How to Choose a Portable Power Station

Most buying guides for this category are product lists wearing a decision framework as a costume. They tell you which unit is “best” and never show you the arithmetic that would let you check them. I have been buying and selling physical products for fifteen years, and I spent a decade running a business from Bali on grid power that failed several times a week, so I have a specific interest in getting this right. This guide from E-Commerce Paradise is the framework, in order, with every number sourced and every assumption stated.

Work through the ten steps below and you will end up with a specification, not a shopping list. Then you match the specification to a machine. That is the correct order, and almost nobody does it.

Step 1: Add up the loads you actually need to run, and for how long

Every mistake in this category starts here. People buy capacity by feel, then discover the unit dies at 2am. The fix is boring: list the devices, write down what each one draws, write down how many hours you need it, and multiply.

Two rules make this list honest. First, use measured or published energy figures where they exist, not the sticker on the back of the appliance. A refrigerator’s plate wattage is what its compressor pulls while running, and the compressor does not run continuously. Second, cut the list to what you genuinely need during an outage. Nobody needs a dishwasher at 3am.

Here is a starting list built only from figures published by the manufacturers themselves.

Load	Published figure	Source	Energy for a typical night or day
Full size refrigerator, 20.8 cu ft bottom freezer (GE GNE21DYR)	487 kWh per year	ENERGY STAR certified product data	1,334 Wh per day
Same class at the maximum the federal standard allows	541 kWh per year	ENERGY STAR certified product data	1,482 Wh per day
CPAP, device plus humidifier (ResMed AirSense 10 AutoSet)	22 watt hours per running hour	ResMed Air10 power consumption sheet	176 Wh over 8 hours
CPAP, device only, no humidifier	7 watt hours per running hour	ResMed Air10 power consumption sheet	56 Wh over 8 hours
Laptop, one full charge	60 Wh battery assumed	BLUETTI Elite 100 V2 runtime table	120 Wh for two charges
Smartphone, one full charge	15 Wh battery assumed	BLUETTI Elite 100 V2 runtime table	60 Wh for four charges
Fan	40 W	BLUETTI Elite 100 V2 runtime table	200 Wh over 5 hours

Add those four household lines together and a realistic outage day is 1,334 plus 176 plus 120 plus 60, which is 1,690 watt hours delivered to the sockets. Hold that number. It is the only input the rest of this guide needs.

Step 2: Convert delivered watt hours to nameplate, and stop skipping the efficiency step

This is the step almost every buying guide leaves out, and leaving it out is why their runtime numbers are wrong. The number on the box is the energy stored in the cells at direct current. Your kettle, your fridge and your laptop brick all run on alternating current, and the inverter that makes that alternating current loses some of the energy as heat.

You do not have to take my word for the size of that loss, because the manufacturers publish it in two different places. BLUETTI’s AC200L page claims “AC efficiency up to 92%”, and the words “up to” are doing real work there, because that is a ceiling under favourable conditions, not an average. More usefully, BLUETTI’s own AC180 product page publishes a runtime table, and you can reverse the arithmetic on it.

Load in BLUETTI’s own AC180 table	Runtime BLUETTI publishes	Energy that implies	Share of the 1,152 Wh nameplate
Microwave, 600 W	1.5 hours	900 Wh	78 percent
Hair dryer, 1,300 W	0.7 hours	910 Wh	79 percent
Heater, 1,800 W	0.5 hours	900 Wh	78 percent
Electric grill, 1,560 W	0.6 hours	936 Wh	81 percent

BLUETTI’s own published runtimes imply that roughly 78 to 81 percent of the nameplate reaches the load. Their figures are rounded to one decimal place and probably include a low battery cutoff, so the true inverter efficiency sits above that. Across this cluster I plan on 85 percent of nameplate reaching an alternating current socket, which is the conservative end of a sensible 85 to 90 percent band, and I state that assumption every single time I use it.

So the conversion is simple. Nameplate needed equals delivered watt hours divided by 0.85. Our 1,690 Wh outage day becomes 1,690 divided by 0.85, which is 1,988 watt hours of nameplate capacity. That immediately rules out the entire class of units around 1,000 Wh for a full outage day, and it tells you a 2,048 Wh or 2,073.6 Wh machine is the right size with almost no margin. If you want a margin, or you want to keep a reserve rather than running the pack flat, add 20 percent and shop at about 2,400 Wh.

Run the same arithmetic on the runtime you get from a specific machine in the refrigerator runtime guide, which works the method on four different fridge types.

Step 3: Continuous output against your largest load, surge against its startup draw

Capacity is how long. Output is whether it runs at all. These are separate specifications and a unit can pass one and fail the other.

Continuous output is the easy half. Find your single largest simultaneous load and make sure the unit’s continuous rating exceeds it. A 1,500 W space heater needs more than 1,500 W of continuous inverter. BLUETTI’s AC60 is rated 600 W continuous, the AC70 is 1,000 W, the AC180 is 1,800 W, the Elite 200 V2 is 2,600 W and the AC200L is 2,400 W. Those are the numbers on their own product pages.

Surge is the half that catches people. Anything with a motor or a compressor draws far more current for the instant it starts than it does while running. Embraco publishes technical data for the compressors that go inside household refrigerators, and its EM3D60HLT sheet lists a power consumption of 139 W at rated conditions and a “Locked Rotor Amperage (LRA) 60Hz 10.2 A” on a 115 to 127 V supply. At 120 volts that locked rotor figure is about 1,224 volt amps of inrush from an appliance that draws 139 W once it is turning.

Locked rotor is a stalled worst case rather than a normal start, and a real start is briefer and smaller. But it is the number the design margin is built against, and it explains the failure everyone reports. A 600 W unit with a 1,200 W surge rating sits right on top of that inrush. A 1,000 W unit with a 1,500 W surge rating has room. That is the entire difference between a fridge that runs and a fridge that trips the inverter on every compressor cycle while the battery still shows 90 percent.

Step 4: Battery chemistry, and why a cycle rating without a capacity is meaningless

Lithium iron phosphate, written LiFePO4, has effectively won this category, and for good reasons: it tolerates heat better, it is far more stable in an abuse scenario, and it lasts several times longer than the nickel manganese cobalt chemistry that used to dominate.

Now the part that matters more than the chemistry. A cycle rating is a meaningless number unless it states the remaining capacity at which the count ends. A battery that is “rated 6,000 cycles” could be rated to 70 percent of original capacity or to 80 percent, and those are very different products. Read the two manufacturers side by side.

Brand and model	Published cycle rating	Down to what capacity
BLUETTI AC180	3,500 or more cycles	“3,500+ Cycles to 80% Original Capacity”
BLUETTI AC70	3,000 or more cycles	“3,000+ Cycles to 80% Original Capacity”
BLUETTI Elite 100 V2	4,000 or more cycles	“4,000+ Cycles to 80% Original Capacity”
Jackery LiFePO4 range	“can last over 6000 cycles”	“the batteries will operate at a capacity of 70% or higher”
EcoFlow, Anker, Goal Zero, Renogy	Check each product page	Do not buy until the page states the capacity basis

Jackery’s 6,000 cycle figure is real and it is on their own site, and their LiFePO4 lineup now runs to nineteen current models. But it is counted to 70 percent remaining capacity while BLUETTI counts to 80 percent, so the two headline numbers are not directly comparable. Neither company is doing anything wrong. The reader who compares 6,000 against 3,500 without reading the qualifier is the one making the mistake.

Once you have cycles and a capacity basis, you can compute the only cost metric that means anything: cost per kilowatt hour delivered across the rated life. It has two terms, and most guides publish only the first one, which is how they end up making a battery look cheaper than the grid.

The hardware term is price divided by nameplate watt hours times 0.85 times rated cycles divided by one thousand. The energy term is what it costs to fill the battery in the first place, which is the grid price divided by that same 0.85, so 17.30 cents divided by 0.85 gives 20.4 cents per delivered kilowatt hour. Add the two terms together and you have the number that actually applies to you.

Model	Price this session	Nameplate	Rated cycles	Hardware term per delivered kWh	All in, plus 20.4 cents of grid energy	Against 17.30 cents grid
BLUETTI AC70	US$359.00	768 Wh	3,000 to 80 percent	18.3 cents	38.7 cents	2.2 times grid
BLUETTI AC180	US$499.00	1,152 Wh	3,500 to 80 percent	14.6 cents	35.0 cents	2.0 times grid
BLUETTI Elite 100 V2	US$499.00 final price	1,024 Wh	4,000 to 80 percent	14.3 cents	34.7 cents	2.0 times grid
BLUETTI Elite 300	US$1,199.00 final price	3,014 Wh	6,000, capacity basis not stated on the page	7.8 cents	28.2 cents	1.6 times grid

Now put that next to the grid. The EIA’s Electric Power Monthly, Table 5.3, puts the 2025 annual average residential price of electricity in the United States at 17.30 cents per kilowatt hour. Every unit in that table delivers energy at between 1.6 and 2.2 times the grid price once both terms are counted, and even the cheapest of them costs about 28 cents where the wall socket costs 17.3.

One note on the Elite 300 row. I have used the US$1,199.00 promotional final price BLUETTI showed this session. At its US$1,499.00 list price the hardware term rises to 9.8 cents and the all in figure to about 30 cents, which is 1.7 times grid. Promotional pricing moves the hardware term and never touches the energy term.

A portable power station is not a device that lowers your electricity bill. It is a device that buys you availability at a price, and that price is roughly double what the wall socket charges. Anyone selling you the first story is either careless or hoping you do not own a calculator.

Step 5: Recharge paths, and how long each one really takes

A battery you cannot refill is a paperweight with a handle. There are four paths and they are wildly different in speed.

Wall charging is the fast one, and modern units are startlingly quick. BLUETTI publishes three AC charging modes for the AC180: silent at 260 W taking about 4.9 hours, standard at around 1,000 W taking about 1.7 hours, and turbo at around 1,440 W taking about 1.3 hours. The headline claim of 0 to 80 percent in 45 minutes is the turbo figure. Note that the silent mode exists because turbo charging is loud, which matters if the unit lives in a bedroom.

Solar is the slow one, and the manufacturers are refreshingly honest about it if you read the fine print. BLUETTI rates the AC60 solar charging time at about 2.5 to 3 hours from a 200 W panel, and adds the qualifier “With prime sunshine, ideal orientation and low temperature”. The AC60 holds 403 Wh, so 200 W of panel would fill it in about two hours if panels delivered their rating. They do not, and BLUETTI’s own best case is 25 to 50 percent longer than the theoretical figure. I go through the full derating stack in the solar generator sizing guide.

Alternator charging from a vehicle is the one that saves van and RV owners. The cigarette lighter socket path is slow, roughly 100 to 200 W depending on the vehicle. A dedicated alternator charger is a different animal and is the right answer if you drive most days. Generator charging is the fourth path and it works exactly like wall charging, which makes a power station a very effective quiet buffer in front of a noisy generator.

Step 6: Pure sine wave against modified sine wave, and what actually breaks

Every BLUETTI unit I checked this session lists “Pure Sine Wave” as its inverter type, including the AC60, the AC70, the AC180 and the AC240. That is the answer you want, and at current prices there is no good reason to buy anything else.

Modified sine wave inverters approximate the grid’s smooth wave with a stepped square approximation. Resistive loads such as incandescent bulbs and simple heating elements do not care. Everything else does, in ways that range from irritating to expensive. Motors run hotter and less efficiently on a stepped wave, which shortens their life. Devices with sensitive timing circuits, including some medical equipment, can misread the waveform entirely and refuse to start. Audio equipment picks up an audible buzz. Variable speed motors, microwaves and anything with a switching supply that expects a clean zero crossing are the usual casualties.

If you are powering a CPAP, a laptop, a router or a fridge, treat pure sine wave as a hard requirement and stop considering any unit that does not state it plainly. The CPAP power station guide goes deeper on the medical device side of this.

Step 7: UPS switchover time, and what it does and does not protect

An uninterruptible power supply mode means the unit passes grid power through to your devices and takes over from its own battery when the grid drops. The specification that matters is how many milliseconds the gap lasts. BLUETTI publishes this for some models and not others.

Model	Published switchover time	Where it is published
BLUETTI Elite 100 V2	10 ms or less	Elite 100 V2 comparison table
BLUETTI Elite 300	10 ms	Elite 300 product page
BLUETTI AC240	15 ms	AC240 specification table
BLUETTI AC180	20 ms	Elite 100 V2 comparison table

A desktop computer or a network switch will usually ride through a 20 ms gap. A desktop with a marginal power supply may not. Understand the limit of what this feature buys you: a fast switchover protects against the outage, not against the surge or the sag that often accompanies it, and it is not a substitute for surge protection on a line that suffers lightning.

If you are running a business on this, and I have, the honest framing is that a fast UPS mode buys you time to shut down cleanly and keep a router and a laptop alive, not immunity.

Step 8: Expandability, and whether you will actually use it

Some units accept external battery packs. BLUETTI’s AC200L pairs with the B300K and the B500K, and BLUETTI lists AC200L plus B300K and AC200L plus two B300K as catalogue bundles. The AC240 lists scalability as itself plus up to four B210 packs, or two AC240 units plus eight packs in parallel.

Expandability is worth paying for in exactly one situation: you are reasonably sure you will want more capacity later and you want to avoid buying a whole second unit. It is worth nothing at all if you buy the expandable base and never buy the pack, which is what usually happens. Expansion packs are not cheap, and a second standalone unit is often better value and gives you redundancy plus the ability to power two locations. Decide honestly which of those you are, and do not pay a premium for optionality you will not exercise.

Step 9: Warranty terms, and the clauses that void them

The headline warranty length is the least interesting part of the warranty. BLUETTI states that “For the product of portable power station, the warranty period varies from 24 months to 72 months”, and that “For the product of solar panel and accessories as cable, charger etc, 12 months is provided for the warranty service”. So the panel you buy in the bundle carries a much shorter warranty than the battery it charges. Read that again before you assume a 5 year warranty covers the whole kit.

The exclusions matter more. BLUETTI’s warranty policy voids coverage for products “Purchased from unauthorized resellers”, for “Unauthorized modification, repair, dis-assembly or operation not in accordance with the official instructions or manuals”, for “Use for special application other than normal consumer use”, and for damage “caused by exposure to excessive heat, cold, liquids or other external causes”.

The clause that catches travellers is this one, quoted from the same page: “The limited warranty is restricted to the country of purchase. The warranty will be invalid on the items once they are taken or shipped to other countries.” If you are a digital nomad who buys in the United States and then relocates, understand that you are buying a unit whose warranty stops at the border. I have watched people learn that the expensive way. There is also a 30 day money back guarantee, which is a genuinely useful window for testing the unit against your actual loads.

Step 10: Weight, and the number nobody publishes

Weight is treated as an afterthought in most reviews, which is strange, because it is the specification that decides whether the machine gets used. The useful metric is usable watt hours per pound, and no manufacturer publishes it because it makes the arithmetic visible. Here it is, computed at 85 percent of nameplate, with weights taken from each product page.

Model	Nameplate	Weight	Usable Wh at 85 percent	Usable Wh per pound
BLUETTI AC60	403 Wh	20.06 lbs	343 Wh	17.1
BLUETTI AC70	768 Wh	22.5 lbs	653 Wh	29.0
BLUETTI AC180	1,152 Wh	35.3 lbs	979 Wh	27.7
BLUETTI Elite 100 V2	1,024 Wh	25 lbs	870 Wh	34.8
BLUETTI Elite 200 V2	2,073.6 Wh	53.4 lbs	1,763 Wh	33.0

The Elite 100 V2 delivers roughly twice the usable energy per pound of the AC60 and beats the physically larger AC180. If the unit has to go in and out of a vehicle, up a flight of stairs, or into a boat, that column is the one that decides whether you resent the purchase. For camping specifically, the tradeoffs are laid out in the camping power station guide.

Match your primary use to a class of machine

With the ten steps done, this table maps the decision. Pick the row that describes what you will actually do most often, not the most dramatic thing you can imagine doing.

Primary use	Capacity class	Output that matters	Non negotiable feature	Representative model
Phones, laptops, camera gear on weekend trips	500 to 800 Wh	1,000 W continuous	Low weight	BLUETTI AC70
CPAP overnight, medical loads	800 to 1,200 Wh	Modest, under 200 W	Pure sine wave and fast UPS	BLUETTI Elite 100 V2
Home outage, fridge plus essentials	2,000 Wh and up	1,800 W continuous, high surge	Surge headroom for the compressor	BLUETTI Elite 200 V2
Van life and off grid living	2,000 Wh and up	2,400 W continuous	High solar input ceiling	BLUETTI AC200L
RV with a 30 amp shore cord	1,500 Wh and up	2,400 W continuous	A TT-30 outlet on the unit	BLUETTI AC240
Job site tools	1,000 Wh and up	1,800 W continuous, 2,700 W surge	Surge rating and durability	BLUETTI AC180
Multi day home backup	3,000 Wh and up	2,400 W continuous	Expandability and high cycle life	BLUETTI Elite 300

The specification checklist, and what a good answer looks like

Print this and hold it against any product page. If the page cannot answer a line, that is information about the manufacturer.

Specification line	What a good answer looks like	Red flag
Capacity	A watt hour figure stated plainly, such as 1,152 Wh	Only amp hours with no voltage given
Cell chemistry	LiFePO4 named explicitly	“Lithium battery” with no chemistry
Cycle life	A count plus the remaining capacity, such as 3,500 or more cycles to 80 percent	A cycle count with no capacity basis
Continuous output	A watt figure that exceeds your largest single load	Only a surge figure quoted
Surge output	At least 1.5 times continuous, stated separately	Surge and continuous conflated
Inverter type	The words “Pure Sine Wave”	Silence, or “modified”
UPS switchover	A millisecond figure, ideally 20 ms or less	“UPS function” with no number
Solar input ceiling	Watts, plus a voltage window and a current limit	Only a watt figure
AC recharge time	A time per charging mode, with the input watts	A single best case number
Weight	A pound figure you can divide into capacity	Weight buried or absent
Warranty	Years, plus a published exclusions list you can read	A number with no policy page

What I would buy at each budget

Prices below are what BLUETTI’s own storefront showed on 5 September 2026. They move constantly on promotion, so treat them as a snapshot and check the current number before you commit. The full pricing picture is in the BLUETTI pricing breakdown.

Under $400. The AC70 at US$359.00. You get 768 Wh, 1,000 W continuous with a 1,500 W surge, 3,000 or more cycles to 80 percent of original capacity, 22.5 lbs, a 500 W solar ceiling and a 5 year warranty. That surge rating clears the compressor inrush arithmetic from step 3 with margin, which a 600 W class unit does not.

Around $500. Two genuinely different answers. The AC180 at US$499.00 gives you 1,152 Wh and 1,800 W continuous with 2,700 W of surge, which is the tool and appliance answer. The Elite 100 V2 at a US$499.00 final price gives you slightly less capacity at 1,024 Wh but the same 1,800 W output, a switchover of 10 ms or less, 4,000 or more cycles to 80 percent, a 1,000 W solar ceiling and 25 lbs instead of 35.3. If your loads are electronics and medical rather than power tools, the Elite 100 V2 is the better machine on every axis except raw surge.

Around $900. The Elite 200 V2 at a US$899.00 final price, 2,073.6 Wh and 2,600 W, or the AC200L at US$847.00, 2,048 Wh and 2,400 W. The AC200L takes a higher solar voltage window of 12V to 145V at 15A and up to 1,200 W, and it accepts the B300K and B500K expansion packs. The Elite 200 V2 is lighter at 53.4 lbs. Buy the AC200L if solar and expansion are the plan, the Elite 200 V2 if the unit moves.

$1,200 and up. The Elite 300 at a US$1,199.00 final price gives 3,014 Wh, 2,400 W continuous with a 4,800 W lifting mode, a 10 ms UPS and a claimed 6,000 or more cycles. BLUETTI does not state the capacity basis for that cycle figure on the product page, so I am not going to pretend I know whether it is to 80 percent or to 70 percent. Ask before you buy. On price alone it is the cheapest energy in the lineup at 7.8 cents per delivered kilowatt hour of hardware cost, and still about 28 cents all in once the grid energy to fill it is counted.

If you want the full brand picture rather than a single model, I go through the range in the BLUETTI review. The main competitor gets the same treatment in the Jackery review.

Selling these instead of just buying one

Portable power is one of the better high ticket categories right now because the average order value is high, the products ship well, and buyers do genuine research before purchase, which rewards anyone willing to publish real arithmetic. If that appeals, the category sits inside a broader set of options in our high ticket niches list. The business model itself is explained in our guide to high ticket dropshipping.

Getting approved by the brands is the hard part, and the process is different from consumer goods. Our supplier sourcing guide walks the approval conversation. Before you contact anyone, form the entity, because most power equipment brands will not open a dealer account with an individual. Read the business formation guide first.

On tooling, I run stores on Shopify because the app ecosystem for shipping quotes on heavy freight items is the deepest there. For the entity itself I use Bizee for the filing and the registered agent.

Battery products carry real product liability exposure, so general liability cover from Hiscox is not optional in my view. And I keep the books in QuickBooks from the first order, because reconstructing a year of freight invoices later is miserable.

Frequently Asked Questions

How many watt hours do I actually need?

Add up the delivered watt hours your loads consume over the period you need to cover, then divide by 0.85 to get nameplate. The worked example above gives 1,690 delivered watt hours for a household outage day covering a refrigerator, a CPAP with humidifier, two laptop charges and four phone charges, which becomes 1,988 Wh of nameplate. Add 20 percent if you want a reserve rather than running the pack to empty.

Is LiFePO4 worth paying more for?

Yes, and the arithmetic in step 4 shows why. At 3,500 cycles to 80 percent of original capacity, the AC180 delivers roughly 3,427 kilowatt hours across its rated life at the 85 percent efficiency this cluster assumes, which works out at 14.6 cents per delivered kilowatt hour of hardware cost, or about 35 cents all in once you add the 20.4 cents of grid energy needed to fill it. A chemistry rated for a fraction of that cycle life costs several times more per delivered kilowatt hour even if the sticker price is lower.

What size inverter do I need for a refrigerator?

Size the continuous rating above the running draw and the surge rating above the compressor inrush. Embraco’s EM3D60HLT data sheet lists 139 W power consumption and a locked rotor amperage of 10.2 A at 60 Hz on a 115 to 127 V supply, which is roughly 1,224 volt amps at 120 volts. A unit rated 1,000 W continuous with 1,500 W of surge clears that comfortably. A 600 W unit with 1,200 W of surge does not have meaningful margin.

Does a portable power station save money on electricity?

No, and any guide that says otherwise is wrong. The EIA’s Electric Power Monthly Table 5.3 puts the 2025 United States residential average at 17.30 cents per kilowatt hour. On an AC180 the hardware amortisation is 14.6 cents per delivered kilowatt hour and the grid energy to fill the battery costs 20.4 cents at 85 percent efficiency, so you pay about 35 cents to deliver what the wall socket delivers for 17.3. You buy one of these for availability, not for savings.

How long do these take to recharge from a wall socket?

Faster than most people expect. BLUETTI publishes three modes for the AC180: silent at 260 W input for about 4.9 hours, standard at around 1,000 W for about 1.7 hours, and turbo at around 1,440 W for about 1.3 hours. The frequently quoted 0 to 80 percent in 45 minutes is the turbo figure, and turbo mode is audibly louder than silent mode.

Can I take my power station abroad?

Physically yes, subject to airline lithium battery rules, but check the warranty first. BLUETTI’s warranty policy states that “The limited warranty is restricted to the country of purchase. The warranty will be invalid on the items once they are taken or shipped to other countries.” That is a live issue for anyone relocating, and it is the sort of clause that only becomes visible when you need it.

The Bottom Line

Ten steps, in order: total your loads in delivered watt hours, divide by 0.85 to get nameplate, check continuous output against your largest load and surge against its inrush, demand a cycle rating with a capacity basis attached, plan realistic recharge paths, insist on pure sine wave, read the UPS millisecond figure, be honest about whether you will ever buy the expansion pack, read the warranty exclusions rather than the warranty length, and divide capacity by weight before you commit.

Do that and you will buy the right machine on the first attempt. Skip the efficiency step in step 2 and every runtime number you calculate will be roughly 15 percent optimistic, which is exactly enough to make a unit that should have worked fail at 3am. The manufacturers publish everything you need. The only thing missing from most buying guides is the willingness to do the arithmetic in public.

If you would rather sell power equipment than write about it, and you want the store, the supplier approvals and the freight setup handled for you, take a look at our done for you high ticket dropshipping build and launch service. It is the fastest route from a category you understand to a store that can actually take an order.

Related Articles

How Long Will a Power Station Run a Refrigerator

How to Size a Solar Generator

Best Portable Power Station for RV Use

Best Solar Generator for Home Backup

BLUETTI vs Jackery Compared