Building a 24/7 Solar-Powered Network Off Grid

A 24/7 solar-powered network for remote locations must be engineered as a complete energy and communications system, not assembled by matching headline wattages. Start with the services that must remain available, measure every continuous and temporary load, then size the battery and solar array against the site’s worst solar month and required recovery period.

The design also needs managed PoE distribution, a suitable backhaul, automatic load shedding, controlled restart logic, remote monitoring and a practical maintenance plan. Continuous operation remains an engineering objective rather than a universal product promise: actual performance depends on site-specific sunlight, shading, temperature, radio conditions, equipment settings and maintenance.

Define network services and an availability objective

Before choosing panels, batteries or switches, define what the remote site must do. A surveillance location, plantation Wi-Fi link, perimeter camera point and environmental monitoring station may all use similar equipment, but they do not have the same service priorities.

List the required services in operational terms:

Video recording at the site
Live camera access from a control room
Wireless backhaul to another location
Local Wi-Fi access
Sensor or alarm transmission
Remote equipment management
Auxiliary lighting or warning devices
Maintenance access to logs and diagnostics

Then classify each service by priority. A practical three-tier model is:

Priority	Service behaviour during low energy	Typical examples
Tier 1: Essential	Remains powered for as long as the design allows	Core router, critical camera, alarm link, system controller
Tier 2: Operational	May be reduced or temporarily disconnected	Secondary cameras, local access point, non-critical sensors
Tier 3: Deferrable	Switched off first and restarted last	Lighting, convenience Wi-Fi, maintenance-only equipment

An availability objective should specify more than “24/7”. Record:

The services that must remain active overnight
The low-solar period the battery must bridge
Which loads may be shed during an energy shortage
How quickly the system should recover after sunlight returns
Whether local recording must continue if the backhaul fails
What constitutes a reportable outage
How technicians will access the site if remote recovery fails

Avoid assigning an uptime percentage unless the design includes the data, redundancy, monitoring and maintenance process needed to support it. For many projects, a clear operational statement is more useful—for example, “keep the essential network tier active through the specified low-harvest period, then restore secondary loads only after battery recovery”.

Inventory continuous, intermittent and startup loads

A reliable load schedule separates equipment by how and when it consumes power. Using only each device’s maximum label rating can result in an oversized system, while using optimistic average figures can leave the site without enough storage.

Continuous loads

These operate throughout the day and usually determine the base energy requirement:

PoE switch electronics
Router or gateway
Wireless backhaul radio
Cameras configured for continuous operation
Network video recorder or edge storage device
Solar controller and monitoring electronics
DC converters and protection equipment

For each item, record its normal measured input power, maximum expected power, operating voltage and hours of operation.

Intermittent loads

Some devices change consumption according to time, traffic or environmental conditions:

Infrared camera illumination after dark
Pan-tilt-zoom movement
Heaters, fans or cabinet ventilation
Alarm sounders
Site lighting
Cellular transmission bursts
Maintenance sockets or temporary equipment

These loads need a duty-cycle assumption. The assumption should be conservative and documented so it can be checked during commissioning.

Startup and peak loads

Startup current may be brief, but it can trigger a controller, converter or protection device if the peak exceeds its limit. Check:

Simultaneous PoE device startup
Radio boot-up
Motor movement in PTZ cameras
Infrared illuminators activating together
DC converter inrush
Recovery after a complete low-voltage shutdown

Sequence high-demand devices where possible rather than starting everything at once.

Build the daily energy budget

For each load:

Daily energy in Wh = operating power in W × operating hours

Add continuous and intermittent energy, then account for losses between the solar array and the final powered device. These can occur in the controller, battery, wiring, DC conversion, PoE conversion and network equipment.

Do not treat battery nameplate energy as energy that can always be delivered to the loads:

Nominal energy is the battery’s voltage multiplied by its rated ampere-hours.
Usable battery energy is the portion permitted by operating limits and the battery management system.
Usable delivered energy is what remains at the devices after wiring and conversion losses.

Cable length matters as well. Excessive voltage drop can make equipment reset even when the battery still contains energy. Check conductor size, cable distance, connector resistance, current and the minimum permitted input voltage of every device.

For a detailed camera-focused approach, see the solar PoE switch guide for remote IP camera networks.

Size worst-month storage, solar and recharge capacity

Solar sizing should use site-specific conditions for the most difficult operating period, not an annual average. In Malaysia, cloud cover, heavy rain, heat, vegetation growth and local shading can materially affect harvest and equipment conditions. Obtain suitable solar resource information for the precise site and assess obstacles at the planned panel position.

Size storage around the required bridge period

Battery storage must cover the selected load tiers for the required low-harvest period. The basic design relationship is:

Required delivered storage = protected daily load × required bridge duration

The nominal battery must then be larger than the delivered requirement because discharge limits, conversion losses, temperature conditions and ageing allowances affect the available energy.

Define two storage cases:

Normal operating case: all intended loads run according to their schedules.
Energy-conservation case: only essential loads remain active after thresholds are reached.

This makes the system’s behaviour predictable. It also prevents a non-critical load from exhausting energy needed for the backhaul or primary camera.

As one architecture reference, the Maxsol CS400150 400W and 1920Wh solar system documents a 400W panel configuration using two 200W modules, a 12.8V 150Ah 1920Wh LiFePO4 battery, MPPT control and four DC12V outputs. These are nominal product details, not a runtime guarantee. Suitability depends on the measured site load, usable delivered energy, solar input, controller limits and required reserve.

Size the array for operation and recovery

The array must do more than cover the current daytime load. After a poor solar period, it also needs enough surplus energy to recharge the battery while the network remains operational.

The daily solar energy requirement therefore includes:

Energy used by the network during the day
Energy needed to replace overnight battery discharge
Conversion and charging losses
Additional recovery energy following low-harvest days
A design allowance based on the project’s uncertainty and operating conditions

A system can have enough storage to survive one difficult period but still fail operationally if the array cannot restore that storage before the next one.

Model recovery explicitly:

Start at the chosen low battery threshold.
Apply the assumed worst-month solar harvest.
Keep essential network loads operating.
Calculate how much charging energy remains.
Determine when secondary loads may safely restart.
Repeat the model across consecutive poor-harvest days.

Panel orientation and placement should also support safe maintenance. Check for future tree growth, mud splash, leaf accumulation, bird fouling, roof shadows and structures that may create new shading.

Common sizing and integration problems are covered in off-grid solar power design mistakes to avoid.

Design PoE switching, backhaul and load priority

The network architecture must match the energy strategy. A large battery cannot compensate for an unmanaged switch that keeps every non-critical device powered until the battery protection system disconnects the entire site.

Select the PoE switch as a control point

Check the following before specifying a switch:

Number and speed of copper ports
Required SFP or uplink interfaces
Total PoE budget
Maximum requirement of each powered device
Supported PoE mode and voltage
Switch input voltage range
Management and remote diagnostic features
VLAN, loop control and restart capabilities required by the project
Environmental installation and enclosure requirements

The Maxsol IOT2802SFG-S managed solar industrial PoE switch documents eight Gigabit PoE ports, two SFP ports, a 120W PoE budget, 90W BT support, PV and battery monitoring, and configurable 24V or 48V PoE on ports 3–4. Its documented PV and battery input is DC22–29.2V.

That input detail is critical: the documented 12.8V CS400150 kit is not a direct compatible input for a DC22–29.2V device. The linked products illustrate alternative architecture components, not a validated bundle. Do not connect them together without confirming voltage architecture, controller limits, interfaces, protection and any required conversion stage.

Also confirm that the planned PoE device combination remains within the total budget. A switch supporting a high-power port does not mean every port can simultaneously provide its individual maximum.

Choose the backhaul based on the real path

A remote site may use fibre, cellular or point-to-point wireless backhaul. Compare them according to access, terrain and operational needs:

Backhaul	Main advantage	Important checks
Fibre	Stable high-capacity connection where infrastructure is practical	Route, termination, surge exposure, civil work and repair access
Cellular	No remote radio endpoint required	Site coverage, antenna placement, data plan, network variability and router power
Point-to-point wireless	Useful where a clear radio path exists	Line of sight, Fresnel clearance, alignment, interference, mounting and endpoint power

The Maxsol CPE295 5.8GHz wireless CPE documents 5.8GHz operation, a 2×2 14dBi directional MIMO antenna, one Gigabit LAN port and 6.5W full-load consumption. Its suitability cannot be decided from distance alone. Verify the complete radio path, local interference, alignment, mounting stability, Ethernet interface and power method.

Apply priority at the port level

Where the switch and control architecture permit it, assign each powered device to an energy tier. A sensible shutdown order might be:

Disconnect auxiliary lighting.
Disconnect convenience Wi-Fi.
Disconnect secondary cameras.
Preserve the primary camera, controller and backhaul.
Shut down the remaining system in a controlled manner before the battery reaches its protection limit.

Use hysteresis between shutdown and restart thresholds. Without it, a load may repeatedly switch on and off as the battery voltage briefly recovers.

Monitor energy, faults and controlled recovery

Remote monitoring should explain both network and energy conditions. A simple “online/offline” indicator is not enough to diagnose whether a failure comes from the radio path, a PoE device, low battery energy or a tripped protection device.

Useful monitoring points include:

PV voltage and charging state
Battery voltage and relevant controller status
Load current or power
PoE port state
Switch and backhaul reachability
Device restart history
Low-energy and overcurrent events
Cabinet temperature where relevant
Repeated link changes or packet loss
Time since the last successful remote report

Set alarms to give operators time to act. A warning threshold should occur before the automatic load-shedding threshold, which in turn should occur before final battery protection disconnects the site.

Plan a controlled restart sequence

When solar energy returns, do not immediately reconnect every load. A staged process reduces simultaneous startup demand and protects the recovery reserve:

Confirm stable controller and battery conditions.
Start the core switch or control layer.
Restore the backhaul.
Verify remote management connectivity.
Start the essential camera or sensor group.
Observe load and battery response.
Restore secondary devices after a defined recovery threshold.
Restart deferrable loads last.

Include retry limits. A faulty PoE camera that continuously restarts should not consume the site’s remaining energy indefinitely. The management plan should isolate the affected port, log the fault and preserve the core network where possible.

Test handover and maintain the remote site

Commissioning should prove the design logic, not merely show that the equipment powers on during a sunny afternoon.

Run a hypothetical worst-month acceptance test

A useful acceptance test can simulate the design’s stated worst-month conditions without claiming that the test predicts universal uptime.

Use this process:

Measure the final installed load. Record normal, night, peak and startup consumption with the actual configuration.
Apply the project’s low-harvest assumption. Limit or simulate charging according to the solar input used in the design model.
Start from the defined battery condition. Use the agreed normal or reduced state rather than an arbitrary full charge.
Operate the planned service schedule. Include infrared operation, wireless traffic and other timed loads.
Trigger energy thresholds. Confirm that Tier 3 and Tier 2 loads disconnect in the correct order.
Verify essential services. Check that the protected network tier behaves as specified.
Restore the simulated solar input. Confirm that charging supports the active essential loads.
Observe staged recovery. Ensure devices restart in sequence rather than simultaneously.
Test individual failures. Disconnect the backhaul, interrupt a PoE device and simulate a controller alarm.
Record the results. Compare measured behaviour with the acceptance criteria established during design.

The pass criteria must come from the project’s energy model and service objective. They should not be replaced with a generic runtime or uptime claim.

Prepare the site for maintenance

Remote installations still require physical service. Design access before handover:

Label every cable, fuse, breaker, port and polarity
Keep an updated wiring and network diagram
Record device settings and approved firmware baselines
Provide safe panel and cabinet access
Document shutdown and restart procedures
Specify cleaning and inspection intervals based on site conditions
Check cable glands, seals, mounting hardware and signs of water entry
Inspect for corrosion, pests, vegetation and physical damage
Keep suitable replacement protection parts and preconfigured critical spares
Record battery and controller alarms during every visit

Spares should reflect the site’s failure impact and travel difficulty. Possible items include fuses, surge protection modules, connectors, injectors or converters, a configured radio, and a compatible switch. Confirm exact part compatibility rather than assuming visually similar components are interchangeable.

Final design checklist

Before approving a 24/7 solar-powered network for a remote location, confirm that:

Every continuous, intermittent and startup load is documented
Nominal battery energy is separated from usable delivered energy
Solar sizing uses site-specific worst-month assumptions
The array has enough recovery capacity, not only daily operating capacity
PoE port demand fits within the total switch budget
Battery, controller, switch and load voltage ranges are compatible
The wireless path has been assessed for clearance and interference
Essential and deferrable loads have defined priorities
Low-energy alarms occur before automatic shutdown
Restart logic is staged and includes retry limits
Acceptance tests verify low-harvest and recovery behaviour
Drawings, settings, spares and maintenance access are ready for handover

As a practical next step, compare the linked Maxsol architecture examples against your required port count, measured load, radio distance, voltage and power design, installation environment, and management functions. Verify every interface and controller limit independently rather than treating the products as a pre-approved system bundle.