Every year, clean, readily available water becomes more and more scarce. Metering water usage is a way to make users more aware of how much water they use, which in turn will increase the desire to conserve water and to reduce their water bill. When meters are tested in their new condition, it is normally performed under ideal laboratory conditions at constant flow rates. Then when the meters are installed in the field, they often are installed in or experience non-recommended conditions that are quite different from the ideal laboratory setting. This study investigated several non-recommended conditions that can exist in a distribution system. The conditions that were simulated were endurance (the study of accuracy as a function of meter throughput), installation (the study of accuracy as a function of upstream piping and meter mounting effects), and flow profile (the study of accuracy as a result of dynamic real world flow variances over time). The meter types that were tested in this study were displacement piston, nutating disc, multi-jet, single-jet, fluidic oscillator, magnetic, and ultrasonic. When comparing the results between the meter types it was found that some meter types were more susceptible than others to the conditions that were simulated. Displacement piston and nutating disc meters had the best overall accuracy performance under the three non-recommended conditions that were simulated.