CHAPTER VII.Pipe and Fittings.

As we continue to follow the dust-laden air in its passage toward the vacuum producer we next encounter that portion of the conduit which is permanently and rigidly fixed in place in the building; namely, the pipe line, its fittings and other appliances.

—The first portion of this conduit which we must consider is the point where the hose is attached to the pipe line; that is, the inlet, or, as it is often improperly termed, the “outlet” valves.

As it is necessary to close the inlets air tight when they are not in actual use, in order to prevent the entrance of air except through the hose lines in use, some kind of a cut-off valve must be provided, as well as a receptacle into which the end of the hose may be connected when desired.

With the earlier systems a high degree of vacuum was carried in the pipe lines and the vacuum producers were of small displacement. Slight leakage would greatly reduce the capacity of the system and the best form of valve was necessary. The valve adopted was the ordinary ground-seat plug cock, on account of its unobstructed air passage and air-tight closing. The hose was connected to these cocks either by a ground-joint, screwed coupling or by a slip coupling similar to those used to unite the sections of the cleaning hose. An inlet cock of this type is illustrated inFig. 51.

These cocks projected about 4¹⁄₂ in. beyond the face of the finished wall and formed a considerable obstruction, especially when located in halls or corridors. In order to reduce the projection into the apartment the manufacturers of the systems using screwed-hose couplings and substituted a projecting nipple closed by a cap screwed in place. The whole projected only ³⁄₄ in. beyond the finished wall line.

These outlets were suitable for use only with hose having screwed connections. When an attempt is made to remove the cap with the vacuum producer in operation, there is a tendency for the vacuum to cause the cap to hug the last thread and render its removal difficult. Also, when the suction is finally broken it is accomplished with considerable hissing noise.

FIG. 51. INLET COCK TO PREVENT AIR LEAKAGE WHEN NOT IN USE.

In order to permit the use of the slip type of hose coupling, a hinged flap valve was substituted for the screwed cap, a rubber gasket being placed under the cap. This was held firmly in place by the vacuum in the pipe line. The interior of the casting inside of the flap was turned to a slip fit for the end of the hose coupling. With this type of valve and the slip hose coupling, described inChapter VI, it is possible to reverse the hose to equalize wear and remove obstructions.

These inlets have been made with valves that are closed only by gravity when there is no vacuum on the system and many are so constructed that when opened wide they will remain open with the vacuum on the piping. This type of valve will often be opened by the inquisitive person when no vacuum exists in the system and as there are no immediate results, they may be left open with the result that there will be a very large leakage of air on starting the vacuum producer. This makes it necessary for some one to make a tour of the building in order to close the valve which is open before the system can be efficiently operated. If the vacuum producer is designed to operate several renovators simultaneously, it may not be discovered that there are any valves open and a considerable amount of power will be wasted.

In order to overcome this difficulty it is necessary to provide a spring on the hinge of the flap valve that will automatically close the valve whenever the hose is withdrawn. When the inlets are located in public places they should be fitted with a lock attachment to prevent them from being opened by unauthorized persons.

A valve of this type is illustrated inFig. 52. This valve has a projection on its inner face which engages with a ridge on the hose couplings, preventing the removal of the hose without slightly raising the cap and making it impossible to accidentally pull the hose out of the inlet.

FIG. 52. TYPE OF AUTOMATIC SELF-CLOSING INLET COCK.

The particular valve here shown is suitable for use only with the all-rubber hose connection described inChapter VI.

We must next consider the material of which the conduit itself is to be made. The commercial wrought-iron or mild steel, screw-jointed pipe, such as is used for water and steam lines, is probably the best suited for this purpose and was the first material used. In earlier installations the pipe was galvanized, but, owing to the tendency for the zinc coating to form irregularities within the pipe, its use has been abandoned in favor of the commercial black iron pipe.

Seamless drawn tubing would undoubtedly make the ideal material for this purpose. However, the ordinary butt or lap-welded pipe is satisfactory and is now generally used.

Sheet metal pipe was introduced by one manufacturer but its use was shortly abandoned in favor of the commercial pipe.

As joints and changes in direction are necessary in the pipe lines, some sort of fittings must be used. The ideal conduit for passage of dust-laden air should be of uniform bore and as smooth on the inside as a gun barrel. Various attempts have been made to accomplish this result in commercial installations, one of which is illustrated inFig. 53. These fittings are made up of three parts for a coupling and four for a branch or change in direction. One of these is screwed on to the end of each piece of pipe, the pipe butting against a shoulder and the end of the pipe made to register with the bore of the fitting by reaming. This piece is faced true and fitted against the face of the casting, forming the bend or branch, or fitted against the piece on the end of the other length of pipe. A thin gasket is placed between them, a projecting ring on one piece fitting into a groove on the other, causing the bore of the two halves to register. The two halves are joined together by the V-grooved clamp, held in place by a small bolt. This is theoretically an ideal joint, but the clamp is not of sufficient strength to withstand the strain of settlement of the building and breakages are frequent. Several instances of this character, particularly on steamers, have come to the observation of the author, and there are several buildings which have been roughed in with this type of fitting, used on concealed piping, which were found to be useless on the completion of the building, due to breaking of the joints in inaccessible places.

FIG. 53. “SMOOTH BORE” PIPE COUPLING.

A modification of this joint which will have ample strength can be made by using standard pipe flanges, screwing the pipethrough the flange and facing the end off in a lathe. Fittings could be made with a bore equal to that of the pipe and proper alignment secured by the use of dowel pins, as illustrated inFig. 54. The cost of making this joint would be high and they would occupy too much space to be easily concealed in partitions, furring or other channels usually provided for the reception of such piping.

FIG. 54. JOINT MADE OF STANDARD PIPE FLANGES.

The standard Durham recessed drainage fitting, having the inside cored to the bore of the pipe and recesses provided for the threads as used in connection with the modern plumbing system, if left ungalvanized and having the inside well sand-blasted to remove all rough places, makes a serviceable fitting. Care should be exercised to cut the threads on the piping of proper depth to allow the end of the pipe to come as close to the shoulder of the recess as practicable and to obtain a tight joint. The end of the pipe should be carefully reamed before assembling.

These fittings have become standard with nearly all manufacturers and are illustrated inFig. 55, which shows the right and wrong way to install same.

Trouble was experienced on some of the earlier systems using high vacuum with the fittings cutting out on the side subjected to the impact of the dust-laden air. To overcome this trouble one manufacturer re-inforced the fittings by increasing the thickness of metal at the point affected. The trouble was undoubtedly caused by too high a velocity in the pipe line, as in the case of the small brass stems, explained inChapter V. With the introduction of vacuum control and larger pipes,this trouble disappeared and the special fittings never came into general use.

RIGHT WAY.WRONG WAY.Use two Y-branches instead of straight or cleanout tees. In case the latter are used the dirt will shoot by into the other branch.Always place Y-branches so they will turn in the direction of the flow.Place the clean-out at right angles to the direction of flow entering the fitting. Otherwise it serves as a pocket to catch passing dirt.Special care must be exercised to see that there is no opportunity for dirt to collect in the basement drops. Above is shown a common wrong way and two possible right ways.FIG. 55. STANDARD DURHAM RECESSED DRAINAGE FITTINGS GENERALLY USED IN VACUUM CLEANING INSTALLATIONS.

RIGHT WAY.WRONG WAY.

RIGHT WAY.

WRONG WAY.

Use two Y-branches instead of straight or cleanout tees. In case the latter are used the dirt will shoot by into the other branch.

Always place Y-branches so they will turn in the direction of the flow.

Place the clean-out at right angles to the direction of flow entering the fitting. Otherwise it serves as a pocket to catch passing dirt.

Special care must be exercised to see that there is no opportunity for dirt to collect in the basement drops. Above is shown a common wrong way and two possible right ways.

FIG. 55. STANDARD DURHAM RECESSED DRAINAGE FITTINGS GENERALLY USED IN VACUUM CLEANING INSTALLATIONS.

While the utmost care should be taken to prevent stoppage of the pipe lines these stoppages are likely to occur in the best-constructedlines and ample clean-out plugs should be provided for the removal of such stoppage. Brass plugs are the most serviceable for this purpose, as they are easily removed when necessary and can usually be replaced air tight.

The brass clean-outs, while most satisfactory, are costly when installed in large sizes. Equally satisfactory results can be obtained at a lower cost by using 2-in. diameter plugs on all lines 2 in. and over in diameter.

FIG. 56. FRICTION LOSS IN PIPE LINES.

Matches are perhaps the most frequent cause of stoppage in pipe lines. Stoppage from this cause can be largely avoided by the use of pipe of sufficient size to permit the match to turn a complete somersault within the pipe whenever it catches against a slight obstruction or rough place in the pipe orfittings. A 2-in. diameter pipe is just large enough to permit this and smaller sizes of pipe should be avoided whenever possible.

—The friction loss in piping follows the same law as that in hose lines and is easily computed by use of the chart (Fig. 56), which is constructed on the same general principle as the chart of hose friction (Fig. 48). The directions for use of the hose chart apply to the pipe chart. In computing this chart the actual inside diameter of the commercial wrought-iron pipes have been used instead of the nominal diameters, resulting in an increased capacity for all sizes except 2¹⁄₂-in. which is less than the nominal diameter.

—Friction in the pipe lines tends to increase the vacuum to be maintained and therefore the power to be expended at the vacuum producer and should be kept as low as possible. The pipe sizes should be made as large as conditions will permit. The limit of size is fixed by the velocity in the pipe. When it is necessary to lift the dirt to any extent, the velocity should not be allowed to fall below 40 ft. per second at any time. When the pipe is a vertical drop, the velocity does not matter as gravitation will assist the air current in removing the dirt. When the line is horizontal a lower velocity than 40 ft. per second is permissible at times, provided that this minimum velocity is exceeded at frequent intervals to flush out any dirt that has lodged in the pipe during periods of low velocity.

If a Type A renovator is used with 1-in. hose and a vacuum of 10 in. of mercury maintained at the hose cock, the minimum air passing, with 100 ft. of hose in use, will be 29 cu. ft. of free air per minute, which is equivalent to 44 cu. ft. at 10 in. of vacuum. The entering velocity in the pipe should be calculated with air at this density. This will give a velocity of 50 ft. per second in a 1¹⁄₂-in. pipe, but only 30 ft. per second in a 2-in. pipe. Therefore, the 1¹⁄₂-in. pipe is the largest that should be used where lifts occur on a line serving but one Type A renovator with 1-in hose. When the renovator is tilted at a considerable angle or lifted from the carpet, as will frequently occur in cleaning operations, the quantity of air passing the renovator will be upwards of 42 cu. ft. of free air, equivalentto 62 cu. ft. at 10-in. vacuum. When this occurs the velocity in a 2-in. pipe will be 44 ft. per second, which will be ample to flush a horizontal line of piping.

If 1¹⁄₄-in. hose is used with a Type A renovator, the minimum quantity of air will be 29 cu. ft. and the vacuum entering the pipe will be 6 in. mercury, giving an equivalent volume of 37 cu. ft. This will produce a velocity of 42 ft. per second in a 1¹⁄₂-in. pipe, which is the largest that can be used where a lift occurs. However, when the renovator is lifted free of the carpet, the air quantity will be 62 cu. ft. of free air, equivalent to 80 cu. ft. at 6 in. of vacuum, and will produce a velocity of 39 ft. per second in a 2¹⁄₂-in. pipe. This would be just about sufficient to flush a horizontal line.

If 1¹⁄₂-in. hose were used the air quantity will be 29 cu. ft. and the vacuum entering the pipe 5 in. mercury, equivalent to 35 cu. ft. This will give a velocity in a 1¹⁄₂-in. pipe of 40 ft. per second. When the renovator is raised from the carpet, the air quantity will be upwards of 90 cu. ft. of free air, equivalent to 110 cu. ft. at the density of that entering the pipe, and will produce a velocity of 33 ft. per second in a 3-in. pipe. This is too low to thoroughly flush a horizontal pipe.

The figures given above are repeated fromChapter VIand show that the use of 1¹⁄₄-in. instead of 1-in. hose, permits the use of a larger-sized horizontal pipe line for serving one renovator, but that the use of 1¹⁄₂-in. hose, instead of 1¹⁄₄-in., will not permit of any enlargement in the pipe size. Since we have seen inChapter VIthat a 1¹⁄₄-in. hose gives the least expenditure of power when used with a Type A renovator, there will be no gain from a reduction in the pipe friction due to the adoption of this hose.

The dependence on the raising of the renovator from the floor to flush out a larger pipe line should not be carried beyond that to be obtained from a single renovator. That is, when the pipe must serve more than one renovator at the same time, the quantity of air that two or more renovators will pass, if they were raised from the floor at the same time, should not be used in determining the limiting velocity in the pipe, as such an occurrence is not likely to be obtained often enough to thoroughly flush the pipe. Furthermore, there will be times when this pipe will have to serve only one renovator and the pipe will not beadequately flushed. When the pipe is serving more than one renovator, the actual air passing the renovators should be used in determining the maximum size of pipe and it is advisable to use this maximum size in nearly all cases where the structural conditions will permit.

These sizes will then be:

TABLE 18.Pipe Sizes Required, as Determined by Air Passing Renovators.

Using these maximum sizes, the friction loss in a pipe line, with carpet renovators in use exclusively, will be:

TABLE 19.Friction Loss in Pipe Lines, with CarpetRenovators in Use Exclusively.

These friction losses are figured with a density of air in the pipe equal to 6-in. vacuum in case of the 1¹⁄₄-in. hose and 10-in. vacuum in case of the 1-in. hose, which will be the density of the air entering the pipe, while the average density should be used in order to give correct results. If the pipe line is not over 400 ft. equivalent length the results will be approximately correct.

These results show, first, that the friction loss in pipe lines is much lower than that in the hose lines used with the same system; second, that the higher vacuum in the pipe causes greater loss, an argument in favor of the use of larger hose.

FIGS. 57-60. DIAGRAMS SHOWING OPERATION OF BRUSH AND CARPET RENOVATORS UNDER DIFFERENT CONDITIONS.

These friction losses are obtained only when carpet renovators are used exclusively and all the renovators are held in the proper position to perform the most economical cleaning. In actual practice this condition will not exist except when one renovator is used. Where more than one renovator is in use simultaneously, some of the renovators will be raised from the floors at the time others are in position to do effective cleaning and will admit a greater quantity of air, increasing the friction. This is not a serious condition as the time that the renovators will be raised is only a small part of the total time spent in cleaning and will merely reduce the efficiency of the other renovators temporarily. However, when brushes or floor renovators are used at the same time as the carpet renovators, there will be a continuous flow of air in greater quantities throughthese brushes, which will permanently increase the friction loss. The use of a single brush or floor renovator with the same sized pipe as is necessary to operate the carpet renovator will not reduce the efficiency of the brush, as a high degree of vacuum at the brush or floor renovator is not necessary or even permissible and a further slight reduction will not affect the operation of these renovators.

When a brush or floor renovator is used on the furthest outlet from the vacuum producer at the same time that carpet renovators are being used on outlets nearer the vacuum producer, the larger quantity of air passing the brush will tend to reduce the vacuum at the hose cock to which the carpet renovator is attached and thereby impair its efficiency. For example, if we have a brush renovator connected through 100 ft. of 1-in. hose to an outlet at the end of a pipe line 400 ft. long, properly designed to serve two carpet renovators, the vacuum at the separators should be maintained at 10-in., plus 2 × 0.20 plus 2 × 0.30, or 11 in. of mercury. Suppose that this vacuum is automatically maintained at this point and a carpet renovator be attached 200 ft. from end of pipe (Fig. 57). The quantity of air passing through the 2¹⁄₂-in. pipe B-C will be approximately 29 plus 40 or 69 cu. ft., and the friction loss in this pipe will be 1.1 in. The vacuum maintained at the outlet B (Fig. 57) will be 9.9 in. or approximately the correct vacuum to maintain 4¹⁄₂-in. vacuum at the renovator “a.” The friction loss in the pipe line from B to A will be 0.7 in. and the resulting vacuum at the hose cock A will be 9.2 in. The quantity of air passing the brush will be 40 cu. ft. Under these conditions there will be no loss in efficiency of cleaning due to the brush renovator being used on the end of the line. If the operator using the brush at the outlet A should use only 25 ft. of hose instead of 100 ft. (Fig. 58) the air passing this brush will be 75 cu. ft. and the vacuum at the hose cock A will be 6.8 in. The vacuum at the hose cock B will be 8.8 in. and the vacuum at the carpet renovator “a” will be reduced to 3¹⁄₂ in. with 25 cu. ft. of air passing, which will reduce the efficiency of the carpet renovator “a.”

If the brush renovator be attached to the hose cock B (Fig. 59), using 25 ft. of hose, the vacuum at hose cock B will be9 in. and the brush renovator will pass 85 cu. ft. of air, while the vacuum at hose cock A will now be reduced to 8.6 in. and the vacuum at the renovator will be reduced to 3 in. mercury and the air passing to 23 cu. ft.

If a brush type of renovator be used at each outlet, with 25 ft. of hose in each case and the vacuum at the separator be maintained at 11 in. mercury the vacuum at hose cock B will be 7 in. and brush “a” will pass 76 cu. ft. of air while the vacuum at hose cock “a” will be 5 in. and brush “b” will pass 63 cu. ft. of air or a total of 144 cu. ft., which will be in excess of the 70 cu. ft. per renovator recommended as the capacity of the plant inChapter VI. This will not result in any loss of efficiency if the vacuum producer be designed to handle but 140 cu. ft. as a maximum, for the vacuum at the separator will then fall to a point where but 140 cu. ft. passes, resulting in a decrease in the vacuum throughout the system. But as only brushes are now in use there will be no loss in efficiency, owing to the reduction in the vacuum at the brushes.

When 1¹⁄₄-in. hose is used with a carpet renovator at the end of the pipe line connected through 100 ft. of hose and a brush at the hose cock B connected through 25 ft. of hose (Fig. 60), the worst case of the three already cited, the vacuum at the separator being maintained at that necessary to carry 4¹⁄₂ in. when two carpet renovators are in use, the vacuum at the hose cock B will be 4.5 in. and brush “a” will pass 116 cu. ft. of air while the vacuum at hose cock A will be 4.4 in. and the vacuum in renovator “b” will be 3.7 in and will pass 24 cu. ft. of air.

These are better cleaning conditions than were obtained when 1-in. hose was used. It will be noted that the total air passing the exhauster is now 140 cu. ft. and this must not be reduced or there will be a falling off in the vacuum at the carpet renovator “b.” It is, therefore, necessary for the exhauster to be capable of handling 140 cu. ft. of air or 70 cu. ft. of air per renovator in order to do effective carpet cleaning when carpet renovators and brushes are used in conjunction.

When two floor brushes are used with the above arrangement of pipe and hose, the vacuum must fall considerably orthe air quantity be greatly increased. However, the reduction in vacuum will not result in serious loss in efficiency when only brushes are in use.

When a larger number of sweepers are used with a system of piping, it is necessary to allow 70 cu. ft. of free air per sweeper in figuring the sizes of pipe to be used, and the total loss of pressure in the piping between the outlet farthest from the vacuum producer and that nearest to same must be limited in order to prevent too wide a difference in the vacuum at the hose cock when all the sweepers for which the plant is designed are in use. The author considers that this loss in pressure should not be greater than 2 in. mercury in order to give satisfactory results.

Before the piping system can be laid out and the sizes of piping determined it is necessary to ascertain, first, the number of sweepers to be operated simultaneously and the number of risers necessary to properly serve these sweepers.

—This is determined by the character of the surfaces to be cleaned, the amount of such surface, and the time allowed for cleaning.

It has been demonstrated in actual practice that one operator can clean as high as 2,500 sq. ft. of carpet when same is on floors of comparatively large areas, and not over 1,500 sq. ft. when the carpets are on small rooms; 2,000 sq. ft. is considered to be a fair average.

Bare floors are cleaned more rapidly. In school house work an ordinary class room has been cleaned in 10 minutes, or at the rate of 7,200 sq. ft. per hour, but time is occupied in moving from one room to another and the writer considers 5,000 sq. ft. per hour as rapid cleaning and 3,500 sq. ft. as a fair average.

The time of cleaning will vary in buildings of different character and used for different purposes. In office buildings the cleaning force work throughout the night or about 10 hours, while in school houses the cleaning is done by the janitor force which has been on duty throughout the school period and the time is necessarily limited to about two or three hours after school hours, the corridors and play rooms being cleaned during the school period and only the class rooms being cleaned after closing time.

Let us assume, as an example, an office building having eight floors each 100 ft. × 150 ft., with a floor plan as shown inFig. 61.

The corridors, stairs and elevator halls will probably be floored with marble which must be scrubbed in order to remove the stain accumulated during the day and they will not be considered in connection with a dry vacuum cleaning system. The area of the floors in the offices on any floor will be approximately 10,000 sq. ft. and one floor can be cleaned by one operator in 5 hours, or two floors during the cleaning period, so the plant must be of sufficient size to serve four sweepers simultaneously.

FIG. 61. TYPICAL FLOOR PLAN OF OFFICE BUILDING ILLUSTRATING NUMBER OF SWEEPERS REQUIRED.

In a school house containing four class rooms, where the janitor cleans the play rooms and corridors during the school period, as can be readily done with a vacuum cleaner since there will be no dust scattered about to fill the air and render it unsanitary, the class rooms can easily be cleaned in one hour by one operator. The author considers that one sweeper capacity for each six to eight rooms is ample for a large school.

Buildings of special construction and used for special purposes must be considered differently according to the conditions to be met, but the size of the plant can be readily determined in each case by use of the rules already given.

—Much difference of opinion exists among the various manufacturers of vacuum cleaning systems as to the maximum length of hose that should be used with a cleaning system, and as this maximum length determines the number of risers to be installed, some fixed standard is necessary. As already stated inChapter VI, the author considers that this maximum should be fixed at 75 ft.; that is, the risers should be so spaced that all parts of the floor of the building can be reached with 75 ft. of hose. Where 50 ft. is used as a maximum, as is recommended by many manufacturers, the number of risers would be increased, incurring a greater cost of installation and requiring the operator to shift his hose from one inlet to another more often than would be the case where fewer inlets were used, and more time would be required in cleaning, with a slight reduction in the power. The author does not consider that this reduction in power would be sufficient to offset the additional time required to change the hose from one inlet to another.

The best and quickest way to determine the number of risers necessary is to cut a piece of string to the length representing 75 ft. on the scale of the plans, and by running this around the plan using corridor doors for access to all rooms, wherever possible, locate the riser so that every point can be reached with the string. In the case of the building illustrated inFig. 61four risers located as shown will be necessary.

—Before we can determine the size of risers to be installed it is necessary to determine the probable number of sweepers that will be attached to any one riser simultaneously. In the case of the building (Fig. 61) it is possible that there may be four sweepers attached to one riser and it is also possible that there may be but one, and two sweepers to a riser is considered to be a safe assumption. The author uses the following rule in determining the size of risers to use:

Where the number of sweepers is double the number of risers, assume that all sweepers will be on one riser simultaneously.

Where the number of sweepers is equal to the number of risers, assume that half the sweepers will be on one riser simultaneously.

Where the number of sweepers is half the number of risers, assume that one-quarter of the sweepers will be on one riser simultaneously.

When no lifts occur a low velocity in the riser is not objectionable and the size of the riser should be made equal to the size of the horizontal branch thereto throughout its length, wherever this branch is not larger than 2¹⁄₂ in. diameter. When larger, reductions in the riser can be made until 2¹⁄₂ in. is reached when this size should be maintained throughout the remainder of its length. No riser should be made less than 2¹⁄₂ in. unless a lift is necessary.

Before finally fixing the size of riser to be used in any case the size of the branch in the horizontal lines serving the same must be approximately determined.

These sizes will be dependent on the location in which it is necessary to install the vacuum producer. In the case of the building (Fig. 61) the most desirable location for the vacuum producer will be in the exact center of the building.

With the vacuum producer centrally located the longest run from any riser will be 55 ft. To this we must add:

5 ft. for each long-turn elbow.

10 ft. for each short-turn elbow.

10 ft. for entrance to each long sweep Y branch.

20 ft. for entrance to a tee branch, except at sweeper inlets on risers, where 10 ft. is ample.

In calculating the riser friction for risers under 150 ft. in length the whole capacity of the riser can be assumed as being connected to a point midway of its length.

In the eight-story building (Fig. 61) the length of the riser from basement ceiling to eighth floor will be 100 ft. and the length to be figured, 50 ft. The equivalent length of pipe line for any of the risers, with the vacuum producer centrally located, will be:

Each riser is to serve two sweepers and must pass 140 cu. ft. of free air per minute. This will give a friction loss in a 2¹⁄₂-in. pipe of 2 in. mercury, if 10 in. mercury be maintained at the hose cock and 1-in. hose used; and 1.5 in. mercury if 6 in. mercury be maintained at the hose cock and 1¹⁄₄-in. hose used. Either of these figures are within the limits set for the maximum friction loss and 2¹⁄₂-in. pipe will be the proper size for the risers and their branches in the basement.

The portion of the main in the basement that serves the two risers on either side of the building (portion “ab,”Fig. 61) must be of such size as will produce the same loss in vacuum with 280 cu. ft. of air passing as the 2¹⁄₂-in. pipe gives with 140 cu. ft. of air passing. This may be determined from any table of equalization of pipes or may be obtained from the chart,Fig. 48, in the following manner:

Find the intersection of the horizontal line “140” with the diagonal representing a 2¹⁄₂-in. pipe and pass on the nearest vertical to its intersection with the horizontal line “280.” The diagonal inclined toward the left passing nearest this intersection will be the pipe size required. In this case a 3-in. pipe will give a slightly greater friction and will be sufficient.

Unfortunately, it is rarely possible to locate the vacuum producer in as favorable a point as that given in the illustration, but an effort should always be made to select a location as nearly central to all risers as possible. The basements of modern office buildings are generally crowded and the space assigned to the mechanical equipment is limited and owing to the necessity of ventilation, the vacuum producer is generally located near the outside of the building.

Probably the best location that could be obtained in this case would be at “d” (Fig. 62). The length of piping to risers 1 and 2 would now be the same as that to all risers in case ofFig. 61, but the distance to risers 3 and 4 will be increased 50 ft. It will be possible to increase the size of the pipe line “bd” to the maximum size to serve four sweepers, or 3¹⁄₂ in., the risers and their branches to remain 2¹⁄₂ in.

The total friction loss to risers 1 and 2 will now be:

When 1-in. hose is used the density of the air entering the 2¹⁄₂-in. pipe is equivalent to a vacuum of 10 in. mercury and the friction loss in the 2¹⁄₂-in. pipe will be 1.9 in. mercury. When 1¹⁄₄-in. hose is used, the density of the air entering the pipe will be equivalent to a vacuum of 6-in. mercury and the friction loss in the 2¹⁄₂-in. pipe will be 1.32 in. mercury.

FIG. 62. ELEVATION OF LAYOUT FOR OFFICE BUILDING, SHOWING BEST LOCATION (AT D) FOR VACUUM PRODUCER.

The density of the air entering the 3¹⁄₂-in. pipe, “bd,” will be equivalent to a vacuum of 11.9 in. mercury when 1-in. hose is used, and to 7.32 in. mercury when 1¹⁄₄-in. hose is used. The friction loss in the 3¹⁄₂-in. pipe will be 0.31 and 0.23 in.mercury, respectively. Total friction loss to inlets on risers 1 and 2 will be 2.21 in. with 1-in. hose in use, and 1.55 in. with 1¹⁄₄-in. hose.

To obtain the friction loss to inlets on risers 3 and 4 the friction loss in the pipe “bc” must be added to the above figures. With 50 ft. of 3¹⁄₂-in. pipe carrying 280 cu. ft. free air the friction loss is 0.6 in. when the vacuum in the pipe is 12 in. and 0.4 when the vacuum in the pipe is 8 in.

The total loss of vacuum to inlets on risers 3 and 4 will be 2.91 in. if 1-in. hose is used and 1.95 in. if 1¹⁄₄-in. hose is used. In this case, the total loss from inlet to vacuum producer is approximately equal to the maximum variation of vacuum permitted at sweeper outlets when 1¹⁄₄-in. hose is used, but is greater than when 1-in. hose is used.

However, it is the variation in vacuum at the hose cock farthest from and that nearest to the vacuum producer that fixes the maximum variation allowable. In this case it will be the difference in vacuum between an inlet on riser 1 or 2 and a similar inlet on riser 3 or 4. The difference in vacuum at the bases of these risers will be the friction loss in the pipe “bc,” and the total difference in friction in the risers will occur when one sweeper is attached to the lowest inlet on one riser, and one sweeper on the eighth and one on the seventh floor on the other riser. The friction loss in the riser having the two sweepers attached to its upper inlets will be:

15 ft. of 2¹⁄₂-in. pipe from seventh to eighth floors, 70 cu. ft. of free air per minute, or 0.051 in. with a density equivalent to 6-in. vacuum, and 0.075 in. with a density equivalent to 10-in. vacuum.

85 ft. of 2¹⁄₂-in. pipe from first to seventh floors, 140 cu. ft. free air per minute, or 0.25 in. with a density equivalent to 6-in. vacuum, and 0.42 in. with a density equivalent to 10-in. vacuum.

The total difference in vacuum at the hose cocks will be:

0.051 + 0.25 + 0.4 = 0.7 in. with 6-in. vacuum at the hose cock.

0.075 + 0.42 + 0.6 = 1.15 in. with 10-in. vacuum at the hose cock.

Either of these values are well within the maximum variation. It is, therefore, evident that when the vacuum producercannot be centrally located that a piping system which will give the most nearly equal length of pipe to each riser will yield the best results.

A vacuum cleaning system for serving a passenger car storage yard will best illustrate the effect of long lines of piping. A typical yard having 8 tracks, each of sufficient length to accommodate 10 cars, is shown inFig. 63. The vacuum producer in this case is located at the side of the yard at one end, which is not an unusual condition.

FIG. 63. VACUUM CLEANING LAYOUT FOR A PASSENGER CAR STORAGE YARD.

The capacity of this yard will be 80 cars which must generally be cleaned between the hours of midnight and 6 A. M., or a period of 6 hours for cleaning.

It will require one operator approximately 20 minutes to thoroughly clean the floor of one car, on account of the difficulty in getting under and around the seat legs. In addition to this, it is also necessary to clean the upholstery of the seats and their backs, which will require approximately 25 minutes more or 45 minutes for one operator to thoroughly clean one car. Therefore, one operator can clean 8 cars during the cleaning period and a ten-sweeper plant will be necessary to serve the yard.

One lateral cleaning pipe must be run between every pair of tracks or four laterals in all to properly reach all cars without running the hose across tracks where it might be cut in two by the shifting of trains.

Outlets should be spaced two car lengths apart in order to bring an outlet opposite the end of every second car. This will make it possible to bring the hose in through the end of the car at the door opening and clean the entire car from one end which can be done by using 100 ft. of hose. The use of double the number of outlets and 50 ft. of hose would require two attachments of the hose to clean one car resulting in a loss of time in cleaning and is not recommended.

In this case, 100 ft. of hose would be the shortest length that would be likely to be used and 60 cu. ft. of free air would be the maximum to be allowed for when using 1¹⁄₄-in. hose.

The simplest layout for a piping system to serve this yard would be that shown inFig. 63.

When the entire yard is filled with cars and the entire force of ten operators is started to clean them it would be possible to so divide them that not over three operators would be working on any one lateral and this condition will be assumed to exist. The maximum size for the laterals between the tracks will be that for three sweepers, or 3 in., and it will not be safe to use this size beyond the second inlet from the manifold, from which point to the end of the lateral it must be made 2¹⁄₂ in., the maximum size for either one or two sweepers. The total loss of pressure due to friction from the inlet at x (Fig. 63) to the separator can be readily calculated from the chart (Fig. 56) as follows:

TABLE 20.Pressure Losses from Inlet to Separator in Systemfor Cleaning Railroad Cars.

This loss will be the maximum that is possible under any condition as it is computed with three sweepers working on the three most remote inlets on laterals “xy” and “vw” and with two sweepers on laterals “tu” and “rs.” The pipes are the largest which will give a velocity of 40 ft. per second with the full load and at the density which will actually exist in the pipe lines with the vacuum maintained at the separator of 20 in. mercury in all cases, except the pipe from “s” to separator. There the size was maintained at 6 in., as it was not considered advisable to increase this on account of the reduced velocity which would occur when less than the total number of sweepers might be working.

As bare floor brushes will be used for cleaning coaches it is not considered advisable to reduce the air quantity below thatrequired by such renovators. However, when carpet renovators are used in Pullman cars and upholstery renovators are used on the cushions of both coaches and Pullmans, the air quantity will be reduced. This condition may exist at any time, also one of these carpet or upholstery renovators may be in use on one of the inlets most remote from the separator at the same time that nine floor brushes are in use on the remaining outlets. In that case a vacuum at the separator of less than 20 in. would result in a decrease in the vacuum at the inlet to which this renovator was attached. The vacuum at the separator must, therefore, be maintained at the point stated.

With such a vacuum there will be variation in the vacuum at the hose cocks of from 6 in. to 20 in. or seven times the maximum allowable variation in vacuum at the hose cocks.

If 1-in. hose be used, the maximum air quantities will be 40 cu. ft. per sweeper. If we start with a vacuum at the inlet “x” of 10-in. mercury, the vacuum at the separator will again be 20 in. and we now have a variation of 10 in. between the nearest and most remote inlet from the separator, or five times the maximum allowed.

Either of these conditions is practically prohibitive, due to:

1. The excessive power consumption at the separator. 50 H. P. in case 1¹⁄₄-in. hose is used, and 33 H. P. in case 1-in. hose is used.

2. The excessive capacity of the exhauster in order to handle the air at such low density, a displacement of 1,800 cu. ft. being necessary in case 1¹⁄₄-in. hose is used and 1,200 cu. ft. in case 1-in hose is used.

3. The great variation in the vacuum at the hose cocks which will admit the passage of so much more air through a brush renovator on an outlet close to the separator as to render useless the calculations already made, or the high vacuum at the carpet or upholstery renovators would render their operation practically impossible.

Back to Index Next