Plumbing, Heating and Gas Fitting Needs? Visit our Affiliate Company:

Case Studies

The following information and case studies is provided to help highlight the utmost importance related to the proper design, installation, inspection, maintenance and repair of fire protection systems and the potential costly consequences of not following these actions.

Case No. 1 - Design Considerations:

This analysis is intended to bring out the efficiency of the various primary water supplies and fire protection delivery systems. The four sources of water supply in the order of the frequency with which they are found are waterworks, gravity tank, pressure tank and automatic fire sprinkler pump. A glance at a fire protection system shows that the automatic fire sprinkler pump was comparatively seldom found, while the waterworks was the source of supply in a little over one half of the fire protection systems. A further study of any fire prevention system will show the relative water pressures found in these fire protection systems. The decidedly higher average for fire sprinkler pressure tanks should be noted and kept in mind when some of the other sprinkler head data is considered. A complete fire sprinkler system also shows the average number of sprinkler heads operating for each fire sprinkler system, and here we find a much better value for the pressure tank than for, the others, while the automatic pump is decidedly the poorest of the four. This fact seems to emphasize the necessity of giving a system depending on an automatic sprinkler system pump a more liberal water supply than is customary for others less efficient sprinkler fire protection systems.

An automatic fire sprinkler system is intended to furnish direct comparison of the different sources of water supply with respect to a fire protection systemís efficiency in the control of fires and prevention of property loss. There seems to be two important points brought out in respect to the control of fires:

(1) The automatic fire sprinkler system with a pressure tank has a decidedly higher percentage of extinguished fires and the lowest value for unsatisfactory fires;

(2) The automatic fire sprinkler system that uses a pump has had by far the largest percentage of unsatisfactory fires. There are several conditions which bring about unsatisfactory results from automatic fire sprinklers in case of fire, among which the proper maintenance of the fire protection system is probably the most important. This sprinkler pipe and sprinkler heads would seem, therefore, to indicate the greater probability of a supply from an automatic fire sprinkler pump being out of order. The reasons for the superior showing of the fire sprinkler system that uses the pressure tank are probably due to several factors, one of which may be the higher average pressure.

The lower half of the fire prevention system brings out practically the same points, when we consider the four sources of water supply, in reference to the amount of fire loss or damage. A little more than half of the fire protection and automatic sprinkler fire protection systems that use pressure tanks fires have resulted in no claim for fire loss, while the probability of a large fire loss with this kind of water supply is only seventy-five percent as great as for the average of all water supplies. The probability for a large loss with an automatic sprinkler system using a pump is over twice as great as with a pressure tank and fully one and one-half times as great as the average for all sources of the entire water supply.

Many new sprinkler heads offer a means for comparing the pressure and number of automatic fire sprinklerís opening for different fire protection systems as well as for fires extinguished, held in check, no claim, small fire loss, etc., for the same system. The first noticeable fact is that in every case the average water pressure of any fire sprinkler system is the lowest value for unsatisfactory or large loss fires. However, the pressure for such fires is not decidedly less, nor does it present such a difference, as exists between the average pressures of the various fire protection systems. The comparatively low pressure of the fire sprinkler system that uses a gravity tank is emphasized. This low average water pressure for the gravity tank fire system presumably must have been somewhat of a drawback to such a system but not as great as one might imagine, as is evidenced by the comparative results shown in the fire sprinkler headís failure. The average number of automatic fire sprinkler heads shows some interesting features. Note the tremendous increase of fire sprinkler heads opening under large loss fires and the generally higher value throughout for the automatic sprinkler system pump.

Case No. 2 -
Fire - Failed Sprinkler System:

A fire occurred on the sixth floor of this building which was occupied as a stock room of a sweater and bathing-suit manufacturer. Fire had gained considerable headway when discovered and had spread over top of wooden table among wrapping cord, burlap, paper, etc., and thence to stock shelves directly in back of table. Fire occurred at a point between non-staggered automatic fire sprinkler s where the distribution of water was obstructed by wooden partitions and shelves. The automatic fire sprinkler s in the immediate vicinity of the fire showed signs of having been painted and were unquestionably slow in operating.

This fact together with obstructions and non-staggering of automatic fire sprinkler s allowed fire to spread rapidly. The heat caused the operation of fifty-three automatic fire sprinkler heads, although the fire was confined to an area covered by four or five. Since this was the top story of the building the automatic fire sprinkler s were further hampered by low water pressure from the gravity tank supply which was only nine pounds on highest line of automatic fire sprinklers. In spite of the fact that the conditions described herein were unfavorable to positive automatic fire sprinkler control, it is of interest to note that the automatic fire sprinkler system held the fire in check and confined it to a small area until the arrival of the fire department.

Two workmen were making repairs to the piping in the engine room basement near the fly wheel. They needed artificial light and a gas torch attached to the gas burner in the basement was used. An accumulation of lint and grease in the wooden enclosure around the fly wheel became ignited and burned rapidly as there\ were no automatic fire sprinklers inside this enclosure. The lack of automatic fire sprinklers in the wooden casing of the fly wheel and the wooden belt boxing in the room directly over it together with the belts furnished sufficient fuel for the fire to make good in dry fire protection systems, particularly in the case of large loss fires. The effect of different water pressures in the two fire protection systems does not seem to have been a factor. Apparently the time element necessary for water to reach a fire through the dry pipe system is an important factor and results in a greater number of sprinkler heads having opened than in the wet system. Flames from the fire spread up through the belt room but were prevented from entering the mill proper on each floor by the operation of automatic fire sprinkler s. The main belting was destroyed while nineteen cards were wet and there was considerable damage to ring spinning frames and to about 200 frames in the first two stories. In the upper stories the counter belts driving the two upper floors of the mill passed out through the wall of the tower without protection for the openings thus formed and were a factor in causing the loss of approximately $180,000.

This fire occurred in a silk shoddy mill and was extinguished with fire pails, chemicals and three private fire hose streams. The automatic fire sprinklers were a failure. Modern 360į heads in glass covers had recently been installed in the dryers and although examination of the heads afterwards showed them to be apparently in good condition, they did not operate at the time of fire. It would appear that the high temperature solder together with the glass tops had raised the operating point too high.

Case No. 3 -
Fire - Improperly Designed Fire System:

A fire occurred in building No. 6 which was occupied entirely for coating and drying oilcloth. There was a space at the west end of each floor occupied by the coating mills and the balance of the floor was occupied as a dry room, being separated from the coating room by a frame partition. Table oilcloth was coated with a paint mixture (containing a little benzene) in the coating mills at west end of the building and then run into the dry rooms and hung in festoons about eight feet high and the length of dry rooms, or about eighty-three feet. Dry rooms were ten feet high and there was a space of ten inches between the bottom of joists and the small sticks carrying the festoons of oilcloth. There was a line of automatic fire sprinkler s over each roll of cloth, these lines being seven feet apart, and the automatic fire sprinkler s were spaced eight feet apart across joists, down the length of the dryer. The automatic fire sprinkler s were on the end of three-inch nipples to one side of the pipes and also raised up a little in order to be more out of the way of the cloth in the racks below. This placed the fire sprinklers slightly at an angle so that the distribution was somewhat interfered with and they were, therefore, only about two inches below the bottom of joists, and the bottom of the deflectors of the automatic fire sprinkler s (fire sprinkler heads were placed pendant) was only six inches above the sticks carrying the festoons of oilcloth. All the automatic fire sprinklers were 286 degree heads. All of the dry rooms on each floor of building were filled with table oilcloth, most of which had been coated and run into dryers the day before the fire occurred. Some of this oilcloth had also been varnished. The fire was discovered at 2 a.m. by the automatic fire sprinkler alarms and the head watchman immediately blew the mill whistle which assembled the men at night work. They at once got out the chemical engine and hurried toward the building which was then a mass of flames. A public fire alarm had also been sent in to the Fire Department. It was apparent that almost from the start the fire was not controlled by the automatic fire sprinklers and must have gained rapid headway even after the first heads opened. Apparently some of the automatic fire sprinklers were obstructed and although they opened early in the start of the fire they did not prevent the flames from growing beyond their control. The fact that all the automatic fire sprinklers were 286 degree heads was presumably a factor in making their operation slow. A second and third alarm was turned in at 3 a.m. and a great many fire hose streams were played on the fire. During the progress of the fire several firemen were on the roof of the one-story building adjoining on the north directing hose streams through windows of building No. 6 when, without warning, over half of the north wall on building No. 6 fell onto this roof, burying the firemen under tons of brick and crushing in the greater part of the roof. Two of the firemen were killed and several were severely injured. The sudden failure of this wall is in part attributed to the tremendous amount of water which was poured into the building and held there by the oilcloth.

The very satisfactory operation of the automatic fire doors in a large measure prevented the fire from gaining access to adjacent buildings. The fire was finally extinguished by the fire department and resulted in a very heavy loss to the building in which it originated. An examination of the ruins showed that quite a number of the automatic fire sprinkler heads in the coating room at west end of building had not opened, and on the second floor of drying rooms, where portions of the ceiling were still standing, a number of automatic fire sprinklers had not opened although the heat in all the portions noted would seem to have been intense. It had been suggested several times that 212 degree heads should be used instead of 286 degree heads.

Case No. 4
- Fire - Lack of Automatic Fire Sprinkler Protection:

Tests made on the eight-inch street main showed that a delivery of 1000 gallons per minute at the level of the plant for fire pump suction could not be had and that the capacity of this main with adequate residual pressure was approximately 500 gallons per minute. The opening of the automatic fire sprinkler heads in No. 6 and No. 4 buildings in itself nearly exhausted the capacity of this main for effective service and left little head available for yard fire hydrant streams and the automatic fire sprinklers opening subsequently. This was promptly demonstrated as soon as the fire from the impregnating building communicated to the areas of the building not protected by automatic fire sprinkler systems being the addition to No. 2 dry house and opened a large number of fire sprinkler heads in the easterly end of the latter.

The effect of automatic fire sprinkler s in the areas that had automatic fire sprinkler heads installed, namely, No. 2 and No. 6 buildings, was to practically extinguish the fires originating in these buildings with the exception of an outside blaze on a small roof structure, which was put out by a hand chemical. The automatic fire sprinklers in No. 2 dry room held the fire in check from the exposure fire to a considerable degree, although the large number which opened, including the twenty heads in No. 4 and No. 6 buildings which were left flowing, bled the system to such an extent that effective yard fire hydrant pressure was not available. The automatic fire sprinkler s in the new dry house building was of no use for when the fire reached this point. The system was bled to such an extent that no water pressure or supply was available for this section.

The total loss will amount to probably thirty percent of the sound values of the property involved. The reason for this large loss can be assigned directly to the lack of automatic fire sprinkler protection in the hazardous impregnating building and the equally hazardous dry room adjoining the No. 2 main dry house in the center of the plant. The water supply and pressure was inadequate to supply the large number of fire sprinkler heads which finally opened, although it would probably have been sufficient to have controlled a fire in the impregnating room had automatic fire sprinklers been installed in that area. The existing fire hazard in this plant is exceptionally severe and the construction of the wooden dry rooms would favor the spread of the fire even under ordinary conditions and good fire protection. The lessons to be learned from this fire are the necessity for the proper automatic fire sprinkler protection for all sections of such manufacturing plants, and the exceptional hazard of not having automatic fire sprinklers in sections exposing or adjoining automatic fire sprinklers portions of buildings of this character, class and occupancy. The necessity for proper outside fire control is also demonstrated in the inability of the fire department to shut off the automatic fire sprinklers in the new dryer building.

Another fire occurred at an oil cloth and linoleum factory. This building was four stories in height with brick walls and open finish, and joist floors. Building was equipped with automatic fire sprinkler s (286 degree sprinkler heads) installed on old standard pipe sizes. The primary water supply consisted of four gravity tanks with a total capacity of 44,700 gallons. Static pressure on automatic fire sprinklers operating was twenty-five pounds. The secondary water supply to automatic fire sprinklers was two fire steam pumps, each having a capacity of 1000 gallons.

Case No. 5
- Fire - Automated Fire Sprinkler System Response:

A fire occurred at 12.10 a. m. from lightning which struck the plant in three places at the same time, namely, the impregnating building, No. 4 treating building and No. 6 bake-oven building. The fire was immediately discovered by employees at work in the plant and an alarm was sent in from a fire alarm box located in the mill yard. Watchman was on duty with good watch-clock service. The inside fire protection consisted of water and sand pails, chemical extinguishers and hand hose, but these were of little value. Outside protection consisted of a paid company of five men with combination chemical wagon, who responded within four minutes of the time the fire alarm was received and found yard fire hydrants in use by the mill private brigade. They therefore laid one line with about 900 feet of hose from a fire hydrant located near the power house of an adjoining plant which was supplied by a four-inch branch from an eight-inch main which also supplied the private fire protection system of the fiber plant. Very little water pressure could be obtained from this line and in the meantime the private fire brigade of the plant had two and, at times, three hose streams from hydrants in mill yard. The fire pump was started, but little pressure appeared to be available. The main plant was equipped with automatic fire sprinklers which covered the greater portion of the buildings, with the exception of the impregnating building and the new dry room on the east side of No. 2 dry house, which were not equipped.

The fire in the impregnating building, which contained the tar dipping tanks, gained rapid headway and almost immediately spread to the areas that did not have automatic fire sprinklers installed; the new frame addition to the east side of No. 2 dry house, passing through numerous openings in stone division wall between this and No. 2 dry house, which contained automatic fire sprinkler heads, and involving the first dryer in this section, opening probably fifty heads. The fires which started in No. 6 and No. 4 buildings were promptly extinguished by ten automatic fire sprinkler heads at each location. A small blaze on the roof of No. 6 was extinguished with hand chemicals in addition to the automatic fire sprinklers. The fire rapidly worked into No. 2 dry house after consuming the entire areas that did not have automatic fire sprinklers installed being the addition on the east end thereof and a large number of open sprinkler heads in this location bled the entire system. Although the pump was maintained at full speed during this time no effective streams could be had, nor was the work of the automatic fire sprinklers here in any way effective other than as a check to the spread of fire in No. 2. The fire by this time had spread from the impregnating building into the loading shed on the north side of the new dry room and from this point both east and west. The roof of the east side of No. 2 dry house had fallen in, thereby assisting in bleeding the entire system. At this time two streams were laid from a 400-gallon rotary pump in pulp mill on the south side of the creek operated either by turbine wheel or electric motor. The latter power was used at this time. Two good streams were obtained from this source. They were carried across the stream on passageway and up to the roof of No. 1 press room and were used to cut off the fire at the frame wall on the east end of new tube building section, which was the limit reached in this building, but the flames swept rapidly eastward through the new dryer building and entirely consumed this section, as the only real protection available at this time was that obtained from rotary pump in the pulp mill. The fire was under control at about 4.30 a.m., after the impregnating building, the east end of No. 2 dry house, and the new dryer building had been completely destroyed.

The primary water supply to the automatic fire sprinkler and yard fire hydrant system is an eight-inch main with a normal static pressure of about 130 pounds in the street or 120 pounds at the level of the plant itself. There is an eight-inch connection from the aforesaid main to a six-inch and eight- inch loop around the plant, and from this eight-inch connection a twelve inch pump suction is taken off for a 1000-gallon steam fire pump in detached building which has an auxiliary suction to two 45,000 gallon tanks. This fire pump is connected directly into the six-inch yard fire hydrant and automatic fire sprinkler loop around the buildings, properly checked.

Case No. 6 -
Fire - Wet and Dry Fire Protection Systems:

Water supplies are analyzed here from the point of view of pressure. More than one half of the fire protection systems had a pressure of fifty pounds or over. The relative number of fire sprinkler heads opening for the different pressures does not vary widely, although the best showing is made by the highest pressures. This chart gives an opportunity to compare the results obtained through fire protection systems using many different pressures. The low pressure of less than fifteen pounds is obviously the poorest both in respect to unsatisfactory and large loss fires. However, the variation between the other groups is not very great, although in general the showing is better for the higher water pressures. The smaller percentage of no claim fires for the pressures of less than fifteen pounds is also an interesting point to be noted. Probably no question in respect to automatic fire sprinklers has been the subject of a greater difference of opinion than the relative merits of the wet and dry fire prevention system. It is hoped that various fire sprinkler heads may serve to answer some of the disputed questions. The fire sprinkler head shows the relatively small percentage of fires which occurred where there were dry pipe fire protection systems. It also brings out the fact that the average pressure in dry fire protection systems was slightly higher and that the average number of fire sprinkler heads opening was much greater, approximately 1.8 times as many as in the wet fire prevention system. The necessity for a liberal water supply in dry fire protection systems is apparent. Most fire sprinkler systems provide an opportunity for directly comparing the two fire protection systems with one another and also with the average for all fire protection systems. The showing of the dry pipe system is decidedly the poorer both from the number of large property loss and unsatisfactory fires. The probability of a large loss fire with a dry pipe system is 1.65 times as great as in the case of a wet system. It should be noted, however, that the no claim fires are appreciably greater in the dry system.

The discussion concerning both wet and dry fire protection systems given has covered only the more prominent points brought out by these systems. The entire blueprint is given in order that the engineer may make a more detailed study of the fire suppression system. There are a few points which have been pretty clearly shown and may be summarized as follows:

(1) The pressure tank fire protection system has the lowest value for large loss fires and number of fire sprinkler heads opening and unsatisfactory fires. It has the highest percentage of no claim fires.

(2) The automatic sprinkler system that uses the pump may be said to represent the other extreme from the pressure tank system. The number of unsatisfactory and large loss fires is greatest in equipments having automatic pumps.

(3) Fire protection systems having pressures under fifteen pounds show a decidedly higher percentage of large property loss and unsatisfactory fires. The showing improves slightly with an increase of pressure but there is no marked superiority between pressures of over twenty-five pounds and those under.

(4) The dry fire sprinkler system has a much higher average number of fire sprinkler heads opening than the wet fire sprinkler system and also a much poorer showing in respect to large loss and unsatisfactory fires. Where automatic fire sprinklers fail the cause is, almost invariably, clear. These fires are selected to indicate such causes, that they may be generally understood and satisfactorily met. A brick building was two stories in height, open finish, joist floors, with a basement extending under the entire length. The basement ceiling was also open-joist construction.

The fire started from a hot shaft bearing in a corner of the unexcavated and non-automatic fire sprinkler part of the basement, and spread to the floor above. The automatic fire sprinklers on the first floor opened and appeared to check the fire, but on account of the large amount of smoke they were allowed to continue operating until the water supply were exhausted. When the water gave out, the fire was still burning throughout the basement and soon the plant was almost entirely destroyed.

   Malden, MA  02198  |  Phone 781-322-0500  |