Foam From Aerated “Wet Water” Reduces Exposure Hazards
Oil Storage Tank Protected by Coating of “Foam” in Carbide Company Tests
A STAFF REPORT
LARGE scale fire tests of a new type of “wet water foam” applied in a new method to materially reduce the exposure hazard of storage tanks to fire, were conducted on May 6-7 by the Process Safety Department of Carbide and Carbon Chemicals Company, a Division of Union Carbide and Carbon Corporation, South Charleston, W. Va. The tests, which were undertaken by the Fire Research Laboratory of the Company, were witnessed by some 70 representatives of the fire service, National Board of Fire Underwriters, Naational Fire Protection Association, American petroleum Institute, and fire equipment and fire insurance companies, as well as larger chemical and petroleum manufacturing companies. About the same number of members of Carbide and Carbon Chemicals, which produces the test-product “Unox” and the Unox Company, a Division of McFarland Manufacturing Corporation, Houston, Tex., which distributes it, handled the numerous details of the field and laboratory tests. Operations were directed by James J. Duggan, superintendent of fire protection for the Corporation. Aiding him were Charles Gilmore, Fred Fisher and Walter Baker.
Since December, 1946, the investigation of wetting agents as an aid to fire extinguishment and/or control has been undertaken by the Fire Research Laboratory. The net result was a formula known and commercially marketed as Unox penetrants 1 and 2. These products have been applied to the extinguishment of fires in both Class A and B combustibles and have had wide commercial acceptance as a penetrant.
As investigation progressed, it was found that by aerating the standard commercial penetrant, an unstable foam was formed which very materially increased the efficiency both as an extinguishing and insulating medium. It was to implement these experiments with the new agent, particularly as an insulator against high temperatures, that the May tests were conducted.
The term “unstable” as referred to above, is used as compared to the orthodox commercially available chemical and mechanical foam. These foams usually contain a stabilizing ingredient, form a heavy viscous mat or blanket which is impervious to vapor flow for a period of time, and leave a residue after the water has drained out. The “wet water foam” returns to the original liquid state when break-down occurs, leaving no residue. The water contained in the product solution is sufficient to absorb most of the heat of an exposure fire, it was disclosed by the tests. The foam, if it may be called that, is opaque to radiant heat and is sufficiently stable to extinguish fire in any flammable liquid that is water insoluble, and can be stored at atmospheric pressure and temperature.
The “foam” is a fluid insulation, the efficiency of which is due to its cellular structure, consisting of myriad minute air bubbles.
On solids, the extinguishing-action of the foam is three-dimensional: (1) the burning material is blanketed or isolated from air by the foam mat or blanket;
(2) the breakdown into water is in direct proportion to the heat to which it is exposed, thereby affording an optimum application rate with sufficient cooling action to extinguish the fire;
(3) through its ability to reduce the surface tension of water, its penetrating qualities tend to cause extinguishment, even on surfaces that are water repellent.
The May tests were intended to provide data for the application of the new wet water foam in large scale systems for the protection of structural members in the plants of the Carbide and Carbon Chemicals Company. It was hoped that the burning tests would provide valuable heat input data by carefully recording temperatures, volumes, etc. Equipment manufacturers were invited to submit their standard equipment which could be evaluated for this purpose, and their representatives, as well as leading authorities in the field of fire protection, were welcomed.
The major tests were conducted on the Fire Research Laboratory grounds, using a regulation 55,000-gallon storage tank, situated in a diked area, to which kerosene (as the burning solvent) was piped under pressure. Three major tests were conducted on this tank, the first two designed to try the efficacy as an insulating and extinguishing agent of 1% and 2% foam solution, and the final test to develop the heat input of the tank contents when subjected to the same type of severe fire, but without any cooling or insulating protection. A series of tests using the foam product in varying solutions was held in a shallow 25-ft. diameter tank, with standard “C” foam chamber, with and without Moeller Tube. In still another out-of-doors test, water spray as well as foam was discharged by sprinklers onto a hot open pit fire.
A number of indoor laboratory experiments were made to demonstrate the manner in which the foam breaks down into tiny bubbles and permeates kerosene: also to illustrate its penetration as an extinguishing agent.
In the first test a dike fire was extinguished with 2% Unox foam by application at one point in center top of a 20 ft. diameter tank. Total dike area: 250 sq. ft.; runoff diked area 380 sq. ft.; net burning area, 2,120 sq. ft. The fuel was kerosene; volume fuel in diked area, 660 gal.; depth of fuel (maintained by pumping), 1/2-in.
The extinguishing agent (2% by volume) was applied at the rate of 0.1 gpm/s.f. of net burning area. Foam expansion ratio determined by previous lab. test was 9.6; extinguishing medium flow rate. 212 gpm.; pressure at inlet (top of tank) 64 psi; orifice diameter in foam maker, 1.251 in.
Pre-burn time was 0.1 min.; foam application stopped at 8 min. 26 sec. Total extinguishment time. 7 min. 26 sec. Total volume extinguishing medium used, 1,600 gal.; total volume concentrate, 26 gal. Total application time at 2% was 4 min. 40 sec. Total extinguishing time, using. 1% solution was 2 min. 26 sec.
Wind at the time was NW 11-15 mph. with gusts up to 30 mph. It was noted that the overflow of the wet water foam first spread out to leeward over the diked burning surface, and that extinguishment of the burning surface first commenced in this area, then extended around the tank working to windward. Some effect of wind was noted on the windward side where drops of foam liquid were thrown into the air. Notwithstanding this fact, there were no areas of the tank left uncovered by the flowing foam.
The second test was the same as No. 1 except that 1% solution was applied. Data on the tank is as follows: diameter, 20 ft.; height 14 1/4 ft.; shell thickness 1/4-in.; nominal capacity 55,000 gal.; total exposed surface area (side and top), 1,829 sq. ft.; surface area wetted by contents, 1,514 sq. ft.; height of water level in tank. 24 1/4 ft.; volume of water in tank, 56,642 gal.; weight of water in tank, 472,394 lbs.
In the opinion of Mr. Duggan, the 1% solution did better than the 2%. Intense heat radiation was noted by the observers and protective sprays and hose lines were employed to cover adjoining property. Temperature rise was taken by instruments surrounding the tank. After 35 min. application in this test, the solution was changed to 2% to note the difference in coverage of tank, and the extinguishing action in the dike. Heat input ratios in all these tests varied but little between the 1% and 2%. indicating the protective cooling qualities of the compound. There was apparently no noticeable difference in the extinguishing action on the diked burning surface. Windage had some effect on this detail. Some observers were of the opinion that the 2% provided the better blanketing action” in extinguishment, and the heavier solution was somewhat less affected by the strong wind.
It was noticed that when the fuel supply was cut off, and the fire allowed to burn out, there was marked boiling action of the dike contents. In spite of this the 1% built up a fairly good blanket in places.
In this test the final average temperature of tank contents was 29 deg. C: total average temperature rise of tank contents, 5.3 deg. C. Average percentage of tank exposed to fire, approx. 50%. Total fuel used, 4,935 gal.; Average area of tank covered by solution, 100%; total volume of water used as cooling medium, 6,300 gal.; total volume of concentrate, 63 gal.
In the final test, the same type fire was built in the dike but no protective coolants were applied. This tost was intended to determine the heat input to the contents of the tank and the destructive effects of fire on the tank itself. The delivery funnel was removed from the tank top.
Temperature readings were taken both at the top of the tank itself, and in the drums around the installation (these drums contained water). It was noted that the vapor temperature rise was rapid and before the test was completed, it went completely off the recording instrument. Rate of water temperature rise was slower. There was no explosion, nor was the top of the tank apparently injured.
Three other tests were conducted to demonstrate the application of the wetwater foam using foam chambers and Moeller tube. These were held in a fire pan, 25 ft. in diameter (491 sq. ft.) into which fuel was pumped.
In the first of these tests, 2% solution was used through a standard “C” foam chamber installed centrally above a 3 ft. deep sump in the fuel pan. The fuel was kerosene. After 1 :10 min. preburn, foam was applied and the fire extinguished in 1 min. 24 sec. The depth of the foam blanket was 1 1/2–in. plus. Total extinguishing medium used 105 gal.
Another test was conducted using this same chamber and the fire extinguished in 1 min. 28 sec. The next test involved application through a standard foam chamber equipped with a Moeller tube. Approximately 500 gallons of kerosene were ignited. About the same pre-burn was allowed, and this fire extinguished in 1 min. 56 sec. Extinguishing medium used was 108 gal.
The possible outdoor use of this type foam on a metal pipe rack located over a pit was shown. Types of sprinkler heads were installed in the rack above this pit, one having 8 jumbo nozzles to discharge about 100 gpm of foam solution. The other system had 29 spray heads which delivered about 229 gpm at 270 psi. No fire was extinguished by the water sprays, but the effort to put out a hot pit fire with the foam solution, using only its 8 heads, was not entirely effective. The heat ignited the aluminum paint on the steel columns and the wind blew away the foam, with the result that two center columns buckled and hand fog lines had to be employed to cool down the metal work. It was disclosed that, whereas the spray sprinkler heads provided coverage for these vertical columns, the foam heads were not so located as to do so.
The technique employed for the tests was simple. The solutions of the liquid wet water were passed through a McFarland proportioner (designed by Mr. Duggan), to produce a foam about ten times the volume of the original solution. The tests demonstrated this coolant provides maximum cooling efficiency when applied to actual fires. It has high insulation and heat reflecting value, since it is white and contains water to absorb most of the heat generated by the fire.
The prescribed rate of application essentially lessens the heat input rate by about 70 per cent and proportionately should lessen the possibilty of damage to equipment and storage tanks that are exposed to fires.
Water was provided by a 500 gpm OCD portable pump through the proportioner; the foam maker was fabricated by Carbide and Carbon Chemicals Company; Foam chambers were National Foam design. Portable nozzles used in the tests were National Foam.