Utility data / field results
One public utility's annual filing recorded 2,140 double check assembly tests, 2,121 passes, and 15 failures. The useful lesson is not a universal failure rate; it is why every assembly needs a documented annual test and a closed follow-up loop.
Primary keyword: double check backflow assembly test results
Key Takeaways
Annual backflow testing can feel routine when most assemblies pass. A 2024 annual summary report filed by Clackamas River Water shows why the routine still matters. The utility listed 2,573 double check backflow prevention assemblies installed in its system and recorded 2,140 as tested.
The same filing listed 2,121 assemblies as passing and 15 as failing their annual test. Those pass and fail figures add to 2,136 rather than 2,140, so the filing leaves four recorded tests outside those two totals. The report also notes that some tests were not recorded in the annual summary because of a report-submission fee issue.
That detail is important. It would be misleading to turn the filing into a precise failure-rate claim when the reported totals do not reconcile exactly. The defensible takeaway is simpler: a large annual testing program still identified failed assemblies, and the quality of the records matters alongside the field work.
A double check valve assembly uses two independently acting check valves to resist reverse flow. Those internal components are mechanical. Debris, wear, damaged seals, pressure conditions, or prior service can affect performance even when the enclosure, piping, and exterior of the device look normal.
A certified field test checks whether the assembly performs within the applicable test procedure. That is different from looking for a leak or confirming that water still flows through the property. Appearance can help identify obvious damage, but it cannot replace the measurements used to document a passing result.
The appropriate assembly type depends on the hazard and the local water supplier's requirements. A double check assembly is not interchangeable with every other backflow prevention method, and property owners should not change device type or configuration without confirming the approved protection for the connection.
California's April 2026 Cross-Connection Control Policy Handbook requires backflow prevention assemblies to be field tested at least annually. It also allows more frequent testing when a public water system, the State Water Board, or a local health agency identifies a higher-risk premise or an increased risk of assembly failure.
The handbook requires testing after installation, repair, depressurization for winterizing, or permanent relocation. When an assembly fails, the public water system must ensure repair or replacement within 30 days after notification, subject to an extension allowed through the system's approved cross-connection control plan.
For an owner, HOA, or facility manager, the annual appointment is only one checkpoint. The complete record should connect the assembly identity, test result, repair authorization, passing retest, and proof of submission to the serving water supplier. That closed loop is what prevents a field test from becoming an unresolved compliance task.
Use one record for the full testing and follow-up chain, especially at properties with several assemblies.
The serving water supplier controls the property-specific assembly, notice, and reporting requirements, but these distinctions help owners plan.
No. These figures come from one utility filing, and the filing's pass and fail counts do not add exactly to the tested total. They should not be presented as a California or national failure-rate benchmark.
California's current policy handbook requires backflow prevention assemblies to be field tested at least annually. More frequent testing may be required for higher-risk premises or assemblies at increased risk of failure.
No. The normal compliance chain includes repair or replacement, a passing retest, and submission through the applicable water supplier process. Keep all three steps tied to the original failed result.
