All major DIY stores have agreed to stop selling ionisation smoke alarms following a request from Consumer NZ.
A recent Consumer NZ test of smoke alarms found ionisation-type smoke alarms performed so poorly that the watchdog called for retailers to pull them from their shelves. Consumer NZ is pleased to report all stores contacted about their ionisation alarms agreed to stop selling them.
Mitre10 and Hammer Hardware said they would “cease selling ionisation smoke alarms, effective immediately”.
PlaceMakers said it would “exit the balance of our ionisation alarms, which we anticipate will be completed within a matter of weeks”.
Bunnings Warehouse said “our intention is to sell through our ionisation smoke alarm stock and no longer sell that product”.
Consumer NZ head of testing Dr Paul Smith welcomed the news.
“Removing a product that doesn’t perform a critical safety task effectively is a major win for New Zealand consumers,” Dr Smith said.
Ionisation alarms give much less warning of smouldering fires (such as those caused by faulty electrical wiring, curtains draped over a heater, or a hot ember igniting upholstery foam) making it less likely you can get out of your home safely. All smoke alarms will respond to a fire, eventually. “The difference is whether they respond to visible smoke. A smouldering fire can fill a home with deadly smoke long before it bursts into flames,” Dr Smith said.
Consumer NZ testing confirmed ionisation models were great at detecting flaming fires, but not so good with visible smoke. “The four ionisation alarms in our test were faster at detecting flaming fires (burning oil and wood) but much slower at detecting smoke from smouldering foam.”
People should check which type of smoke alarm they have in their homes and rental properties. “You can identify an ionisation alarm from a radioactive symbol somewhere on the alarm body – it may be underneath, so you might need to remove it to check.”
Consumer NZ advises people:
For rental properties:
Our test, based on the UL217 standard for smoke alarms, was conducted at an independent lab. Multiple alarms were placed in our “smoke-sensitivity chamber”. Smoke was introduced from flaming wood, flaming oil, smouldering wood chips, and smouldering upholstery foam. We tested three samples of each alarm model and assessed their response to smoke compared to three control sensors.