In Part 1 of this series we pointed out
that water treatment equipment for
different markets differs in design according to:
- reliability of performance
- hydraulic design
- methods of monitoring
- levels of maintenance
System reliability was regarded as being only as strong as its weakest component. Reliability also refers to the confidence level built into the engineering with respect to what level of function the equipment will perform (i.e., the purity). Here in Part 2, we will discuss the physical differences.
It should be noted that the reliability built into a piece of equipment is not solely defined by where that unit is installed. Although a residential softener does not have to produce zero hardness, it must, nonetheless, consistently and reliably produce soft water. On the other hand, there is a different level of performance needed for a residential installation to remove arsenic, uranium, radium or a host of other regulated contaminants. One should immediately recognize that such systems require a more conservative design and some level of monitoring, regular inspections and maintenance to retain proper function. In effect, they liken to the industrial philosophy. Likewise, a large, well-built steel softener for a textile mill must have a reliable function but does not necessarily require extremely low leakage. This system slips into the gray area we call commercial design. Defining residential, commercial and industrial should always include the consideration, “What happens if I fail?”
Since industrial softeners generally regenerate every day, brine efficiency becomes a big factor. One extra pound of salt per cubic foot per day can add up to 365 pounds per year and cost an extra $15-20 per year per cubic foot (cu.ft.) of resin installed. Getting the most out of a pound of salt requires efficient design. To start, capacity and brine efficiency are somewhat sensitive to flow. The slower the flow, the better the capacity; therefore, the more economical the system becomes. Most of the lab and literature values we work with were generated at flows of 1-2 gpm/cu.ft. There is a loss of about six to eight percent in capacity when increasing flow rates from these conservative values to the typical 8-12 gpm/cu.ft. designs of home softeners. Additionally, flow rates of 20-25 gpm/cu.ft. may result in capacity reductions of up to 25 percent. Although single-family homes rarely see these types of flows, apartments, motels and other multi-family establishments routinely will.
A second consideration is the flow itself. Flows of 20 gpm/sq. ft. of surface area will take a higher toll on the resin and result in higher pressure drops than will a design with a maximum flow of 10 gpm/sq.ft. Also consider the typical cycle length for an industrial system used in continuous service. At flows of five to six gpm/cu.ft., softeners can deliver capacities of around 22,000 grains/cu.ft. using seven to eight pounds of salt per regeneration. In our assumed 20 grain feed water, continuous usage would burn a cu.ft. of resin in 220 minutes… less than four hours. It takes longer than that to produce good brine.
Commercial and industrial systems should have at least a six-hour design capacity and many specify eight-hour minimums, with 12-16 hours the norm. If we do the calculations, to achieve a 12-hour capacity run with 20 grain water and a 100 gpm demand (we can assume a capacity of 24,000 grains/cu.ft. here) we will process 72,000 gallons of water and remove 1,440,000 grains of hardness. At the 24 Kgr capacity level, we would burn 60 cu.ft. of resin. Our flow/cu.ft. is then 100/60 or 1.7 gpm/cu.ft. Most industrial systems are designed to flow between one and two gpm/cu.ft.:
- to get the cycle length;
- for better brine efficiency;
- because it is easier on the equipment, which is expected to have a 20-25 year lifespan;
- since it is easier on the resin and
- to achieve a lower energy cost as a result of lower pressure drop.
Part of the hydraulic design for an industrial system deals with the distribution within the system. This is important because the difference between an efficient regeneration using brine flows of 0.25 to 0.5 gpm/cu.ft. versus typical residential small systems saves a lot of salt and a lot of money: about 10-15 percent. In order to achieve good distribution at low flows, there must be a pressure differential built in to the bottom take-off distributor during the brining process. Otherwise, there is no reason for the brine to flow to all portions of the bed equally. This minimum differential is accepted to be 0.1 psi.1 This pressure drop must be built into the distributor and not into the drain line.
Hydraulically, pressure drop (∆p) is directly proportional to the square function of the increased flow. If you run twice as fast, the ∆p goes up by factor of four. If you try to increase the flow by a differential of five, the ∆p goes up by 25 times. Designing a system for 10 gpm/cu.ft. flow with only 5 psi ∆p across the bottom collector dictates that the system be regenerated at a flow rate equal to the square root of 50 (arrived at by dividing the desired pressure drop in service by the assumed drop during brine flow: (5.0/0.1 = 50). The square root of 50 is approximately 7. This means that if we try to minimize the ∆p to gain a higher flow rating, we must increase the flow rate of the brine draw. In this case, to 1.4 gpm (10/7 = 1.4). If we are also trying to meet efficiency numbers of 3,350 grains/lb. of salt by turning the dosage down to 5 lbs./cu.ft., we will complete the brine cycle in only about 3.6 minutes. (Five pounds of salt as a 10 percent brine is about five gallons). 3.6 minutes? No, it takes at least 10 minutes to remove hardness from spent resin during regeneration. Lengthening the brine cycle by using a smaller injector may result in brine channeling at the lower flow rate and poor regeneration. Keeping our 7:1 service-to-brine flow ratio and trying to flow at 0.5 gpm during brining tells us that the ∆p will really start to increase when we flow beyond 3.5 gpm in service. This dilemma also explains the more conservative design of commercial and industrial systems and why residential systems do not deliver dead soft water. In order to achieve greater flow multiples between brine draw and service, many industrial systems use separate bottom distributors for service flow and brine collection (see Figure 1).
Brine injected into the head of a softener with a top-mounted valve will suffer dilution as it trickles down though the freeboard space. This dilution hurts brine efficiency. Industrial systems can employ a number of schemes to compensate: they can partially drain the bed to reduce the dilution; they can use a separate injection distributor located just above the resin bed to avoid dilution and/or they can pump a more concentrated brine into the top of the unit and let it dilute to a proper concentration (10 to 12 percent) on the way down. Most residential systems rely on injectors that cannot do this.
Another option for industrial systems is the counter-flow or up-flow regeneration. In order to utilize up-flow, the bed must be held in place by a blocking flow, or have a separate take-off collector (lateral) just above the bed, or have some other arrangement to prevent the bed from moving. Allowing the bed to float is actually less efficient than co-flow (down-flow). Brine efficiency is all about cost and necessity. To achieve the typical tenths of parts per million hardness leakages it is often necessary for even the industrial softener to operate at fairly high brine levels (in the range of 15 lbs. per cu.ft.). The differences between a well-designed and operated system and one with 20 percent less brine efficiency can be three lbs. of salt (up to 2-3 times per day depending on an 8- or 12-hour cycle). This translates to an added salt consumption of 2,500 lbs. per year per cu.ft. of resin which can cost the buyer an extra $100 per year per cu.ft. of resin. This is why industrial equipment users have engineers writing performance specs and they expect their suppliers to conform to those specification—or else! It is a different mindset. The consequences of failure are simply at a different level.
The newer residential and commercial top-mounted systems with electronic controls can meter the total flow and give readouts for instantaneous flow rates that can be extremely useful. They also facilitate the initial programming for regeneration frequency and total brine. This makes changing the settings very easy if anyone really bothers to check. However, for the user, these features are just bells and whistles. Most owners would not dare to fiddle with them. It is easier to pick up the phone and call for service. There is no readout on the level of hardness coming in or the leakage leaving… and there is no need. If a problem develops, such as running out of salt, it will go unnoticed until a problem shows up days or weeks later. Industrial equipment usually monitors the total flow, hardness leakage and breakthrough through the use of inline monitors. Regenerations can be made automatic on hardness break rather than a fixed volume throughput. Daily checks are logged, as are the salt levels and usage rates. If something doesn’t compute a crew is alerted to look into it and make the necessary adjustments.
Critical industrial installations also keep comprehensive records on the condition of the resin, with periodic analyses on resin moisture content and possible resin fouling. If the chlorine content is too high, many users will install GAC systems or use chemical feed to reduce the chlorine to protect the resin; neither is typical of residential or commercial users. In addition, the influent water is tested to detect changes in the feed water hardness levels and adjustments are made to ensure complete softening. Records are tracked for historical and cyclical changes. The users are more tuned in to the system performance because of the consequences of failure. Industrial systems are operated actively rather than passively, on an hour-by-hour, day-by-day basis. The ability to monitor the system must be included within the design.
Over the lifetime use of a piece of industrial equipment, many components such as switches are used thousands of times. It is imperative that every component function properly. Meters and monitors have to be periodically recalibrated. If a lab test indicates fouling, anti-fouling remedies are performed to right the system. Tanks might be relined, brine tanks cleaned and piping replaced for safety as well as performance. Maintenance is planned to prevent Murphy’s Law from shutting things down at an inopportune time.
Many industrial systems are built to last 20 years or more. During their lifetime, such systems may have the resin and control valves completely replaced periodically to provide the same level of performance as the day they were installed. While commercial systems also need repairs and replacements from time to time, these are usually done when they break rather than during preventative maintenance discoveries. The same can be said of residential systems (see Table 1).
The above chart gives some guide lines for the design and operation of residential, commercial and industrial systems. Remember, it is not the location that dictates the design criteria, it is the risk of failure. Think of these systems as akin to the automobile, a passenger train and an aircraft: then ask yourself, “What happens if the fuel pump fails?” Do you pull over and take a look? Do you inconvenience a bunch of people and risk being hit from behind? Or do you crash and burn?
The differences in the designs of residential, commercial and industrial equipment do not lie in the size of the tank but in the size of the risk should the system fail. If exacting performance must be provided, then exacting design must be employed and maintained. If the system absolutely must work, it must be backed up with a duplicate. The quality of materials of construction must be on a par with the longevity of the system. Knowing that the system is NOT working can be as important as knowing when it IS working. As a licensed aircraft pilot, I have learned to appreciate maintenance and reliability. Nonetheless, I still don’t change the oil in my car every 3,000 miles because there is a different risk level.
The differences between residential, commercial and industrial businesses go beyond equipment design. These markets are sold with different levels of skill, serviced with different levels of staffing, represented with different levels of guarantee and operated with different levels of confidence. Buyers are subconsciously aware of the differences and resist buying what they may not believe in or feel comfortable with. It is just as hard for the industrial filter manufacturer to slip into the residential market as vice versa. It’s more about mindset. Good luck and good growth.
About the author
C.F. “Chubb” Michaud is the CEO and technical director of Systematix Company, of Buena Park, CA, which he founded in 1982. An active member of the Water Quality Association, Michaud chaired the Ion Exchange Task Force (1999-2001) and currently chairs the Commercial/Industrial Section (since 2001). He is a Certified Water Specialist Level VI. He has served on the board of directors of the Pacific WQA since 2001 and chairs its technical committee. He was a founding member of (and continues to serve on) the technical review committee for WC&P and has authored or presented over 100 technical publications and papers. He can be reached at Systematix Inc., 6902 Aragon Circle, Buena Park CA 90620, telephone (714) 522-5453 or via email at [email protected]