perchloride of iron, the subsidence in basins of between 40 and 55 per
cent. of the total organic matter and between 60 and 95 per cent. of the
total suspended matter can be obtained. The bacterial removal is between
80 and 90 per cent., depending upon the character of the sewage. The
objections to this process are great cost of chemicals and labor
required and the difficulty of disposing of a large amount of sludge.
There are a few plants of this kind in operation at the present time and
there seems to be a general agreement among authorities that the process
is now a back number. Fowler says, “It may be doubted whether dilute
sewages resulting from the lavish use of water in American cities lend
themselves generally to economical treatment by this process.” Metcalf
and Eddy in their “American Sewerage Practice” express the opinion that
the quantity of chemicals required for results would be a prohibitive
expense. The sewerage commission report of New Jersey contains the
statement that “on the standpoint of the officials in charge of the
experimental station at Lawrence, Massachusetts, chemical precipitation
is a process of the past.” The experiments of the Massachusetts State
Board of Health showed that it is quite impossible to obtain effluents
by chemical precipitation which compare in organic purity with those
obtained by intermittent sand filtration. About the only plants of any
importance in the United States are those at Worcester, Massachusetts,
and Providence, Rhode Island. According to the report of the
Superintendent of Sewers of Worcester, the experimental plant in that
city has shown that “the cost of operation of Imhoff tanks and
sprinkling filters per million gallons of sewage treated would be much
less than the cost of operation of chemical precipitation or sand
filtration as carried on in Worcester.”
Slate Beds
The equipment for this process consists of tanks with horizontal slabs
of slate separated a few inches by stone blocks. The sewage is allowed
to stand in the tank for about two hours, during which the suspended
matter is deposited on the slabs and is digested by multifarious forms
known as aerobic germs, _i. e._, germs requiring oxygen for the
continuance of their proper vital function. The deposits are thereby
reduced to harmless and inoffensive humus. Slate beds are dosed and
rested alternately so as to give them an opportunity to replenish their
supply of oxygen. Multiple units are therefore necessary. The effluent
must be treated as a tank effluent. Fowler suggests that when filters
are used to purify the effluent, “humus” tanks be provided between the
slate and the filter to retain the solids washed away from the beds and
somewhat to equalize the composition of the effluent passing into the
filter.
Dosing Chambers
After the effluent has passed from a tank after being treated by one or
more of the preliminary processes, it usually flows into a compartment
known as the dosing chamber where it is admitted to the filter for
further purification.
When enough of the liquid has accumulated in the chamber it is
automatically emptied by means of a siphon, thus permitting the
intermittent application of the sewage to the filter bed. When more than
one bed is used the siphons are arranged so that the liquid alternately
flows to different filters or parts of filters.
Contact Filters—Single and Double
The treatment of sewage in a single contact filter is classed as a
preliminary process and when treated in double contact beds or those
arranged in tandem as a final process. A contact filter is a basin
filled with broken stone, coke, slag or coarse gravel, thoroughly
underdrained. The size of stone or other material to be used depends
upon the degree of purification desired, and the manner of operating the
beds. The smaller the stone the more brilliant the effluent will be, but
all reports agree that the cost of operation will be greater and that
there will be a more rapid loss of filter capacity. Experience has
taught the superiority of the coarser material because the interstices
being so large the bed is not so liable to choke. Watson advises a fine
medium bed only when a highly purified effluent is desired, when it
would be difficult to get rid of humus from the filtrate, when a high
cost of maintenance is not prohibitive and when a temporary stoppage of
the whole plant would not be a serious matter. He believes it is not
suitable for installations of any magnitude. Beds have been built with
various depths, the range being between four and seven feet. Some have
been built shallower and have given good results. The method of applying
the sewage is important. Some tanks are overfed and others are underfed.
Francis E. Daniels, Director of Water and Sewage Inspection of the New
Jersey State Board of Health, describes a method which has been found to
be successful in plants in this state. At these plants the effluent is
applied on the top and at one corner of the contact beds. At the point
of application a small area of contact material from 6 inches to one
foot deep is removed from the top of the bed, and fine cinders are
substituted. An embankment about a foot high is constructed of the same
material around this area so that all of the tank effluent applied to
the beds strains through the cinders. Mr. Daniels says that a great deal
of the suspended matter is thus removed from the tank effluent which
reduces clogging and increases the life of the beds. It is Mr. Daniels’
experience that the value of underfed beds is diminishing. If the
effluent is very septic this method has the advantage of reducing odors,
but as Mr. Daniels has pointed out, the practise of reducing the storage
capacity of tanks is becoming prevalent.
In many plants the sewage is distributed by mechanical appliances, some
being motor driven and others cable driven. Springfield, Missouri, which
uses a motor drive, reports a saving in power, first cost, moving
weight, and maintenance, over the cable drive. Another advantage is that
the length of the filter can be increased at will. The total cost of the
distribution per million gallons according to Springfield’s experience
is $1.25 for cable drive and $1.61 for direct motor drive.
After the sewage has been distributed on the beds so that the
interstices are filled, it is allowed to stand for a time. The bed is
then drained and rested. While standing the sewage comes in contact with
a jelly-like film which forms on the surface of the stone, and important
changes occur. As with the septic tanks contact beds require a certain
period in which to ripen. The time of contact and the period of rest
vary in different plants. The rate of filtration varies according to the
construction of the beds, the range is between 600,000 and 1,200,000
gallons per acre per day. The effluent from single contact beds is not
stable but that from double contact beds is non-putrescible and low in
suspended matter, although somewhat turbid. It can be discharged without
offense into small streams. Single contact beds have seldom been used
for final treatment of sewage and fewer filters of this kind are now
being constructed even in conjunction with any preparatory treatment.
The general opinion is that this process is on the wane. Watson says,
“It may now be assumed that percolating filters are being constructed in
England in preference to contact beds wherever the conditions are
suitable.” In America they are not being adopted for large installation
but they are still considered for small disposal works. In their fifth
report the Royal Sewage Commission of England states that taking into
account the gradual loss of capacity of contact beds, a cubic yard of
material arranged in the form of a percolating filter will generally
treat satisfactorily nearly twice as much tank liquor as a cubic yard of
material in a contact bed. Comparing the efficiency of contact beds and
percolating filters it is claimed that the latter are better adapted to
variations of flow and that the effluent is usually much better aerated;
and apart from the suspended solids are of a more uniform character.
With percolating filters the likelihood of odors is greater than from
contact beds and there may be a greater nuisance from flies.
In the report of the City of Leeds, England, the results of very
valuable experiments are given. It says, “Double contact beds give good
results with crude sewage and excellent results with partially settled
sewage or with septic effluent. Single contact beds are insufficient for
dealing with crude sewage but give fair results with settled sewage or
with septic effluent. The real difficulty with contact beds is to
maintain capacity.”
The principal advantages of this process according to reports are low
operating head, and less nuisance from odor and flies, and the
disadvantages are large areas required and cost of maintenance.
Trickling, Percolating or Sprinkling Filters
Trickling or percolating filters consist of beds of coarse grained
material such as pebbles or crushed stone, one-eighth to four inches in
size, from four feet to ten feet deep and well underdrained. The
character and strength of the sewage should determine the size of the
material, the depth of the bed and the rate of operation. Some engineers
give the capacity as about 20,000 persons per acre of stone surface;
others say the rate of flow should be from one to two and one-half
million gallons per acre. In some designs an auxiliary air supply is
inducted into the filter material by tubes connected with the
underground system. The Atlanta plant is equipped with ventilator hoods
having weather vanes so that the mouth of each hood always points toward
the wind. “This form of ventilation is of no particular value and may be
detrimental in cold weather,” says Glenn D. Holmes, Chief Engineer of
the Syracuse, N. Y., Sewer Board. By means of spray jets and moving
sprinklers operated with some device for varying the pressure, such as a
butterfly valve, or by means of an intermittent dosing tank operated by
a siphon, the sewage is sprinkled or deposited on the surface of the bed
in thin films and drops; thus the sewage is freed of objectionable gases
and takes up oxygen as it passes through the air and through the filter.
Sprinkling filters do not produce the best results when crude sewage is
applied. They are most efficient when the suspended matter has been
removed by some preparatory treatment. In some cities the screening
process is first used, in others the sewage receives a preliminary
treatment in tanks. Well designed and efficiently operated filters of
this kind produce an effluent that is stable but not clear. Some plants
are equipped with secondary settling tanks through which the effluent
flows before final discharge and is freed of the humus-like particles it
contains after leaving the filter. Reports agree that the effluent is
not nearly so good in appearance and has a much higher percentage of
bacteria than that produced by good intermittent sand filters. As
compared with the double contact process the general opinion is that
sprinkling filters are superior in respect to the removal of organic
matter and cost less to operate. The chief advantages of a sprinkling
filter are the high rate of filtration and the low cost of operation.
The disadvantages are a possible nuisance, especially during hot
weather, from odor when anything but fresh tank sewage is sprayed; and
the development of insect life. Fowler says, “However economical their
construction and maintenance it cannot be said that such a process meets
all sanitary and æsthetic requirements.” The experience of Worcester,
Massachusetts, at its experimental station was that more than twenty
times as much sewage per unit of area was treated by the sprinkler
filter as could be treated by intermittent sand filtration, and more
than ten times as much per cubic yard of filter. Four times as much
sewage was treated by these experimental filters as could be treated
satisfactorily by experimental contact beds. In order to obtain equal
nitrification with contact beds at least three contacts would be
required.
Intermittent Sand Filters
As a final process of purification in sections where land and filter
material are available at small cost the intermittent sand filter is
superior to any other. This fact has been established by experience and
experiments. The filter material may be clean, coarse sand or any other
porous soil. If a natural area is available the method of construction
is very much simplified and economical. The top soil is removed and used
in embankments between the beds. If the water tables are low the beds
are not underdrained. In artificial beds the size of the sand is
important. While fine sand will give a more brilliant effluent than a
coarser material, the sewage has to be applied in small doses with long
periods of rest. The rate of purification is higher in coarse sand
filters and the effluent while containing more bacteria is non-
putrescible. About twenty-four inches of sand should cover the
underdrains of tile, placed about five feet apart, and surrounded by
small-sized gravel.
In some beds the entire bottom above the underdrain is covered with
about six inches of gravel. In others the bottom is ridged, the
underdrains being placed at the bottom of the valleys which are then
partially or wholly filled with gravel. Risers are constructed at the
head of the underdrain and an intercepting drain completes the system.
The beds vary in size and number according to the amount of sewage to be
treated. The operation of the filter is very important. The sewage must
be applied rapidly in rotation to each bed until the surface is covered
with about three inches of the liquid. The bed is then slowly drained
and allowed to rest. Overdosing and lack of aeration cause clogging. The
surface must at all times be kept clean and loose. To maintain this
condition it is sometimes necessary to break up the surface to a small
depth or periodically to remove the deposit on the surface.
In cold climates the operation of the filters in winter is difficult and
the quality of the effluent somewhat impaired. Several methods have been
adopted to prevent freezing. Some filter beds are ridged so that when
dosed the sewage flows in gutters. The ice which forms at the top of the
sewage remains suspended on the ridges, thus permitting succeeding doses
to flow underneath the ice. In other plants the surface of the filter is
scraped into small piles which form a support for the ice. It is claimed
that by this method the cost of subsequent cleaning is less than when
the beds are ridged.
The effluent in properly constructed and managed plants is clear and
odorless. The bacterial purification is as high as ninety-nine per cent.
The Massachusetts State Board of Health in one of its reports says,
“When sewage filters slowly and intermittently through five feet of
porous earth and sand, an effluent is obtained which is as free from
organic matter, from ammonia and from nitrites as many a natural spring
water.”
The only drawback noted to this process is the cost of treatment in
large quantities where land and filter material are not available.
Francis E. Daniels says that under such conditions the cost is almost
prohibitive. For many cities sufficient area cannot be obtained at any
price, and as population increases the difficulty will become greater.
The New York State Board of Health in general will approve only of the
following rates of operation for different types of filters where
suitable provision for preliminary treatment is made: Intermittent sand
filters, 100,000 gallons per acre per day; contact beds, 100,000 gallons
per acre per day per foot of depth; sprinkling filters, 300,000 gallons
per acre per day per foot of depth. These rates of operation are based
on a sewage contribution of 100 gallons per capita daily and no
variation from these rates of filtration is allowed for any other per
capita contribution of sewage. The allowable effective depths of said
filters will in general range from three to five feet; contact beds from
four to seven feet; sprinkling filters, from five to nine feet.
Broad Irrigation
Broad irrigation, or sewage farming, is the oldest process of sewage
purification, but the constant increase in population has made it
necessary for cities to adopt other methods because of the area of land
necessary for such a plant. Two processes are used, surface irrigation
and filtration, a greater area of land being required for the former.
Sometimes the two are combined into one process. For filtration and
irrigation the sewage is generally first subjected to sedimentation or
screening and then flows on carefully prepared land on which crops are
usually grown. The areas are underdrained and are equipped with
distribution systems.
Local conditions determine the method of irrigation, the ridge and
furrow system being most generally used. The efficiency of the process
depends upon the quality of the soil and proper management. Among the
factors which should enter into the selection of the site are the
quality of the soil, composition of sewage, method of disposal, kind of
crops to be planted, contours and slope of surface, nature of the sub-
soil, sub-soil waters, transportation facilities, nature of streams,
nature of adjacent property, and availability of water supply. The best
lands consist of a fine layer of alluvium overlaying a sub-soil of
gravel, chalk or other porous material. Various kinds of crops are grown
on sewage farms and the revenues therefrom help to reduce the cost of
operation. They also assist in the purification. The principal drawback
are heavy transportation cost and a prejudice against sewage-grown
produce. During the rainy season when the quantity of sewage requiring
treatment is greatest, less sewage can be used for irrigation and the
growing of crops of sewage farms. All evidence points to the fact that
broad irrigation is on a steady decline, although the efficiency of the
treatment under favorable conditions is very high.
Disinfection
When the bacterial efficiency of an effluent from either preparatory or
final treatment is low and the effluent is discharged into a body of
water from which water supplies are derived or shell fish are taken,
disinfection is often found necessary. The purpose is to destroy
objectionable bacteria and disease germs. Hypo-chlorite of lime and
liquid chlorine are the two chemicals most commonly used. The principal
advantages of the liquid chlorine over the hypo-chlorite according to
plant supervisors and operators, are less cost of operation and space
required for both apparatus and storage of materials, no loss of
strength, no lime sludge, and no mixing tanks required. The claim is
also made that it can be better controlled. Chlorine, however, is more
expensive than hypo-chlorite and the control apparatus usually costs
more. There is general agreement among engineers, that except as an
emergency measure or under the above stated conditions, disinfection is
too expensive a process on account of the amount of chemical required.
This varies with the amount, method and degree of previous treatment of
the sewage and the degree of bacterial elimination desired. Tests at the
Cleveland Testing Station indicated that from five to seven parts per
million of available chlorine will effect a bacterial removal of from
eighty-five to ninety per cent.
Activated Sludge Process
Sewage treatment by aeration in the presence of sludge is the latest
development in sewage disposal, and the process is attracting a great
deal of attention in America. Milwaukee has constructed a plant to treat
two million gallons of sewage a day. Houston, Texas, is operating a
plant to treat the sewage for 160,000 persons, and Escanaba, Michigan,
and Jersey City, N. J., have favored the process. Experiments are now
being conducted in Milwaukee, Baltimore, Washington, Cleveland, Regina,
Chicago, Lawrence, Mass., Brooklyn, New Haven, Conn., the University of
Illinois and many other places. The efficiency and economy of the
process as compared with others which have long been in use have not
been completely established. The chief points in dispute are sludge
disposal and cost, but the indications are that these questions will
soon be satisfactorily answered.
The process consists of passing raw sewage through tanks from eight to
twenty feet deep in which a certain amount of activated sludge is always
present. To mix the sewage and the activated sludge air is forced into
the bottom of the tank under low pressure of sufficient volume to keep
the liquor violently disturbed. From this aerating tank the mixture
passes to another or sedimentation tank where the sludge settles and
from which the clear effluent passes over a weir to its final
destination. In order to maintain the proper volume of activated sludge
in the aerating tank a portion of the sludge is pumped back from the
sedimentation tank. The balance of the sludge is pressed and used for
fertilizer base. The Milwaukee experiments indicate that in order to
produce a clear, non-putrescible effluent about four hours aeration is
required, twenty per cent. of activated sludge maintained in the
aerating tank, and about 1.75 cubic feet of free air supplied per gallon
of sewage treated.
The effluent is clear, odorless and practically free from suspended
matter. The sludge will begin to decompose after forty-eight hours and
must be pressed and dried within that time. Chief Engineer, T. Chalkley
Hatton, of the Milwaukee Sewerage Commission, estimates that the sludge
can be reduced to a fertilizer basis for about $8.75 per dry ton,
including overhead charges. Basing the value of the sludge produced upon
a low price per unit, he finds that Milwaukee sludge is worth $12.50 per
dry ton, which represents a clear profit of $3.75 a ton. From ten to
twelve million gallons can be treated upon one acre of ground, which is
about one-fifth the area required for sedimentation tanks and sprinkling
filters. The reasons for the adoption of this process by Milwaukee after
experimentation by competent engineers for more than a year are given by
Mr. Hatton in a recent address before New York State city officials as
follows: “It produces a better effluent than any other known process of
sewage treatment except land treatment or intermittent sand filtration;
it can be built upon a comparatively small area; it produces no
objectionable odors or flies; it produces a sludge of sufficient value
to meet the cost of its reduction to a fertilizer and therefore relieves
the city of the difficult, complicated and wasteful method of sludge
disposal common to all other processes; it is subject to complete and
satisfactory control throughout its operation; it is not materially
influenced by climatic conditions; occupying a small area, its first
cost is less than any other known process from which an equal character
of effluent can be obtained; its operating cost is not prohibitive.”
In a discussion before the Iowa Section of the American Waterworks
Association Dr. Edward Bartow commended activated sludge for its value
as a fertilizer. This has been proved, he said, by its chemical
composition, by its reaction with various solids and by its effect on
the growth of plants. Pot cultures and garden experiments have shown
that the nitrogen is in a very available form.
E. E. Sands, City Engineer of Houston, Texas, bases this statement on
results of experiments conducted for a year: “Our investigation has
demonstrated that sewage can be disposed of anywhere that there is a
vacant tract of land in the city without creating a nuisance and without
any objectionable feature.” The total estimated cost for treatment will
be about $9.14 per million gallons when the plant is run at the rate of
18,900,000 gallons per day. He estimates that the total cost for
treatment by the Imhoff tanks and the sprinkling filters would be not
less than $11 per million gallons.
After an extended investigation by their sanitary engineers, Armour &
Company have concluded that the activated sludge method will
satisfactorily purify the industrial wastes from their Packingtown
factories. Assistant Superintendent, M. D. Harding, estimates that from
data now available the cost per million gallons exclusive of
depreciation, interest and repairs, will be $3.
When considering the applicability of this process to conditions in any
city consideration should be given to the following points. The process
requires competent supervision, which Mr. Hatton claims may be a
blessing in disguise in view of the experiences of cities which, after
having built disposal plants of various kinds, have left their operation
to the kind mercies of Providence with disastrous results. This process
also requires the expenditure for constant power. The cheaper the power
the more adaptable the process is commercially; but if the unit is small
and the power cost high, the operating cost may be too great. The sludge
must be constantly treated to avoid nuisance. The process produces a
high degree of purification. If the local conditions do not demand this
the process might be too expensive in comparison with some other process
which will produce a satisfactory effluent.
Other Processes