Fibreglass dust in your house · every number, one page
How much fibreglass dust ends up in your house, and how it gets there.
I'm Peter Johnson, a Brisbane insulation contractor since 1986. Batt companies scare people about Cellulose Fibre Insulation being loose in a roof — but the respirable loose fibre in fibreglass batts is substantially more than cellulose's tiny 0.1%. This page puts every exposure pathway, every kilogram figure and every dust reading I can source in one place, with the document behind each one linked beside it.
~2–4 kg
of loose respirable fibre, average roof
Every hole is a hole in both directions
The openings between your roof and your rooms — and there are more of them than a downlight.
Everybody talks about downlights, exhaust fans and range hoods, and they are real openings. A downlight cut-out is a hole. A bathroom exhaust fan is a hole with a motor in it. A range hood duct is a hole running straight up over the spot you cook and eat. But that's not the whole list, and the rest of it gets skipped every time.
Wiring runs down from the roof space to every light switch and every power point in the house, and each of those is a hole drilled through the top plate for the cable to pass. Where a plasterboard ceiling meets the brickwork in a corner, there's routinely a gap that was never sealed. And behind your fridge, there's a gap that vents fridge heat and, in plenty of kitchens, microwave and cooktop heat straight up into the roof space — with your cooking sitting directly underneath it.
Downlights
A cut hole straight through to the roof space, popped open every time a globe goes.
Exhaust fans & range hoods
A hole with a motor in it, or a duct running straight up over where you cook.
Wiring holes
Every light switch and power point has a cable running back up to the roof through a drilled hole.
Corner gaps & the fridge vent
Unsealed gaps against brickwork, and the gap behind the fridge that vents straight up — with your cooking underneath it.
Modern LED downlights fix half the problem, not all of it
An IC-4-rated LED downlight runs cool enough to have insulation laid straight over the top, which is a genuine improvement on the old halogen clearance problem. What it doesn't fix is the moment somebody pulls that fitting down to change it or fault-find it. In my own trade experience, next to none of the electricians I see on a roof job are wearing a proper respirator when they do that, and if there are fibreglass batts sitting over an IC-4 downlight, that's the moment they're breathing whatever comes down out of the hole with the fitting — and this may matter more for people with asthma, bronchitis or other breathing issues.
A recessed downlight sitting in a plasterboard ceiling. The trim you can see is the lid — on the other side of that hole is the roof space and whatever is loose in it.
A manhole cover being lifted out. It's normally propped wide open the whole time anybody is working up there, and it's the biggest hole of the lot.
Every hole above is only a problem because of what’s floating around waiting to come through it. Here’s what that actually looks like the moment a batt gets moved — the loose fibres coming off a batt the instant it’s disturbed, on the other side of every hole above.
Loose fibres floating in the air off a fibreglass batt
Loose fibreglass fibres you can actually see floating in the air the moment a batt gets moved.
Read the transcript
Watch the loose fibres you can actually see in the air when I move a fibreglass batt. That's how many loose fibres are floating around in fibreglass batts — it's a big part of why we don't recommend them.
A written distance you're supposed to keep off it
"3 metres back" isn't about installers standing apart. It's about anyone in the roof at all.
Clause 6.14(j) of the national code says that where workable, employees not engaged in the work should not be within 3 metres of an overhead synthetic-mineral-fibre work area, and it repeats the same distance twice more in its removal schedules. NOHSC:2006(1990) cl 6.14(j), plus Schedule 1 cl 31(a) and Schedule 3 cl 17(a) — search the PDF for “3 metres” ↗ The point isn't that installers work in a crowd. The point is that once fibreglass is installed, the fibres are still up there and still moving every time someone climbs into that roof, and without a respirator, that person is standing well inside 3 metres of them. Particles lift off with any turbulence and float with any disturbance — a pulled batt, a dragged cable, a fan starting up.
On a building site, nobody is allowed within 3 metres during install without proper PPE — the batt companies' own paperwork says as much. Once the job is finished and the ladder's put away, that rule doesn't go with it. The plumber running a new line, the electrician chasing a fault, the antenna installer, the pest man — they're all inside that 3 metres with nothing on, because nobody hands them a copy of the code on the way past. And the gap behind your fridge puts your own cooking within 3 metres of the 10 to 20% respirable fibres sitting in the batts directly above it, every day, with no ladder involved at all.
Who's inside the 3 metres, with nothing on
- Electricians chasing a fault or changing a downlight
- Plumbers running a new line through the roof
- Antenna and solar installers
- Pest control, working through the manhole
- Your own family, standing under the fridge vent gap
Worth knowing: that's a 1990 code from the old National Occupational Health and Safety Commission, and NSW has a public review of it open right now, which is the clearest proof it's still a live document, not a dead one.
You're never far from an opening
Walk through your own house and count how close you actually are to the ceiling.
The National Construction Code sets a 2.4 m minimum ceiling height for most habitable rooms (2.1 m in a kitchen, bathroom or laundry). NCC 2025, Housing Provisions Part 10.3, Clause 10.3.1(1)(a) — “in a habitable room excluding a kitchen — 2.4 m” ↗ Most project-built homes today sit at 2.4 to 2.7 m. So as you walk through the kitchen past the exhaust-fan opening, or stand under a downlight, or reach past a power point, you're standing a couple of metres at most from a ceiling that has a working hole in it somewhere close by.
Fibreglass also wraps most air-conditioning ducting run through a roof space, unless the ducting has been specified in polyester instead — another surface for loose fibre to sit on and shed from every time the system cycles on.
Constant airflow, not a sealed box
A house isn't a sealed container with air handling bolted onto it — it has constant airflow running through it, and that constant airflow is exactly what spreads fine fibres far and wide. They're light enough to move with any draft from a fan, an air-conditioner or an open window, and they don't settle the way a heavier dust would. Once a fibre is airborne in that kind of moving air, "where it lands" isn't really the right question — "how long before it moves again" is.
The bucket question
Would you tip a 2 kilogram bucket of loose fibre up into your own roof?
Work the arithmetic through on an average house and it's a real weight, not an abstract percentage.
- 1An average Australian house is around 150 to 250 square metres of ceiling.
- 2Fibreglass batts weigh roughly 1.5 to 2 kilograms per square metre once laid.
- 3So a full job puts somewhere around 225 to 500 kilograms of fibreglass into the roof.
- 4About 2% of that fibre is respirable-size, per the industry's own national code figure for glasswool under 3 microns.
- 52% of that weight range works out to roughly 2 kilograms of loose respirable fibre — call it 3 to 4 kilograms once offcuts and waste are counted.
2% doesn't sound like much on a spec sheet. Once it's converted to a weight sitting loose in your roof space, above your downlights, your exhaust fans and the gap behind your fridge, it's a bucket you'd feel the weight of carrying up a ladder.
Old fibreglass batts pulled back off the ceiling, wiring draped across the top. This is the loose material the kilogram figure is describing — not a single batt, the whole roof's worth.
Working the numbers, not just quoting them
How much boron does an installer actually breathe pumping Cellulose Fibre Insulation?
The fire retardant in our Cellulose Fibre Insulation is boric acid, a natural mineral mined out of the ground like salt. Boron sits on a chemist's shelf as a bone and joint supplement, because a seminal human trial found that 3 mg of boron a day measurably reduced how much calcium and magnesium postmenopausal women were losing in their urine. Nielsen et al., FASEB J 1987;1(5):394-7, PMID 3678698 ↗ The NIH calls that a "might" rather than a settled fact, so I'll call it that too. Nobody's selling fibreglass-dust capsules at the chemist — I asked.
The arithmetic, worked out simply
Worked on respirable dust only — the fraction fine enough to get past a nose and throat and reach the deep lung of somebody wearing no mask. That's the right basis because it's the only fraction a lung actually receives.
- 1. Respirable dust measured in the air on real cellulose jobs: geometric means of 0.11 to 1.53 mg per cubic metre across ten US contractor sites. NTP TOX-74, Table 15, p.49 — area respirable dust, wet and dry ↗
- 2. × 10 cubic metres of air breathed in an eight-hour shift, the standard occupational-hygiene figure = 1.1 to 15.3 mg of respirable dust for the whole day.
- 3. × the boric-acid share of the product, 4.2 to 17% by weight across two manufacturers' own composition tables = 0.046 to 2.6 mg of boric acid.
- 4. × 0.175 (boron is 10.81 of boric acid's 61.83 molecular weight) = about 0.008 to 0.45 mg of elemental boron for the day.
| A mask-less installer's whole day, real worksite readings | 0.008 – 0.45 mg |
| Same model at the legal respirable-dust limit, 5 mg/m³ all shift | 0.4 – 1.5 mg |
| Typical Australian/US daily diet | 1.0 – 1.5 mg |
| One common chemist boron supplement capsule | 3 mg |
| WHO acceptable safe range | 1 – 13 mg/day |
| Government upper intake level, adult | 20 mg/day |
So the top of that range, 0.45 mg, is about 15% of one 3 mg supplement capsule — and the bottom of the range, 0.008 mg, is well under 1% of one. There's no official recommended daily intake for boron, because the nutrition board found the data too thin to set one, but the two benchmarks that do exist put this nowhere near a concern: WHO's safe range starts at 1 mg a day, and the government's upper safety limit sits at 20 mg a day. Even run at the legal respirable-dust ceiling for a full eight-hour shift, the number tops out at 1.5 mg — still half a capsule at most, and about 7.5% of the government limit.
That's also an installer's exposure, not a homeowner's, modelled on a full eight-hour shift with no mask at all. In practice most installers spend around two hours actually pumping on a job before moving to the next one, two to three jobs a day, so a real day's dose sits well under even the low end of that range.
Batt companies point at boron toxicity. Here's the actual comparison
Boric acid and borax sit in the same low-acute-toxicity band as ordinary table salt. Salt's oral rat LD50 is 3,000 mg per kilogram, confirmed by two independent manufacturer safety data sheets. Mallinckrodt Baker MSDS, Sodium Chloride ↗ Fisher Scientific MSDS, Sodium Chloride ↗ Boric acid's is roughly 2,660 to 5,140 mg per kilogram and borax's is roughly 4,550 to 6,000 mg per kilogram, per the EPA-cited figures on the US National Pesticide Information Center's fact sheet. NPIC (Oregon State University / EPA), Boric Acid Technical Fact Sheet ↗ By that single-dose measure, boric acid and borax are actually less acutely toxic than table salt, not more. Nobody worries about the salt shaker on the kitchen table — kids use it themselves. The same LD50 measure doesn't cover boron's chronic upper limit, which is why the 20 mg-a-day ceiling above still matters and why the insulation isn't for eating either way.
Both sides, same page
Fibreglass and cellulose, measured against each other rather than cellulose measured alone.
Showing only the cellulose figure makes it look like cellulose is the only product with a dust number attached. Here's both, on the best real measurement I can find for each.
| Product | What's measured | Reading |
|---|---|---|
| Cellulose Fibre Insulation | Elemental boron inhaled, mask-less installer, whole 8-hr day | 0.008 – 0.45 mg/day |
| Fibreglass batts | Respirable fibres in the air, laying batts | 0.14 fibres/cm³ |
| Loose-pumped fibreglass (like Knauf Jetmax) | Respirable fibres in the air, blowing loose-with-binder | 0.55 fibres/cm³ |
| Unbound loose insulation, worst case in the study | Respirable fibres in the air, blowing without binder | 1.32 – 18.4 fibres/cm³ |
Fibreglass row sourced to a field survey of insulation installers across 107 houses in 11 US states. Lees et al. 1993, Applied Occupational and Environmental Hygiene 8(12):1022–1030, as quoted in the NTP Report on Carcinogens glass wool profile ↗ Cellulose row: NTP TOX-74, Table 15, p.49. NTP TOX-74 ↗
Why these two rows aren't in the same unit, and what that does and doesn't tell you
The cellulose figure is a dust-mass reading, in milligrams — cellulose's fire-retardant content needs a weight basis to convert to a boron dose. Fibreglass occupational-hygiene studies almost universally count fibres per cubic centimetre of air instead, the way glass wool exposure is tracked worldwide — no field study I can find has published a fibreglass dust-mass reading (mg/m³) taken during real batt or blown-fibre installation, only regulatory ceilings, which describe a legal limit rather than what installers actually breathe. So the two rows above are the best real measurement for each product, in that product's own measurement tradition, not a single figure converted both ways. What both readings agree on is the same order of finding: loose-pumped fibreglass measures dustier than batts, and batts measure dustier than cellulose's under-0.1% respirable fraction.
The real numbers, not a rounded-off headline
What the American occupational dust-limit study actually found.
The US National Toxicology Program's worksite survey of ten American cellulose contractors collected 175 personal-breathing-zone total-dust samples. Of those, 26 employees' total-dust eight-hour time-weighted averages exceeded the OSHA permissible exposure limit of 15 mg per cubic metre, and 42 exceeded the ACGIH threshold-limit value of 10 mg per cubic metre. NTP TOX-74, Conclusions, p.9 — “26 employees' total dust 8-hour time-weighted averages (TWAs) exceeded the [OSHA] PEL of 15 mg/m³, and 42 exceeded the [ACGIH] threshold-limit value (TLV) of 10 mg/m³” ↗ That's a real, sourced overexposure finding on cellulose worksites, and it's why the report's own conclusion says cellulose insulation should continue to be regarded as a nuisance dust and that workers should keep wearing protective masks. That's exactly why my crew wears a Sundström respirator on every job, cellulose or otherwise.
That survey covered cellulose only. I haven't found an equivalent US or Australian field study that measured fibreglass batt or blown-fibreglass installation against those same OSHA and ACGIH total-dust limits, so there's no fibreglass exceedance figure to print alongside it. If a comparable fibreglass survey turns up, I'll put the real number here rather than a guess.
The only fair comparison there actually is
The only way fibreglass matches Cellulose Fibre Insulation on dust is to leave it in the bag.
Every dust reading on this page, from the fibre counts to the kilograms in the roof, is what happens once fibreglass is opened, cut, laid or blown. Sealed in its plastic wrap, it sheds nothing. That's the whole comparison, and it's not really one — the moment either product actually does its job, one of them sheds under 0.1% respirable dust and the other sheds 10 to 20%.
I've pulled an unopened R2.0 batt pack straight out of a roof, still sealed in its wrapper, where a previous installer got paid to lay it and simply left it there instead. It wasn't doing that house any good sitting in the plastic. It also wasn't shedding a gram of dust, which is the only circumstance under which that's true.
Anticon blanket, gutters and rainwater tanks
Does the glasswool blanket under a metal roof end up in the gutter and then in the tank water people drink?
This one splits into what I see, what the maker says, and what anybody has actually measured, because they're three different things and mixing them up is how people end up saying something they can't stand behind.
What I see, from forty years on roofs
On plenty of houses the anticon blanket runs right out to the fascia and I can see fibre hanging over the edge into the gutter. The manufacturer's own datasheet says the blanket is ideally suited to sealing ember entry points at ridges, valleys and fascias, so it's meant to be out there. Bradford Anticon blanket datasheet ↗
What the maker says, and it's fair to print it
CSR Bradford's own safety document states that once installed, the product doesn't release dust or fibres, and that the fibre is hydrophobic with no adverse environmental effects expected if it's released into water or soil. CSR Bradford SUIS, FBS-1 Glasswool, §7 p.3 and §12 p.6 ↗ The blanket is also installed sandwiched under the roof sheet, so in a sound, sealed roof the weather never touches it.
What's actually been measured
I went looking for a water authority test or a published study measuring glass fibre in gutter runoff or tank water, and found none. That's the real state of the evidence — the pathway is physically possible on an old roof with torn facing, badly lapped joints or cut edges left open at the eave, and that's as far as the documented evidence goes either way.
What I can stand behind is the settling and the filter arithmetic
Glass is heavier than water, so a fibre in a cup does sink. It just sinks unbelievably slowly. A glasswool maker's own safety data sheet puts the true specific gravity of the glass itself at 2.5 to 4.3 grams per cubic centimetre. Asahi Fiber Glass Co., Ltd, Safety Data Sheet “Glass Wool”, SDS-GW-02, §9, p.5 — “Specific Gravity (Density): 2.5-4.3 g/cm3 (true specific gravity) (HSDB 2005)” ↗ That's the density of the glass the fibre is made of, not the bulk density of a batt or a blanket, which is far lower because most of a batt is air.
Run Stokes' law across that whole density range and a 1-micron fibre takes somewhere between about 15 and 33 hours to fall the ten centimetres to the bottom of a cup, and a 3-micron one takes between about 1.7 and 3.7 hours. In any real glass, with any warmth or movement in it at all, the fibres look evenly mixed through the water rather than sitting in a layer at the bottom. They do sink — it takes the best part of a day, which in a cup you're about to drink amounts to the same thing.
The filter on your tank almost certainly won't stop it
The sediment cartridges that ship as standard on Australian rainwater housings cluster at 1, 5, 10 and 20 micron nominal. Puretec replacement sediment cartridge range ↗ A 5 to 20 micron cartridge is between two and twenty times bigger than the fibre. It won't stop it.
And nominal isn't the same as absolute. A filter manufacturer's own spec sheet defines its micron ratings as 85% or greater removal of a given particle size, so a nominal 1-micron cartridge lets some fraction straight through by definition, not by fault. Pentair/Pentek CBC series cartridge spec sheet, notes ↗
NSW Health's rainwater treatment guide gives the steps that do work: micro and ultrafiltration membranes at 0.1 to 0.01 micron, and reverse osmosis at 0.001 micron. NSW Health, Rainwater Treatment Guide, Filtration Treatment Systems section ↗ Drinking off a roof with loose fibre on it calls for that, or a 1-micron absolute cartridge at minimum, and the loose material off the roof edge either way.
Why this bothers me less with our own product
A borate is a mineral your body is already carrying around and already using — the section above covers that. Glass isn't. At the dilution anything off a roof would ever reach a tank, I haven't lost a minute's sleep over borate, and that's my own view from forty years on roofs rather than a finding out of a document, so weigh it accordingly.
Straight answers
Fibreglass dust in your house, the questions I get.
Is fibreglass dust actually harmful in a house?+
The industry's own national code puts 10 to 20% of glasswool batt fibre under the 3-micron respirable size — fine enough to bypass your nose and throat and reach the deep lung, and fine enough to ride moving air rather than settle. Below about 30 microns, particles are governed mainly by airflow rather than gravity and stay suspended for a long time, and fine particles that have already landed get lifted back off a surface by turbulence. Every hardware-store batt pack's own safety data sheet says unprotected exposure to high levels of that dust may cause discomfort of the nose, throat, and upper and lower respiratory tract — and this may matter more for people with asthma, bronchitis or other breathing issues. I'm not calling it a carcinogen, because IARC moved glasswool down to Group 3, not classifiable, back in 2001. My argument is about where a particle that size ends up once it's airborne in your roof, which is down through every hole in your ceiling rather than staying politely on top of the joists.
How do fibreglass particles get from the roof into the air I breathe?+
Through every hole between your roof space and your rooms — a downlight cut-out, a bathroom exhaust fan, a range hood duct, wiring holes down to light switches and power points, gaps in the corners against the brickwork, and the gap behind the fridge that vents fridge and cooking heat straight up into the roof. Your cooking sits directly under that last one. Warm air moves through a Queensland roof all day, and once a fine fibre is in moving air it travels with the air rather than falling straight down, so it can come back down through any of those openings.
Is there borax in Cellulose Fibre Insulation?+
Ours is treated with boric acid, a natural mineral mined out of the ground like salt — the only fire retardant in it — we make it ourselves at our own factory in Tiaro and I can tell you exactly what goes in the bag. Boron sits on a chemist's shelf as a bone and joint supplement, at a common dose of 3 mg elemental boron. Modelled on real cellulose worksite air readings, a mask-less installer's whole-day boron exposure works out to about 0.008 to 0.45 mg — at the top of that range, roughly 15% of one supplement capsule, and comfortably under the government's 20 mg-a-day upper safety limit.
How much fibreglass dust is actually sitting in an average roof?+
Work the numbers through. An average house is 150 to 250 square metres, fibreglass batts weigh roughly 1.5 to 2 kilograms per square metre once laid, and about 2% of that fibre is respirable-size. That comes out to around 2 kilograms, call it 3 to 4 kilograms once you allow for offcuts and waste, of loose respirable fibre sitting in the roof of an average job. I wouldn't tip a 2 kilogram bucket of loose fibre into my own roof space, and that's the same fibre riding the air down through every downlight and exhaust fan in the house.
Does loose-pumped fibreglass make more dust than fibreglass batts?+
A field survey of installers across 107 houses in 11 US states measured the respirable fibre they actually breathed. Laying glasswool batts averaged 0.14 fibres per cubic centimetre. Loose fibreglass blown in with a binder — the category a product like Knauf's Jetmax sits in — averaged 0.55 for the installer, roughly four times the batt figure. The highest exposure in the whole study, 7.67, came off unbound loose insulation. So yes, loose-pumped fibreglass reads dustier than batts on every measurement I can find, which is exactly why we never blur blow-in fibreglass and pumped Cellulose Fibre Insulation together as the same thing — they are not the same particle, the same size, or the same dust reading.
Can you just leave fibreglass in the pack instead of installing it?+
That's the one way it genuinely matches Cellulose Fibre Insulation for residual inhalable dust — leave it sealed in the bag and don't open it. I've pulled an unopened R2.0 batt pack out of a roof where a previous installer got paid to lay it and didn't. It wasn't doing the house any good sitting there, but at least it wasn't shedding dust either. Once it's opened, cut, laid, or blown, it starts doing what every dust reading on this page describes.
Figures on this page are read from primary government, standards-body and manufacturer documents, cited beside each claim, current as of July 2026. Safety data sheets get revised — check the current version before relying on it. This page is general information, not medical or occupational-hygiene advice.
Do you have fibreglass batts in your roof now? We can quote to remove them or fix your insulation so there are no loose particles floating around up there.
If you'd rather not think about loose particles at all, the two products that don't have the problem are Cellulose Fibre Insulation (a plant fibre) or polyester batts — get a quote.
Had us sort out old fibreglass for you? Worth a mention.
A quick honest review genuinely helps a small family business, and helps the next person decide. Thank you.
Get the loose particles sorted, not just the R-value.
Send us your address and roof type for a fixed-price quote within 48 hours for most houses — removing old fibreglass, topping over it safely, or pumping Cellulose Fibre Insulation or polyester from scratch. Servicing Brisbane, SE QLD and Northern NSW.
Peter Johnson
Owner / installer · Comfort Zone Insulation Team® · Since 1986