It's All in the Family

Nov. 1, 2006
Using family bracketing to save on certification costs

About the author: Rick Andrew is the operations manager of the NSF Drinking Water Treatment Units Program. Andrew is a member of the Editorial Advisory Board of Water Quality Products. He can be reached at 800.NSF.MARK, or by e-mail at [email protected].

We’re all familiar with the story of Goldilocks and the Three Bears. “This bed is too hard. This bed is too soft. But THIS bed is just right,” Goldilocks said as she tested the accommodations chez Papa Bear, Baby Bear and Mama Bear. Just as the Bear family tended to have variety in their offerings based on the tastes and needs of the three family members, water treatment manufacturers often offer families of similar products that vary in size depending on the tastes and needs of their customers.

Water treatment product families also may be structured to include multiple systems that use the same replacement element. Yet, other product lines involve systems with varying numbers of similar treatment or filtration elements designed to operate in parallel or in series.

In all of these cases, we recognize the similarities between the products that make it logical for us to call them families. These recognized similarities, as well as techniques to evaluate the specific members of those families, make it possible to certify an entire family of products by testing only specific family members, as opposed to testing each and every product. This approach results in family bracketing. Family bracketing is beneficial to all stakeholders with an interest in certification. Manufacturers benefit because their testing and certification costs are lower, as is their time to market. Retailers, dealers and consumers benefit because these lower certification costs can be passed on to them in the form of lower prices for products in the family.

The key to certification through family bracketing is development of a sound, scientifically defensible demonstration of conformance to the applicable standards for all non-tested products in the family. NSF pioneered the application of family bracketing and uses it today for certification to every NSF/ANSI standard. Let’s examine various scenarios and the application of these principles for family bracketing decisions.

It’s All Based on the Standards

NSF/ANSI Standard 44 includes family bracketing requirements for pressure drop and softening capacity testing of cation exchange water softeners. The families are based on softeners that use the same control valve, brine system and distributor baskets. Family members must use equivalent cation exchange resin, and vary only in terms of the size of the resin tank and length of the distributor tube. Product families defined by these characteristics may be bracketed together for pressure drop and softening capacity testing. The standard specifies strict limitations in variation of regeneration cycles, salt settings, flow rates and size between family members. In other words, Papa Bear cannot be too much larger than or operate significantly differently from Mama Bear. The standard also includes mathematical formulas used to determine pressure drop and capacity values for non-tested family members.

Based on these requirements and typical softener families, it is not unusual for manufacturers to test only one Mama Bear for every 10 certified Baby Bears and Papa Bears. Obviously, this is a huge cost savings over testing every softener in the family. There are, however, tradeoffs that need to be considered. For example, certain non-tested models would likely have much better capacities if they were directly tested, as opposed to applying the calculated capacities based on other family members. They may have much deeper resin beds and so on. Experienced certifiers can add value for manufacturers by highlighting these situations in advance of testing, allowing the manufacturer to quantify the tradeoffs in cost savings vs. the opportunity to market the full potential of various models.

NSF/ANSI Standard 44 is a great example of family bracketing and guidelines based on fundamental scientific principles. The detail regarding family bracketing fits well in the standard because so many manufacturers produce families of softeners based on the same control valve.

Most other family bracketing scenarios, however, are unique to the manufacturer and its products. Because of the endless possibilities for family bracketing situations, many of them are not detailed in the standards. In these cases, family-bracketing approaches may still be utilized by applying fundamental scientific principles to decision making. Certification bodies developing these families and test models must have the necessary expertise and procedures by which this process is administered, and document their rationale when establishing certification for non-tested products.

An Example—Filters in Parallel

Manufacturers may offer filtration systems that incorporate identical water treatment elements in parallel, as described in Figure 1. Manufacturers may offer a family of systems that includes a Baby Bear with one element, a Mama Bear with two elements, and a Papa Bear with three elements in parallel. These systems can be bracketed into a family group for contaminant reduction testing purposes. There are two basic criteria to consider for this family bracketing:

  1. Each element in parallel must be identical. Different filtration elements in parallel could have different flow characteristics such as the rate at which they clog. Variable clogging rates could cause disproportionate flow to occur through each element.
  2. The system manifold must be of sufficient bore size to have little pressure drop throughout the manifold, thus ensuring equivalent flow to each element.

For families that meet these criteria, it is possible to test a single-element Baby Bear system and establish capacities and flow rates for the additional multi-element, parallel-flow Mama Bear and Papa Bear systems through calculations. An example of these calculations is presented in Figure 2.

Different Family Members for Different Tests

When examining a family of products to determine conservative test models for certification purposes, the selection of the specific model may vary depending on which requirement of certification is being considered. For example, the model determined to be the conservative test model for material extraction testing may not be the most conservative model to select for contaminant reduction testing. In addition, the use of fundamental scientific principles may lead to selection of yet another model in the family to be the most conservative model for structural integrity testing. All of these variables must be considered when developing a plan to ensure the most cost-effective approach to certification of the largest number of products.

Selecting the Correct Family Member for Material Extraction Testing

The surface area to volume ratio is the critical factor influencing the concentration of contaminants leaching from a given material during an extraction test conducted to the NSF/ANSI Drinking Water Treatment Unit (DWTU) Standards. This ratio is the surface area of the material in contact with water compared to the water holding volume of the product.

Consider a family of cation exchange resin tanks being evaluated for conformance to NSF/ANSI Standard 44. The line varies in size only, with the materials of construction being identical for each model. Given the identical materials of construction, there is a scientific basis to group these tanks together as a family for evaluation purposes.

Application of fundamental scientific principles leads to the conclusion that as the tanks get larger, the surface area of the tank liners increases as a squared function, whereas the volume of the tanks increases as a cubic function. So, the larger the tank, the lower the surface area to volume ratio of the liner material. In order to test conservatively, it is necessary to test the family member with the highest surface area to volume ratio. Following this logic leads to the conclusion that the Baby Bear, or smallest, tank will have the highest surface area to volume ratio. This means that testing the Baby Bear tank for material extraction will also establish conformance for the Mama Bear and Papa Bear tanks.

Win-Win Situation

The NSF/ANSI DWTU Standards are very rigorous in their requirements. Residential drinking water treatment systems must undergo extensive, rigorous testing and evaluation to establish conformance to these standards. Because of the number of different attributes and claims of the treatment systems, and the highly sophisticated, data-intensive tests required to properly evaluate them, the process of certifying residential drinking water treatment systems involves considerable cost to the manufacturer.

In order to minimize these testing costs and help ensure these costs are far outweighed by the associated value, family bracketing and selection of conservative test models is imperative. While conformance to the standards is of utmost importance, fundamental scientific principles can be used to limit the number of tests required to establish conformance of an entire family of products based on the testing of one or a few. The savings in testing costs can then be recognized by the manufacturer and passed on to consumers.

In order to take full advantage of the benefits of family bracketing, certifiers must partner with manufacturers to consider the entire family of products prior to initiating any testing. This is key to ensuring that the correct conservative test model is chosen. In other words, before we leap into testing and certification of Papa Bear, we should also consider Mama Bear and Baby Bear. This is important even if there are no immediate plans to certify Mama Bear and Baby Bear because plans can change.

This analysis of family members can be performed prior to any actual manufacturing of a new product line, enabling manufacturers to factor certification costs into their design considerations. NSF frequently engages in this type of analysis with manufacturers, such that test models and costs are known before prototypes even exist. The same is true when new models are being added to an already tested and certified family. The new models can be examined to see if the previously tested models are conservative and appropriate for these new models, or if not, what additional testing will be required to add the new models.

Achieving the full value of certification at the lowest testing costs possible is the ultimate win-win situation for manufacturers, dealers, retailers and consumers. As Goldilocks might say, “THIS family certification approach is just right!”

About the Author

Rick Andrew