AOV Power Budget: How to Verify a Panel and Battery Can Actually Sustain 24/7 Recording

You check it with an energy balance, and it takes four numbers, only one of which is printed on the datasheet. Convert the battery's capacity rating into watt-hours, obtain the camera's average power draw in continuous AOV mode over a full 24-hour cycle, convert the panel's rated wattage into realistic harvest at your site in your worst month, and then test two things: does daily harvest exceed daily consumption with margin, and does the battery alone carry the camera through the longest run of sunless days your location actually gets. A camera passes or fails on those two inequalities. The headline "solar panel + big battery" on a product page tells you nothing, because neither figure means anything until it is placed against the other and against your latitude.

This is the audit that separates an always-on camera that survives a rainy season from one that browns out into PIR-only behaviour every July and quietly stops being the thing you bought.

The two inequalities

Everything below reduces to these. Use consistent units — watt-hours for energy, watts for power.

Test 1, daily balance: panel harvest per day (Wh) ≥ camera consumption per day (Wh), in the worst month, with headroom.

Test 2, autonomy: usable battery energy (Wh) ≥ camera consumption per day (Wh) × the number of consecutive low-sun days your site records.

Test 1 alone is not enough — a system can be balanced on an annual average and still die every wet season. Test 2 alone is not enough either — a large battery that never fully recharges just postpones the failure by a few days. Both must hold, in the same month, using the same consumption figure.

Step 1 — turn the battery headline into energy

Battery capacity is almost always advertised in milliamp-hours, which is not a measure of energy and cannot be compared between products at different pack voltages. Convert:

Energy (Wh) = (capacity in mAh ÷ 1000) × nominal pack voltage (V)

So the first thing to ask any supplier for is the nominal pack voltage, not just the mAh number. Two packs quoted at the same mAh hold different amounts of energy if their cell configurations differ, and a spec sheet that gives mAh without voltage is not a spec sheet. Then apply two reductions the marketing figure ignores:

• Usable depth of discharge. No sane battery management system runs a lithium pack to zero; the controller reserves a portion to protect cycle life. Ask what the low-voltage cutoff is and what fraction of the nameplate that leaves you.
• Cold-weather capacity. Available capacity falls at low temperature, and charging behaviour changes further still. If your sites see cold nights — highland Vietnam, Andean altitudes, a desert winter — the winter number is the one that matters, and it is not the number on the box.

Step 2 — get the number the datasheet hides

This is where nearly every procurement decision goes wrong. Continuous AOV recording has an average draw, and it is not the "standby current" on the spec sheet.

Standby current describes a camera doing nothing: sensor idle, encoder idle, radio asleep. An AOV camera is by definition never in that state. Its real 24-hour average is a blend of several distinct load states, and you need the blend, not the floor.

Two consequences follow. First, night is not free — infrared illumination runs for roughly half of every 24 hours, so any power budget built on a daylight measurement is wrong by a large factor. Second, the radio is the variable you control: the same hardware passes or fails the energy balance purely on how much live viewing your team does. Budget for the viewing policy you will actually have, not the disciplined one you intend.

Why an SoC low-power-mode figure is not your answer

Camera silicon vendors publish impressive low-power numbers for their imaging SoCs, and integrators quote them. Those figures describe the chip in a specific mode on a bench — not the camera. Your device also has an image sensor, an ISP pipeline, IR LEDs, a microSD controller doing sustained writes, a 4G module with its own power-amplifier peaks, and regulators losing energy at every conversion. The SoC figure is a floor for one component. A supplier who quotes it as the camera's consumption has answered a different question than the one you asked.

Step 3 — turn panel wattage into harvest at your site

A panel's rating is measured under standard test conditions: a defined irradiance, a defined cell temperature, a defined spectrum. Your fence post is not a laboratory.

Daily harvest (Wh) = panel rating (W) × peak sun hours at the site × derating factor

Peak sun hours is a local climate value for the specific month you are worried about, taken from irradiance data for your latitude — not an annual average, and never a global one. The gap between a dry-season month and the worst month of a wet season is what kills deployments.

The derating factor covers everything between the sunlight and the cell: panel tilt and azimuth against the sun's path in that month, shading from a shed or a tree at 9am that was not there when you surveyed at noon, dust and bird droppings on the glass, cell temperature above the rated condition, charge-controller and cable losses. You determine it for your installation. Anyone handing you one universal derating number for every site in every country is guessing on your behalf.

The three site errors that break otherwise correct maths

1. Winter sun angle. A tilt chosen for summer loses a large share of its harvest in the month you most need it. Size for the worst month; the rest of the year is surplus.
2. Partial shading. A small shadow across a panel costs far more output than its area suggests. Survey the mounting point at low sun angles, not at midday.
3. Soiling. In dusty agricultural and construction environments, output decays between cleanings — and on a remote fence line nobody will ever clean the panel. That is a permanent derating, not a temporary one.

Step 4 — the bench test that settles the argument

Calculation narrows the field; measurement decides. Before committing to a quantity, take sample units and do this:

1. Charge fully, then disconnect the panel entirely.
2. Configure it exactly as the customer will: same resolution, same AOV settings, same alert sensitivity, same live-view frequency.
3. Point it at a scene with real activity — a busy gate, not a blank wall. A camera aimed at a static wall gives you a beautiful, meaningless runtime.
4. Log the time to low-battery cutoff. That gives you the true consumption per day: usable energy ÷ days survived.
5. Repeat it with night-time hours included, because IR is half the answer.

That single number, measured on your scene, replaces every figure on the datasheet. Feed it back into Tests 1 and 2 and you have a defensible design instead of a hope. Sample orders run 7–10 days by air, which makes this a two-week exercise, not a two-month one.

What to request in writing before you order

• Nominal pack voltage alongside the capacity rating, and the usable fraction after the low-voltage cutoff.
• Average draw over a full 24-hour cycle in continuous AOV mode, at the resolution being quoted, stated separately for day and night.
• Whether the camera reduces bitrate or frame rate under scene analysis when nothing is moving, and how much that saves.
• Panel rating with the test conditions it was measured under.
• Charge-controller behaviour when the battery is deeply discharged and the sun is weak — does it recover, or sit there?
• What the camera does when the battery is low: does it degrade to event-only recording, and does it tell you it has done so?

That last one matters more than it looks. A camera that silently falls back to motion-triggered behaviour when the budget fails looks fine in the app and is not doing the job — the exact failure mode always-on recording exists to prevent. Our AOV solar camera guide covers that difference in the field, and the product catalogue lists all 150 SKUs individually — ask for the power figures on the exact model you are quoting, because they vary within a series and cannot be generalised across one.

The commercial side of getting this right

An undersized system is not a technical embarrassment, it is a warranty and reputation cost — site visits, replacements, a customer who tells other farmers. All outdoor models are IP66, 1080p–2K standard with 4K available as a custom build, and install in 5–10 minutes per camera. The manufacturer, Shenzhen Leksell, has built this category since 2012 with TÜV Rheinland certification plus CE, FCC and RoHS. MOQ is 10 units for a trial and 100 for OEM/ODM; bulk sea freight runs 25–40 days, and 7–10 business days within Vietnam from our Ho Chi Minh City office. Dealer margin starts at 20% from around 10 units and 30% from around 50 — see dealer terms.

Send your site's worst-month irradiance and your viewing policy and we will size against it. WhatsApp +86 188 1869 9870 — https://wa.me/8618818699870 · info@solaguard.net