Industries / Mining and materials / Insulation manufacturing
Performance testing
Performance testing is measuring how well an insulation product slows the flow of heat, in the laboratory and over the product's life, and reporting the result on labels and certificates.
6
Milestones achieved, from 1912 to 2015.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1912
A plate guarded by heat
A guarded hot plate measures how much heat passes through a slab of insulation: a heated plate warms the slab from one side, and a separately heated guard around it keeps heat from leaking sideways, so the measured power is the heat that went through the sample. Hobart C. Dickinson built the first one at the US National Bureau of Standards (NBS, now NIST) in 1912 for the refrigeration industry, and in 1916 he and Milton S. Van Dusen published measurements of air spaces and 30 insulating materials. NIST's own history says Richard Poensgen in Germany had developed the method independently and used it since 1910, and is generally credited as its first developer.
Credit
Richard Poensgen; Hobart C. Dickinson and Milton S. Van Dusen, NBS
1945
One method for every lab
NBS researcher Henry E. Robinson and others produced a draft test method in 1942, and in 1945 ASTM adopted it as Standard C177, which NIST calls the cornerstone of insulation testing in North America. One of the first published comparisons between laboratories showed they needed reference samples to calibrate against, so NBS began supplying measured insulation samples; by 1977 more than 300 laboratories had used them, and from 1979 to 1987 NBS and the US Department of Energy characterized reference boards of fibrous glass and fumed silica.
LessonA shared method was not enough until every lab could check itself against the same sample.
Credit
Henry E. Robinson, NBS; ASTM
1979
R-values on every label
R-value is a number for resistance to heat flow: the higher it is, the better the product insulates. The US Federal Trade Commission's R-value Rule (16 CFR Part 460), issued on 27 August 1979, requires that R-values in labels, fact sheets and ads for home insulation come from named tests. Today these are the guarded hot plate (ASTM C177), the heat flow meter (C518), the hot box (C1363) or the thin heater (C1114), run at a mean temperature of 75 °F. The rule's current test section, from a revision published on 13 May 2019, says polyurethane, polyisocyanurate and extruded polystyrene foams must be tested on samples that fully reflect aging, and loose-fill cellulose at its settled density.
Credit
US Federal Trade Commission
1980
Thick insulation, bigger plates
The 1970s energy crisis raised demand for thick insulation, and the FTC grew concerned that test standards built on 25 mm specimens were not protecting consumers. Its 1980 ruling on insulation labels sped up NBS work on a one-meter line-heat-source guarded hot plate, which heats with circular elements at set positions instead of heaters spread across the plate; it was finished near the end of 1980. Industry later told Congress the better accuracy saved consumers $90 million a year in insulation costs.
LessonA method proven on thin samples had to be rebuilt once the product got thicker.
Credit
M. H. Hahn, B. A. Peavy, F. J. Powell and colleagues at NBS, building on a 1964 design by Henry E. Robinson
2003
Foam aged before labeling
Foam boards lose R-value over time, because air diffuses into their cells and the gas they were blown with diffuses out. In 2003 North American polyisocyanurate (polyiso) makers began publishing long-term thermal resistance (LTTR), predicted by measuring thin slices cut from a board, which age faster than the whole board, under CAN/ULC-S770 or ASTM C1303. The same year their trade group, PIMA, launched QualityMark, a third-party certification of LTTR values administered by FM Global. When the method was updated, the design value for permeable-faced polyiso fell from R-6.0 to R-5.7 per inch for values reported from 1 January 2014, with the products unchanged.
LessonAn accelerated aging test is itself a prediction, and revising it changed the published number.
Credit
Polyisocyanurate Insulation Manufacturers Association (PIMA); FM Global
2015
Walls measured, not assumed
BRE, a UK building research organization, measured the U-value (heat lost per square meter for each degree of temperature difference; lower is better) of about 300 walls in English homes in early 2012, with heat flux plates fixed to the walls as ISO 9869 describes, for the Department of Energy and Climate Change. Uninsulated solid brick walls averaged 1.57 W/m²K, against the 2.1 W/m²K assumed in RdSAP, the method behind home energy ratings. The report, published in March 2015, warned that national insulation programs would then save less than expected.
LessonThe savings promised for insulation depend on the starting value assumed for the wall.
Credit
BRE, for the UK Department of Energy and Climate Change
Watch for
Wet R-value for buried foam
PIMA says that in tests by a third-party laboratory it does not name, polyiso continuous insulation kept 95% of its labeled R-value after 28 days under water, measured wet in a heat flow meter (ASTM C518) with no drying time, after a soak based on ASTM D2842 extended to 28 days. It says polyiso did significantly better than expanded polystyrene (EPS) and about as well as extruded polystyrene (XPS), and calls older short-soak, dry-state methods outdated for foam used below ground.
Achieved whenthe full data are published and a laboratory not paid by a foam maker repeats the 28-day test on retail boards of all three foams.
Claimed by
Michael Rhodes, PIMA, July 2026
Open
01
Real-time foam aging record
A public record, updated monthly, of the R-value of full-thickness polyiso and XPS boards from several makers, measured as they age at room conditions, set beside each product's published LTTR.
The briefCredit · unclaimed02
Wet R-value round robin
The same 28-day soak-and-measure test for foam used below ground, run by at least three independent laboratories on boards from the same retail batches, to show how far the labs' results differ and whether the foams' ranking holds.
The briefCredit · unclaimed
Open 01 · the brief
Real-time foam aging record
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public record, updated monthly, of the R-value of full-thickness polyiso and XPS boards from several makers, measured as they age at room conditions, set beside each product's published LTTR.
Why it is still open
LTTR is a prediction for five years, made from thin slices that age faster than a full board, and the prediction method was revised in 2014. Field checks so far are snapshots of older boards: in a 2006 study for the National Roofing Contractors Association, 17 of 20 polyiso specimens aged under five years fell below their LTTR, RDH Building Science measured boards from roofs in British Columbia in 2009 and 2015, and the National Research Council Canada tested 11 samples taken from roofs after 13 to 31 years, with no clear link between age and R-value because blowing agents and makers varied. An EPS maker argues that polyiso and XPS keep losing R-value after year five. No public month-by-month record of whole boards on sale today, blown with the gases now in use, was found.
Built with what exists
A commercial heat flow meter apparatus built to ASTM C518, checked against a NIST fibrous-glass reference board; boards bought at retail; a temperature and humidity logger; a public data repository such as Zenodo.
First step
Buy two boards each of three polyiso and three XPS products, measure them in the first week, and publish the readings together with the meter's result on the reference board.
Done looks like
At least six products from at least three makers, each measured monthly on full-thickness specimens for 24 months, with raw readings and calibration checks public, and each product's 24-month value compared with its LTTR.
Open 02 · the brief
Wet R-value round robin
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
The same 28-day soak-and-measure test for foam used below ground, run by at least three independent laboratories on boards from the same retail batches, to show how far the labs' results differ and whether the foams' ranking holds.
Why it is still open
The only recent results found come from one laboratory that PIMA, a polyiso trade group, does not name, and the article gives no figures for EPS or XPS. The same article says the US design standard ASCE 32 relies on assumed R-values for polystyrene taken from older testing.
Built with what exists
Water tanks and weighing for an ASTM D2842 soak extended to 28 days; heat flow meters to ASTM C518 in university building-science laboratories; retail polyiso, EPS and XPS boards; a shared written protocol and a public results sheet.
First step
Ask PIMA for its Technical Bulletin 603, write the shared protocol from it, and invite two university laboratories that own heat flow meters to join.
Done looks like
Three laboratories, three foam types, five specimens each, with the retained share of labeled R-value for every specimen and the spread between laboratories published.