The short answer first
For chilled water, refrigerant and hot water lines up to about +125 °C, inside a building, where the pipework is complex and the room is humid: flexible closed-cell elastomeric foam is usually the right default, because it handles condensation and geometry without extra parts. Mineral wool is the right answer above that temperature range, where non-combustibility is mandatory, or on large flat surfaces with room to spare. Polyurethane is the right answer where thickness has to be minimal in a tight space, and the surface is flat or easily prefabricated.
If a supplier tells you one material wins on every count, they are selling rather than comparing. All three are used in buildings for good reasons.
At a glance
| Flexible elastomeric (NBR/EPDM) | Mineral / glass wool | PUR / PIR rigid foam | |
|---|---|---|---|
| Typical λ, W/(m·K) | 0.030–0.040 at 0 °C | 0.032–0.045 at 10 °C | 0.022–0.030 at 10 °C |
| Service temperature | roughly −50 to +125 °C (ours: −57 to +125 °C) | up to +250 °C with a suitable binder | roughly −50 to +120 °C |
| Vapour resistance (μ) | high, typically ≥ 3,000; ours ≥ 10,000 | low — needs a separate vapour barrier | high (closed cell, foil faced) |
| Fire behaviour | B-s3,d0 or Class 0, typical | A1 / A2, non-combustible | B-s1,d0 to B-s2,d0, typical |
| On fittings and bends | cut and glued in place, no extra parts | needs pre-formed sections or cladding | pre-formed sections, joints sealed |
| Water absorption | low, typically ≤ 5 % | high — performance collapses when wet | low, closed cell |
The ranges above are typical published figures for each material family, not a test result from any particular manufacturer. Whenever you compare two products, compare their declaration of performance or test report at the same mean temperature — not their catalogue headline.
How to read a lambda figure, because most comparisons cheat here
Thermal conductivity is not a constant. It rises with temperature, and it rises differently for different materials. Our own data shows it plainly: 0.030 W/(m·K) at −20 °C, 0.032 at 0 °C, 0.037 at +40 °C — the same product, 23 % worse at the top of its range than at the bottom.
So when a datasheet quotes 0.030 and another quotes 0.034, the figures may not be comparable at all: one may be at 0 °C, the other at +40 °C, one at 10 °C mean per EN standards, the other at 20 °C. Two consequences:
- On a hot line, size with the λ at the service temperature, not the cold figure. Otherwise the calculation is optimistic exactly where it matters.
- Between material families, always convert to the same mean temperature before believing a difference. A 0.004 W/(m·K) gap between two datasheets often disappears once both are placed at the same condition.
Condensation is where mineral wool loses first
Mineral wool is not a vapour barrier. It is an open, fibrous material, and the vapour barrier has to be a separate sheet — foil facing, a membrane, or a jacket — installed over it and sealed at every joint, penetration and support. Done properly, that system performs. It is the joints that decide, and there are more of them than on a closed-cell product.
The failure mode is also harsher. When mineral wool gets wet, water displaces air in the fibre structure and the thermal performance drops sharply; it does not dry out quickly inside a closed jacket, and the weight of saturated insulation on a ceiling void is its own problem. Closed-cell foam can get wet on its surface without absorbing much, because there are no open channels.
This is why the choice usually follows the duty rather than the price list: on a chilled line in a humid plant room, the vapour barrier is the product, and the material is secondary. On a hot line above ambient, there is no condensation risk at all and the calculation changes completely.
Where glass wool is the better answer
- Above about +125 °C — steam, condensate return, flue-adjacent services, process lines. Elastomeric foam is simply out of its range.
- Where non-combustibility is a code requirement rather than a preference. Mineral wool at A1/A2 is in a different class from any organic foam, and no foam argument will win that comparison.
- Large flat or cylindrical surfaces with space available, where the cost per square metre dominates and the geometry is simple.
- Acoustic performance on ductwork, where mineral wool's open structure is an advantage rather than a liability.
Where polyurethane wins
- Tight spaces. The lowest λ of the three means the thinnest wall for a given duty, which matters in risers, ceiling voids and retrofits where clearance is fixed.
- Cold rooms and low-temperature plant where rigid sections can be prefabricated off site and installed quickly.
- Flat surfaces and vessels, where rigid boards are faster to install than flexible sheet on a large area.
The trade-off is geometry and joints: rigid sections do not follow a bend, so every elbow, tee, valve and flange needs a fabricated piece, and every fabricated piece is a joint that has to be sealed. On a plant room full of small-bore pipe and fittings, that labour often erases the material saving.
Where flexible elastomeric foam wins
- Chilled water, refrigerant and hot water up to +125 °C — the bulk of building services pipework.
- Humid plant rooms and any line that runs below the dew point for part of the year, where the closed-cell wall with μ ≥ 10,000 is the vapour barrier rather than needing one.
- Complex geometry: elbows, tees, valves, flanges, small-bore instrument lines. The material cuts and glues on site, with no fabricated parts.
- Where the same team installs and maintains it, because the failure points are visible and repairable.
What we do not claim: our material is not non-combustible, it is not the answer above +125 °C, and it is not the cheapest per square metre on a large flat duct. Where a project needs those three things, we say so — and on those projects we are the wrong supplier.
The three questions that decide it
Ask these in order, and the material usually picks itself:
- What is the highest service temperature on the line? Above about +125 °C, the choice is already made: mineral wool, and the conversation is about the vapour barrier and the jacketing.
- Is the line below the dew point of the space for any part of the year? If yes, the vapour barrier is the design driver, and closed-cell material makes that barrier part of the product instead of a separate sheet.
- What fire classification does the authority require? If non-combustible is mandatory rather than a bonus, only mineral wool qualifies and the other two are out — regardless of thermal performance.
If the answers are 'under +125 °C', 'yes, humid', and 'B-s3,d0 or Class 0 is acceptable', that is where we work. Send us the line data and we size it.
What a fair comparison needs from both sides
Two things make cross-material comparisons useless: quoting a competitor's λ at a different mean temperature, and quoting a fire class from a different test standard as if the classes were equivalent. When you compare quotes, ask each supplier for the declaration of performance and the test report, and check the mean temperature and the standard on each. If a supplier will not send them, that is the answer.
Questions engineers ask
Can we use elastomeric foam and mineral wool on the same project?
Yes, and it is common. The usual pattern is flexible foam on chilled and hot water pipework inside the building, and mineral wool on high-temperature services, on ductwork for acoustics, or where a code requires non-combustible insulation. What matters is that the transition points are detailed: a change of material is a change of vapour strategy, and the joint between them needs sealing and jacketing rather than being left open.
Is polyurethane always better because its lambda is lower?
Not always, because λ is one input among several. On a straight run in a tight void, the thinner wall is worth real money in clearance and support. On a plant room with dozens of valves and bends, rigid sections have to be fabricated for every fitting and every one of those is a joint to seal, which is where the saving goes. Compare the installed cost of the finished, sealed system rather than the price per metre of material.
Which material is best for a chilled water line in a humid climate?
Whichever material keeps a continuous vapour barrier, and closed cells make that easier because the barrier is the material rather than a separate sheet. Then size the wall on the real ambient condition, not on a table: at 32 °C and 80 % relative humidity the dew point is 28.1 °C, and a wall sized for a temperate plant room will run wet. The thickness calculation matters more than the material choice on those lines.
Do you supply the same specification as the big international brands?
We supply closed-cell NBR/EPDM foam to our own formulation, with the thermal and moisture data published on our product pages and tested per batch. We do not publish comparisons against named brands, because their published values and test conditions are theirs to state. Compare the declarations of performance at the same mean temperature and the same fire standard, and the differences will be visible without either supplier's marketing.