By Rick Andrew
The NSF/ANSI DWTU Standards include some specifications and diagrams for test apparatus or ‘rigs’ used for contaminant reduction testing of POU/POE systems. There is not, however, a lot of detail included. The Joint Committee prefers to keep these standards from being too prescriptive, both in terms of product requirements and testing lab requirements.
There are certain guiding principles, however, that must be kept in mind when designing and constructing test rigs. Unless these principles are considered and upheld, test-rig design can adversely impact test results.
These adverse effects can include inappropriate flow rates and erroneous failing or passing test results. Some of these guiding principles, why they are important and the potential effects of not considering them when designing and building test rigs are explored and discussed.
NSF/ANSI 53 requires that products whose performance depends on a specific flow rate must include flow restrictors. With this in mind, the standard also requires that contaminant reduction testing must be conducted at the highest achievable flow rate. Test rigs can prevent this highest achievable flow rate if they have flow-restriction issues.
Obvious sources of flow restriction include piping or tubing that is too small in diameter. As a general rule, piping or tubing should be of equal or greater diameter than the inlet of the product being testing.
Other, less obvious sources of flow restriction can include pressure regulators, flow meters and valves. If any of these test- rig components restrict flow, such that the product is not being tested at the highest achievable flow rate, the test-rig design is impacting testing in such a way that challenge water will have more contact time with product media.
Most common flow restrictions in test rigs occur downstream of the tested system. Flowmeters (especially rotameters), cycling valves or throttling valves can cause backpressure that the system would not normally see in use.
All of this can lead to overstatement of product performance.
Dead legs or dead spaces within test rigs must be avoided as much as possible. These areas can trap contaminants from challenge water, especially those of a particulate nature. Any of these particle traps downstream of test units are especially troublesome, because they can shed previously retained particles at some point during testing.
If shedding occurs at a sample point, the result can be false indicators of test particles in the treated effluent. In the worst-case scenario, this can result in a failing test result when, in reality, the product passed.
Slow acting valves
Mechanical filtration systems are more likely to allow particles to pass through at the initiation of flow. The standards acknowledge this by requiring sample collection to occur at the beginning of flow at initiation of the ‘on’ cycle.
If the valve on the test rig is too slow acting, then the effect of initiating flow is minimized. This can result in a less strenuous test than what is intended by the standard and, therefore, the potential for an inappropriate passing test result.
Fluctuating inlet pressure
NSF/ANSI DWTU Standards require test rigs to provide a very stable source of inlet pressure. Real-world, public water supplies have very steady pressure, as there is, in effect, a limitless supply of water provided (the only limitation being the volume and refilling rate of the system’s water tower), with no single home drawing enough water to impact the pressure.
Poorly designed test rigs, however, can cause significant fluctuations in pressure. Pressure can drop when flow is initiated. This can cause issues when testing for mechanical filtration tests, as described above. The effect can be similar to that of a slow acting valve—the impact of initiation of flow on test products is minimized because inlet pressure drops when flow is initiated.
There can be pulses of pressure when pumps turn on and off. This can cause significant effects when testing daily production rate of reverse osmosis (RO) systems.
Pulses can cause automatic shut-off valves (ASOVs) to activate sooner than they otherwise would. The effect of premature activation of ASOVs can be an overstatement of daily production rate and system efficiency.
Inappropriate mixing could be under-mixing or over-mixing, depending on the test being conducted. When testing lead reduction and/or pH 8.5 under NSF/ANSI 53, mixing is critical or the correct particle-size distribution of the particulate lead will not be achieved. Over-mixing can result in a skewed particle-size distribution with particles too large.
Conversely, under-mixing can be an issue with certain organic contaminant reduction tests. If these contaminants are not effectively mixed to form a uniform solution as the challenge water, breakthrough can occur as localized high concentrations of contaminants overwhelm specific portions of the test product’s media.
An opposite effect can also occur. Agglomerations of organics (which are poorly soluble) can reduce the effective chemical challenge into a particulate challenge.
Inappropriate materials may seem fairly obvious, but there are nuances. Certain contaminants tend to adhere to plastics, so testing in rigs utilizing plastic vessels or components becomes problematic.
Contaminants intended to be used for challenging test products can end up adhered to the test rig. They can also be released later, leading to very inconsistent influent challenge levels and cross contamination if the test rig is used for multiple tests.
Other materials tend to corrode, which can lead to multiple problems. Corrosion can flake off, which can cause unintended particulates in the challenge water.
These particles can cause premature clogging of test units or cause issues with influent challenges for mechanical filtration tests. Iron corrosion also can absorb metals onto the iron particulate in a metals challenge test and improve removal by the test device.
Test-rig design—a unique discipline
This discussion is by no means exhaustive. There are many aspects of test-rig design that can impact test results, some of them unique to very specific situations. The point is that a test rig is much more than some garden hose, a sump pump and a spigot.
A very significant amount of work goes into the design and construction of proper test rigs for testing contaminant reduction according to the NSF/ANSI DWTU Standards. If not, the test results can be suspect at best and inappropriate at worst.
About the author
Rick Andrew is Operations Manager of the NSF Drinking Water Treatment Units Program for certification of POE/POU systems and components. He enjoys leveraging his more than ten years of experience in this area to help explain the complexities and details of the NSF/ANSI DWTU Standards. 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: Andrew@nsf.org