By Rick Andrew

The complex landscape of international product standards continues to evolve. Those who cringe at the very thought of even more product standards to learn and demonstrate conformance to may wish to quickly imagine yourself in your happy place, begin chanting your favorite mantra and quickly turn to another article. For the brave souls who crave the masochistic pleasure of learning more, read on!

The last Water Matters column reviewed certification requirements, scope, material safety and structural integrity requirements for a new series of European standards for point of use (POU) and point of entry (POE) devices. This article picks up from there to review the contaminant reduction requirements of these standards.

Technology-specific test methods
Like the NSF/ANSI DWTU standards, the European standards align contaminant reduction protocols with applicable technologies. For a table of applicable claims by technology, see Figure 1. Let’s examine some of these requirements in more detail, starting with mechanical filtration. EN13443-1 and EN13443-2 address particulate reduction in the 80 µm to 150 µm range and the one µm to 80 µm range, respectively. EN13443-2 has a similar range of particle size to NSF/ANSI 42 and EN13443-1 addresses performance of mechanical filtration of particles larger than those addressed by NSF/ANSI 42.

EN13443-2 requires that the device must reduce 99.8 percent of particles at the manufacturer’s claimed particle rating, different from the 85 percent reduction required by NSF/ANSI 42. The test is conducted with either ISO medium (one µm to 25 µm particle rating) or ISO coarse (25 µm to 80 µm particle rating) test dust. This does not exactly align with NSF/ANSI 42, which uses three different test dusts to span the range of 0.5 µm to 80 µm particle ratings.

Testing to EN13443-2 is done under constant flow conditions, with on-line particle counting to determine filtration efficiency. This is very different from NSF/ANSI 42, which requires on/off cycling and collection of grab samples of effluent at the beginning of the ‘on’ cycle.

EN13443-2 also requires measurement of clean system pressure drop, which is limited to the manufacturer’s declared value, measurement of particle retention capacity (i.e., loading), which must be equal to or greater than the manufacturer’s claim and resistance of the cartridge to collapse. NSF/ANSI 42 requires a clean system pressure drop no greater than 15 psig at the manufacturer’s rated service flow for POE systems only. NSF/ANSI 42 does not address loading and addresses cartridge collapse only through the particulate reduction test protocol.

EN13443-1 applies to reduction of particles larger than those in EN13443-2 and NSF/ANSI 42. The protocol uses spherical glass beads sieved into ranges of particle sizes. A maximum and minimum filter rating is determined by the particle sizes at which 90 percent and 10 percent, respectively, of glass beads are captured by the filter system.

Membrane systems
EN14652 includes requirements for POU and POE reverse osmosis (RO), microfiltration, nanofiltration and ultrafiltration systems. This is a departure from NSF/ANSI 58, which includes only POU RO. While NSF/ANSI 58 requires a minimum 75 percent reduction of total dissolved solids (TDS) as 750mg/L sodium chloride, EN14652 requires that systems perform per manufacturer’s claims with reduction of substances appropriate to the technology, as can be seen in Figure 2.

Also, under EN14652, pressurized permeate storage tanks are removed prior to testing, unlike NSF/ANSI 58 which requires the pressurized storage tank to be left in place during testing. Under both standards, daily production rate and recovery rate must be consistent with manufacturers’ claims. However, under EN14652, these measurements are taken without the pressurized storage tank in place. NSF/ANSI 58 requires a specific protocol that incorporates the impact of the permeate storage tank for measuring daily production rate.

Cation exchange water softeners
Requirements for water softeners under EN14743 are similar to requirements under NSF/ANSI 44. Pressure drop requirements are nearly identical, limited to 15 psig under NSF/ANSI 44 and 14.5 psig under EN14743, at the nominal flow rate.

Capacity testing under EN14743 is a bit different from NSF/ANSI 44. The brine delivery system accuracy is evaluated simultaneously with capacity testing, as the system is operated normally when testing capacity. Fifteen conditioning runs are conducted prior to measuring capacity. Capacity is measured over five runs, with variable influent pressure and a flow rate of at least 30 percent of the manufacturer’s nominal flow rate.

This is in contrast with NSF/ANSI 44, where capacity is measured at 50 percent of the manufacturer’s nominal flow rate over a series of three runs at a constant inlet pressure of 35 ± 5 psig. Measured saturated brine is directly introduced to the softener, with accuracy of the brine delivery system being measured under a separate test.

One important point of difference is that there are no conformance-by-calculation procedures included in EN14743. Any certification of non-tested softeners based on similarity to tested softeners would be subject to justification by the certifier.

Active media systems
EN14898 contains protocols for testing reduction of organic contaminants, metals, chlorine and taste and odor by active media filters. Similar to NSF/ANSI 42 and 53, it requires testing over the life of the filter with on/off cycling. It has significant differences from NSF/ANSI 53, however. These differences, with respect to testing reduction of organic contaminants, can be seen in Figure 3.

EN14898 includes requirements for testing reduction of lead, copper and aluminum, but no other metals. Testing is conducted at pH 6.5 and pH 8.5, similar to NSF/ANSI 53. However, like organics, testing is conducted only to the manufacturer’s rated capacity.

Ultraviolet (UV) systems
EN14897, like NSF/ANSI 55, requires a UV dosage of 40 mJ/cm2 for systems making disinfection claims (known as ‘Class A’ systems under NSF/ANSI 55). However, there are also significant differences in the two standards. The test organisms are different, with NSF/ANSI 55 using MS-2 bacteriophage and EN14897 using Bacillus subtilis.

The biggest difference between the standards relates to monitor and flow control requirements. NSF/ANSI 55 requires systems to have a UV sensor to monitor UV intensity through the water pathway that triggers an alarm if intensity is too low and it also requires a flow controller to make sure flow rate does not exceed that at which effective dosage will be achieved.

EN14897 has extensive requirements for the electronic output of the system, such that system performance can be evaluated over an effective operational range limited by UV transmissivity of the water and the flow rate. The system can be allowed to flow at a higher rate if the transmissivity of the water is high.

These differing requirements are manifested in the test protocols. NSF/ANSI 55 requires testing at the maximum achievable flow rate of the system, with UV transmissivity of the water reduced to 70 percent or the system alarm set point, whichever is lower. EN14897 requires measuring irradiance throughout a range of UV transmissivity of the water, variation in flow rate and variation in lamp output, to determine the effective operational range.

EN14897 also requires a dosage of 40 mJ/cm2 for bactericidal systems, whereas NSF/ANSI 55 requires a dosage of 16 mJ/cm2 for such systems (defined as ‘Class B’).

Final thoughts, sources for additional information and a certification option
Fellow masochists, you have now survived an exploration of the contaminant reduction test procedures of the new CEN standards and a comparison to the NSF/ANSI DWTU standards. CEN Committee TC 164, Working Group 13 spent several years developing these new European standards. They took the time to consider and understand the NSF/ANSI DWTU standards as a part of their process. They appreciated the technical expertise and world leadership provided by the NSF Joint Committee on Drinking Water Treatment Units in development of the NSF/ANSI DWTU standards.

But they also modified them, either due to specific technical concerns or for the sake of consistency with other standards, or for other reasons. As a result, the European standards have significant similarities with, but considerable differences from, the NSF/ANSI DWTU standards.

This scenario presented NSF with the opportunity to develop a guide comparing the two sets of standards. In fact, Figure 1 presented in this article is copied directly from that guide. The NSF Guide to the European Standards is available for free downloading on the NSF web site, at Please take the opportunity to obtain your copy today.

NSF is also pleased to announce a partnership with French certifier CSTB. This exciting new partnership provides NSF clients with an opportunity to certify with the NF mark, a premier consumer product certification mark in France that is recognized and respected throughout Europe. Please see NSF’s web site for additional information or contact me directly.

About the author
Rick Andrew is the Operations Manager of the NSF Drinking Water Treatment Units Program. Prior to joining NSF, his previous experience was in the area of analytical and environmental chemistry consulting. Andrew has a bachelor’s degree in chemistry and an MBA from the University of Michigan. He can be reached at 1-800-NSF-MARK or email: [email protected].


Comments are closed.