A 6-inch pot trap similarly arranged lost 2 1/8 inches in one day, 2 5-16 inches, 2 3/8 inches and 2 7-16 inches, in two, three and five days, respectively, after which no apparent further change took place, the experiment lasting several days longer.

A 3-inch pot lost 5/8 inch, 1 inch, 2 inches, 2 inches, 2 7/8 inches and 3 inches in one, six, fifteen, forty-eight, seventy-two, 144 hours, respectively, after which no further change took place.

Figs 288-b and 288-c show the manner in which the un-vented "Sanitas" trap is able to retain its seal under capillary action. The fibrous"matter not being able to raise the water high enough to break the seal. The horizontal extension of the unvented "Securitas" trap combined with the elevation of its outlet, provides similar protection. Moreover, the construction of these traps is such that it would be impossible, without special manipulation, to so weave a mass of fibrous material through the trap that it could connect the lower with the upper bend in such a manner as to place the former within the influence of capillary attraction. This reasoning was corroborated by prolonged tests.

Fig. 288. Pot Trap losing its Seal through Capillary Action.

Fig. 288. Pot Trap losing its Seal through Capillary Action.

Action Forming Short Siphon 298Fig. 288b.

Fig. 288b.

Fig. 288c.

Fig. 288c.

Fig 288e. Securitas Trap.

Fig 288e. Securitas Trap.

Fig. 288d. Securitas Trap Antisiphon Traps Resisting Capillary Action.

Thus we see that the effect of capillary action in traps detached from the drains is similar to that in open vessels, with the exception that in traps unventilated no perceptible loss took place through evaporation, and that after the limit of perpendicular distance at which the capillary force can act has been attained, no further loss of water is perceptible. In open vessels, on the contrary, the draught on the water goes on indefinitely by rapid evaporation aided by the distributing process effected by the capillary action.

Tests of the Effects of Capillary Action in Ventilated and Unventilated S-Traps Fixed in Position.

To test the loss by capillary action on ventilated S-traps as compared with the loss on the same when unventilated I attached an S-trap having a 4 5/8-inch deep seal to a branch waste entering the soil pipe, after having half filled the trap with jute as shown in Figure 288. With the trap unventilated the loss by capillary action was as follows: In the first five minutes inch; in the first forty-five minutes 1 inch; in twenty-four hours 3 inches; in three days 3 inches; in four days 3 3/8 inches. Thereafter no further perceptible change took place. It made no perceptible difference whether the basin side of the trap was opened or closed, showing that evaporation in an unventilated trap is practically imperceptible.

The experiment was then repeated on the same trap, ventilated at the crown, into a cold flue with the following result: In one hour 1 1/8 inches had been removed; in 5 hours 1 7/8 inches; in 22 hours-2l/2 inches; in two days 3 inches; in 3 days 3 inches; in 4 days 3 inches; in 5 days 4 inches. Thus the loss continued at the rate of about inch a day by evaporation, after the outer end of the jute mess had entirely dried up. This rate of evaporation was nearly double what it would have been had it not been assisted by the capillary action. From this we see that ventilation greatly increases the danger arising from capillary action, often rendering the latter dangerous in cases where, without ventilation, the seal would not have been broken.

To test this point still further I placed two ordinary drinking glasses, filled with water side by side. The first was treated as shown in Fig. 284, with a mass of jute hung nearly 5 inches above the surface of the water and having one end immersed in it as shown in the figure, the other extending below the bottom of the glass. Owing to the height from which the jute was suspended the water did not rise to the point of support; consequently the outer arm was dry, and whatever loss of water was observed was, therefore, due to evaporation.

The glass having the water alone lost by evaporation only inch seven days, while that having the jute lost 1 inch, or four times as much in the same time.

Friction.

A consideration of very great importance in trap construction and arrangement is the amount of retardation to the passage of the waste water caused by the friction against its interior surfaces.

In order to obtain the quickest delivery and maximum of scouring action on the waste-pipes below the trap, it is important that the trap should afford the minimum of obstruction to the flow of the water. Many traps, especially gravity-ball and other mechanical traps, are so formed as to greatly retard the flow of the water. With many ball-traps, when the water is permitted to escape from the fixture through the waste pipe "full-bore" above the trap, the ball is so forcibly driven against the outlet mouth of the trap as to very seriously obstruct its further passage, and prevent its exerting its full scouring effect on the pipe below.

An ordinary S-trap offers the least resistance to the flow of the water, the gravity ball trap the most, if we except certain forms of mercury-seal traps.

An ordinary bath tub arranged as shown in Fig. 243 will discharge through 1-inch waste pipe, 9 feet long, descending perpendicularly and without a trap, at the rate of 1.4 gallons of water per second.

An ordinary unventilated 1-inch S-trap, with a seal 1 inches deep, will retard the flow only 23 per cent, or it will discharge at the rate of a little more than one gallon a second.

Another 1-inch unventilated S-trap emptied the tank at the rate of 1.1 gallons a second.

The same trap ventilated prolonged the time of emptying the tank from 90 to 113 seconds, thereby retarding the flow 23 seconds or 26 per cent. The ventilation also created a loud and somewhat terrific roar during the entire duration of the discharge caused by the suction of the air at the vent opening. Without any trap, the tank discharged in 73 seconds. With a hinged valve trap, unventilated, it required 126 seconds. With the same trap ventilated a very much longer time or 163 seconds was required; the ventilation retarding the flow 37 seconds or 30 per cent.

A 4-inch pot trap, unventilated, required 104 seconds. The same ventilated required 144 seconds, or 38 per cent.

A 4-inch bottle trap, unventilated, required 94 seconds. A gravity ball trap required 226 seconds.

In both of these ventilation reduced the flow from 30 to 35 per cent, by calculation.

Thus we see that the average retardation of the discharge from a bath tub, and the consequent loss of scouring effect, caused by ventilation, is very great and amounts to about 30 per cent, or nearly a third of the whole when the outlet is arranged to discharge through a perpendicular waste "full bore," and where the vent pipe is short. When a long vent pipe is used the percentage of loss is somewhat less. The discharge of a wash basin having an outlet large enough to fill the waste-pipe "full-bore" gave similar relative results for different traps.

Summary of Ten Objections to Special Ventilation.

I find, therefore, no advantage whatever in trap ventilation. The disadvantages, however, are very serious, and may be summed up briefly as follows:

(1) It destroys the trap seal by evaporation when applied at or near the crown. With S-traps this position of the vent is necessary to prevent self-siphonage.

(2) It can not always protect the trap from siphonage even when newly applied in the most approved manner.

(3) It increases the unscoured area of the trap, making it a cesspool. It is a very strange piece of inconsistency to condemn the cesspool trap on account of its unscoured chamber and yet adopt in its place a ventilated S-trap, because by so doing the very thing we wish to avoid is reproduced in an aggravated form; the mouth of the vent pipe forming a sediment chamber which is not only greater in extent of surface, more easily fouled and less easily cleansed than that in the pot trap, but one which is far more dangerous in as much as its fouling, even to a limited extent, involves the destruction of the whole system. I have found by repeated tests that the water discharged from a large outlet basin and trap placed where it should be near the floor, is thrown up from 10 to 18 inches into the vent pipe at every discharge. Thus a large sediment chamber is formed which has an area of nearly 100 square inches. Beyond this, congelation of fatty vapor fouls to an indefinite extent, and it is no uncommon thing to hear of a vent pipe filled with grease for several feet above a sink trap.

(4) It retards the outflow of the waste water and its consequent scouring effect about a third when arranged to discharge perpendicularly "full-bore."

(5) It complicates the plumbing and adds to the danger of leakage through bad joining and increased material.

(6) It aggravates the danger arising from capillary action.

(7) It increases the corrosion of branch wastes by retarding the rapidity of flow and scouring effect allowing sediment to collect more rapidly than it otherwise would, and brings soil pipe and sewer air in contact with the branch wastes to take the place of the pure air of the house, which follows every discharge of the fixture. Moreover, as soon as the mouth of the vent pipe begins to get clogged by sediment and grease, the air current it was intended to produce is partially or wholly arrested, and we then have an interior surface of foul piping equally exposed to corrosive action with the unventilated pipe, but more than double in quantity.

(8) Finally it seriously increases the cost of plumbing, an increase which amounts to as much as from five to ten per cent on the total cost of the plumbing in new work, and indefinitely in old work in which the trap ventilation sometimes becomes by far the greatest part of the work to be done.