PPR for Chilled Water: Where It Works and Where It Doesn’t

Ask a pipe supplier whether PP-R suits a chilled water job and you will get a yes. Every page currently ranking for the question is written by someone who sells the pipe, and not one of them states a condition under which the answer changes. That is a problem for the person who has to sign the specification, because the answer does change, and it changes on a number nobody publishes.

Here is the short version. PN20 PP-R is a sound chilled water material on terminal branches and fan coil tails, and it is the wrong material for mains and risers, because the published range stops at 32 mm outside diameter. Temperature is not what rules it out. At a 6.7 °C supply the pipe sits well inside the envelope the standard writes for cold water. What actually decides a PP-R chilled line is three things its own marketing leaves out: the dew point it runs below, the movement it makes when a hot site cools down, and the diameter you can actually buy.

The Short Answer: PP-R Earns Its Place Downstream of the Riser

A chilled water system is not one piping duty. It is four, and they sit in different size bands. The plant room and the primary mains move the whole building’s flow. Risers carry a stack of floors. Floor mains feed a zone. Terminal branches and fan coil tails serve one or two units each. Read the question duty by duty and it answers itself, because the size band of each duty is what puts PP-R in or out.

Only the last two duties, the 20 to 32 mm bands, fall inside a PN20 PP-R range, and only the last one comfortably. Everything upstream of the floor branch is either outside the range or close enough to its ceiling that you are specifying the largest thing available rather than the right thing. For the full segment-by-segment picture of the loop, the chilled water system field guide walks the four segments in order.

Table 1: PP-R verdict by duty on a chilled water system

Duty on the system	Size band (mm OD)	Verdict for PN20 PP-R	Best for
Plant room and primary mains	Above 110	No. Outside every plastics range we publish	Steel, welded and insulated on site
Risers and floor mains	63 to 200	No. Above the 32 mm PP-R ceiling	Steel above 110, HDPE PN16 below it
Floor branches to a zone	40 to 63	No. Above the 32 mm PP-R ceiling	HDPE PN16, or UPVC and CPVC 806
Terminal branches to fan coils	20 to 32	Yes, with movement and insulation designed	PN20 PP-R, series 1103 pipe and 1138 fittings
Fan coil tails and single drops	20 to 25	Yes. This is the duty it was built for	PN20 PP-R, series 1103 pipe and 1138 fittings

Source: IFANNova catalogue range data, September 2026. Size bands are the duties as designed, not a product claim.

The Service Classes Everyone Quotes Are Hot Water Classes

When a datasheet justifies a PP-R pipe by naming a service class, check which class. ISO 10508:2006, the classification standard the EN ISO 15874 system rests on, sets out five of them in Table 1, and every one is a hot water or heating duty. Class 1 runs at a 60 °C design temperature for 49 years and Class 2 at 70 °C for 49 years; Class 3 covers low-temperature under-floor heating, Class 4 under-floor heating and low temperature radiators, Class 5 high temperature radiators. There is no cold water class. The classification system was built for the wrong direction.

Cold water is handled instead as a blanket requirement laid on top of all of them. In the standard’s own words, any system meeting one of the five classes “shall also be suitable for the transportation of cold water for a period of 50 years at a temperature of 20 °C and a design pressure of 10 bar”. That is the condition a chilled line is actually working against, and ISO 15874-5:2013 settles which one wins: Table 2 of that part carries the footnote that “the 20 °C, 10 bar, 50 years, cold water requirement, being higher, determines this value”.

Two things follow. A hot water service class is not a design basis for chilled water, and citing one to justify a chilled line is a category error, however accurately it is quoted. And at 6.7 °C the pipe is running colder than the 20 °C reference, so creep and hydrostatic performance are more conservative than the standard assumes, not less. The pressure side of this decision is the easy side.

What the class system does not cover is where the risk moved. Cold makes polypropylene notch-sensitive: published PP-R technical data states plainly that cold weather weakens its resistance to impact and it becomes fragile. The current general part of the system standard is EN ISO 15874-1:2013+A1:2022, and Amendment 1, published on 13 June 2022, is titled “Impact test”. A standards committee does not add an impact test to a mature standard for decorative reasons.

Diameter Settles It, and Where Our Range Stops

Our PP-R line is series 1103 pipe with 1138 fittings, rated PN20, supplied in 4 m lengths, in three outside diameters: 20, 25 and 32 mm. That is the whole range, and it is a real ceiling rather than a gap in the catalogue. Across every pressure system we make, the range stops at 110 mm outside diameter, and we do not make DN150 to DN400 mains. We say that rather than sub-contracting them, because a buyer who discovers the substitution at the port has lost more than the price difference.

The number that actually sizes the branch is not the outside diameter. A PN20 wall is thick, and it takes the bore with it.

Table 2: Published PN20 PP-R sizes and the bore each one leaves

Outside diameter (mm)	Wall (mm)	Internal bore (mm)	What it serves
20	2.8	14.4	Fan coil tails and single terminal drops
25	3.5	18.0	Terminal branches serving one or two units
32	4.4	23.2	The largest branch this range reaches

Source: IFANNova catalogue wall thickness, September 2026. Bore is outside diameter minus twice the wall.

A 32 mm PP-R branch gives you 23.2 mm of water. Specify it against a steel schedule that assumed a nominal bore of 32 mm and the velocity will land where you did not put it. Work the flow against the bore column and check it on the chilled water pipe size chart; the full designation and dimension tables sit in the PP-R pipe size tables. Anything above 32 mm on a chilled system needs a different material from the same supplier, which is what our chilled water piping range, from branch fittings to insulation practice, is organised around.

Condensation, Not Heat Ageing, Is What Takes a Chilled Line Down

Hot water piping fails slowly, by thermal ageing, and the service classes exist to predict that. Chilled water piping fails differently, and faster. The standard air conditioning design condition is 6.7 °C supply and 12.2 °C return. The comfort condition those temperatures are chosen to hold, 23.9 °C at 50% relative humidity, has a dew point of roughly 12.8 °C. A 6.7 °C line therefore runs 6.1 K below the dew point of the air around it, and bare pipe of any material will sweat.

Supplier pages like to argue that PP-R’s low thermal conductivity reduces that problem. It does not. PP-R sits at 0.23 W/m·K, which is far better than steel and roughly an order of magnitude worse than the closed-cell insulation that actually does the job. The wall is 2.8 to 4.4 mm thick; the insulation is not. Treating the pipe material as part of the thermal envelope is how a vapour barrier ends up under-specified.

The calculation that matters is in EN ISO 12241:2022, whose clause 4.5 is titled “Prevention of surface condensation”: you size insulation so the outer surface stays above the dew point of the space it runs through. That thickness depends on the space, not on whether the pipe under it is plastic or steel. The failure that follows from getting it wrong is documented in our write-up of vapour barrier failures on chilled water lines, and it is always the barrier, never the pipe.

In a 45 °C Ambient a Chilled Line Moves More Than a Hot One

Thermal movement is filed under hot water problems, which is why it gets missed here. Published PP-R technical data puts the coefficient of linear thermal expansion at 1.5 × 10⁻⁴ per kelvin for plain pipe and 0.35 × 10⁻⁴ per kelvin for the glass-fibre composite version. Movement is the coefficient times the length times the temperature change, and on a Gulf site the temperature change on a chilled line is large, because the pipe is installed hot and then runs cold.

Take a 30 m run closed out at a 45 °C site ambient and then commissioned at 6.7 °C. That is a change of 38.3 K, and the run contracts.

Table 3: Movement on a 30 m run, computed from the coefficients

Pipe	Coefficient (per K)	Chilled, 45 to 6.7 °C (mm)	Hot water, 25 to 60 °C (mm)
Plain PP-R	1.5 × 10⁻⁴	172	158
Glass-fibre PP-R	0.35 × 10⁻⁴	40	37

Source: computed from published PP-R coefficients. The 45 °C installation ambient is an input to this example, not a measured site value.

Plain pipe moves 172 mm on the chilled case against 158 mm on the hot one. The chilled job is the more demanding of the two, and it is the one nobody designs supports for. Substituting the glass-fibre composite drops the same run to 40 mm, which is the honest argument for the composite on chilled water, and a better one than any of the thermal claims usually made for it. Support spacing and anchor placement then follow from the number you chose, which we work through separately for thermal expansion on plastic chilled water lines.

Movement against run length on a chilled PP-R branch, 0 to 40 m. Source and method: ΔL = α × L × ΔT with ΔT = 38.3 K, a run closed out at a 45 °C site ambient and operated at the 6.7 °C chilled-water design supply; α = 1.5 × 10⁻⁴ per kelvin for plain PP-R and 0.35 × 10⁻⁴ per kelvin for the glass-fibre composite, both taken from published PP-R technical data (tier T2, confirmed by fresh fetch on 2026-09-13). Values rounded to the nearest millimetre. The 45 °C installation ambient is a stated input of the worked example, not a measured site value.

Run length (m)	Plain PP-R (α 1.5 × 10⁻⁴ /K)	Glass-fibre PP-R (α 0.35 × 10⁻⁴ /K)
0	0	0
10	57	13
20	115	27
30	172	40
40	230	54

Four Ways Buyers Get This Wrong

• Reading a design condition as a warranty. The 50 year figure is a design condition of 20 °C, 10 bar, 50 years from ISO 10508:2006. It is a basis for sizing the pipe series, not a service life anyone has measured on a chilled system, and it says nothing about the insulation around it.
• Sizing from the outside diameter. A PN20 wall costs real bore: 32 mm outside diameter leaves 23.2 mm inside. A branch sized as though it were a 32 mm bore runs over velocity from the day it is commissioned.
• Specifying plain PP-R where the composite belongs. On a 30 m run in a 45 °C install ambient the difference is 172 mm against 40 mm. A short fan coil tail will not notice it. On a long horizontal branch it decides whether the supports hold.
• Treating cold impact as a storage problem. It is a service condition too. Amendment 1 to EN ISO 15874-1:2013 exists because impact resistance in the cold is a property worth testing, and a chilled branch in a plant room is cold for its whole life. Fusion joints made correctly are the strongest part of that line, which is why the fusion time and temperature table is worth holding the crew to.

The joint is the one part of a PP-R chilled branch that is not in question, and it is still the part a crew can get wrong on site. This is the procedure they are being held to.

Conclusion

PP-R buys you a corrosion-free, fusion-jointed branch that a site crew can install quickly, and it charges you for it in bore, in movement and in a hard ceiling at 32 mm outside diameter. Below that ceiling the trade is a good one, and the pressure case is comfortable. Above it there is no trade to make, because the pipe does not exist in our range and we would rather say so than find you a substitute. Design the insulation and the supports as though the pipe were steel, and the material will do its job.

Frequently Asked Questions