Pertex method note

Thermolite vs. Primaloft: The Four Numbers I Check Before Signing a PO

2026-11-10 · Sahana Iyer

What I'm comparing, and what I'm not

I've run fabric and trim procurement for a 240-person workwear brand for seven years — roughly $3.4 million a year across insulation, shell fabrics, linings, and trim. Before that I bought packaging for a much larger company, where the volumes were big enough that a 2% swing actually moved the P&L and small enough that I could still visit every plant. That's probably why I ended up in textiles: the specs are harder.

This is not a “which one is warmer” article. Both Thermolite and Primaloft are legitimate insulations, and the answer to which is warmer depends entirely on thickness, basis weight, batting construction, and which tier of each product you're quoting. What I can actually compare — because I've had to sign for it — is delivered warmth per dollar at a given thickness. That breaks into four dimensions I now run on every insulation RFQ:

  • Clo at a stated thickness and basis weight, not a marketing number
  • Performance after wetting, and how fast it dries
  • Loft retention across real wash cycles
  • Landed cost per clo at our actual order size

In 2019 I learned that the fourth one doesn't mean what most people think it means. My first insulation sign-off was a Thermolite batting quoted 22% under a Primaloft equivalent — same face weight, same thickness, same quilt spacing on paper. Three months later we re-cut 1,400 panels because the first article wouldn't recover loft after the wash test. We were out $9,600 and three weeks. That was the first time I read the construction spec instead of the price column.

Dimension 1: warmth per ounce vs. warmth per millimeter

Thermolite is a hollow-core polyester fiber family — the air trapped inside the filament does the insulating, which is why the pitches tend to lead with weight. The Reactor Plus liner is a decent illustration: it's built on Thermolite specifically because a liner has to add warmth inside a sleeping bag without adding bulk that changes how the bag fits.

Primaloft is a microfiber insulation sold in tiers, and its upper tiers are marketed hard on water resistance. Some of that is a real, testable property. Some of it is copy.

The number to compare both on is clo, measured on a hot plate. ASTM D1518 is the test method I ask for. The trap is that a vendor can quote a perfectly honest clo number at a thickness you can't use. I've had two spec sheets side by side where one quoted clo at 20 mm and the other at 12 mm, and the “warmer” material turned out to be the thinner, denser one at a comparable gauge. When I normalized both to the thickness our pattern could actually fit, the ranking flipped.

Ask for clo at three thicknesses. If a supplier can only give you one, that tells you something useful.

Rule I use: never compare two insulations at different thicknesses, no matter what the spec sheet says. Convert first, then compare.

Dimension 2: wet performance — the dimension that matters less than you'd think

Everyone asks about water resistance in the first meeting. In my experience it's usually the last thing that should decide the spec, because it depends entirely on where the insulation sits.

In a sleeping bag liner, the liner lives inside the bag's shell. Rain never touches it. What does touch it is your own perspiration over eight hours, and then a washing machine. So the property I actually want to test on a Reactor Plus-class liner isn't hydrophobic treatment — it's how fast the batting dries after a wash. A liner that comes out of the machine damp and stays damp in a stuff sack is a problem regardless of how the fibers were treated at the mill.

In insulated bibs and boot liners, the calculus flips. Boot liners sit against a sweaty foot and pack out under compression. Insulated men's denim overalls get worn in slush and then washed hot and dried hot. There, moisture management and compression recovery matter far more than dry-state clo.

So if you ask me, the real question isn't whether Thermolite beats Primaloft when wet. It's where in the garment the batting sits and what it touches. Answer that first and the material shortlist usually writes itself.

Dimension 3: what's left after twenty washes

Workwear gets washed hot and dried hot. Sleep systems get washed less often but stuffed hard. Neither is gentle.

The test I run now is boring on purpose: 20 home launderings at the care-label temperature, tumble dry low, measure loft at 0, 5, and 20 cycles, then re-check clo at the end. I log those numbers in the same spreadsheet I use for freight, which is not where anyone expects them to live, but it means I can pull them the next time a vendor claims “no loft loss.”

What I've learned doing this is that the fiber name explains maybe 40% of the result. The rest is construction — resin-bonded versus stitch-bonded versus thermally bonded batting, whether there's a scrim, and how far apart the quilt lines run. A well-built Thermolite batting beat a badly specified Primaloft one in our 2021 trial by a wide margin. The reverse has also happened. Anyone who tells you the fiber logo decides the outcome has never had to explain a re-cut to a CFO.

Dimension 4: landed cost per clo

This is the one I care about most and the one that's easiest to fake.

Unit price is maybe 60% of the story. The rest is MOQ, lead time, freight, duty, usable width, and the cost of being wrong.

Concrete example from last season. Vendor A quoted $4.20/m for an insulation at 150 cm wide, MOQ 500 m, six-week lead, scrim included. Vendor B quoted $3.30/m — 21% cheaper — at 140 cm, MOQ 3,000 m, twelve-week lead, scrim quoted separately. We needed 1,800 m. B's unit price wins. B's order loses, because we'd have carried 1,200 m of a colorway we might never run again, and the write-off risk priced in at roughly the value of the discount. Three of the last four times I've run this comparison, the lower unit price lost on landed cost.

Then there's the timing problem. In August 2023 our mill announced a 7% increase with three weeks' notice. Normally I'd run three quotes and a wash trial before committing. There was no time. I locked 2,600 m with the incumbent at the old price, based on trust and two prior trials. It probably saved us around $1,900. I never got to prove it, which is the part that still bugs me.

The specs that show up in the same bill of materials

Two questions land in my inbox often enough that I've written canned answers.

“What is rayon fabric made from?”

Rayon is regenerated cellulose. In practice that means dissolving-grade wood pulp — usually eucalyptus, beech, or pine, sometimes bamboo — broken down chemically and extruded into filament. It's not a synthetic in the petrochemical sense, and it's not a natural fiber either. Modal and lyocell are cousins with different process chemistry.

Why a procurement manager cares: the label has to be right. Under the Textile Fiber Products Identification Act, fiber content has to be disclosed accurately, and the FTC has brought enforcement actions against companies marketing rayon textiles as “bamboo” on the strength of the raw material alone. If a supplier's spec sheet says bamboo viscose, ask what the hangtag will say. Verify the current rules at ftc.gov — labeling guidance gets updated.

When rayon shows up as an overall lining, I also ask for pulp origin and a chain-of-custody certificate. FSC is the one most mills can produce on request. CanopyStyle is the one the larger brands ask for. If a mill can't produce either, I note it and move on rather than making it a deal-breaker — for a lining at our volume, it rarely moves the price.

The neoforge vs. fabric performance question

We keep an internal comparison sheet that got nicknamed after its two columns. One side is molded, heat-formed insulation units — one piece, no quilting, no baffles. The other is conventional construction: shell, batting, scrim, quilt lines. The question in that sheet isn't which approach is better in the abstract. It's which one still holds its clo after 20 washes, at our thickness budget, at a MOQ we can absorb. So far the molded units win on consistency and lose on flexibility, which is roughly what you'd expect from a process with less variance built into it by design.

So which one do you spec?

Here's how I'd call it, knowing it depends on your line:

  • Sleeping bag liners and other tight-thickness applications — Thermolite's advantage is warmth at low loft, and Reactor Plus is built around that constraint for a reason. Test dry speed first, clo second.
  • Boot liners and footbed-adjacent insulation — the fiber name matters less than compression recovery. Spec the batting density and demand a packed-out loft test.
  • Insulated men's denim overalls — the denim shell drives the decision. Quilting through 12 oz denim caps your loft, so you're buying warmth per millimeter, not per ounce. Get both vendors to quote at your actual quilt spacing.
  • Exposed mid-layers where moisture actually reaches the insulation — this is where Primaloft's upper tiers earn their price. Pay for it there and only there.

You don't need me to pick for you. You need four numbers per candidate: clo at your thickness, dry time after wash five and wash twenty, loft retention at twenty, and landed cost per clo at your real order size. Ask for the first three in writing from every vendor, then compute the fourth yourself — because nobody is going to compute it honestly on your behalf.

Prices and quotes cited are from our own records and are for illustration only; your landed cost will differ by volume, width, and freight lane.

Sahana Iyer
Sahana Iyer

Sahana Iyer is an interior-textile analyst covering upholstery, curtain, blackout, chenille, velvet, jacquard, bedding, hospitality linen, awning, and outdoor-furniture fabrics. She applies ISO 12947 Martindale methods and ISO 105-B02 lightfastness testing while comparing abrasion cycles, pilling, colour change, seam slippage, pile direction, dimensional stability, usable width, stain response, and dye-lot variation. Her material guides help interior designers, contract specifiers, furniture makers, and hospitality buyers connect test conditions with traffic level, window exposure, cleaning regime, appearance retention, and replacement planning.