The meter swap turns your power off. What that does to a vaccine fridge or a server room.

What actually happens on the day

The technician isolates the supply at the switchboard, removes the old meter, fits the new one, and restores power. It is straightforward work, and the variability is in the site rather than the meter. An old board, a cramped meter position, asbestos backing, or a fault found once the cover is off will all extend the outage.

The quoted durations differ by retailer, which tells you they are estimates rather than guarantees:

• Flow Power says usually less than 30 minutes.
• ActewAGL says usually about an hour.
• Origin Energy says usually up to two hours.

Plan against the top of that range, not the middle. And if you are in an apartment building, a strip of shops or anywhere you share a connection point with neighbours, meters are commonly changed in groups, which makes the outage longer than any of the figures above.

The notice you are entitled to, and why it often goes to the wrong person

Meter providers commit to at least four business days notice of a planned interruption. That is a workable amount of time to move stock, hire a backup, or ask for a different slot.

The practical problem is that the notice goes to whoever holds the electricity account. In a medical practice that is often the landlord or a practice group in another city, not the nurse immuniser who is responsible for the fridge. In a small business it is often an email address nobody reads. The most common way this goes wrong is not that the notice was never sent, it is that it never reached the person who needed it. If you are responsible for temperature-critical stock, ask now who receives the electricity account correspondence, and tell them what to forward.

Two entitlements are worth knowing before the notice arrives:

• You can usually ask to reschedule. A planned interruption is negotiable in a way an unplanned one is not.
• Life support registration changes the rules. Where someone at the premises is registered as requiring life support equipment, the supply cannot be interrupted without you first agreeing to the date and time. If that applies at your address, make sure the registration is current rather than assuming it is.

If you store vaccines

The national standard is the Department of Health's National Vaccine Storage Guidelines, known as Strive for 5. The essentials that matter for an outage:

• Vaccines must be stored between +2°C and +8°C, aiming for +5°C. The name refers to that target.
• They must be in a purpose-built vaccine refrigerator. Domestic and bar fridges are explicitly not suitable, and that includes as an emergency substitute.
• Many vaccines are destroyed by freezing, or by repeated freeze and thaw cycles, and the damage is not visible. A vial that looks fine may be inert.
• Storage outside the range, other than brief excursions during a defrost cycle, is a cold chain breach and triggers a defined protocol.

A purpose-built unit that stays shut will often ride out an hour without drama, because that is part of what you paid for. What nobody can tell you in advance is whether yours will, on that particular day, at that ambient temperature, at that stock level. February in a western Sydney treatment room is a different problem from July in Orange.

So treat it as a managed risk. Before the day: decide whether stock moves to another site or into a validated cold box, confirm your data logger has battery backup and will keep recording through the outage, and print the cold chain breach protocol so it is on the wall rather than in someone's inbox. On the day: keep the door shut, and resist the urge to open it "just to check", because opening the door to check the temperature is the thing most likely to cause the excursion you are checking for. Afterwards: read the log before you use anything.

If you find a breach, the guidance is consistent: isolate the affected stock, keep it refrigerated, label it clearly as not for use, and contact your state or territory immunisation program for advice. Do not discard it, and do not administer it, until you are told which of those applies. Our page on vaccine fridge temperature monitoring and Strive for 5 covers the ongoing monitoring obligations in full.

Last checked: 14 August 2026. Storage range and breach handling from the Australian Government Department of Health National Vaccine Storage Guidelines, Strive for 5. Interruption durations and notice periods from retailer published guidance, current at that date. This page is general information, not clinical or regulatory advice: your state immunisation program and your own policy documents take precedence.

If you run a server or comms room

The failure here is not the servers. It is the cooling.

Almost every small business that has thought about power has a UPS, and almost none of them have the air conditioner on it. The UPS is sized to keep the racks alive and let them shut down cleanly. The consequence is that during the outage the equipment keeps drawing power, keeps producing heat, and the only thing that was removing that heat has stopped. A sealed room with no ventilation and a live rack heats up faster than people expect, and small rooms are worse than large ones because there is less air to absorb it.

The awkward middle case is the one to plan for: a UPS with enough runtime to keep everything powered for the whole outage, in a room with no cooling for the whole outage. That combination produces the highest temperatures. Sometimes the right answer is a graceful shutdown even though the UPS could have carried you, because a controlled outage of non-critical systems is cheaper than thermal damage.

Worth doing before the day: confirm what is actually on the UPS and what is not, decide in advance what gets shut down rather than deciding it at 40°C, prop the door open if the room's security allows it, and make sure someone will see a temperature alert while the building is dark. Our page on server room temperature monitoring for small business covers the safe ranges and sensor placement.

The part everyone skips: proving what happened

The outage itself is usually survivable. What causes the real trouble afterwards is not knowing.

If the power was off for 95 minutes and nothing was recording, then you have a fridge full of vaccines whose history has a hole in it, and no way to demonstrate they stayed in range. In practice that is treated as a potential breach, which means the stock is quarantined and someone spends a day on the phone. The financial loss from discarding stock you did not need to discard is entirely avoidable, and it is avoided by having a record.

That is the whole argument for continuous logging with its own power and its own connection. A sensor with battery backup and a cellular link keeps recording while the building is dark and the internet is down, which is precisely when the reading matters. After the event you have a temperature trace across the outage, and it either shows you stayed in range, in which case you use the stock with confidence, or it shows exactly when and for how long you did not, which is what the breach protocol asks for anyway.

A planned interruption is also a reasonable moment to find out whether your existing monitoring survives a power cut, because a surprising number of installations do not. If the sensor goes quiet the moment the mains drops, it was never going to help you in the outage you actually feared.

Why the meter is being changed at all

The exchange is part of a national program to replace legacy accumulation meters, and the meter itself is not the interesting part. What follows it usually is: a smart meter records consumption in intervals rather than as one number, which is what makes time-of-use and demand pricing possible for your site. If your bill structure changes after the swap, the meter did not raise your rates, the tariff did. Our sister site explains how time of use tariffs work and what the peak and off peak windows are on each network, and for refrigeration specifically our sister site covers what peak and demand charges do to coolroom running costs.

Questions people ask