門鈴攝影機的電池耗電取決於喚醒頻率，而非運作時間

Battery capacity is the number every doorbell camera is sold on, and it is the number that explains the least about how long the product actually lasts between charges. A 5000 mAh cell sounds like it should give you a predictable number of weeks, and then the first month in a busy doorway gives you half of what you expected. Nothing is faulty. The capacity was never the variable that mattered.

The camera is asleep almost all the time

A battery doorbell spends the overwhelming majority of its life in a low-power state where the imaging path is off and only a motion sensor is listening. It is not recording, it is not streaming, and it is not holding a full Wi-Fi session.

This is the only reason a battery doorbell is possible at all. A camera sensor plus an encoder plus a Wi-Fi radio running continuously would drain any reasonably sized cell in hours, not weeks. The entire product category rests on the assumption that interesting moments are rare.

Which means the interesting question is not how big the battery is. It is how often the device decides something is interesting.

Every wake-up costs more than the sleep that follows

When the sensor trips, the device has to come up fully. The imaging path powers on, the processor starts encoding, the Wi-Fi radio has to re-establish its association with the access point, and the clip has to be pushed to the cloud or written to local storage. Then it goes back to sleep.

That sequence is expensive, and the expensive part is disproportionately the network. Bringing a Wi-Fi radio up from sleep, authenticating, and completing a transfer costs far more energy per second than the sensor did during the entire idle period that preceded it. A weak signal makes it worse, because a marginal link means retransmissions and longer transmit time for the same number of bytes.

So the energy budget is dominated by transitions, not by duration. Ten wake-ups cost meaningfully more than one wake-up that lasted ten times as long, because every one of the ten pays the full startup cost again.

What actually trips the sensor

Most doorbell cameras use a passive infrared sensor to decide when to wake, sometimes combined with pixel-change detection on a very low-resolution feed. Both are crude, and both fire on things you do not care about.

The triggers that quietly destroy battery life are usually not people at the door. They are passing traffic, a tree moving in the wind, a neighbour's cat, the sun coming out and warming a surface the sensor can see, and the reflection off a car windscreen as it pulls into a driveway across the street.

A doorway facing a street can see dozens of these an hour. A doorway facing a shared hallway sees far fewer. Two identical cameras with identical batteries in those two positions will have completely different battery lives, and the difference has nothing to do with the product.

Why a bigger battery helps less than the spec suggests

Capacity scales the number of charge cycles you get, but it does not change the ratio between wake cost and sleep cost. If a doorway generates a hundred wake-ups a day, a larger cell gives you proportionally more days, which is useful but not transformative. What it does not give you is immunity from the trigger count.

This is why the honest answer to "how long does the battery last" is always a range rather than a figure. The manufacturer can only quote a number by assuming a trigger rate, and the assumed rate is typically a quiet residential door with a handful of events per day. Real doorways vary by more than an order of magnitude around that.

It is also why the most useful thing a vendor can tell you is not the amp-hour figure but what the device does to reduce unnecessary wakes.

The settings that actually move battery life

Detection sensitivity is the first lever, and it is worth turning down further than feels right. A slightly missed distant event costs less than doubling your charge frequency.

Activity zones matter more than people expect. Excluding the road, the pavement and the neighbour's driveway removes the majority of traffic-based triggers while keeping everything that happens at the threshold itself.

Clip length has a direct effect, because recording and uploading is the expensive phase. Shortening clips from thirty seconds to fifteen roughly halves the upload energy per event.

Human detection, where the device supports it, is the single biggest win available. It moves the decision from "something infrared moved" to "something shaped like a person moved", and it filters out most of the categories that generate false wakes.

Live view is the hidden cost. Every time you open the app and watch the feed live, you are holding the radio and the imaging path awake for the whole session. Two minutes of live viewing can cost more than a dozen recorded events, and people rarely count it.

What a realistic month looks like

Expect the first month to be the worst one. A newly installed camera has no tuning applied, sensitivity is at default, no zones are set, and everyone in the household checks the live view because it is new. Battery life improves after tuning, and it improves again once the novelty wears off.

Plan the charging around the doorway, not around the spec sheet. If the camera faces anything that moves, assume the lower end of whatever range you were quoted and check the trigger log after a week. The log tells you the truth about that specific position faster than any amount of spec reading.

If charging turns out to be genuinely awkward, that is the point at which a wired or solar-assisted unit is worth considering rather than a bigger battery. The problem being solved is trigger count, and if the position cannot be tuned down, the answer is more incoming energy rather than more stored energy.

For peephole-mounted units where the cell has to fit behind the door rather than in a wall box, the 3MP Wi-Fi peephole video doorbell is built around that constraint, and the wider smart peephole doorbell camera range covers the same fitting problem at different resolutions. Where the doorway makes recharging impractical altogether, solar-assisted wireless doorbells address the incoming-energy side instead. For choosing between these trade-offs against the rest of the house, the home security camera selection guide covers the same decisions in wider contexts.

The short version

A battery doorbell is a sleeping device that pays a large fixed cost every time it wakes. Capacity sets how many of those charges fit in the tank; the doorway sets how often they happen. Tune the triggers, draw the zones, shorten the clips, and stop checking live view, and you will get more out of the battery than any spec sheet comparison will show you.