Industry Solutions / Retail and Commercial Sector

Critical Power and Cooling for Retail Chains, Supermarkets and Commercial Buildings in Australia

Retail is the sector where the gap between what a UPS costs and what an outage costs is widest, and where it is most often ignored. A 550 VA to 1500 VA unit protecting a checkout lane is a few hundred dollars. A metropolitan supermarket turning over $250,000 a week loses roughly $1,500 in gross sales for every hour the tills are down, before you count the stock, the staff standing idle and the customers who walk out with a full trolley and do not come back.

The problem is that retail power protection is almost never a single-site decision. It is a fleet decision. A chain with 200 stores does not buy a UPS, it buys a standard, a rollout schedule and a service model, and the difference between a good standard and a lazy one compounds across every site, every battery cycle and every service truck roll for the next decade. Get the model selection wrong and you inherit 200 warranty conversations. Get the battery replacement policy wrong and you pay for 200 emergency callouts instead of eight planned regional visits.

Indigi Power and Cooling is an Indigenous and Veteran owned critical power and cooling specialist, registered with Supply Nation and ICN Gateway. We work across store fleets, shopping centre base building services, quick service restaurant networks and commercial office towers, covering both halves of the problem: the UPS at the checkout and in the comms rack, and the cooling that keeps that comms rack alive in a back-of-house cupboard that was never designed to hold IT equipment.

What a power outage actually costs a store

Most vendor material on retail power stops at "lost sales". That is the smallest and most recoverable part of it. The mechanical damage is more specific than that, and it is worth naming precisely, because each item points at a different piece of equipment that should have been protected.

Payment stops before anything else does

An EFTPOS terminal is the last device the customer touches and the first one to fail. Modern Australian terminals increasingly carry a 4G fallback for network outages, which is genuinely useful and also frequently misunderstood: a 4G fallback protects you when the store's internet drops, not when the store's power drops. A terminal on a dead power point is a dead terminal regardless of how many carriers it can reach. The same applies to the POS till driving it, the network switch it talks through, and the router carrying the transaction out of the building. All four have to stay up together or none of them matter.

Interrupted transactions and reconciliation

A transaction that is cut mid-authorisation is worse than a transaction that never started. The store does not know whether the customer was charged, the bank may hold the authorisation, and the reconciliation lands on a duty manager at close of trade or on head office the next day. Multiply that by every lane that was mid-sale when the power went and you have converted a five minute outage into an afternoon of accounting work across the chain.

Loss of the electronic journal

POS systems write an electronic journal of every transaction, and many hold recent entries in memory before committing them. An unclean shutdown can lose the tail of that journal, which is the record you rely on for refunds, disputes, loyalty accrual and audit. A UPS that delivers even three to five minutes of runtime turns a hard crash into a graceful, scripted shutdown, and that is usually the entire justification for the unit on its own.

Reopening is slower than closing

Stores do not resume trading the moment the power returns. Tills reboot, POS software reconnects to head office, terminals re-authenticate to the acquirer, refrigeration controllers restart in a staggered sequence to avoid tripping the board, and someone has to walk the floor and check the cool rooms. Fifteen minutes of lost supply routinely produces forty five minutes of lost trading, and in a food business it produces a temperature log that somebody has to sign off.

For food retail, refrigeration is the real exposure

In a supermarket, a butcher, a bakery or a quick service restaurant, the tills are not the biggest number on the page. Stock is. Standard 3.2.2 of the Australia New Zealand Food Standards Code requires potentially hazardous food to be held at or below 5 degrees Celsius or at or above 60 degrees Celsius, and the accompanying 2 hour and 4 hour guidance governs what can be sold, what must be used immediately and what must be discarded once food has been between those temperatures. A single medium supermarket carries tens of thousands of dollars of chilled and frozen stock, and the write-off decision is driven by a temperature record, not by a judgement call.

The physical behaviour differs sharply by case type, and this is where honest advice matters. A well-loaded closed cool room with the doors kept shut holds temperature for a useful period, often a few hours, because the thermal mass of the stock carries it. An open multideck display case loses its air curtain the instant the fans stop and climbs out of band far faster, frequently within the first hour. Frozen product in a closed well-stocked freezer is the most forgiving of all. The practical consequence is that the first thing to do in an outage is close the cases, and the second is to have a temperature record that proves what actually happened.

What belongs on the UPS in a store, and what does not

Being explicit about what is not on the UPS is more useful than a long list of what is. Retail sites are quoted extraordinary things by vendors who have not asked what the load actually is. A UPS is a bridging device sized in minutes, not a generator sized in hours, and putting the wrong load behind it is how a chain ends up with a cupboard full of units that alarm every time the compressor starts.

Pure sine wave is not an upsell in retail

This is the single most common cause of unexplained faults in retail UPS fleets, and it is worth being blunt about. Cheap line-interactive units produce a stepped or simulated sine wave on battery, not a true sine wave. Modern POS power supplies, EFTPOS terminal adaptors and refrigeration controllers use active power factor correction, and active PFC front ends interact badly with a stepped waveform. The symptoms are not a clean failure. They are intermittent: a terminal that drops out once a fortnight, a controller that reboots during a brownout, a power supply that buzzes and runs hot and dies eighteen months early.

Those faults are miserable to diagnose because they never reproduce on the bench, and they typically get blamed on the POS vendor, the network or the store manager before anyone looks at the waveform. Specify pure sine wave output on every unit that touches a POS, an EFTPOS terminal or a refrigeration control, and the class of fault disappears. Across a fleet, the diagnostic time you avoid is worth far more than the price difference on the units.

Runtime: minutes, deliberately

Retail runtime targets should be set from what the store actually needs to do, not from a vague desire for more. Five to ten minutes at the checkout lane covers the overwhelming majority of Australian distribution faults, which are momentary, and gives staff time to complete an in-flight sale and shut down cleanly. Fifteen to thirty minutes on the comms rack and the refrigeration monitoring is worth having, because those loads are small and the marginal battery cost is low. Anything beyond about thirty minutes in a retail tenancy is a generator conversation, not a UPS conversation, and quoting it as a UPS is how sites end up with battery cabinets they cannot fit, cannot cool and cannot afford to replace.

Buying for a fleet, not for a store

Everything above describes one site. The commercial reality for a national retailer, a franchise network or a quick service restaurant group is that the per-site engineering matters far less than the per-site economics, repeated hundreds of times. Four decisions carry almost all of the whole-of-life cost.

One standard, two models, no exceptions

Pick a lane unit and a rack unit and hold the line. A fleet running two models can be serviced by any technician with two spare battery part numbers on the van. A fleet running eleven models because each state manager bought locally cannot be serviced efficiently by anyone, needs eleven sets of spares, and produces a maintenance quote that reflects that. Standardisation also makes remote monitoring possible, because one management platform can see the whole fleet rather than four platforms seeing quarters of it.

Replace batteries by cohort, not by failure

This is the biggest single lever. Sealed VRLA batteries have a design life stated at 20 to 25 degrees Celsius, and the well-established rule of thumb is that life roughly halves for every 10 degrees Celsius of sustained temperature above 25. A back-of-house comms cupboard in a Brisbane or Darwin store frequently sits at 32 to 38 degrees. That is not a marginal effect. It is the difference between a four year battery and a two year battery, and it is why fleets that budget on the datasheet number run out of runtime halfway through the budget cycle.

Once you accept those numbers, the planning method follows. Group sites into cohorts by install date and by ambient band, then replace a whole cohort in a planned regional run rather than waiting for individual units to fail. One technician working through eleven stores in a metropolitan cluster in two days costs a fraction of eleven separate emergency mobilisations spread across nine months, and it means no store discovers its battery is dead at the moment it needs it. Use the UPS battery replacement cost calculator to build the cohort budget, and see battery maintenance and replacement for how we run the programme.

Swap out, do not repair

On small single-phase units in a trading store, on-site repair is almost always the wrong answer. The technician is standing in a live retail environment with customers walking past, the diagnostic time is unpredictable, and the store is exposed for the whole visit. Carry a small pool of pre-configured spare units at the state level, swap the failed one out in ten minutes, and deal with the faulty unit off site where time is cheap. For a fleet of any size the spare pool pays for itself in avoided second visits alone. This is a different service model from the large three-phase world, where on-site repair is correct because the units are irreplaceable in a morning.

Remote monitoring across every site

A UPS that cannot tell you it has a problem is a UPS that will tell you at the worst possible moment. Network cards reporting over SNMP, or a monitoring gateway per site, give head office a single view of battery health, load, self-test results and mains events across the entire fleet. That data is what turns the cohort plan from a guess into a schedule, and it surfaces the sites where the mains supply itself is the problem, which is a landlord and distributor conversation rather than a UPS one. It also catches the two stores in the fleet where somebody unplugged the UPS to run a floor polisher and never plugged it back in.

Comms room cooling in retail and commercial tenancies

Almost every retail tenancy comms room in Australia is the same room. It is a back-of-house cupboard, roughly the size of a broom closet, chosen because it was the space nobody else wanted. It holds a wall-mounted rack or a small floor cabinet, the store's switch and router, the NVR, the UPS and usually the alarm panel. It has no dedicated cooling. It often shares a wall with the plant space, the condenser deck or the ovens, and the tenancy air conditioning does not reach it because the door is always shut.

The failure story is exactly the same as the one in every data centre, only smaller and less monitored. Heat accumulates because the room has a continuous load and no heat rejection path. Equipment runs at the top of its temperature band, then above it. The switch throttles or reboots, the NVR drops frames, and the UPS batteries quietly lose half their life as described above. Nobody notices until something stops, and then the fault gets attributed to the equipment rather than the room.

The ASHRAE TC 9.9 recommended envelope of 18 to 27 degrees Celsius at the equipment inlet is the right target here even though nobody would call a retail comms cupboard a data centre. It is the same silicon. The practical options in a tenancy, in ascending order of cost, are a properly sized ducted transfer with mechanical extract, a dedicated wall-mounted or ceiling cassette split system on a separate circuit so it is not switched off with the store, or a small precision unit where the load and the criticality justify it. Refrigerant work is licensed under the Ozone Protection and Synthetic Greenhouse Gas Management Regulations and must be carried out by ARCtick licensed technicians, and the electrical work sits under AS/NZS 3000. In a shopping centre tenancy there is usually also a landlord approval and a make-good condition, which is worth resolving before design rather than after installation.

The two failure modes we see most often are worth naming. First, a comfort split system installed on the store's general lighting and power circuit, which the last staff member out switches off at the board every night, so the room bakes overnight and all weekend. Second, a unit sized on floor area rather than on the actual heat load in the rack, which then short cycles and never dehumidifies. Both are cheap to avoid at design and expensive to fix later. See server room cooling Australia and switchroom cooling for the equivalent work at larger scale.

Shopping centres and commercial building base services

A shopping centre or a commercial office tower is a different customer from a store fleet, with its own set of protected loads that sit on the landlord side of the meter rather than the tenant side. Base building critical power typically covers the building management system, the fire indicator panel interfaces, the security and access control head end, the base building comms room, the car park and directory signage systems, and the lift controllers.

Lifts are the under-served item. AS 1735 covers lifts, escalators and moving walks in Australia, and together with the National Construction Code it drives requirements around emergency operation, lift car lighting and communications, and the behaviour of a lift on loss of normal supply, including automatic rescue operation that brings a car to the nearest landing and opens the doors rather than leaving occupants trapped. In practice that means a UPS or a dedicated battery supply behind the lift controller and car services, sized and maintained for that duty. It is a genuine compliance obligation, it is regularly overlooked in older buildings, and it becomes very visible the first time somebody is stuck between floors during a distribution fault. We cover this in detail on the elevator and lift UPS systems page.

Base building air handling brings AS/NZS 3666 into scope for mechanical ventilation and air conditioning hygiene, which matters for the plant serving common areas and tenancies. On the electrical side AS/NZS 3000 governs the installation work throughout. For centre management, the useful framing is that the base building critical power schedule and the tenant fitout power protection standard are two separate documents that should reference each other, so that a new tenancy fitout inherits a sensible comms room and UPS position rather than inventing one.

What Indigi does for retail and commercial portfolios

Indicative pricing: single-phase UPS installation and commissioning from $850 up to 3 kVA, $1,250 for 3 to 10 kVA and $1,800 for 10 to 20 kVA, with fleet rollout rates quoted per site once the standard and site count are fixed. Three-phase installation for base building and centre applications starts from $3,200 for small systems and $6,500 for medium. UPS maintenance is $720 ex GST for a one-off visit or $1,940 per year ex GST for a standard annual plan, and multi-site portfolios are priced against the schedule rather than site by site. Comms room and server room cooling design and installation runs from $8,000 to $15,000 for small rooms, $25,000 to $45,000 for medium installations and $45,000 to $60,000 plus for large projects.

Frequently asked questions

What is UPS in retail?

In a retail context a UPS is an uninterruptible power supply, a battery-backed device that sits between the mains outlet and the store's critical equipment. It carries the load instantly when mains power fails, and it also conditions incoming power by correcting brownouts and filtering electrical noise. In an Australian store it typically protects the POS till, the EFTPOS terminal, the comms rack, CCTV and the refrigeration controllers and monitoring. It is not the same thing as United Parcel Service, which dominates search results for the same three letters and is a parcel carrier rather than a power product.

Which UPS type is best suited for supporting point of sale kiosks in a retail environment?

A line-interactive UPS with pure sine wave output and automatic voltage regulation is the right choice for a POS or kiosk position. It handles the brownouts and voltage sags that cause most retail dropouts, it is compact enough to fit under a counter, and pure sine wave output avoids the intermittent faults that stepped waveform units cause in active power factor corrected POS power supplies. Capacity of 550 VA is a widely used starting point for a single register, and 1000 VA to 1500 VA is more realistic for a lane carrying a till, screen, scanner, printer and terminal together. Reserve online double conversion topology for the comms rack or for sites with genuinely poor supply quality.

Which is better, AVR or UPS?

They solve different problems and are not alternatives. Automatic voltage regulation corrects voltage that is too high or too low while mains power is still present, which addresses brownouts and sags. A UPS provides battery backup when mains power is absent entirely. Most decent line-interactive UPS units include AVR, so in practice a UPS is the superset and a standalone AVR is only worth considering where the supply is chronically out of tolerance but never actually fails, which is rare. For a retail store that needs to complete a card transaction during an outage, only a UPS will do it.

What is a good backup power source?

It depends entirely on the load and the duration required. For seconds to minutes on electronics, a UPS is the correct answer and nothing else comes close on response time, because the transfer is effectively instantaneous. For hours on large loads such as refrigeration compressors, lighting and cooking equipment, a generator or a permanent standby supply is the answer, because batteries sized for that duty are not commercially sensible in a retail tenancy. Battery energy storage sits between the two and can be worthwhile where there is also a tariff or solar case. Most retail sites need a UPS on the small critical loads and, if the stock exposure justifies it, a generator connection point for refrigeration.

How to get power without a generator?

For a retail site the practical options are a UPS with an extended battery pack, a battery energy storage system, or a generator inlet with a portable unit brought in only when needed. A UPS with extra battery modules can carry a small critical load such as a comms rack and refrigeration monitoring for one to several hours, which is often enough for a distribution fault. Battery storage can carry larger loads and pays for part of itself through tariff arbitrage or solar self-consumption. What none of these do cheaply is carry refrigeration compressors for a full day, which is why sites with high stock exposure usually end up with a generator changeover point even if they do not own the generator.

Does a server room need AC?

Yes, if the room holds a continuously running load and has no other heat rejection path, which describes almost every retail comms cupboard. IT equipment converts essentially all the electricity it draws into heat, so a rack pulling 500 W is a 500 W heater running around the clock in a small enclosed space. Without dedicated cooling the room will settle well above ambient, and in a back-of-house cupboard adjoining a plant space it will exceed 35 degrees Celsius routinely. Target the ASHRAE TC 9.9 recommended range of 18 to 27 degrees Celsius at the equipment inlet, and put the cooling on its own circuit so it is not switched off with the store lighting.

What happens if a server room gets too hot?

Three things happen in sequence. Equipment first throttles, so switches and recorders slow down or drop packets and frames without alarming. Then components fail early, because electrolytic capacitors and semiconductors age faster at elevated temperature. Meanwhile UPS batteries degrade fastest of all, since sealed VRLA life roughly halves for every 10 degrees Celsius of sustained temperature above 25, so a cupboard at 35 degrees turns a four year battery into a two year one. Eventually equipment thermally shuts down, which in a store means the network, the tills and the recorded CCTV all stop at once. The damage that matters is usually done long before that point.

Where we work

Related services and equipment

• Single phase UPS installation and commissioning for checkout lanes and store comms racks
• UPS maintenance plans priced against a multi-site schedule
• Battery maintenance and replacement run by cohort across a store fleet
• UPS battery replacement cost calculator for building the fleet budget
• Server room and comms room cooling Australia
• Switchroom cooling for base building electrical rooms
• High temperature UPS systems for uncooled back-of-house and outdoor positions
• UPS for medical and vaccine fridges, the pharmacy equivalent of refrigeration monitoring power
• Elevator and lift UPS systems for AS 1735 and NCC obligations in centres and towers
• Warehousing and logistics sector for the distribution centres behind the stores
• Banking and financial services sector for ATM fleets and branch networks
• CRAC unit maintenance services for base building precision cooling

Rolling out a store standard, or fixing a fleet you inherited? Send us your site count, the states you trade in, what is currently installed and the age of the batteries if you know it. We will come back with a store standard, a cohort replacement schedule and a per-site installed price. Indigi Power and Cooling is Indigenous and Veteran owned, and registered with Supply Nation and ICN Gateway. Contact Indigi Power and Cooling UPS Maintenance Plans