Relief Line Loads: Thrust, Sensors & Valves

When a fire pump relief valve opens, the pipe behind it takes a load that most spec sheets describe in one number and never explain. We supply grooved couplings and butterfly valves for pump rooms at LEDE, and the same three questions come back from engineers: where does the force come from, what should the instruments read, and how do you size the valve. Here is how we work them out.

• Most of a "45 kN recoil" figure is usually pressure thrust on the bend, and the flow momentum is a small part
• A sensor is only as fast as its tapping line and logger, so a high sampling rate alone proves little
• Size a butterfly valve from flow loss and torque, not from a single Cv number

What Force Does a Pressure Relief Valve Put on the Pipe?

Two forces get mixed up. The first is pressure thrust. At a bend or a cap, internal pressure pushes on the full bore area: F = P × A. For a DN200 line with a 206 mm bore, the area is 33,300 mm². At 1.4 MPa the thrust is about 46.7 kN. That is where a figure like 45 kN usually comes from.

The second is flow momentum. Take 2,500 gpm, which is 0.158 m³/s, in the same bore. The velocity is 4.7 m/s, and the momentum term ρ × Q × v is only about 750 N. So the flow in the pipe adds little.

A free jet is different. If the relief valve discharges to atmosphere through a nozzle, the reaction is close to 2 × ΔP × A of the jet in the ideal case. For a 100 mm jet at 1.4 MPa that is about 22 kN. For a 150 mm jet it is about 49 kN. Real valves open over seconds and have losses, so these are upper bounds, not predictions.

Now check the coupling. A force of 46.7 kN acting at a 0.35 m arm makes a moment of 16.3 kN·m. That is right at the 16.5 kN·m bending capacity we list as a target for the angel pad rigid coupling. So a rigid coupling reduces angular play, but it is no substitute for anchors that take the thrust. We tell buyers to restrain the bend and to keep the coupling for alignment. Our target play of 0.15° is a design value to be confirmed by a test report.

Check	LEDE Rigid Grooved Relief Line	Flat-Pad Grooved Couplings	Welded Flange Spool	Threaded Pipework
How it resists bending	Angel pad wedge, rigid joint	Joint allows some angular play	Welded or bolted flange	Thread and pipe wall
Angular play	About 0.15° (target)	Often 1° or more	None if welded	Small, but thread can crack
Needs thrust restraint	Yes, anchor the bend	Yes, and more of it	Yes	Yes, and limited by thread
Field assembly	Bolt, check pad contact	Bolt	Weld and test	Thread and seal
Inspection access	Open joint, quick check	Open joint	Weld inspection	Hard to inspect
Paperwork to ask for	Test report and listing letters	Same	Weld records	Same

"A coupling carries moment, not magic. Work out the thrust and the arm, compare the result with a rated capacity, and anchor the bend. Then use the rigid joint to keep alignment, which is the job it is good at."

— Guo Wei, Chief Metallurgical & Critical Infrastructure Piping Specialist, LEDE

Do a Pressure Transmitter and a Magnetic Flow Meter Beat Gauges?

For recording and trending, yes. For protection, no. A relief valve is a mechanical safety device, and an electronic signal does not replace it.

Start with the surge. A pressure wave in a 100 m line with a wave speed of 1,000 m/s has a period of 4L ÷ a = 0.4 s, or 2.5 Hz. If the front rises over 50 ms, its bandwidth is about 0.35 ÷ 0.05 = 7 Hz. A rule of thumb is to sample at ten times that, so about 70 Hz is enough. A pressure transmitter that samples at 1,000 Hz is therefore ample, and it is not the limit. The weak points are an air pocket in the tapping line, a long impulse tube, and a logger that records once a second. Bleed the line and check the logger rate.

A Bourdon gauge is not 2 seconds slow. Its limit is that a needle cannot follow a fast transient, and damping fluid hides the peak. That is a good reason to add a transmitter for commissioning, and a good reason to keep the gauge for local reading.

Flow measurement is where the head loss is clearer. A magnetic flow meter has no obstruction in the bore, so its loss is close to zero. It needs a full pipe, conductive water and the straight runs the maker lists. An orifice plate does cost head. At 1,500 gpm in a 154 mm bore the velocity is 5.1 m/s. With a loss coefficient of 2.5, ΔP = K × ρ × v² ÷ 2 is about 32 kPa. To reach 0.12 MPa the coefficient would have to be about 9.3. So the loss is real, but it is smaller than some brochures say.

How Do You Size a LEDE Butterfly Valve for a Bypass?

Use two checks. The first is flow loss. A valve with Cv = 1,950 (gpm per root psi) on DN150 is a target value we list, so check the LEDE gate valve catalogue for the size you plan to use. The loss coefficient is K = 891 × d⁴ ÷ Cv², with d in inches. For d = 6.065 that gives K ≈ 0.32. At 1,500 gpm, ΔP = (1,500 ÷ 1,950)² ≈ 0.59 psi, or 4.1 kPa. That is small, so the valve will not be what limits your flow.

The second is torque. Hydraulic torque on a butterfly disc is roughly T = Ct × ΔP × D³. With an assumed Ct between 0.05 and 0.1, a 154 mm disc and 1.4 MPa across it, T is about 260 to 510 N·m. That is the same order as the 350 N·m we list as a reverse-torque target for the gearbox. So ask for the actuator rating, and compare it with your own calculation at the worst case of opening and pressure.

A worm gear holds position when its lead angle is smaller than the friction angle. We do not publish a drift figure, since we have none to quote. Instead, check that the handwheel or actuator stays put under load, and ask for a seat test report. A bubble-tight test at 1.5 times rated pressure is a target, not a result, until the report says so.

Three-step relief line acceptance checklist:

1. Calculate pressure thrust and any jet reaction, then compare with the coupling capacity and fit anchors
2. Bleed the tapping lines, set the logger above 70 Hz, and record one relief valve test
3. Compare the butterfly valve's rated torque with your calculated worst case, and ask for the seat test report

LEDE (ledefittings.com) supplies grooved couplings, butterfly valves and gate valves for fire protection pipework. All figures here are design targets or worked examples unless a test report says otherwise; documentation is available on request.