PP-R Pressure Derating: What to Specify by Service Class

Search for the pressure a PP-R pipe holds at 95°C and page one will give you three different answers for the same pipe. One page says a PN20 pipe holds 6 bar. Another says 5 bar. A third describes a PN20 pipe holding 30 bar at 20°C, which cannot be true, because PN20 means 20 bar at 20°C. None of the three shows where its number came from.

That matters more than it looks. Pressure–temperature derating is the calculation that decides whether a hot-water riser lasts fifty years or splits in its fourth winter, and it is the one number on a technical datasheet that buyers almost never audit. This article rebuilds the derating figures from the two standards that actually define them — ISO 15874-2:2013 for polypropylene pipe and ISO 10508:2006 for the service classes — so that every value below carries a table number you can check.

One thing to be clear about before the numbers start: IFAN supplies PP-R to ISO 15874 and DIN 8077/8078, and the figures on this page are the standard's, not in-house laboratory measurements. Where the standard and a supplier datasheet disagree, this article follows the standard and says so.

The short version

The rest of this page works through where those figures come from, how to turn them into a wall thickness you can put on a purchase order, and how to audit a supplier's derating table in about two minutes.

Why Published PP-R Derating Figures Disagree

Almost every derating figure in circulation is quoted as a pressure at a temperature: PN20 at 70°C, PN16 at 60°C. That format is the source of the contradictions, because a temperature on its own is not enough information to calculate anything.

Polypropylene does not fail at a pressure. It fails after an accumulated duration under a combination of stress and temperature, and the standard treats that explicitly. ISO 15874-2 derives the design stress using Miner's rule in accordance with ISO 13760 — a cumulative-damage method — applied across the whole temperature–time profile of an application class. Ask "what does PN20 hold at 70°C" and you have not specified how long, how often, or what the system does the rest of the time. Three writers make three different assumptions, publish three different numbers, and all of them look authoritative.

The pipe itself gives the game away. ISO 15874-2 requires conformance to the reference lines to be demonstrated at 20°C, at 60°C to 70°C and at 95°C, with at least three failures falling in each of the intervals 10 to 100 hours, 100 to 1000 hours, 1000 to 8760 hours, and beyond 8760 hours. The qualification programme is built around time bands because time is a variable of the same standing as temperature.

The Service Class Sets the Pressure, Not the Temperature

ISO 10508:2006 Table 1 classifies service conditions into five classes, each written for a 50-year design period. Each class is a sequence of temperatures with a duration attached to every step, plus a short excursion allowance and a malfunction allowance. Read one and the derating logic stops being mysterious.

Source: ISO 10508:2006, Table 1 — Classification of service conditions. Class 3 is omitted here because its service-temperature profile is the low-temperature underfloor-heating duty, which is not the hot-water and radiator service this article is about; check ISO 10508 Table 1 directly if your project sits in that class.

Two things in that table do most of the work in a real design.

The malfunction temperature is a lifetime budget, not a rating. ISO 10508 defines T_mal as the highest temperature reached when the control limits are exceeded, and notes that this can occur up to a total of 100 hours over a period of 50 years. One hundred hours across five decades is roughly two hours a year. A system whose thermostatic control lets the cylinder run away for an afternoon each month is not experiencing an excursion; it is running a different class from the one it was priced for.

There is a second, harder edge to that budget. If a specifier writes a design life shorter than 50 years, ISO 10508 reduces all the times in Table 1 proportionately — except the malfunction time, which stays at 100 hours. Halving the design life does not buy twice the excursion allowance per year. It buys none.

Every class also has to pass a cold-water duty. Whatever hot-water profile a system is designed for, ISO 10508 requires it to be suitable for cold water for 50 years at 20°C and a design pressure of 10 bar, demonstrated by extrapolation to ISO 9080. This is why a pipe sized for a modest hot duty is sometimes thicker than the hot calculation alone demands — the cold check governs instead, and the next section shows exactly where that happens.

The Design Stress Table That Sets Every PP-R Derating Number

Once the class is fixed, ISO 15874-2 converts it into a single number: the design stress σ_D, in megapascals, that the pipe wall is allowed to carry for that class. Annex A publishes the results for each polypropylene type, and for PP-R they are these.

Source: ISO 15874-2:2013, Table A.2 — Design stress. The final column is calculated from those values and shows what fraction of the cold-water design stress survives in each class.

The last column is the honest version of "how much does PP-R derate". In Class 2 hot-water service, PP-R is working at 31% of its cold-water design stress — and that single figure explains why a pipe comfortable on a cold main needs a visibly heavier wall on a hot riser.

Why Class 4 beats Class 2

Look at the order of the numbers. Class 4 permits 3,29 MPa, which is more than Class 1 and substantially more than Class 2 — yet Class 4 reaches a malfunction temperature of 100°C, hotter than Class 2's 95°C. A derating model based on peak temperature would rank these two the other way round.

The class profiles explain it. Class 2 runs at 70°C for 49 straight years. Class 4 spends 2,5 years at 20°C and 20 years at 40°C before it ever sees 60°C, and it never runs continuously above 60°C at all. Cumulative damage tracks the whole profile, so the underfloor circuit — hotter at its worst moment, far cooler across its life — ends up with more allowable stress than the permanently hot riser. This is the single most useful consequence of the class system, and it is invisible to anyone reading derating as a function of temperature.

The safety factor is already applied

ISO 15874-2 Table A.1 sets the design coefficient C used in deriving those stresses. For PP-R it is 1,5 at the design temperature, 1,3 at T_max, 1,0 at the malfunction temperature and 1,4 for cold water. In other words the margin is spent inside the table, deliberately, and it is smallest exactly where the standard expects the pipe to be in trouble anyway.

The practical consequence is a warning. If a consultant derates to Table A.2 and then applies a further factor of 1,5 "for safety", the system is being designed at roughly a third of the stress the standard intends and the pipe schedule will be heavier and dearer than the project needs. Use the table's numbers as they are.

Working Out the Allowable Pressure for a Real Pipe

The standard does not publish a bar figure per pipe. It publishes a stress and a geometry rule, and the two combine through the pipe series S — the ratio that links diameter to wall thickness. Working it in the standard's own direction takes about a minute.

ISO 15874-2 clause A.3 defines the maximum permitted series value as the smaller of the hot-class result and the cold-water result:

A worked example: a 70°C riser at 8 bar

Take a domestic hot-water riser running at 70°C with a design pressure of 8 bar — Class 2, p_D = 0,8 MPa.

• Hot-class result: 2,12 ÷ 0,8 = 2,65
• Cold-water result: 6,93 ÷ 1,0 = 6,93
• The smaller governs, so S_calc,max = 2,65, which ISO 15874-2 Table A.5 prints as 2,6 after rounding.
• S2.5 is the thinnest series permitted here — S3.2 is thinner than this limit allows and is out. Thicker series such as S2 remain available; the limit sets a ceiling on thinness, not a single answer.

For a DN32 line, ISO 15874-2 Table 5 gives S2.5 a minimum wall of 5,4 mm. Check it: S = (32 − 5,4) ÷ (2 × 5,4) = 2,46, which is below the 2,65 ceiling. The pipe qualifies with a little room to spare. That is the whole calculation, and it is the same three lines whatever the diameter.

The reason to trust this method is that it reproduces the standard's own published answers. Running it across all four design pressures and all four classes returns ISO 15874-2 Table A.5 essentially cell for cell — including the footnote cases below, where the results carry the standard's marker "based on σ_cold". The one divergence is Class 5 at 6 bar, where the calculation gives 3,15 and the standard prints 3,2; that is the standard's own rounding to one decimal, not a disagreement.

Where the cold-water check quietly takes over

At a design pressure of 4 bar, Classes 1 and 4 both return S_calc,max = 6,9 in Table A.5 — and the standard footnotes both cells "based on σ_cold". The hot calculation at that pressure would permit a thinner wall than the cold-water duty allows, so the 20°C requirement becomes the binding constraint.

This is worth knowing before you argue with a supplier. On a low-pressure hot line, the wall thickness you are quoted may have nothing to do with your hot-water temperature at all. It is the 50-year cold-water case setting the floor, and no amount of relaxing the hot duty will make that pipe thinner.

Choosing the Pipe Series for Your Service Class

Here is the selection table the calculation produces — the thinnest pipe series ISO 15874-2 permits for each combination of application class and design pressure, with the standard's S_calc,max value beside it.

S_calc,max values from ISO 15874-2:2013, Table A.5 (PP-R). Series selection applies clause A.4: choose a series whose S is not greater than S_calc,max.

The bottom-right cell is the one to notice. Class 5 at 10 bar demands S_calc,max = 1,9, and the thickest series in the dimension tables is S2 — still thinner than the limit requires. Plain PP-R has no compliant answer for high-temperature radiator service at 10 bar. If a supplier quotes one, the quote is outside ISO 15874, and this is precisely the case PP-RCT exists for: its Class 5 design stress is 2,92 MPa against PP-R's 1,89 MPa, which brings the requirement back inside the available range.

One related trap sits in the dimension tables themselves. Table 5 lists series S8 and S6.3, and both carry a footnote reading "Only valid for PP-RCT". Those two columns are not available to PP-R at any class. A datasheet offering PP-R in S6.3 has either mislabelled the material or misread the table.

Wall thickness by diameter

Once the series is chosen, the wall thickness is a lookup. These are the minimum wall thicknesses from ISO 15874-2 Table 5, dimension class A, for the sizes that move in volume.

Minimum wall thickness e_min from ISO 15874-2:2013, Table 5, dimension class A. Published supplier tables occasionally show 8,4 mm at DN50 and 18,4 mm at DN110 in this series; where a datasheet differs from the standard by a tenth of a millimetre, specify to the standard and ask the supplier which figure their tooling actually holds.

Note how quickly the hot duty spends money. A DN50 line in S5 needs 4,6 mm of wall; the same line in S2.5 needs 8,3 mm. That is 80% more wall, which works out to about 66% more polypropylene per metre once you take it round the annulus, and on a container of pipe it is the difference between a competitive quote and a lost tender — which is why the pressure to under-specify the class is commercial, not technical.

How Derated Systems Actually Fail

Under-derating rarely produces a burst on commissioning day. That is what makes it dangerous commercially: the pipe passes its pressure test, the project is signed off, and the consequence arrives years later when nobody connects it to the specification decision.

The mechanism is in the standard's own qualification data. ISO 15874-2 states the reference line for PP-R as log t = −19,98 + 9507/T − 4,11 log σ, where t is time to failure, T is temperature and σ is hoop stress. The shape matters more than the constants: time to failure is logarithmic in stress and temperature, so a modest, permanent overstress does not shift failure from year fifty to year forty-nine. It moves it by an order of magnitude.

Three ways it goes wrong

• Spending the malfunction budget early. The 100-hour allowance at T_mal covers 50 years. An uncontrolled solar thermal loop or a failed mixing valve can burn through it in a single season, after which every further excursion is consuming design life the pipe was never sold.
• Specifying a hot-water class for a heating circuit. Class 2 and Class 5 look adjacent on a datasheet and are not: 2,12 MPa against 1,89 MPa, with Class 5 also reaching 100°C at malfunction rather than 95°C. Pipe bought to Class 2 numbers and installed on high-temperature radiators is under-walled from day one.
• Reading a processing figure as a service rating. ISO 15874-2 Table 11 specifies the longitudinal reversion oven test for PP-R at 135°C. That is a dimensional-stability check on the extrusion, held for one hour on a thin-walled test piece — it is not evidence the pipe conveys anything at 135°C, and it should never appear in a temperature-capability argument.

The recognisable field signature of a class mismatch is failure away from the joints, in the hottest continuously running section, some years in, with the pipe wall showing brittle rather than ductile fracture. Where the same system also shows failures at fusion joints from the first months, the cause is usually installation rather than derating — a different problem, covered in our guide to PP-R installation practice.

What to Require From a PP-R Supplier

Everything above turns into four questions you can put to any supplier, and the answers are checkable against the tables on this page in about two minutes.

• Ask for the class, not just the PN. "PN20" states a 20°C rating and says nothing about service conditions. "PN20, suitable for Class 2 at 8 bar" is a specification you can verify.
• Audit their derating table against Table A.2. If a supplier publishes derated pressures, back-calculate one row. A table that ranks Class 4 below Class 2, or that offers PP-R in S6.3, was not built from ISO 15874-2.
• Check the marking carries the class. ISO 15874-2 requires marking at least once per metre, legible after handling and installation. Pipe whose print gives only a diameter and a PN leaves the site with no way to prove what was installed.
• Separate declared conformity from test evidence. "Manufactured to ISO 15874" is a declaration. A batch certificate, a third-party report from SGS or BV, or independent lab verification of the delivered batch is evidence. Ask which one you are being offered.

The two-minute datasheet check

This is the sequence to run on any PP-R quotation before it goes into a tender file. It uses only figures published in this article, so it needs no access to the standard itself.

1. Find the class. Look for an application class (1, 2, 4 or 5) anywhere on the datasheet. If the document gives only a PN and a temperature, the specification is incomplete — ask for the class before comparing prices.
2. Check the class ranking. If the supplier publishes derated pressures, confirm that Class 4 sits above Class 2. A table that ranks them the other way was built from peak temperature rather than ISO 15874-2 Table A.2.
3. Check the series offered. Any PP-R quoted in S6.3 or S8 is wrong on its face; ISO 15874-2 Table 5 restricts both series to PP-RCT.
4. Match the wall to the table. Take one diameter and compare the quoted wall thickness against the Table 5 figures above. A wall below the minimum for the stated series is the cheapest place for a supplier to save money and the hardest for a buyer to see.
5. Ask what the marking will say. The class, the standard and the dimensions should be printed on the pipe itself, because that is the only evidence that survives to site.

Where the answer to any step is missing rather than wrong, that is not necessarily a bad supplier — but it is an unpriced risk, and it should be resolved in writing before a container is committed.

On the commercial facts a first inquiry needs: IFAN's published PP-R supply range runs DN20–DN160 in PN12.5, PN16, PN20 and PN25, supplied to ISO 15874 and DIN 8077/8078 with a material certificate per batch. Minimum order is structured per container with mixed sizes accepted rather than as a fixed figure per SKU.

Pricing is quoted against a size-and-quantity list rather than published as a rate card, because wall thickness — which is exactly what the class decision changes — is the main driver of the per-metre cost. Regional certification such as SASO, SONCAP or NOM is arranged per destination market, and lead time is quoted per order against the size mix and the certification route, because those approvals can govern the schedule more than production does.

If you are still upstream of this decision, our PP-R pipe size guide covers diameter selection and the PN system, and the complete PP-R guide sets out grades, jointing and where the material fits. For conformity documents specifically, see how we handle PP-R certification.

Specify the Class, Then the Pipe

The practical sequence is short, and following it makes the derating question answer itself.

• Name the application class first from ISO 10508 Table 1 — hot water at 60°C or 70°C, underfloor, or high-temperature radiators — and write it on the enquiry.
• Take the design stress for that class from ISO 15874-2 Table A.2: 3,02 / 2,12 / 3,29 / 1,89 MPa for PP-R.
• Divide by your design pressure and compare against the cold-water result, then pick a series no thinner than the answer.
• Read the wall thickness off Table 5 for the diameters you are buying, and put both the series and the millimetre figure on the purchase order.
• If the answer is Class 5 at 10 bar, stop — no PP-R series complies, and the conversation should move to PP-RCT.

A supplier who can follow that sequence with you is quoting engineering. One who answers with a single derating percentage and no table number is quoting a marketing figure, and page one of the search results shows how far apart those two things can be.

Frequently Asked Questions