Most crews that get called back to a PP-R job did not weld badly. They welded well, in the wrong place, on a run that was never allowed to move. The joint holds; the pipe pushes the fitting off a bracket, or bows out of a chase, or drags a brass insert until the thread weeps eighteen months later. Fusion is the part everyone practises. The install around the fusion is the part that fails.
This guide covers that envelope end to end: setting out, cutting and marking, the fusion sequence, absorbing thermal movement, supporting and anchoring, testing, and handing over a system somebody else has to maintain. It is written for contractors and MEP buyers working with a matched PP-R pipe-and-fitting range in OD 20–110 mm.
- Movement, not fusion, is the usual cause of a callback. Standard PP-R expands 0.15 mm per metre per kelvin. A 10 m hot run installed at 30 °C and running at 70 °C grows 60 mm — roughly the length of your thumb.
- Fibre-reinforced PP-R cuts that to 0.05 mm/m·K, so the same run grows 20 mm. Specifying the variant is a design decision made before the first cut, not a substitution made on site.
- Never run fusion parameters from memory. Take temperature and times from the chart of the pipe you actually bought. Published PP-R tables to DVS 2207-11 sit around 260 ± 10 °C, but times change with diameter, wall and ambient.
- Fuse in a sequence that leaves joints reachable. Every joint you cannot see at test time is a joint you are trusting blind.
- Test to a written procedure and hand over the record. BS EN 806 puts the test at a minimum of 1.5 × the allowable maximum operating pressure, and EN 806-4 procedure C is the plastic-pipe method.
On this page
- Set out the run before you cut anything
- Thermal movement: the number that decides your bracket plan
- Cutting, chamfering and marking insertion depth
- The fusion sequence, and the parameters you must not invent
- Supports, anchors and guides
- Where PP-R meets metal, valves and appliances
- Inspect before anything gets concealed
- Commissioning, testing and handover
- The end-to-end install checklist
- Frequently asked questions
Set out the run before you cut anything
Set-out on a PP-R job is not the same exercise as on copper. With copper you can push a bend into a run to take up an error. PP-R socket fusion gives you a rigid, homogeneous joint with no adjustment left in it once it has cooled — the pipe and the fitting have become one wall. Every millimetre of tolerance has to exist in the layout, because none of it will exist in the joint.
Start by reading the print line on the pipe. A PP-R pipe carries its outside diameter, its pressure class and the standard it is made to, repeated along the length. That line is the only thing on site that tells you whether the coil in the corner is the PN 20 you specified or the PN 16 that arrived as a substitute.
The difference is real. For a typical PP-R range, PN 20 (SDR 6) is rated 30.9 bar at 20 °C and 10.2 bar at 70 °C, while PN 16 (SDR 7.4) is 24.5 bar and 8.1 bar at the same temperatures, both on a 50-year basis to DIN 8077 Table 9 at a safety factor of 1.25. A heating circuit designed around the first number does not have the margin you think it does if the second pipe went in.
Decide the three things that cannot change later
Before cutting, fix three decisions in the layout, because all three are extremely expensive to revise once fittings are fused:
- Where the fixed points go. A fixed point is a clamp that grips the pipe hard enough to stop it moving axially. It divides a run into sections and decides which way each section grows. Place them first; everything else follows.
- Where the run is allowed to move. Between fixed points you need somewhere for the growth to go — a change of direction, a purpose-made loop, or an axial compensator.
- Which joints stay accessible. Unions and valves need to be reachable for the life of the system, not just until the plasterboard goes on.
A short worked example: a 25 mm PP-R hot-water riser runs 12 m up a service shaft, installed on a cold morning at about 15 °C, serving a system that runs at 60 °C. That is a temperature difference of 45 K. On standard PP-R the run wants to grow roughly 81 mm. If both ends are clamped hard and there is no relief between them, that 81 mm does not disappear — it turns into compressive load in the pipe wall and shear at the brackets, and the run bows sideways in the shaft until something gives.
Thermal movement: the number that decides your bracket plan
This is the single calculation that separates a PP-R install that lasts from one that gets a callback, and it takes about fifteen seconds:
ΔL = α × L × ΔT
ΔL = change in length (mm) · α = expansion coefficient (mm/m·K) · L = run length (m) · ΔT = working temperature minus installation temperature (K)
The coefficient is where the specification decision bites. Standard PP-R has a linear thermal expansion coefficient of 0.15 mm/m·K — a mean coefficient of 1.5 × 10⁻⁴ K⁻¹ across 0–80 °C. Glass-fibre reinforced PP-R (PPR-FB-PPR, the type carrying the identification stripe) comes in at 0.05 mm/m·K, one third of the plain material.
| Run | ΔT | Standard PP-R (0.15) | Fibre-reinforced (0.05) |
|---|---|---|---|
| 10 m | 40 K | 60 mm | 20 mm |
| 12 m | 45 K | 81 mm | 27 mm |
| 6 m | 30 K | 27 mm | 9 mm |
| 3 m cold branch | 10 K | 4.5 mm | 1.5 mm |
The 10 m / 40 K row is the worked comparison published in PP-R technical data: 60 mm against 20 mm on identical geometry. The other rows apply the same two coefficients to lengths you meet on real jobs. Note the last row — a short cold branch moves so little that the whole argument evaporates. This is why blanket rules like “always use expansion loops on PP-R” waste money on cold work and why blanket rules like “PP-R is fine, it’s flexible” put callbacks into hot risers.
Where the movement goes
Once you know ΔL, you give it somewhere to go. In order of preference on a building services job: use a change of direction that already exists, add a purpose-made expansion loop, or fit an axial compensator where there is genuinely no room.
A flexible leg or loop is sized from the diameter and the calculated movement, in the form Ls = K × √(d × ΔL). The constant K is manufacturer-specific — one published PP-R system uses 15 for metric sizing — so take it from the chart of the pipe you bought rather than from a number you remember from another brand. Getting K wrong makes a loop that looks right and is too stiff to absorb anything.
Cutting, chamfering and marking insertion depth
Three preparation steps decide whether the joint you are about to make is sound, and all three take seconds. Skipping them is the most common reason a joint that looked fine leaks at test.
Three steps that take seconds and decide the joint
Cut square. Use a proper PP-R shear or wheel cutter, not a hacksaw. An out-of-square cut means the pipe bottoms in the socket on one side while a gap remains on the other, and that gap becomes a void in the weld. If you are working near the top of the range, an out-of-round pipe end is worth re-rounding before it goes anywhere near a heater.
Chamfer the outside edge. Published PP practice is an external chamfer of approximately 15°, 2 mm wide for diameters up to 50 mm and 3 mm for larger sizes. The chamfer stops the leading edge of the pipe scraping molten material off the socket wall as it is pushed in — that scraped material is what forms an internal bead big enough to restrict a 20 mm branch.
Mark the insertion depth. Measure the socket depth on the fitting, transfer it to the pipe and mark it. The mark is what tells you the joint went fully home and, just as importantly, that it did not go too far. Both errors are invisible once the joint cools.
Clean both surfaces immediately before welding. Dust, plaster, PVC swarf and hand grease all end up in the fusion zone. Cleaning is done directly before welding, not at the start of the shift.
The fusion sequence, and the parameters you must not invent
Socket fusion heats the outside of the pipe end and the inside of the fitting socket at the same time, then pushes them together so that the two melts become one wall. There is no gasket, no O-ring and no separate sealing element inside a correct PP-R socket joint — which is exactly why the parameters matter so much. There is nothing else in there to compensate for a bad weld.
The sequence itself is short: bring the tool to temperature and verify it with a surface or infrared thermometer rather than trusting the indicator lamp; push the fitting onto the spigot to the stop and the pipe into the bush up to your mark; hold for the warming time; withdraw both in one movement; push them together to the mark without twisting or tilting; then hold the joint still while it sets.
Do not twist the joint as you push it home. Rotating the pipe into the socket shears the melt instead of fusing it, and it produces a joint that passes a short pressure test and fails under thermal cycling months later. Straight in, hold, let go only when the fixing time has elapsed.
What published tables actually say
Hitze does not publish a size-by-size socket-fusion chart, so this article will not print one as though it did. What can be shown is a clearly identified third-party benchmark. The figures below are SIMONA’s own published values for their PP-H and PP-R products, tabulated to DVS 2207-11 as at 2023, at 20 °C ambient with moderate air movement:
| OD (SDR 11) | Warming | Change-over max | Hold fixed | Total cooling |
|---|---|---|---|---|
| 20 mm | 5 s | 4 s | 6 s | 2 min |
| 25 mm | 7 s | 4 s | 10 s | 2 min |
| 32 mm | 8 s | 6 s | 10 s | 4 min |
| 50 mm | 18 s | 6 s | 20 s | 4 min |
| 63 mm | 24 s | 8 s | 30 s | 6 min |
Read three things out of that table rather than copying the numbers onto your wall. First, the heated tool sits at 260 ± 10 °C and does not change with size — only the times do.
Second, warming time rises steeply with diameter, from 5 seconds at 20 mm to 24 seconds at 63 mm, while change-over stays tight throughout: you have 4 to 8 seconds to get the parts off the tool and together, which is why a fumbled 63 mm joint is nearly always a cold joint. Third, the “total cooling” column is when the joint may be loaded by other laying operations — 2 minutes at 20 mm but 6 minutes at 63 mm. Crews that hang a big line off a fresh 63 mm joint at two minutes are stressing a weld that is still soft.
Two structural warnings come out of the same source. At 63 mm and above, a welding jig is required for PP — hand-holding a large joint square for 30 seconds is not realistic. And thin-wall PP-R at SDR 17 is marked “not to be recommended” for socket fusion below 63 mm because the wall thickness is inadequate; if a thin-wall pipe turns up on a job specified for socket fusion, that is a supply problem to resolve before welding, not a technique problem to work around.
Ambient matters too. Those figures are for 20 °C with moderate air movement. Cold, wind and strong sun all change the outcome, and published PP practice is to protect the weld area below +5 °C and to shield pipe that the sun has heated unevenly. For the full parameter discussion, machine setup and defect catalogue, see our detailed guide on how to weld PP-R pipe and the comparison of socket, butt and electrofusion methods by size.
Supports, anchors and guides
Three different components get called “a bracket” on site and they do three incompatible jobs. Getting them mixed up is how a correctly calculated expansion allowance ends up doing nothing.
| Component | What it does | Where it goes | Failure if wrong |
|---|---|---|---|
| Fixed point / anchor | Grips the pipe so it cannot move axially; splits the run into defined sections | Ends of a section, and adjacent to heavy items like valves | Two anchors with no relief between them: the run bows or loads the joints |
| Guide | Lets the pipe slide lengthways but keeps it in line | Between anchors, along the moving section | Clamped too tight and it becomes an accidental anchor |
| Support / hanger | Carries weight and stops sag between points | At the interval given by the pipe manufacturer’s chart | Too far apart on a hot line: visible sag between brackets |
Why this article gives you no spacing number
Support intervals for PP-R depend on four variables at once: outside diameter, SDR or wall thickness, whether the pipe is fibre-reinforced, and the water temperature. Hot lines need closer support than cold ones in the same size, because the material softens as it warms. Published support-interval tables exist from pipe manufacturers, but they are specific to that manufacturer’s product, and We transcribed one manufacturer’s full set into our guide to PP-R support spacing and thermal expansion, with the source named on every table.
So take the interval from the chart supplied with the pipe you actually bought, at the design water temperature, not at room temperature. A number remembered from a cold-water PVC job will be too generous for a 60 °C PP-R line every time. If your supplier cannot give you a spacing table against temperature, that absence tells you something about the supplier.
Two rules do hold regardless of chart: closer supports reduce sag but do not reduce total growth — the run still expands by ΔL and still needs somewhere to put it — and a support directly against a fitting concentrates load exactly where the wall section changes. Support the pipe, not the joint.
Where PP-R meets metal, valves and appliances
The fused side of a PP-R system rarely leaks. The threaded side is where callbacks live, because a brass-insert transition fitting brings a completely different failure mode into an otherwise homogeneous system: a thread that can be overtightened, a sealant that can be under- or over-applied, and a thread standard that can be quietly mismatched.
Three sequencing rules save most of the trouble. Make the threaded joint first, then fuse — trying to spin a threaded connection tight while it is already fused into a rigid run twists the fusion joint you just made. Support the fitting body while tightening, so the torque goes into the thread and not into the pipe. And put a union or a serviceable joint at every appliance, because a boiler or a water heater will be replaced long before the pipework is, and a fused-in connection turns a two-hour swap into a cutting job.
Confirm the thread standard before you order
Thread standards deserve a specific warning. A tapered BSP thread to ISO 7-1 and an NPT thread will start together and feel tight, then weep under thermal cycling because the thread forms and taper angles do not match. Confirm which standard the metal side of the job uses before ordering the transitions, not on the day. Hitze supplies brass-insert metal transition fittings from 1/2″ to 2″; our full treatment of the subject is in the guide to connecting PP-R to metal pipe, which covers the four transition SKU families, thread standards and the overtightening failure in detail.
Inspect before anything gets concealed
There is a window on every PP-R job between the last joint being made and the first chase being filled. Everything you did not check in that window becomes a demolition job later. Walk the run with a torch and look for five things:
- An even bead all round each socket. A bead that is fat on one side and absent on the other means the pipe went in at an angle, or the parts were not fully seated.
- The insertion mark sitting where it should. Visible mark well clear of the socket means the joint is short; the mark buried far inside means it was forced past the stop and the bore may be restricted.
- Scorch marks or deformed sockets, which point to a tool running hot or a part left on the heater too long.
- Pipe touching structure at the wrong point. Anywhere the run rubs a joist edge or a duct is a wear point once thermal movement starts cycling it.
- Anchors and guides actually installed as drawn, not improvised because the right clamp was not in the van.
The reason inspection carries so much weight on PP-R specifically is visible in that cross-section. On a compression or push-fit system, a marginal joint has a seal that may still hold. On a fused joint the weld is the seal. It is either continuous or it is not, and once it is behind a wall you will find out which by the stain on the plaster.
Commissioning, testing and handover
A pressure test is not a formality at the end — it is the only evidence you will ever have that the system was sound when you left it. Two things make it defensible: testing to a named procedure, and writing down what happened.
For potable installations, BS EN 806 sets the test pressure at at least 1.5 times the allowable maximum operating pressure (PMA). EN 806-4 Section 6 gives procedure C as the hydrostatic method for plastic piping systems, and it judges watertightness on how much pressure is lost across a defined sequence of timed intervals rather than on a single glance at a gauge. The interval lengths and the permitted drop at each one are set out in the standard text itself, which is paywalled — take them from your own copy of BS EN 806-4 rather than from any secondary summary, this page included. That staged structure exists because plastics creep: pressure falling slightly as the pipe relaxes is expected behaviour, and a test design that does not account for it either fails good systems or passes leaking ones.
Four practical points that decide whether a test result means anything. Let the joints reach full cooling before pressurising — the “total cooling” column above is the minimum before a joint is loaded, and a test is a load. Fill and vent properly, because trapped air compresses and masks a slow leak. Watch the temperature: EN 806-4 requires the material derating factor fT to be taken into account if the system’s equilibrium temperature is above 25 °C, so a test run in a hot plant room is not the same test as one run in a cold shell.
And test in sections that match how you built it, so a failure tells you where to look.
What to hand over
The handover pack is what turns your work into something maintainable, and it costs almost nothing to assemble as you go:
- The test record — date, section, test pressure, start and end readings at the defined intervals, ambient temperature, and who witnessed it.
- Marked-up drawings showing where fixed points, guides and expansion provisions actually ended up, which is rarely exactly where they were drawn.
- The pipe and fitting identification — manufacturer, variant (plain, fibre-reinforced, aluminium composite), size and pressure class per circuit, so a future repair uses matched components.
- The manufacturer’s own fusion and support charts for the material installed, so whoever alters the system in five years works to the same parameters you did.
- Certification scope for the material, where the job requires it for sign-off.
What we check before a PP-R batch leaves the plant
That last item is worth a word on how to ask for it. A credential covers a defined scope, and reputable suppliers will tell you what that scope is — and what was measured to earn it. On the Hitze PP-R line the documented factory sequence is: resin verified before extrusion; outside diameter, wall thickness at multiple points and ovality checked against the S-series tables; and pipes, fittings and fused joint assemblies pressure-tested on in-house hydrostatic benches against DIN and EN ISO thresholds. That last step matters to an installer specifically — it means the joint geometry, not just the pipe, was proved under pressure before the product shipped.
Hitze’s PP-R range holds SKZ testing in Germany on both PP-R pressure pipe and PP-R fittings, a DVGW type examination for PP-R drinking-water pipe, and an NSF laboratory BS 6920 report for hot (≤65 °C) and cold drinking-water contact. When you request documents for a submittal, ask which models each certificate covers and for the certificate reference itself — a supplier who names a scope but cannot say which SKUs it applies to has told you something useful. Our PP-R specification and submittal checklist sets out what a complete pack contains, and the Hitze PP-R pipe and fittings range lists the sizes and classes those documents cover.
The end-to-end install checklist
One pass through the whole job, in order. Print it, or work down it on a phone.
| Stage | Check | Go deeper |
|---|---|---|
| Material check | Print line matches the spec: OD, pressure class, standard, variant | Sizes, SDR and PN ratings |
| Set-out | Fixed points, movement relief and serviceable joints all located before cutting | — |
| Expansion | ΔL calculated per section at the real design ΔT, relief provided | Hot-water temperature ratings |
| Preparation | Square cut, 15° chamfer, depth marked, both faces clean | Choosing a cutter by size |
| Fusion | Tool temperature verified; times from the manufacturer’s chart; jig at 63 mm+; no twisting | Full welding procedure |
| Transitions | Thread standard confirmed; threaded first then fused; unions at appliances | PP-R to metal connections |
| Support | Anchors, guides and hangers installed as three distinct things, spacing per chart at design temperature | — |
| Inspection | Every joint seen before concealment: bead, mark, scorching, contact points | — |
| Test | Cooling complete, system vented, named procedure, readings recorded at the intervals | EN ISO 15874 service classes |
| Handover | Test record, marked-up drawings, material identification, manufacturer charts | Submittal checklist |
Who this sequence is for, and who it is not for
It fits contractors running PP-R risers and distribution in residential, hotel and light commercial work in OD 20–110 mm, and MEP buyers who need one document a crew can follow. It is not sufficient for industrial process pipework, compressed air or chemical service, where the material selection and joint qualification are a different exercise; nor does it replace a manufacturer’s own installation manual for the specific product on your job, which always wins where the two differ.
Conclusion
PP-R rewards planning and punishes improvisation, because the joint you make cannot be adjusted afterwards. The crews who never get called back do the same four things: they calculate ΔL for every section instead of assuming the pipe will cope, they take fusion parameters from the chart that came with the pipe rather than from memory, they look at every joint before it disappears, and they hand over a written test record. None of that is difficult. It is just done before the work rather than after the complaint.
One practical note on procurement that affects all four: mixing brands across a system makes every one of those steps harder to defend. Different manufacturers publish different fusion times, different support intervals and different expansion constants, and a warranty on a matched system does not extend to a joint made between two suppliers’ parts. Buying pipe, fittings and transitions from one range is not a loyalty question — it is what makes the charts you are working from apply to the joint in front of you.
Frequently asked questions
How long before I can pressure-test a newly fused PP-R system?
Wait for full cooling on the largest joints, not the smallest. Published PP-R tables give total cooling of 2 minutes at 20 mm rising to 6 minutes at 63 mm before a joint may be loaded, and a pressure test is a load. Use the manufacturer’s chart for your pipe.
What temperature should a PP-R fusion tool be set to?
Published PP socket-fusion practice to DVS 2207-11 puts the heated tool at 260 ± 10 °C, and the temperature does not change with pipe size — only the times do. Verify it with a surface or infrared thermometer rather than trusting the machine’s indicator lamp.
How far apart should PP-R pipe supports be?
There is no single figure. Spacing depends on diameter, wall thickness, whether the pipe is fibre-reinforced, and water temperature — hot lines need closer support than cold ones in the same size. Take the interval from your pipe manufacturer’s chart at the design temperature.
Do I need expansion loops on every PP-R run?
No. Calculate ΔL = α × L × ΔT per section first. A 3 m cold branch at 10 K moves about 4.5 mm on standard PP-R and usually needs nothing special, while a 12 m hot riser at 45 K moves about 81 mm and certainly does.
Can I socket-fuse thin-wall PP-R pipe?
Not in small sizes. Published PP-R tables mark SDR 17 as not recommended for socket fusion below 63 mm because the wall thickness is inadequate for the joint. If thin-wall pipe arrives on a job specified for socket fusion, treat it as a supply problem.
What is the minimum order for PP-R pipe and fittings for a single job?
Hitze has no minimum order quantity, and samples are available for validation before committing to a range. Lead time and any customisation-specific constraints are confirmed per quote rather than published, since they depend on the size mix.
Can I mix PP-R fittings from a different brand?
Mechanically it often appears to work, but socket depths, tolerances and published fusion times differ between manufacturers, and a matched-system warranty does not cover joints made between two suppliers’ parts. Keep pipe, fittings and transitions within one range.
