Every machine supplier ships a parameter chart, and they differ slightly. What does not differ is the physics: a PE100 joint is made by melting two faces to the same depth, bringing them together inside a time window, and holding them still while the melt cools. Get any of those three wrong and the chart numbers stop meaning anything.
The four numbers on the chart
| Parameter | Typical value (PE100) | What it controls |
|---|---|---|
| Heating plate temperature | 200 – 220 °C, commonly 210 °C | Melt viscosity at the interface |
| Heat soak time | ≈ 10 × wall thickness (s) | Depth of the molten layer |
| Changeover time | As short as possible, ≤ 10 s (small) / ≤ 8 s (large) | Whether the melt is still fluid at contact |
| Cooling time | ≈ 10 × wall thickness (min) at ambient, longer above 30 °C | Whether the joint is disturbed before it is strong |
The values above are the rules our machines are set to for PE100 at 20–25 °C ambient. Always confirm against the chart supplied with your machine and the standard your project cites — ISO 12176-2 practice is the usual reference.
Worked example: DN315 SDR17
Wall thickness 18.5 mm. Heat soak ≈ 185 s, cooling ≈ 18 min, changeover under 10 s. In practice the operator is watching two things: the bead that forms against the heating plate, and the clock.
The three things that decide the joint
1. The pipe end is clean and round
PE pipe arrives with an oxide layer and handling dirt. Scraping is not cosmetic — fusion bonds to freshly exposed material, not to the surface that has been in the sun since it left the extruder. A dull scraper is worse than no scraper because it smears instead of cutting.
2. The faces are parallel
The planer cuts both faces in one pass, which is the only reliable way to get them parallel. A joint made on faces that are 0.5 mm out of parallel has an uneven bead, and the side with the thin bead is where it will fail under pressure.
3. Nothing moves during cooling
Most joint failures we see are not wrong temperature — they are a joint that was lifted, twisted or put under load while still warm. Cooling time is not negotiable because the pipe "looks set" after a couple of minutes.
What the bead tells you
- Even bead all round, both sides — parameters and alignment were right.
- Bead larger on one side — misalignment or uneven heating plate contact.
- Bead too small overall — plate too cold, soak time too short, or pressure applied too late.
- Bead with a visible groove or burnt ring — plate too hot or contaminated with old material.
Cold weather
Below about 5 °C the pipe surface takes longer to reach fusion temperature and the melt cools faster after changeover. The usual adjustment is a longer soak and a longer cooling time, not a hotter plate. If the machine cannot extend changeover, build a windbreak rather than raising the temperature.
Pressure testing
Do not pressure test a line the same day it was fused if the ambient is cold. Give the joints the full cooling time plus a margin, then test. A joint that was disturbed during cooling usually passes a short test and fails in service — which is the expensive kind of failure.