Home / Science / Engineering / Thermal and fluid engineering / Pipe network pressure dropTool #660

Pipe pressure drop and pipe network solver: Darcy-Weisbach with Colebrook, minor losses, Hardy Cross and Newton

Work out the head loss and pressure drop in a pipe with the Darcy-Weisbach equation, the Colebrook-White or Swamee-Jain friction factor and minor losses (K), or the flow a given head will push. Then solve a looped pipe network with reservoirs, tanks and demands by Newton's method or Hardy Cross, with flows, velocities, head losses and pressures at every node.

Keep it one tap away for everyday use.

Bookmark Pipe network pressure drop?

Press Ctrl + D now. Your browser opens its own bookmark box, where you choose the name and folder.

  1. Pipes
  2. Friction
  3. Solve
  4. Pressures
Nothing is uploadedWorks offline once open

Liquid and friction formula

Single pipe

K adds up the fittings: very roughly 0.5 for a square pipe entry, 1 for the exit into a tank, 0.3 to 0.9 for a 90° bend, about 0.2 for an open gate valve and several for a globe valve. Take K values for your own fittings from the maker or a hydraulics handbook. Roughness and K values here are rounded examples.

Pipe network

A positive flow runs from the "from" node to the "to" node; a negative one runs the other way. Demands are what leaves the network at a node (use a negative demand for an inflow). A reservoir or tank holds its water level (fixed head) whatever flows in or out. The example network is made up.

Network results

Pipes

Nodes

Download

What it does

Water (or any liquid) loses energy as it rubs along a pipe wall and squeezes through bends, valves and fittings. This tool works that loss out two ways.

Single pipe. The Darcy-Weisbach equation gives the head loss

h = (f L / D + ΣK) × V² / 2g

where f is the friction factor, L the length, D the inside diameter, V the mean velocity and ΣK the sum of the minor-loss coefficients of the fittings. The friction factor comes from the Colebrook-White equation (solved exactly by iteration) or the Swamee-Jain formula, an explicit approximation within about 1 % of Colebrook. Laminar flow (Reynolds number below 2,000) uses f = 64/Re; between 2,000 and 4,000 the tool blends the two, because real flow there is unpredictable. You get the velocity, Reynolds number, friction factor, the friction and fitting losses, the pressure drop including any rise, the hydraulic power, and a system curve of head against flow. Or give the head you have and it finds the flow.

Pipe network. List the nodes (junctions, reservoirs and tanks) and the pipes joining them. The tool finds the flow in every pipe and the head and pressure at every junction, so that flow balances at each junction and head losses balance around each loop. Two solvers are offered: Newton’s method on the node heads (the global gradient method used in modern network software, usually 5 to 10 iterations) and the classic Hardy Cross method, which corrects one loop at a time. They give the same answer.

How to use it

  1. Choose the liquid: water at a temperature, or type the density and viscosity of another liquid (1 mPa·s = 1 cP).
  2. For one pipe, give the flow in L/s, the length, the inside diameter in mm, the roughness ε in mm and ΣK. The “rise” is how much higher the outlet is than the inlet; it adds to the pressure drop but not to the friction loss.
  3. For a network, type or paste the node and pipe lists (or open a CSV or Excel file). A node with a number in the head column is a reservoir or tank: its water level stays fixed. Every other node is a junction, with a demand that leaves the network there. Add x and y columns to get a sketch.
  4. Read the tables. A pipe’s direction shows which way the water really flows. Pressure head is head minus ground level; a negative value means the water cannot reach that node under pressure.

Worked example

One pipe: 25 L/s of water at 20 °C through 250 m of 150 mm steel pipe (ε = 0.05 mm) with ΣK = 3.5, rising 5 m. The velocity is 1.41 m/s and the Reynolds number 211,500, so the flow is turbulent; Colebrook gives f = 0.01781. Friction loses 3.028 m and the fittings 0.357 m, a total head loss of 3.385 m. With the 5 m rise the pressure drop is 82.09 kPa.

Network (made up): a reservoir R at 60 m feeds five junctions through two loops, and a tank T held at 52 m hangs off junction J5. The demands add up to 60 L/s. Newton’s method settles in 8 iterations: R supplies 66.27 L/s, and 6.265 L/s flows on into tank T. Junction J1 sits at a head of 58.71 m, 38.71 m of pressure above the ground. Switch to Hardy Cross and the same flows come out after a few dozen loop corrections.

Tips

  • Keep velocities in water mains to roughly 0.5 to 2 m/s; faster wastes energy and risks water hammer.
  • Roughness grows as pipes age and scale up. Try a larger ε to see the effect of old pipes.
  • For a pump in the network, model its outlet as a reservoir at the pump’s delivery head, or use the pump and system curve tool.

FAQ

Why Darcy-Weisbach and not Hazen-Williams? Darcy-Weisbach works for any liquid, temperature and pipe size, and is the physically based formula. Hazen-Williams only suits water at ordinary temperatures.

Can I trust the result for design? It is for study and planning. The answer is only as good as the roughness, K values and demands you type; check important designs with a qualified engineer.

What does “loops” count? The independent loops plus one path between each extra reservoir: pipes minus junctions.

Where the numbers come from

  • Darcy-Weisbach equation, Colebrook-White (1939) friction factor and the Swamee and Jain (1976) explicit approximation: standard fluid mechanics, for example White, F. M., Fluid Mechanics, and Streeter and Wylie, Fluid Mechanics (equations only).
  • Hardy Cross (1936) method of balancing loop heads, and the global gradient (Newton) method of Todini and Pilati (1988), as described in hydraulics textbooks (methods only).
  • Water density and viscosity at a temperature: smooth general-purpose formulas (a Thiesen-type density fit and a Vogel-type viscosity fit), checked against the IAPWS-95 formulation in the Python iapws package.
  • Roughness and K values in the hints are rounded, general figures chosen by fibub for a first estimate, not copied from any table. The example pipe and network are made up.
  • Checked against an independent SciPy solver (fsolve on all the network equations at once) written for the tests.

Standard scientific values, such as constants, atomic weights and melting points, are facts found in every textbook; we name where ours were checked. If you are an author or publisher and want a value, a credit or a table changed or removed, tell us and we will act quickly.

Last updated 2026-11-09.

Related tools

Manning and weir flow calculator #140 Work out the discharge or normal depth in a rectangular, trapezoidal or triangular channel or a part-full pipe with Manning's equation, with velocity, Froude number, critical depth, specific energy, bed shear and a rating curve; and the flow over sharp-crested, V-notch, broad-crested or Cipolletti weirs, or the head for a given flow. SI or US units. Darcy flow calculator #128 Groundwater flow from Darcy's law: flux, discharge, seepage velocity and travel time. Find the hydraulic gradient and flow direction from three or more wells, convert K to intrinsic permeability, and estimate K from a sieve curve (d10, Cu, Cc and Hazen). Uncertainty propagation calculator #25 Type your formula and your measurements with their uncertainties, or paste repeated readings. Get the result with its uncertainty, which input matters most, and a Monte Carlo check. Steam tables and Rankine cycle #661 Look up water and steam properties from pressure and temperature, saturation by pressure or temperature, or from pressure with enthalpy, entropy or quality, using the IAPWS-IF97 industrial formulation. Then work out a simple or reheat Rankine power cycle: turbine and pump work, heat in, thermal efficiency, exit quality, steam rate and a temperature-entropy diagram. Pump and system curve #662 Fit a pump curve to points read off the maker's chart, build the system curve from static head and pipe friction, and find the operating point with its efficiency and power. Try pumps in parallel or series, a new speed or a trimmed impeller (affinity laws), and check NPSH available against NPSH required. Room heating and cooling load #663 Estimate the peak heating and cooling load of a room from its walls, windows and roof (U-values), sun through the glass, air leakage and ventilation (sensible and latent), and the heat from people, lighting and equipment. Results in watts, BTU per hour and tons of refrigeration, with a breakdown. A simplified estimate, not a full ASHRAE or CIBSE method.