on globe valves, sometimes on both sides controlling this section as a separate of the valve. LEAKS IN VALVE BONNETS. Leaks often occurred in the bonnets of valves at the radiators. These were very annoying as they did not remain as corrected. This was due to valve stems which had become bent or twisted. When the valves were operated the bent stems would work the packing loose. Formerly any return which was pocketed would freeze up if the radiator to which it was connected had its supply valve shut off, as at the close of working hours when the men left the offices. Evidently the steamfitters objected to the amount of attention which they were called upon to give these frozen lines and had applied a wrench to the valve stems in order to unit. TROUBLES FOUND AFTER VACUUM HEAT ING SYSTEM WAS INSTALLED. After the system had been changed to operate as a vacuum return system a number of further troubles developed, which were not the fault of the new system itself, but due rather to incorrect methods of piping. A 4-in. line carrying steam at 50 lbs. pressure extended from the boiler house to the yard. A connection had been taken off of this main and a line run to the radiators in the building nearest the boiler house. The returns from these radiators were connected into the returns of the vacuum system. This occurred at a vital point in the system, as all other Supply. Radiator Angle Branch Floor Line Ordinary Straightway FIG. 2-CONDITIONS IN OLD SYSTEM AND AS CHANGED IN NEW SYSTEM. lock the valves in an open' position, and thus keep the steam flowing through the radiators at all times. In this way the valve stems were bent. Another scheme which was worked by the steamfitter was to remove the discs from the globe valves at radiators. The writer has noted with amusement the apparent satisfaction which some of the tenants of the offices derived from operating these valves, being perfectly oblivious to the fact that the disc had been removed. The supply and return branches to the radiators in offices at the north section were in an exceptionally poor condition, in the way of numerous pockets, uncovered pipes, etc. It would have taken considerable time to place this piping in good condition. Therefore, all of this portion of the system was removed and replaced by mains of larger size, thus eliminating the former method of supplying the radiators in roundabout. ways. A differential valve was installed returns are beyond this point, and being the farthest from the pump the flow through them was greatly retarded. As the differential valves operate on a difference in pressure, it may be readily seen that those which were nearest the pump would open and pass steam continuously on account of the high pressure back of them, thus destroying the vacuum in the main return and causing the whole plant to operate unsatisfactorily. The radiators in this building were finally connected to the low pressure supply main and the trouble avoided. A number of low points were formed in the old return piping by the installation of impulse valves of the angle type at radiators. The available pitch for the returns above the floor was limited to a few inches. Had the return from a radiator or coil been started with an impulse valve of the straight-way type there would have been available more of the headroom which existed between the return connection of the radiator and the floor. The following sketches will make this point clearer. The sketch at the left shows the conditions as they existed in the old system, and one at the right the conditions as existing after the system had been changed over. (Fig. 2). The contents of a strainer at the pumps was removed. There were a few pebbles and a large cinder in this deposit which had collected during a period of two months. Strainers should be cleaned more frequently, however. The deposit was formed chiefly by the combination. of oil with the rust in the pipes and resembled a sort of putty or red lead. TROUBLES DUE TO PRESENCE OF OIL AND pressor, as the locomotive boilers, when forced, supply wet steam to the same. The oil separator is close to the compressor and its discharge is trapped. Exhaust steam for the feed water heater is taken off at right angles from the main, and this seems to prevent the oil from being carried into the heater. A frequent trouble with oil separators is imperfect drainage, or in the use of an oil which is too volatile, thus allowing some to go through the separator without striking the baffles at all. The baffles should be frequently cleaned, and the efficiency of the oil separator should then be from 95 to 99% elimination of the total quantity of oil used. The troubles arising from the mixture of oil and rust continued throughout the first heating season. This system was formerly supplied by live steam only. When exhaust steam was used for heating, as is usual with vacuum systems, the oil in the steam loosened up the scale which adhered to the pipes. Instances can be recalled where boilers have become leaky and crown sheets overheated from loosening scale whenever an open feed water heater was installed. The exhaust steam used in these heaters contains oil which does not harm the boilers by causing the leaks directly, but indirectly by loosening the scale over these leaky spots. The troubles from scale in old installations are an annoyance at first, but this also occurs in any new heating system until the piping has been entirely flushed. out. FIGS. 3-5-TYPICAL EXAMPLES OF IMPROPER METHODS OF PIPING. A-Differential Valve. B-Globe Valve. C-Branch Return. D-Main Return. TYPICAL CASES OF IMPROPER AND PROPER The following photographs (Figs. 3-5) are submitted as typical examples of improper methods of piping which usual practice with the type of system here installed, it was necessary that differential valves should be connected to the main return at various points in order to receive the discharge from the branch returns and regulate the flow through each. Some better methods of piping are shown in the following photographs. Note the conditions existing under the the floor and the few feet of headroom. (Figs. 6-11). The use of so many fittings in connecting the branch returns to the main return is undesirable as they unnecessarily impede a free flow, increasing the liability of leaks, also increasing the maintenance; in other words they have the inherent potentiality of trouble. FIGS. 6-11-EXAMPLES OF BETTER METHODS OF PIPING. occurred mainly C-Branch Return. Some of the connections which are not shown in these photographs were particularly bad. In one instance a return rose in order to connect to the top of an old tee which happened to be convenient, in the main return. This tee might have been turned to allow its ed by the use of shorter lengths of pipe, shorter nipples, or by turning a tee or fitting on its side when making a connection thereto. Some of these returns had a grade in the opposite direction to the flow of condensation through them. The photograph (Fig. 12) shows the high point which existed in one of these. returns caused by the hanger supporting this pipe being too short in length. The branches should connect direct to the main and not parallel the same for some distance before connecting to it. The branch returns cannot strictly comply with this unless there is a differential installed for each, otherwise a number of branches must be connected together and run to one valve. One of the sketches (Fig. 13) illustrates how one of these long runs parallel to the main was finally avoided by installing another differential valve. This also divided one large unit controlled by one valve, into two smaller units each controlled by a differential valve, thereby affecting better control over each, and distributing the work involved. The cost for two valves of smaller size is about the same as for the one valve of larger size. 2"Main Return to Pump In order to divide one large unit into two smaller units, the Differential Valve and Piping shown in dotted was installed FIG. 13-METHOD OF DIVIDING ONE LONG RUN INTO TWO SMALLER RUNS BY USE OF DIFFERENTIAL VALVE. Reduction or Elimination of Noise Attending the Operation of Mechanical Ventilating Machinery (Presented at the mid-summer meeting of the American Society of Heating and Ventilating Engineers, Cleveland, O., July 9-11, 1914). By R. W. As there does not seem to be any law which could be laid down to govern oneself on the installation of ventilating machinery, I will endeavor to illustrate some different phases of the subject with the remedies used to correct the trouble. Below are given some experiences: A 60-in. steel plate fan, with overhung blast wheel, was operating in a concreted basement. This basement was walled-in in a substantial manner, and entering into this basement where the fan was located, were some conduits. The fan was operating at a pressure of something over 1 oz. and there was no perceptible noise immediately in the vicinity of the fan. However, about a half a block away there was a decided rumble, which was traced to the operation of this fan. The fan was covered with 6 in. of hair felt. This helped to deaden the sound somewhat, but did not entirely correct it. It seemed as though the fan or something started a vibration of the air in this chamber, the sound of which was projected or carried through the conduit tubes to the location where the sound became audible. I believe this is one of the most elusive problems the engineer has to contend with. Another instance was the case of two 24-in. motor-driven propeller wheels, which were located in the rear of a counting room of a bank, discharging the air into a brick areaway. These fans made a howling noise, which might be termed windage. Due to the air rushing through the blades, the fact that they discharged into this brick areaway, aggravated the condition. These propeller wheels were changed to flat blade type, the speeds being approximately the same. in either case, but the amount of air handled by the flat blade wheel is about 75% of that handled by the curved blade. The result was, that the improvement was very noticeable and the noise practically eliminated. The speed of the first PRYOR, JR. fan, and the velocity of the air through it, was too great for the size of the areaway and the sound was the result of vibrations due to the high velocity in a confined space. OVERCOME OBJECTIONABLE HUM OF APPARATUS. In a certain building consisting of a hall approximately 48x54x17 ft. high, with an open attic space above, there was located a 36-in. disc wheel with directconnected single-phase motor running at 600 r.p.m. This fan had no immediate connection with any duct work, being mounted on a special steel support held firmly in the brick wall. This fan drawing on the attic space, effected its ventilation of the hall through openings leading from the hall into the attic space. The hum from this outfit was so objectionable that the apparatus could not be used. The location of this outfit was subsequently changed so that it did not pull its air through quite so great a distance, thereby changing the proportion of the cubical contents through which it was drawing to the volume of air handled. The outfit was supported in a similar manner as before, and to one of the same brick walls; the same material being used as far as possible to make the change, and the fan running in the new location gives the desired ventilation without the noise which was so objectionable. This seems to bear out the theory that we have to consider the question of sound produced by the vibration of the air independent of the sound produced by the friction of the air against the walls of the duct. I think this condition is further exemplified by the fact that sound travels approximately 61,000 ft. a minute, which on ventilating systems, such as we have been considering, gives the velocity of sound anywhere from 60 to 100 times greater than that at which the air is traveling. |