In this guide, you’ll find everything you need to know to get started with water meter testing programs, including:
- Water meter measurement basics.
- What is sample testing and why it’s important?
- Regulatory compliance and background.
- Benefits of testing random samples of your meter population.
- Preparation, testing, and interpretation of the results.
- Best practices for meter population management.
- Practical case study.
Water Meter Measurement Basics.
All mechanical utility meters (water, gas, electric) will decrease in accuracy over time. Water meters from 5/8” to 2” in size leave the factory with a measurement accuracy of 100% +/- 1.5% registration as per AWWA standards. Positive displacement meters which are typically used in residential and light duty commercial settings do not over register by design and cannot be calibrated. Water meters in general tend to register less usage (slow down) as they get older. Over time mechanical water meter accuracy can be affected by three main components:
- Water turbidity.
- Water hardness.
- Water volume.
What is Sample Testing?
The practice of statistical sample testing of water meter populations has been around for quite some time and is also used in the gas and electric meter industry as well. It involves the testing of meters that are a randomly selected from the overall meter population. The batch of meters selected is generally a statistically relevant sample size such that the number of water meters chosen will properly represent the overall population of meters. Sample testing is important because the tests results of the group of meters selected will help a utility to map out a meter replacement strategy for the remainder of the metering system.
Regulatory Compliance & Background
The American Water Works Association (AWWA) is the current standard for cold potable water meter use in North America to date. Measurement Canada (MC) enacted legislation in 1985 that governs the units of measure used in Canada for the custody transfer of products. MC has the right to regulate any measuring device such as weigh scales, gas pumps and even utility meters. Currently MC regulates natural gas and electric meters in Canada and have chosen to leave water meters as status quo. This means that water utilities throughout North America continue to do what they have been doing for the last 120 years which is to self-regulate water meter sizing, selection, testing, and replacement cycles based on the AWWA standards.
Benefits of Sample Testing
The benefits of statistical meter sampling for water purveyors are:
- Provides a snapshot of the metering system health.
- Helps take the guess work out of deciding when to replace meters.
- Forms part of the ongoing meter population management plan.
Typically, water billing is the second largest component of a municipality’s cash register next to tax revenue. This makes it super important for water utilities to keep the water meters as close to a hundred percent accurate for as long as possible. As water meter accuracy declines over time so does the annual water budget revenue. In most cases a cost justification can be made to replace the water meters on new meter revenue increase alone.
Preparation, Testing, and Interpretation of the Results
Starting a dry meter on the low flow test may lead to false representation of actual low flow accuracy. It is recommended that after removal from service that water meters be capped at each end with a small amount of water left remaining in the meter. Starting at the highest flow rate helps the meter to acclimate to measurement again since the meter may have been sitting for a couple of weeks between removal of service and arrival at test facility.
Test rates vary by meter size however all meters are tested at a minimum of three flow rates high, intermediate, and low. Water demand in a home is not constant, so testing at different flow rates gives us a better representation of the meter’s overall or weighted average accuracy.
All water meters decline in accuracy over time due to water quality factors such as turbidity and hardness as well as years in service, including amount of consumption metered. Positive displacement water meters are typically designed to sustain reasonable accuracy over a 20-year period. Historical test data shows us that we can expect somewhere between 3-5% accuracy decline within this period mostly occurring in years 15-20. Since a considerable amount of residential consumption occurs at low and intermediate flow rates much of the accuracy decline throughout the meter’s installation period will occur at these rates. Meters beyond 20 years old can expect test results to show accuracy decline anywhere between 5-10%.
Best Practices for Meter Population Management
Know your water meter system! A water meter database is only as good as the quality of data entered into it. Some practices to consider when creating an overall testing plan:
- Are all your water meters the same age? If you have zones with varying age ranges, then you will have a plan of plans.
- Are they all the same size and type? Likely not, so create groups like residential, industrial/commercial/institutional (ICI) and large meters to help with priorities.
- Largest meters and largest revenue generators (not necessarily largest meter) should have priority in your meter testing and maintenance plans. A large meter or consumer that is 2% inaccurate does not have the same impact on the municipal cash register that a residential consumer has with the same inaccuracy.
- Understand your annual budget and determine if it is more conducive to the “small bite approach” or large-scale replacement programs.
- Once you have arrived at residential meter testing on your priority list choose a representative sample number from a specific area of the distribution system and have them tested.
Most statisticians will agree that 100 water meters would be the minimum sample size for any meaningful data conclusions. Many say that the maximum would be 10% of your meter population (up to 1,000 meters) would be the maximum representative sample. Going above 1,000 meters tested won’t add much to the accuracy profile you are creating and would add significantly to the meter testing budget.
Bottom line is that there is no perfect plan, start somewhere as it’s better than not starting at all. If uncertain about the best approach, contact an industry professional for a consultation. It will help you to determine the best overall fit for your requirements.
Practical Case Study
Evans Supply Limited helped the City of Port Colborne’s business case to replace 6,000 water meters twenty years after their original installation. The city removed 170 water meters from their system and testing determined that 40% of the randomly selected batch were not registering accurately. The meter test results played an important role in their business case to replace their aging water meter population.
Get Started Today
Start building your meter test database today. Whether you’re replacing a meter for a technology upgrade or any other work order, don’t scrap the meter immediately. Set it aside for batch testing. Once you’ve accumulated a box or two send it to your local test facility. Over the duration of your meter population life cycle, you will have accumulated test data that fairly represents the state of your system. This data will enable the utility staff to confidently justify their meter replacement line item in the budget. Get in contact with us today to arrange for some sample testing!