EVALUAT1ON OF BACKFLOW PREVENTION DEVICES SUITABLE FOR SOME SCHEDULE A RISKS (ED 9009) Final Report to the Department of the Environment
Report No DWI0263

Sept 1994

SUMMARY

Backflow occurs when flow in a water distribution system within premises is reversed, thus creating the possibility of drawing in contaminated water. Backflow can lead to degradation of the quality of the water supplied to other points in the premises and, in extreme cases, the public supply. The water byelaws (see byelaw 25) require a type A air gap or an interposed cistern to protect against backflow at a point of use where "backflow or backsiphonage is, or is likely to be, harmful to health by reason of a substance which is continuously or frequently present in the contaminated water". Schedule A of the Water Supply Byelaws Guide lists examples of such points of use or delivery of water, drawn from the Backsiphonage Report of 1974. The type A air gap necessitates being open to atmosphere thereby allowing the possibility of debris, bacteria etc., entering the water supply. Another drawback of air gap systems is that water supplied via a type A air gap to, for example, a food process may not necessarily be wholesome because prolonged storage can adversely affect the water quality. European standards for water supply systems, as well as utilising air gaps, include the use of certain mechanical backflow prevention devices as protection against some Schedule A risks. The UK has little experience in the use, operational requirements and reliability of these devices. This report was commissioned by the Department of the Environment to evaluate their suitability and appropriate installation, maintenance and inspection requirements.

The main objective of this work was to evaluate mechanical backflow prevention devices that are potentially suitable for some Schedule A risk categories and evaluate their performance against selected international specifications and standards. Following this work recommendations as to their suitability as an alternative to air gap type installations have been made.

Two types of mechanical device were identified as possible alternative backflow protection to a type A air gap or interposed cistern: the RPZ valve and the pipe disconnector. Both are essentially double check valves with additional 'fail safe' mechanisms to protect against and identify malfunction of the check valves. The RPZ valve has an intermediate diaphragm operated relief valve which vents to atmosphere under backsiphonage and backpressure conditions, thus preventing backflow. The pipe disconnector also consists of a double check valve mechanism and an intermediate disconnection tube. Under no-flow conditions the tube remains disconnected, providing a protective air gap between the potential pollutant and the water supply. When water is drawn off, an integral servo control valve activates both check valves and enables the disconnection tube to bridge the air gap, permitting flow through the device. Under no-flow or backpressure conditions, the spring loaded disconnection pipe retracts to reinstate the air gap.

A programme of work was conducted to select mechanical devices from around the world and international test specifications were evaluated to facilitate the development of a test protocol. Laboratory test rigs were then constructed for the evaluation of 22 mm mechanical backflow preventers. Thirteen devices were subjected to laboratory performance and endurance testing. Larger devices were evaluated in field studies over 20 months. Three separate locations were provided through the generous co-operation and assistance of Southern Water Services plc.

The main conclusions are:

• Mechanical backflow prevention devices cannot be intrinsically reliable, whereas the type A air gap is. However, the air gap/interposed cistern can introduce other operational problems and prejudice the microbiological quality of the water delivered.

• Only three of the 13 devices tested in the laboratory met the full performance requirements of the test procedure. The specification of mechanical devices and the associated testing are therefore of great importance.

• Mechanical devices must be installed correctly and thereafter require regular inspection and maintenance by trained personnel if security against backflow is to be achieved.

• All three devices installed in the field functioned reliably during the trial period. There was, however, evidence of biofilm growth on the upstream diaphragm faces of two of the devices.

• The Montout definition of classes of risk set out and to be recommended in the European Standard Working Document N663 (prEN Draft, March 1994) needs to deal adequately with the risks posed by toxic non-biological fluids e.g. pesticides, currently classified as Category 4. These could pose a higher risk than many biological fluids classified as Category 5.

The report makes the following recommendations:

• The potential operational and water quality advantages, in certain circumstances, of mechanical backflow prevention devices over conventional air gaps should be recognised and taken into account.

• The concepts of the Montout method of classification are well established and should be regarded as the likely basis for future European standards relating to backflow prevention. Effort should be directed to improving the classification of backflow risks with particular reference to the classification of highly toxic substances.

• The current draft European standard for RPZ valves could be improved by the inclusion of a test with the valves fouled, after the style of the US test ASSE.1013.

• Further consideration is needed to establish whether or not the stagnant zones of RPZ valves are a health hazard. Manufacturers should be encouraged to develop backflow devices that remove the problems of stagnant zones and offer the mechanical and hydraulic characteristics of self cleaning of major components thus enhancing protection and reliability.

• Mechanical backflow prevention devices must be regularly serviced. Experience elsewhere suggests that installation and servicing should be overseen by a competent authority.

Copies of this report may be available as an Acrobat pdf download under the 'Find Completed Research' heading on the DWI website.