Backup Power for Communications: Batteries, Generators, and Runtime Math

Power is the silent single point of failure

Most conversations about communications reliability focus on coverage, frequencies, and equipment. Those matter, but they all sit on top of one assumption: that power keeps flowing. A repeater with no power does not transmit. A dispatch console with a dead battery does not answer. A portable radio that will not hold a charge is a paperweight clipped to a turnout coat. The uncomfortable truth is that the largest failures of a radio system are rarely about radios at all. They are about the electricity feeding them.

What makes power dangerous as a failure point is that it is invisible until it is gone. A weak antenna shows up on a coverage map. A failing battery in a backup string can sit quietly at a site for months looking fine, and then collapse under load in the first minute of a real outage. The people who depend on the system never see the backup until the day it is called on, and by then it is too late to fix.

The events that stress communications hardest are exactly the events that also take down the grid. Severe weather, wildland fire, ice storms, and flooding all knock out power and simultaneously drive call volume through the roof. So backup power is not an edge case to plan for someday. It is the condition under which your system will be doing its most important work.

The layers of backup power

Backup power is best understood as a stack of independent layers, each covering for the one below it. No single layer is enough on its own, and a gap in any one of them can leave the whole chain silent.

• Utility (grid) power. The normal source. Reliable most of the time, and the thing every other layer exists to replace.
• Site and equipment batteries. At a repeater or tower site, a battery bank carries the load instantly when the grid drops, with no interruption. Consoles and network gear often have their own battery-backed supplies for the same reason. This layer is the bridge that keeps the system live during the seconds and minutes before anything else spins up.
• Generators. A generator is the layer that turns a short battery window into a long one. It cannot start instantly, so it is designed to take over from the batteries and then run for hours or days as long as it has fuel. Some sites have permanent standby generators; others rely on portable units that get delivered and connected when needed.
• Portable and subscriber batteries. The handhelds, mobile radios, and spare battery packs carried by crews. This is the layer closest to the person doing the work, and it is the one most often overlooked in a power plan. A perfectly powered tower site does no good for a crew whose portables all died three hours into an operation.

The right way to read this stack is from both ends. From the infrastructure side, ask whether the site stays up. From the user side, ask whether the person in the field can still talk. A resilient plan answers yes to both, and treats the two as separate problems that happen to share a name.

How to think about runtime

Runtime is where a lot of planning goes wrong, usually because people reach for a single number and treat it as gospel. It is more useful to think about runtime as the relationship between three things, and to keep them separate in your head.

• The load. How much power the equipment actually draws. A transmitter keying up under heavy traffic draws more than one sitting idle. Cooling, lighting, and network gear at a site all add to it. The load is not a fixed value; it rises and falls with how hard the system is being used, and it tends to be highest during the exact events that also cut the power.
• The stored energy. How much energy the battery bank actually holds right now. Note the phrase "right now." A battery's usable capacity is not its rating on the label. It shrinks with age, with temperature, and with how deeply it has been drained in the past. A bank that was sized correctly when it was installed may hold far less today.
• How long you need to run. The window you are actually trying to cover. Covering a brief flicker is a different problem than covering a multi-day event, and pretending one plan handles both is how sites go dark.

The math itself is simple in principle: more stored energy and a lighter load buy you more time, and a heavier load burns through it faster. What matters is refusing to trust a comfortable assumption at any of the three points. Do not assume the load is the idle load. Do not assume the battery holds its rated capacity. Do not assume the outage will be as short as the last one. The honest planning question is not "what is our runtime," as if it were one number, but "under a heavy load, with batteries as tired as they actually are, does our stored energy carry us long enough for the generator or the fuel truck to take over." If the answer depends on everything going right, you do not have a plan. You have a hope.

Brief outages versus extended events

Two very different problems hide under the word "outage," and conflating them is one of the most common planning mistakes.

A brief outage is a flicker, a switching fault, or a storm cell passing through. It lasts seconds to a couple of hours. Here the battery layer is the hero. The whole job of stored energy is to ride through the gap with no interruption, either until the grid returns or until a generator picks up the load. For brief events, if your batteries are healthy and your automatic transfer works, the users may never even notice.

An extended event is a different animal entirely. A multi-day ice storm, a regional grid failure, or a wildland event can keep the grid down far longer than any reasonable battery bank can bridge. No battery string is sized to run a site for days. Once you cross from hours into days, batteries stop being the answer and become the bridge to the real answer, which is a generator that keeps running. The planning center of gravity shifts from stored energy to fuel and to the logistics of keeping fuel flowing.

The practical implication is that you should plan the two cases separately. Ask "will the batteries carry us through a short outage cleanly," and then ask the completely different question, "can we keep a generator fueled and running for as many days as a worst-case regional event might last." A site can pass the first test and fail the second, and the second is the one that decides whether you stay on the air through the events that matter most.

Fuel and refueling dominate long events

Once an event runs long, the story stops being about electrons and starts being about logistics. A generator does not run on being present. It runs on fuel, and fuel has to be stored, protected, delivered, and replenished, all while the same disaster that took down the grid is also blocking roads and stretching every supplier thin.

Several hard realities show up during extended events, and they are worth thinking through before one happens:

• On-site fuel is finite. A tank runs down. The question is not whether you will need to refuel, but when, and whether anyone can reach the site to do it.
• Roads may be impassable. The event that killed the power often makes the site hard to reach. A refueling plan that assumes clear roads and normal delivery times is a plan for a good day, not a bad one.
• Fuel supply is contended. During a regional event, every hospital, shelter, and critical facility is competing for the same deliveries. Standing arrangements made in advance matter far more than phone calls made during the crisis.
• Fuel degrades. Stored fuel does not last forever. A tank that has sat untouched for a long time may not perform when it is finally called on, so stored fuel needs a rotation and inspection routine of its own.
• Someone has to physically do it. Refueling a running generator is a task that takes a trained person, the right equipment, and safe conditions. During a stretched multi-day response, personnel are the scarcest resource of all.

The takeaway is that runtime on a long event is not really a property of your generator. It is a property of your fuel logistics. A generator with a plan to keep it fed can run indefinitely. A generator with a full tank and no refueling plan runs exactly as long as that tank lasts, and then the site goes quiet.

Battery health, testing, and replacement

A backup battery has one job, and it does that job on rare occasions, usually with no warning. That combination makes batteries uniquely prone to silent failure. They sit fully charged, looking healthy, while their real ability to deliver under load quietly erodes. The only way to know a backup battery will perform is to test it before you need it, not to assume it based on the fact that it looks fine.

A few principles keep a battery layer honest:

• A resting voltage is not a health check. A tired battery can show a normal voltage sitting idle and then collapse the instant a real load is applied. Meaningful testing checks how the battery behaves under load, not just what it reads at rest.
• Age and temperature are the quiet killers. Batteries lose capacity as they age, and heat accelerates it. A battery in a hot enclosure ages faster than the calendar suggests. Track install dates so you are replacing on a schedule, not waiting for a failure.
• Test on a schedule and log the results. A single good test tells you the battery was healthy that day. A trend of tests over time tells you when capacity is sliding and lets you replace before it becomes a failure. Testing you do not record is testing you cannot learn from.
• Replace proactively, not reactively. A battery that fails during an outage failed at the worst possible moment. Replacing on a defined interval, and sooner if testing shows decline, moves that failure out of the crisis and into a controlled maintenance window.
• Do not forget the subscriber batteries. The packs in portables and spares age the same way. A charging and rotation routine for field batteries is as important as the bank at the site, because it is the layer closest to the person on the call.

The theme running through all of this is that a backup you never test is not a backup. It is an assumption. The difference between the two only becomes visible on the day the grid drops, and that is the worst possible time to find out which one you had.

Safety: generators and battery handling

Backup power hardware carries real hazards, and the people asked to operate it during an event are often tired, working in bad conditions, and under pressure. That is exactly when shortcuts happen and when they hurt someone. The safety basics here are not optional background reading. They are the difference between a backup that saves the day and one that causes a second emergency.

None of this is exotic. It is the routine discipline that keeps a power plan from becoming a hazard of its own. Write these steps into your procedures, and make sure the people who will actually operate the equipment during an event have been trained on them, not handed a manual for the first time in the dark.

Building resilience into planning and maintenance

Backup power resilience is not a purchase. It is a habit. Buying batteries and a generator is the easy part. The hard part is the ongoing discipline of testing, records, fuel rotation, and drills that keeps all of it ready on a day you did not choose. The departments that stay on the air through the worst events are not the ones with the most hardware. They are the ones who treated backup power as a living part of their maintenance program rather than a box checked once at installation.

Fold power into the same maintenance rhythm you use for the rest of your critical equipment. Give every battery bank, every generator, and every fuel supply a record with install dates, test results, service history, and replacement targets. When those records live in one place instead of scattered notebooks and memory, the trends become visible and the failures become predictable, which is another way of saying preventable.

Use this checklist as a starting point for a power resilience review:

• Map every layer: grid, site batteries, generators, and subscriber batteries. Identify the gaps where a single failure goes dark.
• Confirm the battery layer can bridge cleanly to the generator, and that the transfer actually works when tested, not just on paper.
• Separate your short-outage plan from your multi-day plan, and stress-test the long one against a worst-case regional event.
• Write a real refueling plan: on-site capacity, resupply arrangements made in advance, road-access contingencies, fuel rotation, and who physically performs it.
• Test batteries under load on a schedule, log every result, and watch the trend rather than the single reading.
• Set proactive replacement intervals for site and subscriber batteries, and replace sooner when testing shows decline.
• Establish a charge-and-rotate routine for field portables and spare packs.
• Document generator safety, fueling, and battery-handling procedures, and train the people who will run the equipment during an event.
• Keep every battery, generator, and fuel record in one maintained place so history and due dates are visible at a glance.
• Run a drill. A plan that has never been exercised is a theory, and an event is a bad place to test a theory.

Backup power is the least glamorous part of a communications program and one of the most consequential. It earns its keep on the rare, terrible days when everything else is failing at once. The work of keeping it ready happens quietly, months or years ahead of the moment it is needed, and that is precisely why it is so easy to let slide. Do not let it slide.