(832) 899-4040

NFPA 25 annual flow test using a clamp-on meter

The measurement sequence, the setup that decides whether your numbers are any good, and the mistakes that produce a plausible but wrong flow figure.

Before test day

  1. Identify the measurement point. Discharge riser, ten diameters downstream of the check valve and five upstream of the next fitting. Pump rooms are tight; find this on a survey visit, not on test morning.
  2. Record pipe data. Measure the outside diameter rather than reading it off a drawing. Measure or confirm wall thickness. Note any lining.
  3. Check surface condition. The transducers need clean metal or sound paint. Heavy rust or loose coating needs wire-brushing back at the measurement band.
  4. Confirm the test arrangement. Closed-loop test header, hose valves, or flow meter loop — the clamp-on reads flow either way, but it changes where you clamp.

Setting up on the day

  1. Clean a band of pipe roughly the sensor spacing plus 100 mm.
  2. Enter pipe outside diameter, wall thickness, material, lining and fluid into the meter. Let it compute the spacing.
  3. Apply coupling gel to the transducer faces, mount on the clamping rail at the computed spacing, and clamp firmly.
  4. Check signal strength and measured sound speed before you trust anything. Sound speed should be close to 1482 m/s for water near 20 °C. If it is not, the wall thickness entry is almost certainly wrong.
  5. Run the pump at churn for a few minutes to clear air and settle the reading.

The test sequence

  1. Churn (no flow). Record pressures. The meter should read effectively zero; if it reads a meaningful flow at churn, something is passing that should not be.
  2. 100% rated flow. Open to the rated point, let it stabilise, record flow and pressures together.
  3. 150% rated flow. Same. This is the point that determines whether the pump meets its curve.
  4. Log continuously through the whole run rather than writing spot values. The log shows stabilisation and gives you defensible data if a number is queried later.
  5. Return to churn, shut down, download the log by USB.

Errors that produce a plausible wrong answer

  • Wall thickness assumed from schedule tables. Old risers corrode; the wall may not be what the schedule says.
  • Cement lining not entered. Common on ductile iron. Shifts the effective bore and the flow with it.
  • Sensors too close to the check valve. The most frequent installation compromise in a tight pump room, and it biases the reading systematically.
  • Reading taken before the flow stabilises. Watch the trace, not the number.
  • Air in a riser that has stood for a year. Run the pump first.

What the meter does not replace

Suction and discharge pressure readings, the pump curve comparison, the controller and transfer switch checks, exercising hose valves where the AHJ requires it, and your judgement about whether the pump passed. The clamp-on meter supplies one number in the test record — an important one, obtained without dumping water, but one number.

Rent a meter for the annual test

Ultra ProLite, $699 per week, shipped overnight. Covers typical 4 to 10 inch fire pump risers.

Request a rental quote(832) 899-4040