PPR Pipe Flow Rate Chart: Size a Line Without Losing Pressure

A contractor calls to say the top-floor showers in a nine-storey hotel go weak whenever the ground-floor laundry runs. The riser was specified as “PP-R 32 mm” on a drawing that never named a pressure series. The pipe on site is SDR 6. Its bore is 21.2 mm, not the 26.2 mm the designer had in his head, and at the design flow the water is moving at 2.36 m/s — past the limit EN 806-3 sets for a riser, and loud enough that the client has already complained about noise.

Nothing was installed wrong. The fusion joints are sound, the pipe meets its standard, the pressure rating is more than adequate. The sizing was done on a number that does not size a PP-R line: the outside diameter.

Why an OD Number Cannot Size a PP-R Line

Steel and copper are ordered by a nominal size that stays close to the bore. PP-R is not. Under ISO 15874-2:2013 the outside diameter is fixed for a given nominal size and the wall carries the pressure rating, so raising the rating thickens the wall inward and takes the flow area with it.

The standard expresses this as a pipe series S, related to the more familiar SDR by SDR = 2S + 1. A lower S number means a thicker wall, a higher pressure rating and a smaller bore. Three series cover almost all plain PP-R plumbing work, and they are not interchangeable hydraulically.

Look at what that does to a single nominal size. These are the standard’s own minimum wall thicknesses for dn32, with the bore and flow area that follow from them:

A third of the flow area disappears between the lightest and heaviest dn32 on the market, and every one of those pipes is legitimately labelled 32 mm. This is why a bill of quantities that says “PP-R 32 mm” transfers the sizing risk to whoever fills the order. If your supplier stocks S2.5 and your calculation assumed S5, the system is undersized before anything is cut.

One further trap: ISO 15874-2 Table 5 also lists series S8, S6.3 and S4, and footnote (a) restricts all three to PP-RCT. If you are buying plain PP-R, those columns are not available to you regardless of what a supplier’s brochure shows.

The Velocity Limits That Actually Decide the Size

Most people reach for pressure rating when sizing a pipe. In building services that is rarely the constraint — a PN20 pipe will hold the pressure long after the system has become unusable for other reasons. What decides the size is velocity, and velocity is capped for reasons that have nothing to do with burst strength.

Fast water is loud, it erodes fittings, and it makes water hammer violent enough to split a joint when a solenoid valve closes. EN 806-3 clause 4.4 states the limits its sizing tables are built on:

• 2.0 m/s — header pipes, rising pipes and floor service pipes
• 4.0 m/s — connection pipes to a single fitting (dead legs)

The standard adds a note that national regulations may require lower velocities to avoid water hammer and noise, so treat 2.0 m/s as a ceiling rather than a target. Where a duty runs continuously, the German companion rule is stricter still: DIN 1988-300 Table 5 allows 5 m/s in a supply section with low-resistance fittings only for operating periods up to 15 minutes, and drops to 2 m/s for anything longer. A recirculation loop or a process line that runs all shift is a “beyond 15 minutes” case.

The PP-R Flow Rate Chart

Here is the chart the sizing actually needs. For each nominal size and pipe series it gives the bore from ISO 15874-2:2013 Table 5, the flow at three design velocities, and the pressure gradient the run will lose per metre at each of them.

Three declared inputs produce every number, so you can rebuild any cell: internal diameters from the standard, an absolute roughness of 0.007 mm (the pipe friction coefficient published on the technical data sheet of a named PP-R manufacturer), and the Colebrook-White friction factor at water properties for 50 °C. Flow is in litres per hour; gradient is in pascals per metre of straight pipe.

Two patterns are worth reading off the chart before you use it. First, the flow columns scale with the square of the bore while the gradient columns scale roughly with velocity squared over bore — which is why stepping up one nominal size cuts the loss dramatically. Going from dn32 to dn40 in S3.2 at a fixed 3,000 L/h drops the gradient from about 1,660 Pa/m to 560 Pa/m, a two-thirds reduction for one size step.

Neither of those figures is a column in the table above, and that is the point: at a fixed flow the velocity falls as the bore opens, so the dn32 runs at 1.97 m/s while the dn40 drops to 1.26 m/s. Comparing two sizes at constant flow means reading diagonally across the chart, not down one velocity column.

Second, the series penalty compounds along a run. At 1.5 m/s a dn50 in S2.5 loses 645 Pa/m against 505 Pa/m in S5 — 28% more, on every metre, forever.

Step 1: Establish Your Design Flow

The chart answers “what will this size carry”. It cannot tell you what you need to carry, and that is where most sizing errors originate rather than in the hydraulics.

Design flow is not the sum of every outlet. Real buildings never run all fixtures at once, so codes apply a diversity or simultaneity function to convert total connected demand into a peak design flow. EN 806-3 does this through loading units and a simplified table; DIN 1988-300 uses a peak-flow function against total flow. Use whichever your project’s authority recognises, and use it consistently for the whole system.

• Take the flow from the code, not from the pump. Sizing to a pump’s rated output rather than the building’s diversified peak is the fastest route to oversized, stagnant pipework — a hygiene problem as well as a cost one.
• Size each segment to its own flow. A riser carries the diversified demand of everything above the take-off, and it drops at every branch. A riser held at one diameter for nine floors is over-sized at the top and probably over-velocity at the bottom.
• Add recirculation to the hot flow. A domestic hot-water return imposes a continuous duty that the fixture-based diversity calculation does not include.

Step 2: Read the Chart Against Your Velocity Ceiling

With a design flow in litres per hour, go to the chart’s block for your series and read down the 2.0 m/s flow column until you find the first size whose figure exceeds your flow. That is your minimum size on the velocity criterion. If the pipe is a riser or a header in a noise-sensitive building, do the same in the 1.5 m/s column instead and take the larger answer.

Then read the matching Pa/m cell and multiply it by the run length to get the friction loss for that segment. Compare that against the pressure you have available: incoming main pressure, less static lift, less fixture minimum flow pressure, less any meter, filter or softener loss, is the budget the pipework has to fit inside.

Step 3: Pick the Series Before You Pick the Size

This is the step the drawing usually skips, and it must come first because it changes every number in the chart. The series follows from the service condition, not from preference.

ISO 15874-2 handles this through application classes and a maximum design series. For PP-R at a 10 bar design pressure, the standard’s Table 3 gives an S_calc,max of 3.0 for Class 1, 2.1 for Class 2, and 1.9 for Class 5. A lower permitted S means a thicker wall, and the comparison is a straight numerical one: the nominal series number must not exceed S_calc,max.

So a Class 2 hot-water duty at 10 bar rules out both S3.2 and S2.5 — 3.2 and 2.5 are each above 2.1 — leaving only S2 for that combination. Drop the design pressure and the options return: at 8 bar the Class 2 limit rises to 2.6 and S2.5 becomes permissible; at 6 bar it rises to 3.5 and S3.2 is valid again. Settle the design pressure before the bore, because it decides which columns of the chart you may read at all.

The practical order of operations, then, is: fix the application class and design pressure, derive the permitted series, and only then take the flow reading. Reversing those steps produces the hotel riser in the opening paragraph.

One caution about pipes that are not plain PP-R. An aluminium-composite or copper-lined PP-R has a different internal geometry from the base pipe, and the standard permits the manufacturer to state dimensions that differ from Tables 5 to 9. For those constructions the chart above is a starting estimate and the manufacturer’s own dimension table governs.

Fittings Are Not Free: Adding Local Losses

A pipe run is straight pipe plus the things that interrupt it, and the interruptions are not a rounding error. Each fitting is assigned a dimensionless resistance coefficient (zeta), and the pressure it costs is Z = ζ × ρ × v² / 2, with Z in pascals.

What surprises people is that the coefficient is not constant across sizes. Published values for PP-R heat-fusion fittings show a step change once the moulding geometry changes at around dn50:

Read the elbow row again. A 90° elbow in dn32 carries a coefficient of 2.0, four times the 0.5 of the same fitting in dn50. Note also that below dn40 the published 45° values sit above the 90° ones, which is the opposite of the fluid-mechanics intuition. That is what the manufacturer’s catalogue reports, and it reflects moulding geometry in the small sizes rather than the turn angle — another reason to take these coefficients from a published table rather than from a textbook rule of thumb. Small-bore, fitting-dense pipework — the manifold cupboard, the plant-room header, the tight riser offset — is where local losses stop being a correction and start dominating the segment.

A Worked Example You Can Copy

Take the hotel riser branch from the opening. Step 3 has already been settled for it: the branch is a Class 2 duty at a 6 bar design pressure, so by the Table 3 rule above the permitted S calc,max is 3.5 — S3.2 is in bounds, S5 is not, and the S5 columns of the chart are out of play before any flow is read. The design flow after diversity is 3,000 L/h at 50 °C, the run is 28 m of straight pipe, and there are two 90° elbows. The question is which size and series to specify.

The instructive rows are the two dn32 entries. The same nominal size passes in SDR 7.4 and fails in SDR 6, on identical flow and identical route, because 2 mm of extra wall pushes the velocity from 1.97 to 2.36 m/s. A drawing that says “32 mm” cannot distinguish between them, which is precisely how the hotel ended up with weak showers.

The chosen dn40 costs one size step and returns 54.2 kPa down to 18.6 kPa — about 36 kPa of pressure handed back to the system, which on a nine-storey riser is roughly three-and-a-half metres of static head recovered for the price of slightly more expensive pipe. That trade is almost always worth making on a riser, and almost never worth making on a short dead leg.

Where This Chart Stops Being Valid

A chart used outside its assumptions is worse than no chart, so here is where this one stops.

• Temperature. The gradient columns assume water at 50 °C. Cold water at 10 °C is more viscous, so a cold line loses slightly more per metre at the same velocity than the table shows. The flow columns are geometry only and are unaffected.
• Roughness. The 0.007 mm figure comes from a named PP-R manufacturer’s technical data sheet and is representative of the material, not a value published in ISO 15874-2. It is a declared assumption. Because PP-R is hydraulically smooth, the result is not very sensitive to it — but if your project’s authority mandates a different roughness or the Hazen-Williams method with a specified C value, use theirs and expect small differences.
• Composite and lined pipes. Aluminium-composite, fibre-reinforced and copper-lined constructions may carry manufacturer-stated dimensions that differ from the standard’s tables. Take the bore from that manufacturer’s data.
• Dimension class. These are dimension class A sizes. ISO 15874-2 also defines class B1 on copper pipe sizes, where a “22 mm” pipe is a genuinely different object.
• Ageing. PP-R does not tuberculate like steel, which is why a 20-year-old PP-R line still flows close to its original figures. That is a genuine advantage of the material, but it is not a licence to size with no margin.

What to Check Before the Spec Leaves Your Desk

Sizing failures are rarely arithmetic failures. They are almost always a mismatch between what the calculation assumed and what arrives on site, so the checks that matter are documentary.

• Does every pipe line item name a series or SDR? If a BOQ line reads “PP-R 32 mm”, it is not yet a specification. Write “dn32, series S3.2 (SDR 7.4), EN ISO 15874”.
• Does the submitted product’s declared wall thickness match the series you sized on? This is a two-minute check against the standard’s Table 5 and it catches the substitution that causes the failure in the opening paragraph.
• Is the application class on the submittal the one you designed to? Class 2 at 10 bar and Class 1 at 10 bar permit different series for PP-R, and a supplier optimising for cost will offer the thinner wall the lighter class allows.
• Do the fittings come from the same system? Fusion fittings are matched to the pipe’s dimensions; mixing systems is a joint-integrity question before it is a hydraulic one.
• Have you asked for the manufacturer’s own dimension table? For composite or lined pipe this is mandatory, because the standard explicitly allows those dimensions to differ.

What a manufacturer does to keep the bore you sized on the bore you receive is the link between the table and reality — but as a specifier you cannot audit anyone’s extrusion line, so the question worth asking is narrower: who has independently tested this pipe, and against what.

For the Hitze PP-R range that is SKZ, which holds testing on the PP-R pressure pipe and PP-R fittings, and DVGW, which issued a type examination for the PP-R drinking-water pipe. Third-party marks against a named scope are a different class of evidence from a supplier describing its own QC. Ask for the certificate reference and the models each covers, then verify it in the issuing body’s directory rather than accepting a brochure logo — one email, and it separates a tested range from a claimed one.

One dimensional caution specific to buying against this chart: Hitze’s PP-R range runs OD 20–110 mm (DN15–DN90), so every size in the table above exists — but a size existing is not the series existing, and S5, S3.2 and S2.5 are three different products at the same dn. Name the series on the enquiry, not just the diameter. There is no minimum order quantity and samples can be requested with the spec sheet, which for a sizing decision is the point — a caliper on a real bore settles in a minute what a catalogue argument will not.

If you are assembling the wider submittal rather than just the sizing, the PP-R specification and submittal checklist covers the document set that goes with these numbers, and the EN ISO 15874 service classes explainer covers the class-and-design-pressure decision that fixes your series in the first place. For the dimensional background, see the PP-R sizes, SDR and PN ratings guide.

Sizing PP-R, in Order

Fix the application class and design pressure, and let ISO 15874-2 Table 3 tell you the heaviest series you are allowed to use. Take the bore for that series from Table 5. Establish the diversified design flow for each segment from your project’s code. Read the chart down the velocity column your pipe’s role permits — 2.0 m/s for risers and headers, less where noise matters — and take the first size that carries the flow. Multiply the gradient by the run, add the fitting coefficients, and check the total against the pressure you actually have.

Then write the series on the drawing. Everything above is undone by a bill of quantities that says only “32 mm”.

Frequently Asked Questions