Choosing a UPS for a remote or unmanned site is less about picking a model and more about engineering years of unattended, autonomous operation, because when the grid drops there is no one on site to babysit the load. The right specification starts from the duty cycle and ends with a management plan that pages you before a failure, not after. Get those two ends right and the hardware in the middle is straightforward.
The first number is the real load in watts, measured at the site, not estimated from a nameplate. A remote site often carries a mixed load: radios or controllers, a small server, lighting and a heater, each with its own inrush. Record the peak, because UPS capacity is set by the worst minute, not the average. If you only have the average, size for at least 1.3 times it to leave headroom for motor start and future growth.
The second number is how long the load must keep running with no grid. A site with a reliable generator needs only five to ten minutes of battery to cover start and stabilisation. A site with no generator needs enough autonomy to ride out the longest realistic outage, which can mean thirty minutes to several hours and a much larger battery.
Convert the peak load to the UPS rating with this sequence. Take peak watts W, divide by the output power factor pf (typically 0.8 to 0.9) to get volt-amperes: kVA = W / (1000 x pf). Then add redundancy: for N+1, divide the required kVA by the module size and round up one module. As a worked example, a 40 kW load at 0.8 power factor needs 50 kVA; with 25 kVA modules that is two modules for the load plus one spare, so a 75 kVA frame.
Battery runtime follows from energy, not from the inverter. The cell capacity in ampere-hours is roughly Ah = (W x T) / (60 x V x efficiency x depth_of_discharge), where T is minutes, V the string voltage, and depth of discharge about 0.5 for lead-acid or 0.8 for lithium. The same 40 kW load for thirty minutes at 240 V with 0.9 efficiency and 0.5 discharge needs around 580 Ah - a large string, which is why autonomy longer than an hour pushes operators toward a generator or photovoltaic support.
An unmanned site without monitoring is a silent failure waiting to happen. Specify an SNMP card or an integrated cellular modem so the UPS reports status, battery state of health, alarms and self-test results to your network or a cloud portal. The value is not the dashboard; it is the early warning. A battery that fails its weekly self-test should open a ticket before the next outage, not after it.
Management also supports planned service. With trend data you dispatch a technician only when needed, which matters when the site is three hours down a rough road. Treat remote visibility as part of the specification, equal to the inverter, not as an accessory.
A remote site rarely sits in a clean, air-conditioned room. It may be a cabinet on a tower, a shed in a field or a housing at a pumping station. The UPS must tolerate the envelope temperature and the dust load. For harsh or outdoor placements favour IP-rated, temperature-controlled enclosures and wide-temperature electronics, and confirm the cooling capacity at the worst-case ambient rather than the lab figure. The same rules that govern an outdoor cabinet apply here, because the remote site is effectively an outdoor site that happens to have a UPS in it.
Decide what backs the UPS up. Where a generator is present and starts reliably, keep battery small and trust the generator for duration. Where the grid is weak but the sun is strong, a photovoltaic generator feeding the same DC bus can cut diesel runtime by sixty to eighty percent and shrink the battery needed for daytime events. Where neither is available, the battery alone carries the load, so runtime directly drives cost and footprint. Match the source to the site instead of buying one configuration for every location. Our telecom and off-grid power configurations are detailed at https://www.upsboss.com/telecom/.
Reliability at a remote site is set as much by logistics as by hardware. Define a mean-time-to-repair target, then place spares and a trained contact within that window. Favour a supplier with local reach and a stated response time, because a cheap unit with no support is expensive during the first incident three hundred kilometres from the nearest town. Lithium's longer service life helps here, since it stretches the interval between visits and reduces the number of battery truck-rolls over the asset's life.
Give us the peak load, the required autonomy and the environment, and our engineers will return a sized, remotely managed configuration. Review the full range of UPS and energy-source models at https://www.upsboss.com/products/.
Key takeaway: for remote and unmanned sites, specify from the measured duty cycle and the autonomy window, build in remote management as a first-class requirement, and match the energy source to the location - the goal is years of unattended operation with a warning before any failure.
How much runtime should a remote site carry?
Only as much as the gap to the next source. With a reliable generator, five to ten minutes covers start and stabilisation; with no generator, runtime is a business decision that can extend to hours and drives battery size directly.
Can one UPS serve several remote sites?
Not one unit for several sites, but one proven configuration rolls out across them. Re-size each deployment to its own load and environment rather than copying a template, then manage them all from the same monitoring portal.
Is a generator still needed at a solar-backed site?
Often yes for long outages, but a photovoltaic generator can cut diesel runtime by sixty to eighty percent for daytime events and shrink the battery needed, which lowers cost and the number of fuel deliveries.
Contact: Frank Zhang
Phone: +86-135 5688 8641
Email: frank@upsboss.com
Add: Jufeng Road, Guangming Street, Guangming District, Shenzhen City, Guangdong Province, China
