78 



THE CIVIL ENGINEER AND ARCHITECT'S JOURNAL. 



MAuni, 



Mr. PiiiM.ips, C.E., e:ave the following evidence as to the system 

 of sewafie iidojited by him : — 



'•Solidity of execution in construction, economy of niiiterials and 

 liihoiir, coniliiiied «itli ^^trenjrth to hear the latenil and vertical 

 )Fressures of the j^round, and efficiency in atfordinjj;thehest channel 

 fur ijuickly convcyinf; away the seivape, are the essential reiiuisites 

 fur a sewer. The circle affords the most capacious area of all 

 plane figures havin-r the same circumferciu^e, and conversely its 

 I'ircuiiiference is less than any other tiijure of the same capacity. 

 It, therefore, sup]ilies the greatest capacity for receiving the 

 water, w itli the smallest frictional surface, and the least consump- 

 tion of materials. As regards strength : when the ])ressure from 

 t]ie ground around a circle is the same, it is equally distributed 

 throughout the entire thickness composing the arch ; for, as the 

 extradosal length is greater than the intradosal length, the arch 

 is necessarily made up of a series of wedges all pointing to the 

 centre of the circle ; hence the circular form prevents the earth 

 outside of it from forcing it in, and from disturbing it, provided 

 the pressure be e(i\ial, while upright walls in tlie same circum- 

 stances would most probably be unable to withstand tlie pressure. 



The removal of sewage and prevention of deposit of matter in 

 sewers are entirely dependent on the tpiantity and velocity of 

 the water running through tliem. In order therefore to keep them 

 well washed out and cleansed, the utmost scouring force should be 

 imparted to the streams. A semicircular, or still narrower and 

 deeper-curved channel of a semi-elliptical or catenarian form, 

 concentrates the flow on a small area of friction, heaps it up, and 

 so increases its velocity, and makes it more powerful in lifting, 

 holding in suspension, and carrying away all matters which may 

 find their way into the sewers, than a wide and flat channel. A 

 sewer, therefore, having an arched crown, curved side walls, and a 

 narrow and deeply-curved bottom, which, combined together, give 

 Uie shape of an rgg irith the i-imill end phiced dnwiiwcirds, is, in my 

 opinion, the best and most efficient form for all branch sewers. It 

 would ai)|)ear liovvever from what has been stated, that the circle, 

 from ha\ing a more capacious area and less rubbing surface than 

 any other ligure, is the best shape for all sewers. But this is not 

 the case ; for although the surface of contact of the egg-shaped 

 sewer is somewhat greater than a circle of the same area, yet by 

 contracting the channel and so raising the height of the stream, 

 the ratio of velocity and consequent power to scour is increased 

 thereby, as will be evident on experiment being made. It is the pre- 

 rogati\'e of the egg-shaped sewer, therefore, to combine in its 

 form, capacity, economy, sti'ength, and efficiency. 



For the short collateral branches of the sewers in street, courts, 

 &c., the smaller tliey are, (provided they be large enough to 

 recei\e and carry off storm waters in addition to the ordinary run), 

 the less chance will there be for them to choke up. In the course 

 of my experience I have examined hundreds of drains, and I have 

 always found small drains and sewers which had a moderate fall, 

 and anything like a good supply of water, quite clean and perfect 

 in that respect. I anticipate, indeed I confidently entertain an 

 opinion, that with a combination of the water supply and a tubular 

 system of sewerage and house-drainage, the whole of the annoy- 

 ance now experienced by the public from defective drains and 

 sewers may be made to cease. 



If constant currents of water be carried through the drains and 

 sewers, though the currents may he small, yet provided they be 

 constant and concentrated on very narrow and smooth bottoms, 

 they M ill keep the sewers clean. Where the supply is intermittent, 

 tlie matter discharged from the house-drains, meeting with no 

 <'urreut, accumulates. In order to prevent deposit in drains and 

 fcewers, there must be a certain degree of velocity and force given 

 to each current, so as to produce agitation equal to, or rather 

 gi'cater than the ins inertiip, or weight, mass, figure, and superficies, 

 of the sand, silt, mud, and other substances, to be lifted, and kept 

 always moving, or united and incorporated with the running 

 water, added to the friction of the bottom and sides of the channel. 



The chance of any sewer keeping itself clean is dependent on 

 four things, — namely, its capacity, its ft>rm, its fall, and the 

 quantity and force of the water running through it. It is only from 

 oltservation and experience, and the application of rules deduced 

 therefrom, to the proportioning the capacity, the form, and the 

 fall, as also the quantity and force of water requisite to prevent 

 deposit, that we can ho])e to arrive at perfection in sewerage. 

 From observation and experiment, I fiiul that it requires a con- 

 stant velocity of current to be running through the sewers equal 

 to about 2ij feet per second, or l| mUe per hour, to prevent the 

 soil from depositing within them. 



There is less water ruaning in the sewers on Sundays than on 



other days of the w cek ; and most on Saturdays. The height of 

 the flow every day goes on increasing from an early hour in the 

 morning until about noon, when it is highest; it tlien gradually 

 subsides to its lowest level. The period of the greatest flow es^ry 

 day is between 11 a. m. and 1 p. nu 



The fall of sewers should be proportioned to the quantity of 

 water that is to pass through them. For, with the same fall, the 

 greater the body of water the greater will be the velocity ami 

 scour ; and conversely, the less the body of water the less will lie 

 the velocity and scour. Again, a large body of water will, with a 

 little fall, run with the same velocity as a small quantity will with 

 a great fall. Hence the recipient of many branch sewers may hav« 

 less fall than the branches themselves. A fall of a quarter of an 

 inch in 10 feet has been considered the least fall that should be 

 given to branch and summit-level sewers; but this fall is not 

 enough to keep the sewers clean. No ; such sewers should, in my 

 opinion, have not less fall than half an inch in 10 feet. In some dis- 

 tricts it is found impossible to get even so much fall as a quarter 

 of an inch in 10 feet. In districts where proper fall cannot he 

 obtained, it is necessary to resort to flushing to keep the seivers 

 free of deposit and clean. 



"When a main stream receives a branch stream, the united 

 body of water causes the height of the main stream to increase, 

 consequently the surface rises somewhat higher than the surface of 

 the divided streams; hence the water flows back, producing tie- 

 posits of heavy substances about the junctions, which deposits 

 draw back and impede the flow of the two streams. Now, in 

 ortler to remedy this evil, the bottom of the main sewer, immedi- 

 ately below the junctions should be made some inches deeper than 

 the bottoms above the junctions. By this mode of forming the 

 bottoms, the surface of the main and branch streams w ill ha^^e a 

 uniform inclination, and the acceleration of this fall will prevent 

 regurgitation and deposit, and the united streams will flow on- 

 wards with increased speed. 



In order to determine the depth below the junctions, it is ne- 

 cessary to calculate what height the body of water falling from 

 the branches will increase the stream in the main. Tlie capacity 

 of the united stream is very much less than the sum of the) 

 capacities of the divided streams, and the velocity in the former is 

 considerably greater than either of the latter. The ratio of 

 increase of velocity follows the ratio of decrease of capacity. It 

 follows, therefore, that a graduallj' accelerating velocity takes 

 place immediately below the confluence of the sewers throughout 

 the ramified system from their sources to their outfalls, and such 

 I have found to be the case. 



Egg-shaped sewers, varying in capacity according to the area, 

 the number of houses to be drained, and the quantity of water to 

 be discharged, from 9 inches wide by 1 ft. 3 in. high, to 1 ft. 6 in. 

 wide by 2 ft. 6 in. high, would suffice for sewers on summit levehs, 

 and also for branch or collateral sewers which had to receive the 

 drainage of from one to twelve or more ordinary-sized streets. 

 Of course the secondary mains which would have to carry off the 

 water from these branch or collateral sewers, as well as the prin- 

 cipal main lines into which the secondary ones would discharge 

 themselves, must be larger in proportion ; but under a proper ar- 

 rangement, fewer principal lines would be required. 



Instead of discharging a large body of water uselessly, as t» 

 any power of sweep, I would, under the system of constant and 

 concentrated supplies and smaller sewers, economise the water by 

 using it to scour several small sewers instead of one large one. 

 For this reason I would prefer having more outlets, or at least 

 more catch-water sewers, instead of discharging all the drainage 

 by one large main sewer throughout, although at or near the 

 outlet, I might probably be obliged to lead the whole of the water 

 into one main line ; but I should not like to part with it into a 

 main line until I had made it serviceable in sweeping as many 

 sewers as possible. As the keeping of all sewers thoroughly 

 washed out is necessarily dependent upon an abundant supply of 

 water, the principle which I have thought it best to follow for that 

 purpose is to tie and connect all the sewers together upon a uniform 

 system of levels so as to use the water running along sewers on 

 high levels for washing out those on low levels. For this purpose, 

 as will be seen by the plans, (Plate V., figs. 2, 3, and t), I would 

 connect the heads of adjoining sewers below with the superior 

 sewers above them, and arrange the connections so that, as the 

 currents of water running along the latter sewers arrive opposite 

 the connections, they may divide and subdivide themselves by the 

 ridges or groynes formed by the meeting of the inverts. By this 

 means the water would traverse from one sewer to another, and a) 

 keep up a perpetual flow throughout the entire system. There 

 can be no doubt that with much smaller sewers than those now i4i 



