Pressure Loss in a Dead-End Loop
A distribution system serves 6,200 connections from a 500,000-gallon elevated tank. The tank floats on the system at a normal high-water level of 148 ft. A neighborhood at the far end of a 12-inch dead-end main reports low pressure at hydrant flow tests: 32 psi static, dropping to 14 psi at 900 gpm. The main is 4,800 ft long and has no looping. A new 60-unit subdivision is proposed at the end of this main. The operator also notices that during high-demand afternoons the tank drains faster than the model predicts, and a chlorine residual sample at the dead end reads 0.2 mg/L versus 0.9 mg/L at the plant. There is one 8-inch main feeding the dead end, and the only storage is the elevated tank. The engineer's model shows adequate pressure, but the field data disagree.
1. Explain why the field pressure data and the model disagree, and what that implies for the proposed subdivision.
Read the worked answer
The model is probably built on assumed roughness, actual pipe diameters, and demand allocation that do not match the field. A 12-inch main that is actually tuberculated, partially valved down, or reduced in effective diameter will show far more friction loss than a clean-pipe model. The static pressure of 32 psi at a 148-ft tank (roughly 64 psi at the tank, less elevation and losses) suggests either significant elevation difference or a restriction upstream of the test point. The residual drop from 0.9 to 0.2 mg/L at the dead end is a strong secondary signal of long detention time and low velocity, consistent with a long dead-end main.
Implication: the model must be calibrated to the field test before the subdivision is approved. Adding 60 units of demand to a main already dropping to 14 psi at 900 gpm will worsen both pressure and water quality. The proposal should be conditioned on looping the main, upsizing, or adding storage/booster capacity, and the model should be recalibrated with the hydrant test data as the basis.
2. What field investigations would you perform to reconcile the model with reality?
Read the worked answer
- Re-run hydrant flow tests at multiple points along the main, not just the dead end, and record static, residual, and pitot readings with the correct coefficients.
- Confirm valve status: check that no partially closed valve or a missing open valve is throttling the main.
- Verify actual pipe material, age, and internal condition (C-factor) by records and, if possible, by inspection or coupon.
- Check for unidentified demands or leaks: compare tank drawdown with metered production and model demand over a 24-hour period; a fast-draining tank suggests leakage or unaccounted use.
- Confirm elevations at the tank and test points, since a small elevation error changes static pressure materially.
- Measure chlorine residual and detention indicators along the main to confirm low-velocity, long-residence conditions.
- Update the model with measured roughness and demands, then re-check predicted versus observed pressures.
3. Recommend operational and capital measures to protect pressure and water quality at the dead end now and for the subdivision.
Read the worked answer
Near-term operations:
- Flush the dead end on a scheduled basis to control detention time and restore residual; document volumes and residuals.
- Consider a temporary booster or pressure-sustaining arrangement only if it can be done without masking the real restriction.
- Investigate and repair any leaks or unaccounted demand that is accelerating tank drawdown.
Capital/planning:
- Loop the main or upsize the 8-inch feed so the dead end is no longer a terminal branch.
- Recalibrate the hydraulic model to the field data before approving the subdivision; require the developer to fund looping or storage improvements as a condition.
- Evaluate additional storage or a booster to maintain minimum pressure under fire flow and peak demand.
- Set a water-quality monitoring plan (residual, and where relevant, DBP precursors) for the extended dead end until looping is complete. The core recommendation: do not approve added demand on a main whose field performance already fails the pressure and residual tests.