A Watt-Hour Budget for Your Pet's Equipment: The Math, and the Four Things the Division Leaves Out

This page ranks no product and recommends none. Our portable power station comparison is where the which-unit question lives. This one is the arithmetic underneath it, built the other way round: you start from the equipment already sitting in your house, read its numbers off its own data plates, and end up with a watt-hour budget that belongs to your household instead of to a category average. It is also the page that tells you what the arithmetic cannot do, because a division that looks like engineering turns into a safety promise the moment an animal or a person is depending on the answer.

Here is the shape of the problem. A power station’s headline number is watt-hours, which is stored energy. Your equipment’s headline number is watts, which is how fast it spends that energy. Divide the first by the second and hours fall out. That is genuinely the whole calculation, and it is genuinely not the answer, because both inputs are softer than they look and the quotient sits above the truth on both ends. The rest of this page is about where the softness lives, what the manufacturers and the standards actually publish about it, and the two places where we stop calculating and hand the question to a named human being.

What This Page Will Not Do

Stating the refusals first is not throat-clearing. Each one is a place where a page like this normally overreaches, and knowing which ones apply saves you reading for an answer that is not here.

It will not tell you a battery is enough. Not for an aquarium, not for a terrarium, and above all not for anything keeping a body alive. Enough is a guarantee about a future outage of unknown length in a room of unknown temperature with equipment we have never seen. We can show you the arithmetic and every assumption inside it. We cannot underwrite it.

It will not supply your equipment’s wattage. Not a typical aquarium heater, not a representative reptile heat mat, not an average air pump. The correct number is on your device, and a number borrowed from a similar product is wrong in a way that compounds across every hour of the calculation.

It will not rank a human medical device against an animal’s equipment. There is no shed order on this page containing a CPAP, an oxygen concentrator, a nebulizer or an insulin cooler. That is not squeamishness. It is that the ranking would be a clinical decision dressed up as a spreadsheet, and this site is not competent to make it. Human life outranks animal welfare wherever the two compete for the same battery, and the decision route for the human load is named further down.

It will not publish the hour at which you move the animal instead of powering it. We looked for a published threshold. The section near the end of this page records what we searched and what came back.

Step One: The Number Is On Your Equipment, Not On This Page

Every honest power budget starts at a data plate. That is the small printed panel, sticker or moulded label carrying the model number, the voltage, and either the wattage or the current draw in amps. On a refrigerator it is usually inside the cabinet or on a side or rear panel. On an aquarium or terrarium accessory it is often on the plug-in transformer rather than on the device in the tank. On a fan it is typically on the motor housing or the base.

Manufacturers will tell you this directly if you ask them. Danby’s own product page for its 1.6 cubic foot compact refrigerator carries a customer FAQ whose first question is “What are the standard Wattage and amps used?” The published answer is one sentence: “Information pertaining to watts and amps can be found on the rating plate located on the side of the unit.” That is a manufacturer declining to publish a single number for the model on its own product page, and pointing you at the physical label instead, because the label is where the figure that governs your unit lives.

Some manufacturers do publish the figure, and where they do it is because wattage is the product’s defining specification. Exo Terra’s Heat Mat page states the line is “Available in 4, 8, 16 and 25 Watts.” Zoo Med’s printed instruction sheet for its ReptiCare Ceramic Heat Emitter opens its instructions with “Use appropriate wattage Heat Emitter for your terrarium. General recommendations are as follows: 60 watt for 10-20 gallon terrarium, 100 watt for 30-40 gallon terrariums, and 150 watt for 50-100 gallon terrariums or large wire Iguanariums®.” Read the sentence that comes next in that same paragraph, because it is the qualifier that a wattage table would drop: “Because of variables like ambient room temperature and ventilation, more or less heat may be required. Zoo Med recommends the use of the ReptiTemp 500R thermostat for accurate and consistent temperature control.” The manufacturer is telling you two things at once. Here is the rated wattage, and here is why the wattage that suits your room may not be the one on the shelf.

What to write down, per device

One line per powered item, filled in before a season rather than during an outage:

• Model number, exactly as printed. Sizes within one product line often differ in wattage by a factor of five.
• Volts and either watts or amps. If the plate gives volts and amps rather than watts, multiplying volts by amps gives volt-amps, not watts. Volt-amps equals watts only for a purely resistive load, such as a heat mat or a heating element: a heat mat plate reading 120 V and 0.5 A is a 60 W device by that multiplication. For anything with a motor or a compressor (an air pump, a filtration pump, a box fan, a refrigerator) the volt-amp product is an upper bound on real watts, not the watts figure itself, unless the plate or the manual separately publishes a watt rating or a power factor. Using the volt-amp figure in place of an unpublished watt figure makes the runtime you calculate later conservative rather than optimistic, because it treats the load as drawing more than it likely does.
• Where the figure came from: the plate, the manual, or the manufacturer’s page. A year from now you will not remember, and the provenance is what lets you re-check it.
• Whether the device is thermostat-controlled or runs continuously. This is the single biggest fork in the next section.
• Whether the manufacturer publishes a starting or surge figure. Most do not, which is its own finding.

There is no federal energy label on pet equipment, and that matters

For a household appliance you may be used to the yellow EnergyGuide label doing this work for you. It does not exist for tank and terrarium equipment. The Federal Trade Commission’s Energy Labeling Rule, at 16 CFR Part 305, sets label content product type by product type, and reading the current part on August 11, 2026, its label-content sections cover refrigerators, refrigerator-freezers and freezers, clothes washers, dishwashers, water heaters, room and portable air conditioners, pool heaters, central air conditioners, heat pumps and furnaces, ceiling fans, lighting products, plumbing products, and televisions. Aquarium heaters, air pumps, substrate heaters, ceramic heat emitters and box fans are not among the labelled categories.

So for the pet half of your budget there is no standardized, test-derived energy figure to look up. There is a nameplate and there is the manufacturer. That is the whole universe of sources, which is precisely why the borrowed figure is such a tempting and such a bad shortcut.

A note on how we checked: while preparing this page we tried to open manufacturer specification pages for several aquarium heaters and air pumps and received server errors and not-found responses on more than one of them. That is a retrieval failure on our end, not evidence that those manufacturers publish nothing. Your unit’s own label and manual carry the figure regardless of whether we could reach a web page on a given afternoon.

Step Two: Nameplate Watts Is a Ceiling, Not a Consumption Rate

Here is the distinction that quietly wrecks most outage math. A device’s rated wattage describes how much it draws while it is drawing. It does not describe how much energy it consumes over an hour, because a great many of the loads in a pet household are switched off for part of every hour by a thermostat or a compressor control.

Zoo Med’s instruction sheet makes the mechanism explicit when it describes the thermostat it recommends: “This thermostat will control all your terrarium heating devices and shut them off if your terrarium gets too warm.” Shut them off. A 100 W emitter under a thermostat that holds it off for half the hour has consumed 50 watt-hours in that hour, not 100. Exo Terra’s Heat Mat page points the same direction, noting the mats “Can be used with an Exo Terra Thermostat and/or in conjunction with additional heat sources for higher temperature species.”

Refrigeration behaves the same way and its own manufacturer says so. Danby’s product FAQ answers the question “My compressor comes on often. Is this normal?” with “This is normal.” The compressor cycles. It is not drawing its running watts continuously, which means the running figure is the wrong input for an energy budget and the right input for a capacity check.

The one category where a test-derived energy figure exists

For covered appliances, the EnergyGuide label gives you the energy answer directly. Under the Energy Labeling Rule, section 305.14 requires that labels for refrigerators, refrigerator-freezers and freezers “contain the model’s estimated annual energy consumption as determined in accordance with this part.” The rule defines that term at section 305.2 as “the energy or (for plumbing products) water that is likely to be consumed annually in representative use of a consumer product, as determined in accordance with tests prescribed under section 323 of the Act (42 U.S.C. 6293),” and it defines the unit as “Kilowatt-hour use per year, or kWh/yr.”

Worked on one named model, because the method only becomes usable when you can see it run. The EnergyGuide label Danby publishes for its DCR016A3WDB compact refrigerator, label revision code 202402, states an Estimated Yearly Electricity Use of 207 kWh. Turning that into the units a power budget uses:

• 207 kWh is 207,000 watt-hours per year.
• Divided by 365 days, that is about 567 watt-hours per day.
• Divided by 24 hours, that is about 24 watts averaged across the whole day, cycling included.

Notice how far below any running-watts figure that average sits. That gap is the duty cycle, made visible.

Now the caveats, all of which are on the label or in the rule rather than invented here. The label’s own printed text reads “Compare ONLY to other labels with yellow numbers. Labels with yellow numbers are based on the same test procedures”, and it adds “Your cost will depend on your utility rates and use.” The rule’s definition says representative use, determined by a prescribed test. An outage is not that test. The room is warmer than a laboratory, the door gets opened because someone is checking on medication, and a cabinet that has warmed up works harder to pull back down than one holding steady. So the label figure is a floor for daily energy in the best case, and your budget should sit above it rather than on it. It is also a figure for that model in that configuration, and yours is a different appliance until you have read its own label.

If you want your own number rather than a published one

The only way to know what a specific device draws in your specific room is to measure it, with a plug-in energy meter that reports watt-hours accumulated over time rather than instantaneous watts. Run it for a full day on the equipment you actually intend to back up, in the season you actually expect an outage, and write down the watt-hours. That number beats every published figure on this page, including the ones we sourced, because it is the only one measured on your equipment. We are not naming or ranking a meter here.

Step Three: The Division, and the Four Things It Ignores

With a load in watts and a pack in watt-hours, the arithmetic is one operation. Capacity divided by load equals hours.

Manufacturers publish their own version of it. Jackery’s knowledge article on watt-hours presents the calculation as “Working Time = Capacity Wh × 0.85 / Operating wattage of the device” and works it through with a named example on one of its own units. That page does not state what the 0.85 factor accounts for. It also does not present the result as a promise. The same article closes with a disclaimer worth reading in full, because it is the manufacturer describing the limits of its own formula: “Disclaimer: The runtime mentioned for appliances powered by Jackery is for reference only. Actual runtime may vary under different conditions. Please refer to real-world performance for accurate results.”

That is the honest frame for every number below. Here are the four conditions that “different conditions” is doing the work of hiding.

1. Conversion loss between the cells and the outlet

Household equipment runs on alternating current. The cells store direct current. The inverter in between converts one to the other and some energy is lost doing it, which is why a planning factor below 1.0 exists in the first place.

What nobody we checked publishes is the actual figure for a given unit. On August 11, 2026 we read EcoFlow’s RIVER 2 Pro product-page specification table and its own English-language user manual, Jackery’s knowledge article carrying the 0.85 factor, the UL 2743 scope, and Anker’s SOLIX C300 product page, which did not render its specification block for us. None of the specification tables we could read published an inverter efficiency percentage. Jackery’s 0.85 is the only conversion factor any of these manufacturers put in print in front of us, and it appears as a rule of thumb in a consumer article rather than as a specification for a named model. Treat it as a manufacturer’s planning convention, not as a measured property of your unit, and note that a retrieval failure on one product page is not proof that the manufacturer publishes nothing.

2. The capacity on the box is not the capacity you can withdraw

A pack’s published watt-hours describe the cells. What you can actually pull out is decided by the battery management system, which stops the discharge before the cells are empty in order to protect them. EcoFlow states in a footnote to its RIVER 2 Pro specification table that the system “includes over-voltage, overloading, over-temperature, short circuit, low-temperature, low voltage, and overcurrent” protections. Low voltage is in that list, which is the cutoff that ends your run.

We did not find a published usable-capacity or depth-of-discharge figure for the units we checked. What manufacturers publish is nominal capacity and rated output. That means the honest treatment is not to apply an invented derate, but to say plainly that the withdrawal figure is lower than the nameplate by an amount the manufacturer has not disclosed, and to build the plan so that the gap does not matter.

One thing not to do under any circumstances: do not attempt to bypass, disable or work around a low-voltage cutoff, a battery management limit, or an over-temperature protection in order to squeeze out the last of a pack. Those limits are the reason lithium packs are allowed indoors at all.

3. Temperature, which is published, and which is not one number

This is the factor most likely to bite a pet owner specifically, because backup power tends to live in a garage, a utility room or a shed rather than in the heated part of the house.

EcoFlow publishes four separate temperature rows for the RIVER 2 Pro, and they do not agree with each other:

Published row	Figure
Discharge Temperature	14°F to 113°F
Charge Temperature	32°F to 113°F
Optimal Operating Temperature	68°F to 86°F
Storage Temperature	14°F to 113°F (68°F to 86°F is best)

Read the first two rows together. There is a band, from 14°F up to 32°F, in which that unit is inside its published discharge window and outside its published charging window at the same time. A pack in a freezing garage may run your equipment and then refuse to take a charge, which is exactly the wrong failure to discover on the second night of an outage. EcoFlow’s troubleshooting table describes the behaviour in its own words, listing a “Battery charging low temperature” indicator whose stated solution is “Normal operation will resume automatically after the battery temperature is restored to above 3°C.”

The third row is the one people skip. An optimal operating range of 68°F to 86°F is the manufacturer saying that outside that band the unit is working, but not at its best. It does not publish how much capacity you lose at 20°F, and neither do we.

The manual states the risk on both ends: “When using the product, please strictly follow the operating environment temperature specified in this user manual. If the temperature is too high, it may result in a fire or explosion; if the temperature is too low, the product performance may be severely reduced, or the product may cease to work.” These figures are for one named model. Yours will publish its own, and its own are the ones that govern.

4. Duty cycle, running the other way

Covered above, and worth restating here because it is the one factor that pushes in your favour. A thermostat-controlled load consumes less energy per hour than its nameplate suggests, which means a nameplate-based division understates your hours. That does not cancel out the first three. It means your real figure is bracketed rather than known, and the bracket is why an outage plan is built on margin rather than on precision.

A Worked Budget, With Every Assumption On the Table

The point of the table below is not the numbers. It is that every cell has a named provenance and a named limitation, which is the format your own worksheet should copy. These are figures published by three named manufacturers for three named products. They are not stand-ins for your equipment.

Item	Published figure	What kind of figure it is	Where it came from
Exo Terra Heat Mat line	“Available in 4, 8, 16 and 25 Watts”	Rated wattage: a ceiling while energised, and these mats can be run under a thermostat	Exo Terra Heat Mat 4W product page, read 2026-08-11
Zoo Med ReptiCare Ceramic Heat Emitter	60, 100 and 150 watt sizes, recommended by terrarium size	Rated wattage, published with the manufacturer’s own qualifier that room temperature and ventilation change what is required	Zoo Med instruction sheet CE-60/100/150/250, read 2026-08-11
Danby DCR016A3WDB compact refrigerator	Estimated Yearly Electricity Use 207 kWh	Test-derived annual energy under a prescribed test procedure, cycling included; not a running-watts figure	Danby EnergyGuide label, revision 202402, read 2026-08-11
EcoFlow RIVER 2 Pro	Capacity 768Wh; AC Output “Pure Sine Wave, 800W total (surge 1600W)”	Nominal cell capacity and rated output; the manufacturer publishes no usable-capacity or inverter-efficiency figure	EcoFlow RIVER 2 Pro product page, read 2026-08-11

Now the arithmetic, written so that you can see every assumption it rests on rather than only its result.

Assumptions, stated: capacity is the manufacturer’s nominal watt-hour figure and not a usable figure, because no usable figure is published. Conversion loss is represented by Jackery’s published 0.85 planning factor, applied here because it is the only manufacturer-published factor we found, and not because it has been verified for any specific unit. Ambient temperature is assumed inside the manufacturer’s stated optimal band, which a garage in winter is not. Duty cycling is assumed away: every load is treated as drawing its full rated wattage every minute, meaning a 100 percent duty cycle, which is deliberately pessimistic for anything thermostat-controlled. Every one of those four assumptions can be wrong in your house. The first worked example below holds to the duty-cycle assumption throughout. The second and third do not, on the refrigeration side, and each says so where it applies.

One 25 W substrate heater, treated as continuous, on a 768 Wh nominal pack. The raw quotient is 768 divided by 25, which is 30.7 hours. With Jackery’s 0.85 planning factor applied, 768 times 0.85 divided by 25 is 26.1 hours. That is a ceiling under the assumptions above. It is not a floor, it is not a guarantee, and it says nothing about whether 26 hours of substrate heat is the right plan for your animal, which is a veterinary question and not an electrical one.

The same pack against a refrigeration load described by its EnergyGuide figure. Using the Danby label’s 207 kWh per year, the daily energy is about 567 watt-hours. Against 768 nominal watt-hours, with the 0.85 factor applied, the pack holds about 653 watt-hours of planning energy. Set against 567, that is a ratio of about 1.15, which says the pack’s planning energy is a little larger than one day of that named model’s test-procedure consumption with nothing else drawing. Then read the caveat that governs the whole comparison. The label figure is a standardized-test annual average. An outage is warmer, the door gets opened, and a warm cabinet consumes more, so the real daily energy during an outage is higher than 567 watt-hours by an amount neither the label nor this page can quantify. A ratio of 1.15 has no room in it for that. At the label’s own average, 653 watt-hours of planning energy against 567 watt-hours a day works out to about 27.6 hours, and that ceiling gets shorter, not longer, once the room is warm and the door opens. Anyone whose plan turns on where inside that day the pack actually runs out needs a measured figure rather than a calculated one.

Loads add before they divide, and this example mixes two different kinds of number on purpose so the mismatch has to be named rather than hidden inside a sum. Two devices running at once are one combined number in the denominator. A 25 W mat, held to the full-rated-wattage assumption stated above, alongside the refrigerator’s 24 W EnergyGuide average, is 49 W, and 768 times 0.85 divided by 49 is 13.3 hours. That is not a ceiling on the same basis as the 26.1-hour figure above it. The mat’s 25 W already assumes full rated wattage every minute. The refrigerator’s 24 W does not: it is a standardized-test annual average with the compressor’s duty cycle baked in, and the caveat two paragraphs up applies here just as it did there, that an outage’s real daily energy runs higher than that average by an amount neither the label nor this page can quantify. So one half of the 49 W denominator is a pessimistic ceiling and the other half is an optimistic floor, added together and divided as if they were the same kind of number. Treat 13.3 hours as bracketed rather than as a ceiling, and if your plan turns on the difference, measure the refrigerator’s own draw with a plug-in meter rather than trust the label average for outage conditions. Adding a third load does not shorten the runtime a little. It shortens it in proportion, and the same caution about mixed bases applies to whatever you add.

Surge Is a Separate Question, and We Will Not Multiply Your Way To It

Watt-hours decide how long. Watts decide whether it starts at all. Anything with a motor or a compressor demands more current for the first moment than it does once running, and if that demand exceeds what the inverter can deliver the load never starts, no matter how full the pack is.

You will find rules of thumb everywhere claiming a starting draw of some multiple of running watts. We are not restating one here in our own voice, because the only multiplier we found published by a manufacturer arrives as an explicit rule of thumb rather than as a specification for a named appliance. The Westinghouse WGen5500 generator owner’s manual publishes one, a starting power multiplier of 1.2 for small hand-held or portable devices and 3.5 for larger stationary devices, and it states in the same paragraph why the multiplier exists at all: “Manufacturers of such devices rarely publish this starting power demand and so it’s often necessary to estimate it.” That is one generator manufacturer’s rule of thumb, published in one manual, presented by its own author as an estimate rather than as a specification for your appliance; our generator sizing for pet loads page quotes it in full where it actually applies, against a generator’s starting-watts rating. We are not carrying it onto a power station calculation here, because a rule of thumb stripped of its hedge starts being read as a specification for your device, which it is not. Multiplying your fridge’s running watts by a number you read on the internet, including this one taken out of its context, produces a figure with no source you can stand behind.

What you can do instead is read what is published. Power stations publish a rated output and, separately, a peak or surge output: EcoFlow states the RIVER 2 Pro’s AC output as “Pure Sine Wave, 800W total (surge 1600W)” on its specification table. On the appliance side, look for a published starting wattage, a locked-rotor current figure, or an inrush specification in the manual or on the plate. If your appliance’s manufacturer does not publish one, the answer is to ask the manufacturer, not to estimate. And if the pack simply will not start the appliance, that is information about a pairing, arrived at by testing before an outage rather than during one.

One related compatibility question belongs to the manufacturers as well. Some inverters produce a pure sine wave and some produce a modified sine wave, and whether a particular pump, controller or medical device tolerates the latter is a statement only that device’s manufacturer can make. We do not make it here for any device.

The Refrigerator Warning That Is In the Manual and Nowhere Else

This one is worth the price of the whole page if you keep refrigerated pet medication, because it is a failure mode that no amount of correct watt-hour arithmetic prevents.

EcoFlow’s own RIVER 2 Pro user manual carries this as a numbered safety instruction: “When the power supply is connected in the normal mode to a refrigerator, the power supply may be automatically shut down due to the power fluctuation property of the refrigerator. If the refrigerator stores medicines, vaccines or other high-value items, it is highly recommended that you set the AC output to ‘Never Turn Off’ through the APP when connecting the power supply product to ensure continuous power supply. Users should pay attention to the power consumption of the power supply.”

Read what that says. The station can shut its own output off because a compressor’s intermittent draw looks, to the station, like nothing is plugged in. The manufacturer’s remedy is a setting, named in the manual, that has to be changed in advance. And the manufacturer’s closing clause puts the monitoring burden back on the owner rather than on the device.

If your household refrigerates insulin or another pet medication, that paragraph is the reason to plug the appliance into the pack on a quiet afternoon and watch it for an hour or two rather than assuming a full battery equals a running fridge. Your own unit’s manual may describe the same behaviour with a different setting name, or may not describe it at all, which is its own answer. Everything about whether the medication remains usable after a temperature excursion is a separate question with a separate owner, and our pet medication and refrigeration in a power outage page works it through with the storage guidance quoted directly.

The Shed Order This Page Refuses To Publish, and the Worksheet That Replaces It

The plan for this page called for a ranked triage: life support before comfort before convenience, applied to a finite battery. Half of that is defensible and half of it is not, so here is exactly where the line falls.

A human medical load is not in the ranking. Not at the top, not anywhere. If someone in your household uses a powered medical device, the question of what that device needs, whether it can run on battery at all, and what happens if it stops is settled by three named parties and not by this page: the device’s own labeling and instructions, the durable medical equipment supplier, and the prescribing clinician.

Both the device manufacturer and the federal regulator route it the same way. EcoFlow’s RIVER 2 Pro manual states that the product “is not recommended for powering medical emergency equipment related to personal safety, including but not limited to medical grade ventilators (hospital version CPAP: Continuous Positive Airway Pressure), artificial lungs (ECMO: Extracorporeal Membrane Oxygenation).” The sentences immediately after that one matter as much as the sentence itself, so here they are: “It can normally work with a home ventilator (home version CPAP), and does not have to be continuously monitored by professionals. Please follow your doctor’s instructions and consult with the manufacturer for restrictions on the use of the equipment. If it is used for general medical equipment, please be sure to monitor the power status to ensure that the power does not run out.” The manufacturer permits a category, then hands the decision to your doctor and to the device’s own maker.

FDA’s guidance page for medical devices during natural disasters, current as of April 16, 2024, gives the reader-side version. Under Power Outage it says to “Notify your electric company and fire department to let them know you have a medical device that needs power (e.g., ventilator, apnea monitor)” and to “Read your user instructions or call your distributor or device manufacturer to find out if your device can be used with batteries or a generator.” Its General Safety section carries the sentence that outranks everything else on this page: “If you depend on your device to keep you alive, seek emergency services immediately. If possible, notify your local Public Health Authority to request evacuation prior to adverse weather events.”

FDA also publishes a fill-in booklet for exactly this planning, and its Section 2 is a list of questions to put to the device manufacturer or medical provider rather than to a calculator. Among them: “Can my device operate on another power source? If yes, what type?”, “Could I be harmed if my device stops for a short period of time? If yes, what is that time period?” and “How long will the device last on battery power?” Those are the right questions. They are addressed to a person who knows the device, which is the point.

The practical consequence for a pet household is unsentimental and worth saying in plain language, because it is genuinely hard. If a person in your home depends on powered equipment, the backup power in that house is spoken for. The animals’ plan is then built from whatever capacity is left over, and if there is not enough left over, the answer is not to shave the human margin. It is to solve the animal’s problem a different way: relocating the animal, arranging boarding, insulating an enclosure instead of heating it, or leaving earlier. Our board or evacuate together decision and the aquarium and reptile outage guide both cover routes that do not consume watt-hours at all.

The animal side, and why it is a veterinary question

Which of your animal’s equipment counts as life support is not a property of the equipment. It is a property of the species, the individual animal, its age and its condition, and the person qualified to say so is your veterinarian. A filtration pump that is a convenience on one system is the oxygenation path on another. A heat source that is optional in July is not in January.

So instead of a ranking, here is the worksheet the ranking would have to come from. Fill it in once, in writing, with your vet’s input on the first column, and keep it with your outage plan:

1. What changes for this animal if this device is off for one hour, and who told me so? An answer with no source attached is a guess.
2. Is that change reversible when power returns? A tank that cools and rewarms is a different problem from a tank whose biological filtration has died.
3. Is there a non-powered substitute? Insulation, a pre-chilled or pre-warmed mass, a smaller enclosure, moving the animal to the warmest or coolest room in the house, a manual water change. Every load you can answer this way is a load that no longer competes for the battery.
4. What is this device’s measured or published draw, and where did the figure come from? From step one above.
5. At what point does this animal need to be somewhere else rather than here on backup power? See the next section, because the honest answer is that this is a question for your vet and there is no published number.

The AVMA’s list of animal emergencies requiring immediate veterinary consultation includes “Choking, difficulty breathing or nonstop coughing and gagging” and “Heat stress or heatstroke”, and the same page states “The bottom line is that ANY concern about your pet’s health warrants, at minimum, a call to your veterinarian.” When an animal is in distress, the power budget has stopped being the problem you are solving.

Where the Battery Stops and the Generator Starts

A watt-hour budget sometimes ends with the conclusion that the load is bigger than any battery in the house. That conclusion opens a door onto a completely different product with completely different and completely non-negotiable rules, so the boundary gets stated here rather than blurred.

A generator’s rules are the generator’s rules. The US Consumer Product Safety Commission’s safety alert on generators and carbon monoxide, publication CPSC468 0322, states in its danger panel: “NEVER use inside a home or garage, EVEN IF doors and windows are open.” The same panel continues: “Only use OUTSIDE and far away from windows, doors, and vents.” Its recommended-practice list includes this instruction, quoted whole because the second half of it is as load-bearing as the first: “DO operate portable generators outside only, at least 20 feet away from the house, and direct the generator’s exhaust away from the home and other buildings.” The distance in that sentence is measured from the house, and it arrives attached to an exhaust-direction instruction that a shortened quotation would throw away. Its don’ts include “DON’T operate a portable generator inside a home, garage, basement, crawlspace, shed or on a porch; and never leave a car running in an attached garage” and “DON’T use a portable generator near openings to your home including doors, windows or vents.”

No enclosure, canopy, tent, cracked door, large room, short run or fan makes indoor operation acceptable. CPSC does recommend a carbon monoxide shutoff feature, stating “DO ask retailers for a portable generator with a safety feature to shut off automatically when high CO concentrations are present. Some models with CO shut-off also have reduced emissions”, and the never-indoors instruction sits on the same page unchanged. A shutoff feature is an additional protection, not a permission. Our generators, carbon monoxide and pets page carries this ground in full, and quiet generator versus battery power station covers the choice between the two categories.

The carbon monoxide alarm instruction, because this page named a generator. CPSC’s same alert states: “DO install battery-operated CO alarms or CO alarms with battery backup outside sleeping areas, and on each floor of your home. Test alarms monthly.” Where the alarm physically goes on the wall or ceiling is set by the alarm’s own manufacturer instructions and by the code your jurisdiction has adopted. NFPA’s storefront lists the current edition of its fire alarm code as “the 2025 edition of NFPA 72®, National Fire Alarm and Signaling Code®,” and the chapter that governs household single- and multiple-station alarms is Chapter 29, which is what NFPA’s own published committee report for its household signaling committee shows section 29.1.1 applying to. We are describing that document rather than quoting it, because it is a draft record marked up with proposed changes rather than adopted code, and we could not retrieve the published code text itself, which sits behind NFPA’s account-gated viewer. So we quote no requirement from the code and we publish no mounting height in our own voice. Follow the instructions that came in the box with your alarm, and ask your local fire department or building department which edition your jurisdiction enforces.

Connecting a generator to house wiring is not a do-it-yourself topic and no procedure appears here. The Portable Generator Manufacturers’ Association’s release of September 12, 2025 names the two legitimate pieces of equipment and the person who installs them: “A transfer switch serves as a connection point that prevents a generator from feeding electricity back into the power lines. Interlock devices, when installed by a qualified electrician, enable homeowners to safely connect their portable generators to their home’s electrical panel while ensuring the main breaker is disengaged. These are the only safe ways to use a portable generator to power an entire residence.” The same release explains why the shortcut is lethal to people who are nowhere near your house, and why it is dangerous to your own house as well: plugging a generator into a wall outlet “is extremely dangerous and presents an electrocution risk to utility workers and neighbors served by the same utility transformer. It also bypasses some of the built-in household circuit protection devices and can cause short circuits and overloads that lead to electrical fires, prompt fluctuating or excessive voltage that can damage appliances and electronics, and damage a property’s wiring and circuit breaker panel.” That is the whole of what this site will say about the subject. A licensed electrician and your local authority having jurisdiction own the rest.

A battery pack is a different product, and the rules do not transfer either way. ANSI/CAN/UL 2743, the Standard for Portable Power Packs, Third Edition dated September 24, 2025, classifies packs by permitted location: “These requirements cover power packs suitable for outdoor use, temporary outdoor use, or indoor use only. Outdoor use packs are intended to be used outdoors with no restrictions. Temporary outdoor use packs are intended to be used outdoors in limited wet conditions and always stored indoors. Indoor use only packs are intended to be stored indoors and used indoors and are not intended to be used outdoors at any time. A power pack with a booster function is not considered indoor use only under any conditions.” A jump-starter or other pack with a booster function is therefore never an indoor-use-only pack under the standard, regardless of what else its listing says. Which of those classifications your pack otherwise carries, and what its manual says about operating and charging it, is settled by your pack’s own documentation. Nothing a generator’s placement rule says gives a battery pack permission, and nothing a battery pack’s indoor listing says gives a generator permission. EcoFlow’s manual adds two instructions for its own unit that a pet household should read twice: “Store the product in a dry and ventilated place” and “Keep this product out of reach of children and pets.”

An indoor-use listing answers which room the pack may sit in. It does not answer where in that room, and for a household running tank or terrarium equipment, that second question is the one with a tank next to it. A tank, sump or reservoir is a standing water source with a splash risk, a siphon risk and a hose or pump line that can fail. Neither UL 2743 nor EcoFlow’s manual names a tank, so what follows is this page’s own reasoning, not a published protocol: treat the manual’s instruction to store the unit in a dry place as a siting rule for that exact spot, above and away from the tank, any sump, any drip loop and any hose run, on a surface a leak or a siphon cannot reach, not merely somewhere inside a dry-classified room. The same instruction to keep the product out of reach of pets applies to its cords as much as to the unit itself, since a cord running past a paw or a beak is not out of reach just because the pack sitting at the end of it is.

For what it is worth on scope, UL 2743 also states that its requirements “do not cover power banks which are covered in the Standard for Power Banks, UL 2056” and do not cover packs that are “permanently connected to a premise wiring system, or having a function to supply power to a premise wiring system, or have utility interactive function, which are covered by the Standard for Energy Storage Systems and Equipment, UL 9540.” If you are looking at a product that connects to your house wiring, you have left the portable-power-pack category entirely, and the electrician conversation above applies.

What Nobody Publishes: The Hour You Move the Animal

The plan for this page originally called for a cutoff, expressed in hours, at which an owner stops powering an animal’s enclosure and relocates the animal instead. We went looking for a published one.

What we searched, on August 11, 2026: web searches for a published hour threshold or point at which a pet should be moved to another location rather than kept on backup power during an outage, including phrasings built around veterinary guidance and around outage duration. What came back was utility-company advice, veterinary-practice blog posts and pet-retailer articles, several of them recommending boarding an animal during an extended outage, and none of them publishing a threshold in hours. We also read the current federal and standards material cited throughout this page. None of it addresses animal relocation timing, which is expected, because carbon monoxide alerts, battery standards and appliance labeling rules have no reason to.

So the honest output is an empty cell rather than an invented number. There is no hour count here because we did not find one published, and a threshold assembled from species comparisons, body mass or interpolation would be a fabrication wearing a decimal point.

What replaces it is a conversation you can have before a season starts. Ask your veterinarian, about your animal and your species: what are the signs that mean this animal needs to be somewhere else rather than here on backup power, and where is that somewhere else. Write the answer next to the worksheet above. Our aquarium and reptile power outage guide covers relocation as a decision in its own right, including what moving an animal costs it, and sheltering in place with pets through power outages sets the wider picture the relocation decision sits inside.

Two related timing questions have their own homes. Whether a battery-backed feeder or fountain is holding up is answered in how long an automatic feeder lasts in a power outage, which works from named products’ published backup figures rather than from a general calculation. And whether refrigerated medication is still usable after an excursion is a cold-chain question, not a power question, covered in pet medication and refrigeration in a power outage.

The Version of This You Do On a Quiet Afternoon

Nothing above works if it is attempted for the first time in the dark. The whole exercise takes an evening and it is durable for years.

• Walk the house with a notebook and read every data plate. Model number, volts, and watts or amps, for every powered item any animal depends on. Add the plug-in transformers, which is where the tank and terrarium figures usually hide.
• Mark each item as thermostat-controlled or continuous. That single column changes the arithmetic more than any other.
• For any covered appliance in the list, find its EnergyGuide label or the manufacturer’s published estimated yearly electricity use, and convert it to watt-hours per day. That is the energy figure a nameplate cannot give you.
• Look up your power station’s published capacity, rated output, surge output, and its four temperature rows. Write down where the pack is stored and what temperature that place reaches in your worst season.
• Read your pack’s manual for its statements on operating location, on medical equipment, and on refrigerators. Change any setting the manual names before you need it, not during.
• Run the arithmetic for your real combination of loads, write down the assumptions you used alongside the answer, and then plan under the answer rather than on it.
• Test the pairing. Plug the actual equipment into the actual pack for an hour or two and watch what happens, especially anything with a compressor.
• If the pack sits near a tank, sump or reservoir, site it above and away from any splash, siphon or hose-failure path, on a surface a leak cannot reach, and route its cords where neither water nor an animal can reach them. An indoor-use listing is about the room, not the exact spot in it.
• Put the human medical question to the prescribing clinician and the equipment supplier, using FDA’s own list of questions, and settle it independently of the animals’ plan.
• Ask your veterinarian which of your animal’s equipment is genuinely time-critical, and what the signs are that the animal needs to be moved rather than powered.
• Confirm you have carbon monoxide alarms per CPSC’s instruction if any combustion appliance or generator could ever enter the picture, and test them on the schedule CPSC states.

Where to Go Next

This page is the arithmetic spoke of our sheltering in place through power outages hub. The rest of the backup-power problem splits four ways.

Choosing hardware. Portable power stations compared for pet loads is the buying-guide answer with manufacturer specifications, and Jackery versus EcoFlow for pets narrows it to two brands. At the small end, where the load is a tracker or a low-draw device rather than a tank, solar charger versus power bank is the comparison that fits.

Fuel instead of cells. Quiet generator versus battery power station for pet owners is the category comparison, and generators, carbon monoxide and pets is the safety ground that has to be read first.

The specific loads. Aquariums and reptiles in a power outage covers tank and enclosure timelines, pet medication and refrigeration in a power outage covers the cold chain, insulin cooler cases for a diabetic pet covers the go-bag version of that problem, and automatic feeders and fountains without power covers the machines most owners forget are electrical.

Heat and cold without the grid. Dog cooling gear during a power outage and battery-powered fans for a pet carrier both cover the loads a summer outage creates, sized in watts rather than in hope.

Then do the one thing that makes all of it real: read three data plates this week and write the numbers down. The household that knows its own watts before an outage is not doing better arithmetic than everyone else. It is doing arithmetic on real inputs, which is the only kind that survives contact with a dark house.