By C.F. Chubb Michaud CWS-VI

During my tenure as Chair of the WQA Industrial Section, we spent countless hours in hot debate as to the definition of an ‘industrial’ system. Many will argue that size (above a certain diameter) constitutes the birth of industrial. Others say it has to do with the size of the pipe and still others will embrace the complexity as the criteria…in that is uses more than one technology.

While industrial water treatment equipment is usually large and complex, I prefer to think of industrial treatment more philosophically; it may be neither large nor complex. The key is in the system’s design for reliability. If it absolutely has to work, it’s probably an industrial design.

There is little argument as to what constitutes a ‘residential’ piece of equipment. But, perhaps, there should be.

Residential equipment
We seem to think that if it goes into someone’s home, it is ‘residential.’ Residential equipment tends to be of the ‘one-size-fits-all’ category. That concept worked well when all we had to sell was a water softener or, perhaps, an under-the-sink RO. What was the worst that could happen if the system didn’t work properly? We got spots on the stemware and cloudy ice.

As a result, these devices took on a commodity envelope and were sold, installed and rarely checked on to see if they actually worked. The owners probably lost the manual after the first spring-cleaning. Their only reminder to ‘service’ the unit occurred when the spots could not be removed from the shower door.

These devices had no alarms to tell the owners that the power had gone off and the timer needed to be reset. They had no salt level monitors to tell the owner he/she had to add salt. These devices had no pressure gauges that said, ‘I’m plugged.’ They were purely owner controlled and maintained appliances—just another piece of kitchen equipment.

But is this still adequate? With the ‘ruralization’ of America, more homeowners are opting for the country life, gaining some elbowroom by building new homes away from the city and removed from the city supplied water and sewer. They are finding chemicals stranger than chlorine and contaminants more intimidating than a little iron. Many are discovering perchlorates, arsenic, radium, uranium and more in these rural wells.

Dealers know how to treat these contaminants and gladly sell more equipment. Philosophically and morally, this is not a ‘one-size-fits-all’ type of sale and definitely not a ‘set-it-and-forget-it’ device. You are dealing with the health and well being of people and the consequences of failure are at a different level than with a softener.

A softener used for the removal of radium or heavy metals cannot be allowed to run out of salt. Nor can a nitrate removal system. One cannot let the detection of a failure of these systems be dependent solely on the observations of the owners or users. These systems must be sold with a service contract that facilitates regular testing and maintenance.

These applications are protecting lives. The consequences of failure are dire. Anyone selling ‘life saving’ treatment equipment has both the liability and the obligation to also provide the service (or a service provider). This is not purely ‘residential’ equipment and we might want to look at how we define a ‘residential’ system.

Commercial equipment
Commercial equipment is often defined as simply a large version of residential. However, commercial equipment may also have to meet very high performance standards such as low hardness leakage, continuous (24/7) production, warnings and alerts and other instrumentation to give a higher degree of reliance for continuous and quality production. Most of these systems do not use chemical additives other than salt.

Most commercial equipment depends upon an operator to do the sampling and record readings to check on the day-to-day operation. Failure to maintain proper operation could have serious financial consequences on product quality downstream. Most ‘commercial’ equipment is straightforward. It may include media filtration, granular carbon, softening, RO, UV and portable exchange DI but rarely would it involve on-site regeneration of deionization equipment.

Commercial systems may be sized more conservatively than residential but it is often not necessary. Commercial systems may have built-in redundancy but usually this is for continuity and not strictly for reliability. A hotel, for instance, might install softeners for laundry and kitchen and may have three or four huge vessels to handle the flow.

At this point, this is commercial equipment. However, the hotel may also use this water for a moderate pressure boiler and faces serious repair bills if the boiler tubes clog with scale. The soft water requirement is not simply the less-than-one-grain level considered adequate for laundry and kitchen but must maintain <0.5 ppm of total hardness to protect the downstream system.

This part of the requirement implies monitoring, redundant equipment, daily operator checks on performance, etc., in order to avoid the consequences of failure. Note that not all of the water has to be ‘industrially produced.’ One might best consider two systems here: one commercial for the routine soft water and an industrial system for the boilers.

The original premise here was that ‘industrial’ equipment is not a matter of size and for that matter, not even complexity. Equipment becomes ‘industrial’ when it crosses a certain threshold of need for dependability and reliability.

High-tech example
I received a call a little over a year ago from a high-tech company in need of cooling water for a very exotic reactor. The water had three different specifications. One was for a low hardness 200-ppm cooling loop (feed water was around 450 TDS with 19 grains of hardness). This loop also had to be void of chlorine. The recirculation loop on this part of the reactor was 1,150 gpm.

The second requirement had to provide water for a 280-gpm loop and maintain a minimum of 10 megohm (0.1 µS/cm2 ). A third requirement was that this recirculation loop also be de-oxygenated to reduce corrosion in the loop. My initial sketch and layout is shown in Figure 1.

City water would be fed to a GAC filter which would remove organics, take out chlorine and protect the softeners, which would follow. A 5 filter would follow as pre-treatment to the RO. The RO would provide process water far too aggressive for the specification, so a bypass bleed line would be installed to blend the RO water with higher TDS softened water.

This would meet the <200 TDS soft water requirement (Storage Tank 1). RO water would then be the feed for the mixed bed deionization system that would supply the 10-megohm requirements for the high purity loop (Storage Tank 2). The high purity loop would also recirculate through mixed bed polishers and de-oxygenation vessels (strong-based anion resin operating in the bi-sulfite form) to maintain the third requirement of the specifications.

Foolproofing the system
How do we make this system foolproof? First of all, GAC and the softener would be twin units that would backwash or regenerate on a given thoughput gallonage so that there would always be one unit on line. Next and just to be safe, both GAC and softeners could supply water with a bypass, even when one unit was out of service and the other was in regeneration.

A vacuum breaker was supplied in the softener line…just in case the RO pump decided to come on with the feed supply to the system turned off. The RO unit was also equipped with a low-pressure feed cut off so it would not start without minimum pressure. Pressure gauges on the in- and out-flow of the 5µ cartridge filter would tell us when they needed a change.

We decided to let the flow output of the RO unit control flow through the DI makeup. The RO output could, in part, be easily controlled by regulating backpressure on the reject line. The RO would also go through a three-minute, fast- forward flush every time the unit started and would have a recirculation provision so some of the reject would be saved to minimize water usage. So far, so good.

At the customer’s request, the RO and softened water blend provision would be manual. The idea was that he would initially fill the storage tank with water at about 100 TDS and then make up evaporative losses with RO water only. Now we had two streams that met the needs. How do we get the system to automatically turn on and off on demand?

It was decided that a pressure switch located after the RO outlet would do the job. Each line providing make up for storage would be equipped with an electric solenoid valve controlled by a float mechanism in the respective storage tanks. If either tank demanded water, it would automatically open the solenoid valve for the respective feed line that would drop the line pressure and start the system.

Water could flow from RO to storage or could pass though the DI to storage. When the tank was full, the float would close the solenoid valve and the pressure switch would shut down the system. It was decided that in order to maintain pressure in the feed line (to keep the pressure switch in the ‘off’ mode), we would install a check valve before the switch and add a small bladder tank after it. (See Figure 1).

DI loop system
Starting up a DI system, especially a mixed bed that has sat for a while, would necessitate a quality rinse. An in-line conductivity meter would be provided to measure the mixed bed quality.

Water conducts electricity in proportion to its salt content. We measure conductivity in micro-siemens (µS/cm2), which is equal to micromhos. Water that is high in conductivity has a higher salt content than one that is low.

The opposite of conductivity is resistance and it is measured in ohms. Very high resistance (very low salt) is measured in megohms (millions) or ‘megs.’ Figure 2 shows the relationship between micromhos and salt content and it represents both the effluent from a DI system (red line) or one from an R/O (green line). The green line can also represent feed water. (Figure 2)

If the water were less than 10 megs it would be diverted to the drain. If better than 10 meg, it would be sent to the DI storage tank. Most DI quality meters are capable of doing this but the meter must be requested with the necessary outputs to control the solenoids. For this particular water requirement, the DI water-cooling loop had to be non-conductive. Hence, the 10 meg requirement. Chemical biocides could not be used. To control biological buildup in the system, the customer opted to use UV in a small secondary recirculation line external to the main system.

To maintain the water quality in the DI loop, the system would be provided with a 20-gpm recirculation loop in a secondary line. This loop would also be equipped with an oxygen scavenger resin system to keep the dissolved oxygen low for corrosion control. Since the only contaminants in the DI recirculation loop would be metal corrosion from the system itself, it was decided to offer this portion as a ‘load-and- dispose’ system to avoid a full-blown waste treatment plant.

The recirculated DI quality would be monitored with an in-line monitor that would signal an alarm if 10 meg could not be maintained and the customer would monitor dissolved oxygen via wet chemistry on a daily basis. The customer would be responsible for rebedding his recirc loop tanks and disposal of spent resin.

By way of redundancy, make up water to the non-DI loop could be supplied by softened water only in case the RO was down (providing the initial make-up was well below the 200 ppm limit). The non-DI tank is only 5,000 gallons and this has to provide for an 1150 gpm recirculation loop.

This system design would allow for soft water make-up. DI make up would be supplied with dual tanks but in a pinch, could use a small amount of RO water injected into the DI recirculation loop.

Ordering the system
After many hours of design study, this customer finally decided to go ahead and order the system. The system was delivered and the start-up was flawless. The RO delivered less than 10 ppm water and the DI delivered >15 meg. That’s always a great confidence builder for the builder and the customer alike.

The customer was very pleased with the redundancy and reliability of the design. Even the electrical control panel was split so that the system logic and timers stay alert even when the power is shut down.

What makes this an industrial system? It’s the reliability, the back up, the monitoring, total NEMA 4 enclosures, a thorough training program, an easy-to- understand and well-illustrated manual and a very compact skid. Did I mention that this system only makes up at one gpm and sits totally on a 30 by 120 inch skid?

Like I said, size doesn’t matter.

Municipal distribution
Usually we don’t think of municipal water distribution in terms of ‘industrial’ in design. However, since water needs for any given residence come out of the same pipe as does water for hospitals, dialysis centers and fire hydrants, perhaps we should.

Fortunately, most municipal water is supplied via gravity feed. Gravity never fails. That’s why many countries utilize water tanks on their roofs. They fill the tanks as catch can.

Water is not always available via pump and pipe when it is needed. But sometime during the day, they can fill their roof top tanks from pump-supplied water. Gravity takes over after that.

Water that is supplied only via pump is unreliable. A prime example of that are the devastating fires that swept Southern California in November 2008. The City of Yorba Linda, CA supplies most of its residents with a gravity fed water supply. The exception is Hidden Hills Estates, where water is supplied solely by electric and gas pumps that push water uphill to homes (and hydrants).

These are the very pumps that failed. Why? Because fire had destroyed the power supply. In addition, fire destroyed the communications network that then shut down the backup pumps. 19 of the 110 homes in the area were destroyed because the water delivery system was not failsafe.

Industrial systems have to be designed and built around a failsafe mindset. If it absolutely has to work every time or all the time, it has to incorporate an industrial design.

Let’s go back to those ‘life-saving’ systems installed by the thousands in homes across the country to protect the residents from the dangers of nitrates, perchlorates, arsenic, uranium and radium ad nauseum. The consequences of failure are pretty severe. Yet, few, if any, provide any sort of backup or redundancy. Few, if any, provide any sort of automatic monitoring. And few, if any, are regularly checked by the seller or installer to confirm they are even working.

These are not ‘residential’ systems. Their designs have to fall into the same reliability category as a prescription drug.

About the author
C.F. ‘Chubb’ Michaud is the CEO and Technical Director of Systematix Company, Buena Park, CA, which he founded in 1982. An active member of the Water Quality Association, Michaud has been a member of its Board and of the Board of Governors and past Chair of the Commercial/Industrial Section. He is a Certified Water Specialist Level VI. He serves on the Board of Directors of the Pacific WQA (since 2001) and Chairs its Technical Committee. A founding member of WC&P’s Technical Review Committee, Michaud has authored aor 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


Comments are closed.