Flowrates and the NSF/ANSI DWTU Standards
By Rick Andrew
Flowrate seems to be such an elementary concept that it might strike readers as odd that an entire article could be devoted to the subject. Although the concept is elementary, the implications of flowrates on POU/POE drinking water treatment are critical. In general, for a given product configuration, treatment is more effective when flowrates are lower. At some point, most products will become ineffective as the flowrate increases. It is very important to qualify performance by describing the associated performance. Additionally, consumer protection can be significantly enhanced through the use of flow restrictors to prevent flowrates from exceeding the threshold at which performance decreases to unacceptable levels.
It may also seem that measurement of flowrates is hardly worth mentioning. There can be, however, significant inaccuracy if inappropriate or improperly calibrated flow measurement devices are used. The most accurate method of flowrate measurement is a capture-and-weight technique. Here, water is collected in a vessel on a calibrated balance for a period of time, such that the flow per time then describes the flowrate.
This technique is impractical for many applications in the laboratory. Typically, rotary flow meters are used. Care must be taken to select flow meters designed for the flowrates being used for testing. If higher flowrates occur, the impeller may not be able to keep up and the resulting measured flowrate will err on the low side. If lower flowrates occur, the impeller may not move freely and the resulting measurement will be reported too low. These meters must be calibrated frequently at the flowrate to be used for testing, using capture-and-weigh or other highly accurate techniques, or accuracy may be less than acceptable.
NSF/ANSI 42 Drinking water treatment units – Aesthetic effects and NSF/ANSI 53 Drinking water treatment units – Health effects defines rated service flow for plumbed-in POU systems as follows: “For systems connected to a pressurized line, the rated service flowrate shall be equal to or less than the minimum, initial clean-system flowrate obtained during contaminant reduction testing at an inlet pressure of 410 ± 20 kPa (60 ± 3 psig) and a water temperature of 20 ± 3°C (68 ± 5°F).”
Essentially, the certifiable flowrate for these systems cannot exceed the lowest flowrate used for contaminant reduction testing. This requirement is included to ensure that these devices will be used by consumers at flowrates low enough that their contaminant reduction performance has been verified. If flowrates exceed those used for contaminant reduction testing, all bets are off, be- cause at some point of increasing flow, the system will no longer meet the criteria for performance.
NSF/ANSI 53, because it addresses contami- nant reduction for health effects, has an additional requirement that the system must have a flow con- troller: “If the performance of a system is depen- dent on a specified flowrate, an automatic fixed flowrate control shall be provided as an integral part of the system to prevent excessive flow.”
Taking into account the requirement for a flow controller, NSF/ANSI 53 further requires that contaminant reduction testing be conducted at the highest achievable flowrate with an inlet pressure of 410 ± 20 kPa (60 ± 3 psig) and a water temperature of 20 ± 3°C (68 ± 5°F). Health effects contaminant reduction tests are conducted in this manner to ensure optimal consumer protection; essentially, at 60 psig, the flow controller prevents flowrates higher than those used to confirm contaminant reduction performance in the laboratory.
Testing laboratories must consider this requirement carefully when designing test stands. This is a very important consideration that can significantly impact flowrates achieved for contaminant reduction testing, and therefore it can also have a large impact on contaminant reduction test results themselves. Test stands must be designed to not restrict flow or cause pressure drop such that the test is conducted at a lower flowrate than could be achieved with different plumbing. There are many test-stand design issues that can reduce NSF/ANSI 53 contaminant reduction test flowrates, including:
- Piping or tubing that is too small in diameter (a general rule of thumb is that it must be at least as large in diameter as the inlet to the product being tested)
- Flowmeters, cycling valves or pressure regulators that cause pressure drop
- Excessive lengths of tubing downstream of the test product
Essentially, if two different laboratories with accurate flow measurement techniques are measuring the flowrate of the same product, and one laboratory is generating flowrate measurements lower than the other, then the laboratory that is generating the lower flowrate measurements has some type of excessive pressure drop or flow restriction in their test apparatus and is not conducting the test according to NSF/ANSI 53.
Minimum service flow
NSF/ANSI 42 and NSF/ANSI 53 also specify minimum service flowrates. These specifications are intended to assure consumers acceptable flowrates from systems conforming to the standards. Minimum service flowrates are not based on any technical criteria, but rather on suitability for end use. As opposed to being measured at 60 psig, minimum service flowrates are measured at 30 psig, to further help assure adequate flows for consumers. Figure 1 summarizes minimum service flow requirements by system configuration for POU systems. POE systems have additional considerations beyond minimum service flowrates. POE systems without flow controls must be able to achieve a minimum service flow of four gpm at 30 psig, and must have no more than 15 psig pressure drop at their rated service flow. POE systems with flow controls must have a pressure drop of no more than 15 psig at four gpm flowrate, when tested with 30 psig inlet pressure.
Ultraviolet (UV) system disinfection performance is highly dependent on contact time with the UV reactor. The longer the contact time, the higher the UV dosage that is achieved. With this in mind, NSF/ANSI 55 Ultraviolet Microbiological Water Treatment Systems requires that systems must include a flow controller. Further, the standard requires that prior to disinfection performance testing, UV systems are first evaluated for maximum achievable flowrate. This involves flowrate measurement at 15 psig, 20 psig, 30 psig, 40 psig, 50 psig, 60 psig, 70 psig, 80 psig, 90 psig, 100 psig and the system’s maximum working pressure. The highest flowrate achieved is then used for disinfection testing. As with testing under NSF/ANSI 53, test-stand design is critical (such that it does not cause pressure drop or reduction in flowrate) to ensure that measured flowrates are indeed the highest achievable.
Often, long conversations about the NSF/ANSI DWTU Standards begin when someone approaches me and says: “Excuse me, but I have a simple question about NSF.” I have learned that there are very few simple questions or simple answers about the highly developed NSF/ANSI DWTU Standards. This high level of standard development is a result of the complexities of water treatment itself, and the sophisticated equipment that the POU/ POE industry has invented to perform the function so well. Even concepts such as flowrate (that may appear to be quite simple, straightforward and maybe even self-evident) can become worthy of in-depth discussions. Flowrate is important because of its impact on system contaminant reduction performance and consumer acceptance. In addition to the intricacies of measuring it and not unintentionally impeding flow through test stands, we must ensure that consumers are protected even under the highest flowrate conditions they could encounter.
About the author
Rick Andrew is the General Manager of NSF’s Drinking Water Treatment Units (POU/POE), ERS (Protocols) and Biosafety Cabinetry Programs. He has previously served as the Operations Manager, and prior to that, Technical Manager for the pro- gram. Andrew has a Bachelor’s Degree in chemistry and an MBA from the University of Michigan. He can be reached at (800) NSF-MARK or email: Andrew@nsf.org.