
A bank branch keeps its teller terminals, ATM controller, network switch, surveillance recorder and safe lock alive through a grid fault by placing those loads behind an online UPS sized to at least the measured critical load plus thirty percent headroom. The branch avoids a dropped transaction, a frozen ATM and a dark lobby because the protected equipment never sees the interruption, and that continuity is what turns a power event from a customer-visible failure into a non-event.
A branch is not a quiet office. Cash-handling equipment, encryption devices and transaction networks all expect a steady supply, and a momentary interruption can abort a transfer, desynchronise a terminal or take an ATM offline mid-transaction. Beyond the immediate failure, a visible outage during business hours erodes customer trust faster than almost any other fault, because the customer is standing at the counter when it happens. Regulators in many markets also expect financial premises to demonstrate resilience, so the power design is both an operational and a compliance matter rather than a convenience.
The first step is to meter, not to guess. A typical branch carries a genuine critical load in the range of two to five kilowatts: the core switch and router, the teller and back-office terminals, the ATM controller, the surveillance network video recorder, the safe and alarm circuit, and the branch server or thin-client host. Lighting, air conditioning and staff PCs are usually excluded from the protected budget. Metering the real draw for a week shows the true figure, and that figure, not the nameplate of one device, sets the UPS size. A branch measuring 3.2 kW critical load would land on a 5 kVA / 4 kW unit running at about eighty percent, leaving headroom for growth and for battery recharge.
For transactional loads, online double-conversion is the safe default. The rectifier and inverter hold the load continuously, so grid sags, swells and brief interruptions never reach the protected equipment, and transfer time is effectively zero. Standby or line-interactive units that switch to battery in a few milliseconds are acceptable for back-office desktops, but the systems that move money deserve the continuous conditioning an online topology provides. The small efficiency cost of double conversion is paid back by the absence of transaction aborts and the quiet confidence that every event is absorbed.
Runtime is set by the branch continuity plan, not by a default. Most branches need ten to twenty minutes: enough to ride through a utility fault, enough for an orderly shutdown if the fault persists, and enough to bridge to a site generator where one exists. Ten minutes at a 3.2 kW load requires roughly a 24-volt battery string of about fifteen ampere-hours, which fits inside the UPS cabinet. Branches that must stay open through longer regional outages extend runtime with external battery cabinets rather than by oversizing the inverter.
A network management card turns the UPS from a silent box into a reporting asset. SNMP monitoring logs every event with a timestamp, tracks battery impedance, and pages the duty engineer when a test fails or a cell drifts. For a bank, that log doubles as the audit evidence regulators and internal risk teams ask for, proving that protection was exercised and healthy on a given date. Scheduled self-tests keep the battery honest without a site visit, and remote alarm delivery means a weak battery is replaced on a planned window instead of during a real outage.
A branch UPS is a heat source as well as a power source. A 5 kVA unit at full load sheds several hundred BTU per hour, so the cabinet needs ventilation and must not be tucked into a sealed cupboard with the network gear. Plan the connection path before delivery: external battery cabinets add a second enclosure and a DC cable run, and the management card adds a network drop. Sites that resolve the heat and cabling question during planning avoid the all-too-common outcome of a perfectly good UPS that nobody dares to switch on because the room is the wrong shape.
Our branch and transaction-grade UPS ranges are listed at https://www.upsboss.com/products/, and sizing and topology questions are answered in our https://www.upsboss.com/faq/ section. Send us your metered critical load and your continuity target and we will return a specification sized to keep the branch open through the next grid event.
Key takeaway: a bank branch protects its money-moving loads by metering the real critical draw, sizing an online UPS to that load plus headroom, and setting runtime to the continuity plan, then proving the result with monitored event logs. Continuity, not kVA headline, is the deliverable.
How big a UPS does a typical branch need?
Most branches sit between two and five kilowatts of genuinely critical load. Meter that load for a week, add thirty percent headroom, and the kVA follows; a 3.2 kW measured load lands on a 5 kVA unit running near eighty percent.
Is online double-conversion worth the efficiency loss?
For transactional equipment, yes. The cost of one aborted transfer or a frozen ATM far exceeds the small extra heat of double conversion, and the zero-transfer behaviour is exactly what those loads require.
Do I need a generator as well?
Not for a normal fault. Ten to twenty minutes of battery runtime rides through almost all utility events; a generator only becomes necessary where regional outages routinely exceed the battery window, and then external battery cabinets extend that window first.
Contact: Frank Zhang
Phone: +86-135 5688 8641
Email: frank@upsboss.com
Add: Jufeng Road, Guangming Street, Guangming District, Shenzhen City, Guangdong Province, China