How the two differ in kind
Sizing a water supply pipe needs two questions answered, and the two methods here answer one each.
The FIXTURE-UNIT method answers how much flow the pipe has to carry. Each fixture is assigned a loading value reflecting how much water it uses and how often, the values served by a given run are added up, and a published table converts the total into a design flow rate. The crucial property is that the conversion is not linear: it has DIVERSITY built into it, because not every fixture runs simultaneously and the larger the number served the smaller the proportion that realistically do. A building with fifty fixtures is not sized for fifty simultaneous demands, and a pipe sized as though it were would be enormous, expensive, and slow enough for the water in it to stagnate.
The FRICTION-LOSS calculation answers whether enough pressure survives the journey. Water moving through a pipe loses pressure to friction, at a rate depending on the flow, the pipe's diameter and its internal roughness, and the calculation totals that over the run, adds the losses through every fitting and valve, and accounts for the static height gain or loss. What remains at the fixture is the residual pressure, and that is the criterion that actually governs — a fixture needs a minimum flowing pressure to work as intended, and a shower on a long run through a small pipe can be starved while the pipe's capacity is nominally fine.
So the sequence is: fixture units to establish the design flow, then a friction-loss check to confirm the size delivers it at an acceptable pressure and an acceptable velocity. The second step is the one that gets skipped, because the first produces a number that looks like an answer.
Velocity is the third constraint and it binds from the other direction. Too fast and the system is noisy and prone to erosion and water hammer; too slow and water stagnates, which matters for hygiene.
The factors that actually differ
Show
| Fixture-unit method | Friction-loss calculation | |
|---|---|---|
| Question it answers | How much flow must this pipe carry? | Will enough pressure be left at the fixture? |
| What it accounts for | Fixture types and counts, with diversity built into the conversion table. | Flow, diameter, length, roughness, fittings and static height. |
| Effect of run length | None. A fixture unit total is the same at two metres and two hundred. | Directly proportional. This is the whole point of it. |
| Effect of fittings | None. | Substantial — bends, tees, valves and meters can exceed the straight pipe's loss on a compact system. |
| Diversity | Built in, and it is why the design flow is far below the sum of the fixtures' rated flows. | Takes whatever flow it is given. |
| Governing criterion | Capacity. | Residual pressure at the fixture, which is what actually decides whether the outlet works. |
| What it misses alone | That a long run or a congested one may not deliver the flow at a usable pressure. | Nothing about sizing — but it needs a flow figure to start from, which fixture units supply. |
| Velocity | Not addressed. | Falls out of the calculation, and it is a design limit at both ends — noise and erosion above, stagnation below. |
| Where it is used | First, on every branch and main, to get the design flow. | Second, along the index run — the path with the greatest total loss. |
| When it decides | Short runs with generous pressure, where capacity is the binding constraint. | Long runs, high buildings, low incoming pressure, and anywhere the fixture is far from the supply. |
Which one, and when
Choose fixture-unit method when…
- Establishing the design flow for a run, which is where every sizing exercise starts.
- Sizing a water meter or a service connection, which are specified in these terms.
- Working to a code that prescribes the method, which most do.
- A quick check that a proposed pipe size is in the right range for the fixtures it serves.
Choose friction-loss calculation when…
- The run is long, or the building is tall, where pressure rather than capacity governs.
- The incoming supply pressure is low or unreliable.
- A specific fixture is under-performing and the cause is in the distribution rather than the fixture.
- Choosing between two pipe sizes that both have adequate capacity — which is the normal case.
Now run your own numbers
This page holds no prices on purpose — a national average is wrong for almost every real project. Quantify both options with your dimensions and your local quotes.
Frequently asked questions
- Why is the design flow so much less than the sum of the fixtures?
- Because of diversity, and the effect is large. Every fixture in a building has a rated flow, but they do not all run at once — and the more fixtures there are, the smaller the proportion that realistically run simultaneously, because the probability of everyone opening a tap in the same minute falls as the number of people rises. The fixture-unit method encodes that statistically: each fixture contributes a loading value reflecting its demand and frequency, and the table converting the total into a design flow flattens out sharply, so doubling the fixture count adds far less than double the flow. Sizing to the arithmetic sum of the rated flows would give pipework several sizes larger than necessary — expensive, and genuinely harmful, because water moving too slowly in an oversized pipe stagnates.
- What actually decides whether a shower works?
- The flowing pressure at the fixture, which is what a friction-loss calculation produces and what a fixture-unit total says nothing about. Every fixture has a minimum pressure at which it delivers its rated flow, and thermostatic showers and combination appliances are among the more demanding. The pressure available at that point is the incoming supply pressure, minus every loss between the supply and the outlet: friction along the pipe, losses through each fitting and valve, the meter, any filter or softener, and the static loss of lifting the water to that floor. On a long run of small pipe with several bends and a rise, those losses can consume most of the available pressure while the pipe's capacity was never in question — which is exactly the case people diagnose as a faulty shower.
- How much do fittings matter?
- Often more than the straight pipe, particularly on a compact system, which is why a calculation covering only pipe length underestimates the loss badly. Each bend, tee, valve, meter and appliance connection imposes a loss, conventionally expressed as an equivalent length of straight pipe that would cause the same drop, and those equivalent lengths are added to the actual run before the friction calculation. A run of ten metres with a dozen fittings can behave hydraulically like a run several times longer. The quality of the fittings matters too: a swept bend costs far less than a sharp elbow, and a full-bore valve far less than a globe valve. This is also why repiping a house in a smaller-bore material with many push-fit fittings can produce worse performance than the larger, less congested pipework it replaced.
- Why is there a maximum velocity?
- Because fast-moving water is noisy, erosive and dangerous when it stops. Noise is the symptom people notice first — velocity in a pipe produces a rushing sound that carries through the structure. Erosion-corrosion is the durability issue: water moving too fast scours the protective film from the inside of the pipe, particularly in copper and particularly at bends and fittings, thinning the wall over years. And water hammer is the acute one: a fast-moving column of water stopped abruptly by a quick-closing valve produces a pressure surge that can damage fittings and appliances, which is why arrestors are fitted and why velocity is limited. Codes and guidance state maximum velocities, usually lower for hot water than cold because erosion-corrosion accelerates with temperature.
- Is there a minimum velocity too?
- Effectively yes, and it is a hygiene requirement rather than a hydraulic one. Water sitting still in an oversized pipe stagnates: it warms toward room temperature, any disinfectant residual decays, and the conditions favour bacterial growth including legionella — which is precisely why the temperature bands for hot and cold water matter, and why dead legs are a recognised hazard. An oversized system is effectively a distributed dead leg. This is the constraint that stops the sizing exercise resolving into simply choosing generous pipes: bigger is not safer. It is also why the length and the frequency of use of individual branches matter, and why a rarely used outlet on a long branch is a design problem rather than a harmless one.
- What is the index run?
- The path from the supply to the fixture that suffers the greatest total pressure loss, and it is the one the friction calculation is performed along — because if that fixture has enough pressure, every other one does. It is usually the longest run, but not always: a shorter run with a large static rise, many fittings, or a high-demand fixture at the end can lose more. Identifying it means considering length, height, fitting count and the fixture's own pressure requirement together rather than reaching for a tape measure. Once the index run is sized to deliver adequate pressure at its end, the rest of the system is sized by its own fixture-unit loads, and any branch that proves marginal is checked in the same way. Missing the true index run means sizing everything against the wrong worst case.
- Does pipe material change the sizing?
- Yes, in two ways that both matter. Internal roughness varies between materials and affects the friction loss directly — smooth plastics generally lose less than metal for the same bore, which is one of their advantages. More significantly, the INTERNAL bore for a given nominal size differs between materials and systems, and plastic push-fit and press systems in particular can have noticeably smaller bores at their fittings than the nominal size suggests, because the fitting inserts into the pipe. A run repiped like for like in nominal size can therefore have meaningfully less capacity and more loss than the original. The calculation should use the actual internal diameters for the system being installed rather than nominal sizes, and this is a common cause of a repipe that performs worse than what it replaced.
- Which do I do first?
- Fixture units, then friction loss, and the order is not interchangeable because the second needs the first's output as its input. Establish the design flow for each section from the fixture units it serves; propose a pipe size; calculate the friction loss along the index run at that flow, including every fitting's equivalent length and the static height; compare the residual pressure at the critical fixture against what it needs; and check the velocity is within limits at the same time. If the residual is inadequate, increase the size and repeat. The step that gets skipped is the friction check, because the fixture-unit table produces a pipe size that looks authoritative — and it is authoritative about capacity only, which is not the criterion the occupant will judge the system by.