was treated with various amounts of bleach from 0.05 to 0.5 grains per gallon of water. To another set of samples were applied the same amounts of bleach, but to each jar of water was added sulphate of alumina at the rate of 0.5 grain per gallon. Several of these tests were made during the year.

The color of the raw water at Avalon ranges from 5 to 65 parts per 1,000,000, most of which is not in true solution, hut in a colloidal state. The water is more highly colored during spring and fall floods than at other times of the year. The effluent from the mechanical filters is colorless at all times, also the effluent from the slow sandbeds, when the color of the applied water is below 20 parts per 1,000,000. When the color of the raw water is greater than this, there is always some color in the filtered water from these beds.

By the use of hypochlorite of lime it has been possible to reduce this organic stain from 5 to 15 per cent., the percentage reduction depending on the character of the color, the amount present, and the quantity of bleach used.

The hypochlorite of lime removed color that was in true solution and not in colloidal form. The quantity of hypochlorite of lime as used here, as in other experiments carried on in the laboratory, was in excess of that which was used at any time at the filtration plant. The reason for using to much was simply to demonstrate the action of the hypochlorite of lime which was the only object of this special work.

An attempt was also made to determine the effect of hypochlorite upon organic matter as measured by the required oxygen. These experiments, as those upon color, were done at the laboratory and consisted of treating liter samples of raw water with bleach varying from 0.05 to 0.5 of a grain per gallon and determining the organic matter before and after treatment.

From the data at hand, the reduction of organic matter by the use of hypochlorite was hardly noticeable even with water high in organic matter, and when compartively large quantities of the chemical were used.

To determine the efficiency of the bleach on the operation of the mechanical filters, a two months’ test was run. The filters were operated during the month of November, using only alum. During December, lrom the fourth to the end of the month, hypochlorite of lime was applied with the coagulant. There was but little change in the character of the raw water during this period, so the results can be compared fairly well. By the use of a very small amount of the hypochlorite (averaging 0.087 grains per gallon) it was possible to reduce the alum from 0.87 to 0.58 of a grain per gallon.

The percentage of water used in washing the filters was reduced from 4.1 per cent, to 2.9 per cent., at the same time increasing the length of time between changing the filters one hour and ten minutes. This reduction in coagulant of 0.23 of a grain per gallon amounts to 31 pounds per 1,000,000 gallons: with the cost of alum at 1 3-10 cents per pound f.o.b. Avalon, the saving was 41 cents per 1,000,000. Deducting from this amount 11 cents for the amount of hypochlorite used, leaves a net saving in the cost of coagulant of 30 cents per 1,000,000 gallons. The actual saving in the operating expenses really amounts to much more, as to this must be added the saving in wash water and the increased amount of water passing through the filter between cleaning.

The average number of bacteria of all samples of filtered water taken during November was 18 per cubic centimeter, while in December, when the bleach was applied with the coagulant, the average was 8 per cubic centimeter.

Tests also were made to determine the effect of hypochlorite of lime upon water of high turbidity. The results from these experiments have shown that as the turbidity increases, the bacterial efficiency of the hypochlorite decreases. This is partly due to the increase in organic, which invariably increase with the turbidity, and partly to the bacteria being mechanically protected by the particles of matter in suspension. With water having a turbidity greater than 50 parts per 1,000,000, the hypochlorite of lime gave better bacterial results, when part or all of the turbidity was removed before treating with the chemical. This work has also shown that with the raw water at Avalon, increasing the dose of hypochlorite about 0.10 of a grain per gallon, but slightly increased the percentage reduction in the number of bacteria.


The use of hypochlorite of lime at this plant has shown that:

The effluent from both the slow sand and mechanical filters can be made practically sterile, with the total elimination of B. coli, reducing the color from 5 to 15 per cent.

To reduce the amount of alum and to increase the length of runs between changing the filters, and at the same time reducing the amount of wash water.

Also that hypochlorite of lime is more efficient when used with alum than when used alone.

That high turbidity in the raw water reduces the bacterial efficiency of the hypochlorite of lime.

And that the hypochlorite of lime even when used in large quantities causes but a slight reduction in organic matter.

Waterworks Litigation.

Shenandoah, Pa., has agreed to settle with six owners of mills on Catawissa creek for water rights, of which it is claimed the borough in question had deprived them. The water was impounded in a reservoir that Shenandoah built, and the mills had to put in steam power. The cases had been prepared for presentation in court, Inti the borough settled them, paying, it is said, the lull amount ot the claims, which is more than $30,000.

the attempt of the city of Gainesville, lex., to secure the torieiture oi the franchise ot the Gainesville W ater Company, on the grounds that the company had not fulfilled its contract and was m an insolvent condition, making such fulfillment impossible, has been defeated. The court decided that the company was not insolvent to the extent ot being incapable of carrying out its contract and that the evidence of non-lullillment was conflicting.

The executive committee of the sewerage and water board of New Orleans, La., has agreed to settle the $60,000 suit ol the New Orleans Mater Supply Company tor $15,000, which the company is willing to accept. The report goes to the lull hoard for approval.

Suit has been commenced against the Springfield Consolidated Mater Company by the owners of 21 small houses in Media, Pa., used as null tenements, the water rate of which has been advanced, under company’s new minimum schedule, from $10 to $18 per year. The company threatened to cut off the water if the bill was not paid, but a preliminary injunction to prevent such action was granted. The company, which supplies about 75 towns and villages in the outskirts of Philadelphia, seeks to have the injunction removed, on the ground .that the new minimum rate is neither unjust nor unreasonable, in view of the fact that no limit is set as to the quantity of water that may be used, and that it has the right to cut off the water where the bills are not met. The property owners seek to have the injunction, which was granted under a bond of $500, made permanent. In the course of a hearing the judge intimated that the case looked like “unreasonableness against the small buyer.” The court reserved its decision, the injunction being in the meantime continued, pending the outcome of the suit instituted to prevent the increase in rates.

The Roaring Creek W ater Company, which supplied I crndale and hairview, suburbs of Shaino kin, Pa., when its o nlract expired recently, did not renew it and discontinued the supply.

A case in progress at Oakland, Cal., promises to furnish a liberal crop of precedents for use in litigation relating to subterranean water rights. The Pleasanton Hop Company had extensive and productive hop fields on tin upper part of a mountain and had sunk wells from which they obtained a supply of water for their irrigation. The Spring Valley Water Company, at a point lower down tile mountainside, also sunk wells from which to obtain a supply of water for use in its business, and as soon as it began to pump them the hop company’s wells went dry. Thereupon the hop growers, with other property owners, who claimed to have suffered damage, commenced suit against the water company, asking damages for their direct loss and injunction to prevent further interference with their water rights. A jury will try the case, which is expected to occupy the attention of the court for a long time, but as a preliminary there are a number of points of law to he settled, and on these the lawyers are now arguing.

Because it could not furnish the Seventh ward of the city of Norfolk. Va„ with an adequate supply of water for fire protection at 30 pounds pressure under an ironclad contract with the city, and at the same time supply all its other customers, the Norfolk County W ater Company ‘lint off the supply from Ocean View and Willoughby. The Norfolk &* Portsmouth Traction Company and the Chesapeake Club, of Ocean View, for the benefit of themselves and others similarly situated, went into court and obtained a temporary injunction, compelling the turning on of the water They will endeavor to have the order made permanent.

New Cumberland. Pa., had a contract with the Mountain Water Company under which at the end of ten years the city could purchase the plant. In the meantime the Mountain company sold out to the Riverton Consolidated Water Company, and when at the expiration of the ten years New Cumberland wanted to buy the plant the new owner refused to recognize the option. Suit was instituted to compel the water company to observe the contract, hut a demurrer was filed to the hill of complaint, which the court has sustained, and the case has been dismissed. New Cumberland, it is asserted, will carry the case to a higher court.

No posts to display