Professional Engineeing Publication

10) Chlorine Residual Analyzers

Capital Control's chlorine residual analyzers (Series 1870E) provide state-of-art continuous, on-line, amperometric measurements of free or total residual chlorine in water. This analyzers are very helpful to measure the residual chlorine in the hot water return line of cooling tower. Also, it is possible to record the measured chlorine level via a separate recorder. For the details, refer to Bulletin A1.11870E of Capital Controls Company.

11) Cylinders and Ton Containers

(1) 100-and 150-lb Cylinders

The pertinent dimensions and tare weight of these containers are shown in below. Minor variations in these dimensions depend upon the cylinder age and the maker. The 150-lb cylinder is so popular that the 100-lb size may be considered obsolete. Some packagers have available a 35-lb cylinder, which is suitable for laboratory or test work on a small scale.

The packager fills these cylinders with liquid chlorine to approximately 85 percent of total volume; the remaining 15 percent is occupied by the chlorine gas. There must be strict adherence to these figures in order to prevent hydrostatic rupture of the cylinder in the event of abnormally high ambient temperatures. As the temperature rises, the liquid chlorine expends. Theoretically, the cylinder could get hot enough to completely fill the remaining 15 percent occupied by gas and therefore rupture the cylinder. However, the outlet valve on these cylinders is fitted with a fusible plug, the core of which will melt at approximately 158^oF, thus preventing rupture of the cylinder during instances of abnormally high temperatures. When the plug melts, the liquid chlorine discharged through the core opening (1/8" diameter) cools so rapidly that it freezes, momentarily halting the flow of liquid chlorine. By this time the danger of cylinder rupture is over, and a trained operator wearing air or oxygen breathing equipment can apply the chlorine safety kit, stop the leak, and remove the cylinder from the area.

The gross weight of a full 150-lb cylinder varies from 250 to 285 lb. Therefore these cylinders are best handled by a two-wheel cylinder hand truck. Never use slings, or try to pick up cylinders with hoisting equipment to the protective cap or valve. The water volume of a 150-lb cylinder is 14.4 gallons (1.93 cubic feet). The most important design considerations are as follows:

• Direct sunlight must never reach the cylinder.
• The maximum withdrawal rate should be limited to 40 lb/day/cylinder.
• Minimum allowable room temperature is 40^oF.
• Heat must never be applied directly to the cylinder.
• Sufficient space should be allowed in the supply area for at least one spare cylinder for each one in service.

Provisions should be made so that the operator can determine the amount of chlorine left in the supply system. This is most effectively done by weighing scales.

(2) Ton Containers

A. General Discussion

Unlike the 150-lb cylinders, either gas or liquid may be withdrawn from ton containers; consequently each container has two outlet valves. Also unlike the small cylinder, the ton containers have six fusible plugs - three in each of the dished heads. They are transported either by truck or by multiple-unit tank cars (TMU). A truck can carry a maximum of 14 containers; a TMU, 15. The water volume of a ton container is 192 gallons (25.67 cubic feet).

The gross weight of these containers (3500 lb) dictates that proper handing equipment must be used. The container is designed for use in the horizontal position. Each container must be positioned so that the outlet valves line up in the vertical before being connected to the supply system. In other words, the container must be positioned so that one eductor tube is in the gas position and the other in the liquid when the container is full.

B. Handling Equipment

Proper handling equipment includes the following:

• Two-ton capacity electric hoist
• Lifting bar
• Container trunnions
• Monorail for hoist
• Cinch straps (Where seismic forces are of concern, plastic cinch straps should be used to prevent the container from becoming dislodged from the trunnions.)

Not only must the containers be moved from transport to supply position; it is most important that each container to be connected and to be easily rotated in order to align the outlet valves vertically. This is accomplished by a pair of trunnions, which serve not only as a method of positioning the outlet valves but also for the spacing and support for each container. Further, if the trunnion is properly designed, it should function to contain the container in the event of a collision with an incoming container on the traveling hoist, and to prevent an empty container from rotating without an external force of at least 15 ft-lb. (A recent earthquake and 5.8 - 6.2 Richter force failed to move banks of ton containers at two wastewater treatment plants nearby. Plant container alignment was 90^o different from each other. Similarly, the October 1989 San Francisco quake known as the Loma Prieta 7.6 quake, did not affect ton containers on trunnions without cinch straps.)

One of the critical design dimensions for ton container installations is the distance from the bottom of the monorail to the floor of the container room. The monorail must be high enough to pick up a container off the truck and also high enough to lift one container over another that is on the floor. Usually the governing distance is the height of the truck bed above the container floor.

C. Ton Containers: Reliability and Safety Considerations

a. Design Strength: The enormous strength of ton containers can be best described by the fact that when they are hydrostatically tested at 500 psig, it is not uncommon for a 3 percent physical expansion of the container to occur without any damage whatsoever. This would easily permit a temperature of 160^oF in a "skin full" condition without any fear of rupturing. These containers also have an additional expansion factor in the dished heads, which will reverse to a convex posture before rupturing. During an investigation of nitrogen trichloride explosions at a chlor-alkali plant near Bogota, Colombia, South America, we examined more than a dozen ton containers with the dished heads reversed. None of these containers exhibited any damage from the explosions. The damage occurred in the gas header immediately down-stream from the evaporator discharge piping.

b. Metal Fatigue: Many times in the past 25 years metal fatigue in chlorine cylinders has been mentioned as a hazard to the life of ton containers. To date there has only been one known failure by container rupture, which occurred when a container owned by the U.S. Army split at the seam between the dished head and the container body. The container had been in use about 45years. This brings up the question of whether or not containers should be scrapped after a specified length of service, as now a great number of these containers, made during World War II, are more than 45 years old.

Many such containers are still in use. That they have survived this long is quite remarkable. Compared to a bulk storage tank, one of these containers undergoes countless cycles of pressurization and depressurization.

c. Fusible Plugs: Their use on these containers is a questionable practice because of the hazard quotient. Fusible lug failure is the leading cause of major leaks when ton containers are used. It is for this reason that the American Water Works Association passed a resolution in 1935 asking that these plugs be banned from ton containers. There are three fusible plugs on each of the concave container heads. The three plugs on the head where the container outlet valves are located are a major cause for concern. These valves and their associated flexible connectors prevent easy access to a leaking plug with the emergency kit. These plugs were originally incorporated in the container design as a safety pressure relief device in case of a fire in the container area. However, flammable material or combustibles have long since been banned from the storage area of any compressed gases; so the need for fusible plugs has never materialized.

After a further look into the logic of fusible plug performance, the following interesting anomaly was discovered:

• When properly filled, a ton container becomes "skin full" at 155.1^oF.
• These plugs are made with a brass body and a core that melts somewhere between 158 and 165^oF.
• To bridge the gap 3 to 10^oF, the dished heads in the container are supposed to pop out, going from concave to convex! Presumably to containers were propositionally designed to account for the above-described temperature gap.

In spite of this anomaly, the design might have worked to the user's advantage, as the container designers had to take this bizarre set of circumstances into account by making the containers much stronger than would have been required otherwise. When NCl₃ is formed during the production of liquid chlorine, it remains soluble only in the liquid form of chlorine. Therefore, at the moment of liquid exhaustion in the container, the NCl₃ is released as a vapor and explodes. In this particular case the presence of NCl₃ was due to ammonia in the electrolytic cell water.

Fusible plugs are a liability because their brass body disintegrates rapidly if there is the slightest amount of moisture in the chlorine above the allowable limit. The resulting corrosion finds its way easily through the steel of the container because of the inability of brass to resist corrosion by moist chlorine. The resulting leak is the primary cause of major chlorine release when ton containers are used. Next in severity and frequency is the failure of the flexible connectors.

If fusible plugs are inevitable because of an industry attitude, they should be limited only to the head that does not hold the outlet valves. Moreover, these plugs should be used only in the area where chlorine vapor is present when properly aligned. This would greatly reduce the leak rate. If plugs must be used on both heads, then consideration should be given to using the 150-lb cylinder outlet valves that are fitted with a fusible plug.

d. Fire Protection: If the local fire marshal insists upon additional protection case of a fire, there is good reason why sprinklers cannot be used to keep the containers cool and the pressure down during a fire. It must be remembered that the air in any closed room housing multiple containers of chlorine suffers from general corrosion, no matter how adequate the ventilation. Off-gassing is inevitable when containers are being changed. Cumulatively, these small amounts of chlorine emissions will quickly damage any modern sprinkler used for fire protection. In other words, if sprinklers were installed, they would become inoperative in a short time and be unable to release water at the time of a fire.

Spraying water on a container during a fire to keep the pressure from escalating is a perfectly safe procedure. Failure of metallic components due to chlorine corrosion always occurs from within the container when an external leak is not in progress; for example, consider the "sweating" of any container subjected to vapor withdrawal in excess of its allowable rate. Also see below, discussion of the "Arizona Blanket", under the heading "Chlorine Storage Pressure Control System."

e. Manifolding Containers: The Chlorine Institute recommends that for liquid chlorine withdrawal, the only acceptable way is to manifold the containers in accordance with their drawing. This arrangement eliminates the potential hazard of connecting a full container with a higher vapor pressure than that of the containers already "on-line." Such a situation could lead to overfilling of those containers already connected. This overfilling scenario never has been reported, nor has anymore ever seen an installation that confirms to the CI drawing at either a water or a wastewater treatment plant installation.

In manifolding ton containers for vapor withdrawal, it is obvious that the overfilling potential is nonexistent because containers of unequal pressure will automatically come to the same pressure.