TABLE II-COMMERCIAL SIZES AND AREAS OF FLUE TILE. In fact, a square chimney can hardly be figured as having its full area effective, a deduction of 10% to 15% being necessary on account of this very movement leaving the corners dead and inactive. In an oblong shape the effect is worse, a deduction of 25% being none too much. The use of tile flues will not only aid this slightly-owing to their rounded corners-but will also safeguard much of the fire risk otherwise due to mortar falling out between the bricks as time passes and leaving openings through which a spark might pass. As the tile serves to retain the gases and smoke it also prevents leakage which can spoil a chimney action at any time, no matter how perfectly the flue may otherwise be built. The height of residence chimneys, as before mentioned, is not so important where the larger sizes of anthracite coal are used. The draft, however, ought to approximate in inches of water the various values given in Table II-A after being in use twenty-four hours and with all the inlets suddenly closed. The figures given are about 90% efficiency of the theoretical draft which an absolutely tight chimney would show and a properly constructed chimney should approximate these figures. METHOD OF USING DRAFT GAUGE. Measurement of the draft in a chimney is made by a draft gauge which in its simplest form is nothing but a glass tube of U-shape, one end of which is connected to the base of the chimney and the other end is left open to the air. The suction or draft of the chimney will raise water placed in the tube so TABLE II-A-DRAFT IN CHIMNEYS FOR RESIDENCES. that the water in the leg connected to the chimney stands higher than that in the leg open to the atmosphere, this being due to the unbalanced pressure and proportional thereto. The difference in level between the two columns of water is measured in inches, this difference being termed the draft in "inches of water." Some draft gauges have the legs of the U inclined only a few degrees off the horizontal so that a rise of a small amount causes a large movement horizontally along the inclined tube on which finer graduations can be read. To take a theoretical draft reading (which is the most accurate method to pursue) see that the temperature of the flue gases is known at the time of taking the reading and that a tight damper can be closed in the smoke pipe. Connect the draft gauge to the chimney, take the flue gas temperature and then quickly close the smoke pipe damper and read the draft gauge as soon as it reaches the highest fixed point. For points above sea level the readings will be less than that shown in the table the percentages being for Leakage may take place not only through the joints of the chimney itself but through the connections to the chimney. For this reason a separate chimney flue for every connection is advocated. A stove on the first floor and a heating apparatus in the basement may be connected to the same flue, if it is large enough, and both may work satisfactorily; but put out the fire in the stove and open the doors-the inrush of cold air into the flue will spoil the draft for the heating apparatus, this inrush of air being simply excessive leakage into the flue all at one point. The leakage may not even come from the outside air but may come from an adjoining flue in which case it is just as detrimental in its effect. For this reason chimneys with more than one flue should never be built without a 4-in. brick wall between the flue tiles. The common practice of placing flue tile close together without this brick wall later leads to trouble in many chimneys. Several boilers may be connected into one flue of sufficient size if they all are properly dampered in the smoke pipe with tight dampers and if these dampers are kept closed when the boilers are out of commission. USE OF TWO FLUES FOR ONE APPARATUS. Questions are often asked about using two old flues for one apparatus when it is desired to make alterations in an existing building. In this regard it can be said that reasonably satisfactory operation can be secured from such an installation if the dividing partition is knocked out 5 or 6 ft. down from the top of the chimney and up to 5 or 6 ft. above the point where the smoke pipe enters the chimney. There must be no other connections to the chimney, however, and the chimney must be in good condition otherwise to serve its purpose; the best results will be obtained where the smoke pipe enters the middle of the enlarged chimney so as to distribute the hot gases as equally as possible into both flues higher up. Of course, the center partition remaining exposes just so much more rubbing surface to the gases, giving a greater friction loss and the chimney will give better serviceotherwise a stronger draft if the center partition is removed entirely. OTHER CHIMNEY DEFECTS. Oftentimes chimney troubles are entirely due to outside causes and not to the construction of the chimney itself. The most frequent of these is the termination of a chimney below the level of the adjacent ridge of the roof or other nearby obstruction. Even nearby buildings or mountains may interfere with the operation of a chimney, the cause in all cases being the wind coming over the higher obstruction and dropping down onto the top of the chimney. As can be easily realized, chimney draft (after deducting the friction in the chimney, in the heating apparatus, and in the fire and smoke pipe) does not possess a very great force, and a high wind, with the air dropping down toward the top of the chimney, could readily hold back the action and, in some cases, even result in "back-drafts." The best remedy for such trouble is to extend the chimney higher so as to get above the nearby ridge or building if possible. This at the same time will both prevent the wind dropping down onto the top of the chimney and will give the chimney (due to the increase of its height) a stronger draft to resist such action. If this is impractical one of the many types of chimney tops having side or downward outlets will prevent "back-drafts" but these will at Top of Chimney FIG. 3-GOOD TYPE OF CHIMNEY TOP. the same time slightly retard the chimney action, due to the greater number of turns the gases are forced to make and the additional friction thus produced. A very desirable type of chimney top, where "down-drafts" are not likely to be encountered, consists of finishing off the top at a 45% angle all around from the flue outlet to the side of the chimney, as shown in Fig. 3. This results in deflecting horizontal winds upward when blowing across the top of the chimney and gives a slight ejector action, FIG. 4-HOW CHIMNEY AREA MAY BE RESTRICTED BY SMOKE PIPE. experienced on certain days when the cellar door comes on the leeside of the building or when the cellar is so air tight that sufficient air does not leak in to replace that lost up the chimney. This results in a partial vacuum in the cellar which vacuum reacts against the draft that would otherwise be obtained. The remedy for this is to open a cellar window on the windward side, close the opening to the leeward, or else provide some means of allowing air to enter from some other source where the wind action is not encountered. The The connection to the chimney commonly termed the smoke pipe is another factor which must be considered. smoke pipe should never be reduced below the size of that on the heating apparatus; if more than one smoke pipe is connected into a common pipe before going into the chimney the area of the common pipe should have an area equal to the combined areas of the two entering pipes. A loose connection where the pipe enters the chimney, a loose fitting cleanout door or other openings leading into the chimney will act the same as a leakage from any other source. Neither should the smoke pipe be shoved into the chimney so far as to project into the interior opening as shown in dotted lines-Fig. 4-since this restricts the area and if carried to an extreme will shut off the draft entirely. Every smoke pipe should have a tight damper operated by hand for closing down during high winds, for testing and for use when the apparatus is out of service. Chimneys which have given good serv ice for years and which then gradually fail to have sufficient draft usually are found to need cleaning and chimneys with offsets at a sharp angle are usually troubled in this way. A cobble stone tied to a rope and dropped down gradually from the top of the chimney, working this up and down as it goes down, will clean out many such troubles at slight expense. Gun powder in small charges ignited in the base of the chimney has also been used to jar the soot loose but this method does not come within the "safety first" limitations. Next month the author will take up designing data for chimneys and power plant stacks. His discussion will be accompanied by friction loss charts for brick and steel stacks and a theoretical draft chart for stacks. Remodeling the Ventilation of a Theatre Changes Made to Overcome Overheating, Local Cold Drafts and Unsatisfactory Ventilation. By SAMUEL R. LEWIS. (Presented at the annual meeting of The American Society of Heating and Ventilating Engineers, New York, January 16-18, 1917.) The purpose of this paper is to describe some experiences with the heating and ventilating of a very large theatre. The original plant was not satisfactory. It was changed at slight expense into a very satisfactory system by performing a comparatively small amount of labor. The theatre is used mainly for concert purposes, being the home of an endowed orchestra. As originally installed, the plant was arranged for downward ventilation. The fresh air, taken from a point about 20 ft. above the ground, was warmed to prevent freezing, drawn through an air washer and tempering heaters to a fan, and then under pressure was delivered through an underground masonry tunnel, as shown in Fig. 1, to reheaters placed approximately under the center of the auditorium. These reheaters were placed in connection with plenum chambers in such a way that single metal ducts provided with automatic double mixing dampers, radiated from them, leading to the various flues placed around the walls. The mixing dampers in these ducts were controlled by thermostats placed in the auditorium. The air was to be removed through approximately 800 5-in. floor tubes through the first floor, which connected into chambers formed by the concrete first floor as a ceiling, the circumferential floor beams as sides and the galvanized steel bottom. These chambers, at each side of the basement, connected with metal ducts to two large vent flues, one at each end of the stage, running from the basement to the exhaust fan ehamber in the attic. The illustration, Fig. 1, which is a center line section, shows the location of the inlets and outlets as follows: Flues A: One of these is placed at each side of the stage, runs to the attic and across the attic to the rear of the house. Each has openings as follows: at A, 10 ft. above the lowest point of the first floor; at A-1, opposite the balcony, and at A-2 where each duct supplies two openings discharging diagonally through the ceiling over the gallery. Flues B: Two on each side under the balcony discharging toward the center of the house. Flues C: One on each side under the balcony discharging toward the stage. Flues D: Two on each side underneath FIG. 1-LONGITUDINAL SECTION OF THEATRE BUILDING, ON CENTER LINE, SHOWING ARRANGEMENT OF AIR DUCTS AND OUTLETS. tended to prevent the escape of heated air when the auditorium was unoccupied. The very tortuous duct systems in the basement and attic, comprising both fresh- and foul-air conveyors presented a very complicated appearance, and caused a very heavy frictional resistance. Very little heating effect was apparent when operating the plant without running the fans. HEATING The lobbies had entrances and stairways for the first floor and balcony pleted the following changes were made: The basement ducts leading to flues E were cut off from these flues and connected so as to deliver fresh air into the former foul air chamber M, between the rear-most circumferential beam and the rear wall of the first floor. This chamber was closed off from its former connection with the foul-air collecting ducts at the side walls. The flues, E, however, were connected with these foul-air ducts. Thus, the seats in the rear panel across the house on the first floor were supplied with air by |