A gateway can remain powered during an outage and still lose its connection to the application. The reason is often one step further along the network: a switch, router or internet terminal has stopped working. A PoE UPS protects the equipment connected to its supported output, not automatically the entire communication path.
Start by deciding what must continue operating and for how long. That makes it easier to choose a suitable PoE battery backup, identify other power dependencies and test the result without relying on a runtime claim alone.
What does a PoE battery backup actually protect?
Power over Ethernet carries electrical power alongside data over a suitable network connection. A compatible UPS can maintain that power when its normal supply fails. The precise installation depends on the UPS design and the equipment it supports.
Keeping the endpoint powered can preserve local operation and avoid an immediate restart. It does not guarantee that upstream networking, internet access or a remote service remains available. Those are separate dependencies with their own power arrangements.
For a monitoring gateway, draw a simple list from the gateway to the destination application. Include every local device needed along that route. The first unprotected component may determine what remains useful during the outage.
Check PoE compatibility beyond the voltage label
A label saying 48 V is not enough to prove that two PoE products are compatible. Standards-based PoE and passive power arrangements differ, and supported power negotiation, wiring and output limits matter.
Check the exact UPS output specification against the receiving device’s requirements. Do not connect equipment based only on the shape of its Ethernet socket. A matching connector does not establish a matching electrical interface.
The PoE-03 in the store lists specific RAK gateway compatibility and a 48 V, 0.23 A PoE output. Use the current documentation for the exact combination rather than assuming that this makes it suitable for every PoE camera, access point or switch.
Output power and battery capacity answer different questions
The output rating determines whether the UPS can supply the connected load. Battery energy helps determine how long it can do so. A large battery cannot make an undersized output support a device whose demand exceeds the port’s limit.
For the PoE-03’s stated 48 V and 0.23 A output, multiplication gives approximately 11 W at that stated output rating. That is a useful compatibility check, not a promise that every 11 W device will work under all conditions.
Startup demand, supported power method and the manufacturer’s operating limits still need checking. Where a UPS provides several outputs, do not assume all maximum ratings can be used simultaneously without confirming the total supported budget.
PoE UPS runtime: begin with watt-hours
A first estimate divides usable battery energy in watt-hours by the connected load in watts. If 20 Wh were usable at a steady 5 W load, the simple estimate would be four hours. This is an illustrative calculation, not a rating for a particular UPS.
Real runtime also depends on conversion losses, the UPS’s own consumption, battery condition, temperature and the load’s changing demand. Rated battery energy is not necessarily the energy delivered to the device’s input.
The PoE-03 is listed with 29.6 Wh rated capacity and an up-to-two-hour backup description. Treat the published runtime as conditional on its test circumstances, not as a guaranteed duration for any connected equipment. Verify the actual installation against the duration you need.
Milliamp-hours need a voltage to become an energy comparison
Battery products are often advertised in mAh. That charge capacity does not by itself tell you how much energy reaches a load at another voltage. Comparing two UPS units only by their mAh figures can therefore be misleading.
Watt-hours provide a more useful basis when the rating is properly defined. Even then, compare the supported outputs, usable capacity and operating conditions. A battery pack designed for one DC interface is not automatically a PoE power source.
For example, the Milesight UPS01 versions in the store use particular Type-C or M12 arrangements. Those interfaces need their own compatibility check; they are not interchangeable with a PoE output merely because all the products provide backup power.
Protect the communication path that matters
Consider a gateway connected through a local switch and router. If only the gateway has backup power, it may keep listening locally while losing its onward connection. Whether it can retain information for later delivery depends on its actual supported behaviour.
If continued remote visibility is required, include the necessary switch, router and internet termination equipment in the backup design. Check their combined load and the available backup time rather than assuming one small unit covers the whole site.
A cellular alternative can change the dependency list, but it does not remove every dependency. The gateway, modem, service and local coverage still need to support the intended operation. Test the path you actually plan to use.
Test a controlled outage, including restoration
Once the installation is correctly assembled, plan a controlled test with the people responsible for the equipment. Observe whether the endpoint stays running when normal power is removed and whether data continues to reach its destination.
Record the load, battery state, conditions and observed duration. A brief changeover test answers a different question from a sustained runtime test. Both can be useful, but neither should be presented as evidence of the other.
Then restore normal power and check recovery. Confirm that charging resumes as expected, communications remain stable and any affected devices recover correctly. A backup arrangement is only fully understood when its return to normal operation has also been checked.
Repeated outages can arrive before the battery is ready
The first outage may begin with a fully charged battery. A second outage shortly afterward may not. Recharge time matters when the site experiences several interruptions close together.
Include this possibility in the operating plan. Decide how battery condition is checked and how reduced reserve is communicated to the people relying on the system. Where status monitoring is supported, verify what the reported state actually means.
Battery ageing also changes the available reserve. A successful commissioning test is a starting reference, not a lifetime guarantee. Periodic checks should reflect the importance of the supported equipment and the manufacturer’s maintenance guidance.
A UPS is not a substitute for every protection measure
Backup power addresses interruptions within its design limits. It does not make an installation immune to damaged cables, failed networking, water ingress or every electrical disturbance. Any surge protection included with a particular package has its own stated scope.
Use the required mounting, environment and cable protection for the complete installation. Do not place an indoor backup unit outside simply because the powered gateway is rated for outdoor use.
The useful outcome is a clear statement: which devices remain powered, which functions remain available and for approximately how long under tested conditions. That statement is a stronger basis for choosing a PoE UPS than battery size alone.
Compare the UPS and battery backup options against your endpoint’s power requirements and the complete communication path you need to preserve.