foot cylinders vertically installed, filled with three-quarter-inch tubes. We force the cold sludge, which is about 45 to 55 degrees temperature through these tubes, and from there it passes to the condenser, what we call the heat balancing unit, simply a turbine generator, the exhaust steam from which passes through the condenser outside the tubes and the sludge passes through the tubes until it is raised to any temperature desired, up to 190 degrees. From the condenser it goes into the filters and is filtered by a vacuum filter of the Oliver type, and the effluent of the filters is then about 120 to 130 degrees temperature, accord. ing to the temperature of the sludge when it goes into the filter, this filtrate is brought back to the receivers, from there to the vacuum condensers, and then back into the heat exchangers, and thus we transmit and save the heat of the filtrate to first heat up the sludge; otherwise we would never be able to do it. We have one very important thing which we feel very proud of. With the Allis-Chalmers engineers and our own engineers we have designed an automatic governor on the steam turbine unit by which we can automatically take as much exhaust steam out of the steam turbine unit into the condenser as we need to raise the heat of the sludge to the temperature we desire, the balance of the steam going into the other condenser, thus making it possible to run the steam turbine unit from atmosphere to 281⁄2 inches of vacuum. That is a thing that enabled us to save thousands of dollars during the winter time, by running the steam turbine unit, either condensing, partially condensing or atmospheric. With that in mind we will try the next film.

(Shows second part of films.)

PRELIMINARY TREATMENT OF SEWAGE AS AFFECTING RATES ON TRICKLING FILTERS

By Almon L. Fales, of Metcalf & Eddy, Consulting Engineers, Boston, Mass.

The application of crude sewage to a trickling filter is likely to result in rapid clogging of the filter, even though the filter medium be relatively coarse and the rate of application relatively low. Grit chambers and coarse screens alone cannot ordinarily be depended upon to prevent filter clogging.

Fine screens will afford sufficient preparatory treatment if the filter medium be coarse enough to permit of the unloading of the remaining solids which accumulate in the filter and if the rate of operation be low enough to produce a satisfactory effluent. More complete removal of suspended solids will make possible higher rates on trickling filters, for two reasons— first, because of the reduced load of organic matter, and second, because it makes possible the use of finer media which for the same cubical contents, present a greater surface area for bacterial action and afford a longer period of bacterial contact with the sewage applied.

Sedimentation tanks will remove solids which are too small to be removed by the finest screens which it is practicable to use. The percentage removal of suspended solids by fine screens will vary according to the strength and character of the sewage, particularly as regards its freshness and the extent of comminution of suspended matters. The greater the proportion of relatively coarse matters, the greater the efficiency of the screens. The efficiency of sedimentation is also greater with fresh sewage containing a relatively large proportion of uncomminuted suspended matter, but the advantage is not so great as in the case of fine screens. In other words, the comminution of suspended solids does not reduce sedimentation tank efficiency to the same extent that it does the efficiency of fine screens. The advantage of sedimentation over fine screening will vary greatly with different sewages, but under average operating conditions sedimentation will probably remove about five times as much suspended matter as will be removed by fine

screens, leaving in the effluent about half as much suspended matter as in the effluent from fine screens.

The comparative loads with effluent from fine screens and from sedimentation tanks cannot be based solely on the content of suspended solids. The dissolved and colloidal matter, the amount of which is substantially the same in the two effluents, must be taken into consideration. On the other hand, the loads cannot safely be based on the total organic matter irrespective of the proportion in suspension, because the suspended matter accumulates largely in the upper part of the filter, tending to reduce the permissible rate of operation without clogging. As previously pointed out, with a smaller amount of suspended solids smaller filter media can safely be used and the rate of operation correspondingly increased.

George W. Fuller, in a paper before the Indiana Sanitary and Water Supply Association, March 28, 1918, stated as follows:

"I assumed in my Indianapolis investigations that with the same area of sprinkling filters receiving the effluent of Imhoff tanks and fine screens, respectively, the depths of beds should be in the ratio of 7 feet to 10 feet. In other words, per unit volume of filtering material, fine screen effluent should be applied in the proportion of 7 volumes as compared with 10 volumes. for the more completely clarified tank effluent."

Mr. Fuller recommended fine screens and relatively deep filters and proposed cutting out units in rotation, for drying and hosing, to facilitate the unloading of accumulated solids. Few, if any, fine screen-trickling filter plants have been built up to the present time.

Septic tanks are less efficient in the removal of suspended solids than Imhoff tanks or sedimentation tanks from which the sludge is removed at frequent intervals. The decomposition of the sludge in septic tanks produces gas which causes masses of sludge to rise, and portions of this may be carried away in the effluent. This condition was so marked in the case of an experimental septic tank at Worcester, Mass., that the experimental trickling filter receiving this effluent became clogged to such an extent that it was necessary to remove and wash the filtering material to a depth of about two feet.

Another disadvantage of the septic tank for preliminary treatment is the foul odor likely to be given off by the tank effluent when it is sprayed on trickling filters. Further, it is quite generally believed that for the desired bacterial action septic sewage is not as favorable as fresh sewage.

Chemical precipitation tank effluent was applied to an experimental trickling filter for over a year at Worcester, Mass., and the results were compared with those obtained by similar experiments using the effluent from plain sedimentation. The conclusion reached was that about 30 per cent more of the chemical effluent than of the corresponding settled sewage could be applied to trickling filters and produce an effluent of like quality, but that the increase in rate possible by the chemical precipitation was more than offset by the additional cost of this treatment.

In England the use of chemical precipitation as a preparatory treatment for trickling filters is still quite common, and in most cases where this preparatory treatment is not used thorough sedimentation is afforded. In this connection it should be borne in mind that as a rule English sewage is much stronger than American sewage, the water consumption in England averaging less than half as much as in this country.

At Birmingham, England, an activated sludge plant has been operated for preliminary treatment of a portion of the sedimentation tank effluent applied to the trickling filters, and it has been found that with an aeration period of one hour, and reaeration of the sludge returned, the rate of application to trickling filters can be increased two to three times and still obtain a better filter effluent. The conclusion reached is that it will be cheaper to add activated sludge units for such preparatory treatment than to build additional trickling filters.

Aeration at Birmingham is secured by mechanical means and this also creates sufficient velocity to maintain the sludge in suspension while passing through the aeration channels. The aeration tank is called a "bio-flocculation tank" and the sludge reaeration tank is called a "sludge bio-generation tank." The entire aeration plant is called a "deodorizing and partial purification plant," one great advantage claimed for this plant being that it eliminates the smell that would otherwise be created by spraying sedimentation tank effluent on the trickling filters.

John D. Watson, in his booklet on "The Works of the Birmingham, Tame and Rea District Drainage Board," July, 1923, explains the advantage gained by this preparatory treatment, as follows:

66

it is shown that flocculated sewage, which has been deprived of colloidal matter, is much more readily oxidized, . . . The bacteria beds will thus be relieved of work for which they are not well fitted. They are no doubt capable of dealing with humus which arises from the particulation of colloidal matter, provided sufficient time and favorable conditions are given, but it is not profitable to keep them employed on work for which they are not well fitted."

It does not follow from the Birmingham experience that the activated sludge process can profitably be employed for preliminary treatment of the relatively weak sewage in this country, but the possible advantages of such treatment should not be lost sight of. It seems probable that in the majority of cases in this country efficient sedimentation will be the most economical preliminary treatment for trickling filters.

DISCUSSION

PRESIDENT-ELECT HATTON: Is there any discussion on this paper, gentlemen? This is a very valuable paper. I suppose a lot of you here have your troubles with trickling filters. I want to say that I visited that Birmingham plant of Mr. Watson's last year, and I was very, very much struck and impressed with the lack of odor on that portion of the plant of his trickling filters in which he was treating this partially aerated sewerage and that portion upon which he was treating the sewerage from sedimentation alone. There was not any odor at all about the four or five acres of the treatment of the first named filters; whereas, just on the other side of the alley where he was treating the precipitated liquor it was quite odoriferous; and yet the cost, as his engineer explained to me, for this re-aeration, was so small that he could afford to carry it on at less expense than the annual interest charges on the cost of extending his trickling filter plant. I do not quite agree with Mr. Fales that it is not economy. He suggests that it would not be economy