How We Research
Trek4Tech exists to help you buy the right portable power gear without wading through marketing copy. Here is exactly how we arrive at our numbers, prices, and recommendations — and, just as importantly, what we don't do.
What we do
- Calculate runtime from published specifications. We compute estimated runtime from each unit's rated capacity (Wh) and a standard 0.85 inverter-efficiency factor, compared against real-world usage scenarios (e.g. a 60W CPAP, a cycling fridge).
- Verify products against live retailer listings. ASINs, prices, images, and availability are checked against current retailer pages, and refreshed on a recurring schedule. Prices change constantly, so we show the source and treat every figure as a snapshot.
- Compare on the specs that actually matter. Capacity, continuous and surge output, battery chemistry and cycle life, solar input, and weight — the factors that determine whether a station fits your use case and how long it lasts.
- Assign editorial ratings. Our ratings reflect how well a product serves its intended buyer. The full rubric and weights are below, so you can check our arithmetic rather than take the number on faith.
How we score
Every rating is built from four weighted factors, scored 1–5 against the product's own class. Power stations and solar panels use different rubrics, because the specs that decide a good station are not the specs that decide a good panel.
Power stations and solar bundles
| Factor | Weight | What we look at |
|---|---|---|
| Longevity | 25% | Battery chemistry and rated cycle life — the spec that decides whether a station lasts three years or twelve |
| Output headroom | 20% | Continuous and surge output relative to capacity class, and whether it starts real compressor loads |
| Solar intake | 15% | Maximum solar input relative to capacity — how fast you can refill it off-grid |
| Recharge speed | 15% | Time to a full charge from a wall outlet using the power supply included in the box. Where a manufacturer sells a faster supply separately, we score what you get by default and note the upgrade in the review. |
| Value | 25% | Capability per dollar at typical street price, not list price |
Solar panels are scored on four factors instead of five — a panel has no battery to recharge.
Solar panels
| Factor | Weight | What we look at |
|---|---|---|
| Real-world yield | 35% | Cell efficiency, bifacial gain, and the wattage we'd expect in the field rather than at STC |
| Build and weather resistance | 25% | IP rating, hinge and stand durability, deployment format |
| Compatibility | 15% | Connector standards and how well it works outside its own brand |
| Value | 25% | Watts per dollar at typical street price |
What the bands mean
- 4.5–5.0 — Class-leading. Hard to beat for its intended buyer.
- 4.0–4.4 — Strong. A clear recommendation with a known tradeoff.
- 3.5–3.9 — Competent but compromised. Buy it for one specific reason.
- 3.0–3.4 — Specialist or outclassed. Buy it for the one thing it does that bigger options don't, or on price.
- Below 3.0 — Dated. We still list it only where it leads on a single genuine strength, and we say what that is.
Factor anchors
Weights tell you how much each factor counts; anchors tell you what a 3 or a 4 means on that factor. Each factor is scored 1–5 against the tables below, and the weighted sum is the published rating — so you can rebuild any of our scores from the specs in the review.
Longevity — rated cycle life to 80% capacity
| Score | Anchor |
|---|---|
| 5 | 5,000+ cycles |
| 4.5 | 3,500–4,999 cycles |
| 4 | 3,000–3,499 cycles |
| 3 | 2,000–2,999 cycles |
| 2 | 1,000–1,999 cycles |
| 1 | Under 1,000 cycles |
We use the manufacturer's published cycle rating. Chemistry isn't scored separately — it's already expressed in the cycle count.
Output headroom — continuous and surge output relative to capacity class
We score both an absolute-watts table and a proportionate table (continuous watts ÷ rated watt-hours), take the higher of the two, then apply the surge modifier. A station qualifies either by driving large loads outright or by being well-proportioned to its own battery — a small station shouldn't be penalized for being small.
| Score | Absolute output |
|---|---|
| 5 | 3,000W+ continuous, or 120/240V dual voltage for hardwired circuits |
| 4 | 1,800–2,999W |
| 3 | 1,200–1,799W |
| 2 | 600–1,199W |
| 1 | Under 600W |
| Score | Proportionate output (continuous W ÷ rated Wh) |
|---|---|
| 5 | 1.80 W/Wh or higher |
| 4 | 1.40–1.79 |
| 3 | 1.00–1.39 |
| 2 | 0.70–0.99 |
| 1 | Under 0.70 |
Surge modifier, applied after taking the higher of the two scores above and capped at 5.0: +0.3 if surge is 2.0× continuous or better, −0.3 if surge is under 1.5×.
Solar intake — max solar input ÷ rated capacity (W per Wh)
| Score | Anchor |
|---|---|
| 5 | 0.80 W/Wh or higher |
| 4 | 0.55–0.79 |
| 3 | 0.40–0.54 |
| 2 | 0.25–0.39 |
| 1 | Under 0.25 |
Recharge speed — AC time to full on the in-box power supply
| Score | Anchor |
|---|---|
| 5 | Under 60 minutes |
| 4 | 60–90 minutes |
| 3 | 91–150 minutes |
| 2 | 151–300 minutes |
| 1 | Over 300 minutes, or the manufacturer publishes only an accessory-gated figure |
Where an advertised time requires a supply or cable sold separately, we derive the in-box floor (rated Wh ÷ in-box AC input watts), state the arithmetic in the review, and score the floor. We never score the advertised figure.
Value — rated lifetime energy per dollar
(capacity Wh × rated cycles) ÷ 1000 ÷ street price, in kWh per dollar. Street price, never list. If a product has no current street price, we use the last price recorded on a dated check and say so.
| Score | Anchor |
|---|---|
| 5 | 6.0 kWh per dollar or higher |
| 4 | 4.0–5.9 |
| 3 | 2.5–3.9 |
| 2 | 1.5–2.4 |
| 1 | Under 1.5 |
Adjusted ±0.3 for value that isn't energy: a bundled panel, retail-returns availability, an ecosystem expansion path, a best-in-class app. The review says which adjustment was applied and why.
Panel factor anchors
Real-world yield — cell technology and published efficiency
| Score | Anchor |
|---|---|
| 5 | Bifacial and published cell efficiency 23%+ |
| 4 | Bifacial, or single-sided with published efficiency 23%+ |
| 3 | Single-sided with published efficiency 20–22%, or efficiency not published |
| 2 | Single-sided with published efficiency under 20% |
| 1 | Documented underperformance in the manufacturer's own spec or warranty terms |
Rated wattage isn't scored here — a 100W panel isn't a worse panel for being 100W, it's a different category. An absent efficiency figure is an information gap, not evidence of poor performance, and scores a 3. Instead, each review states which station class the panel suits (a 100W panel suits sub-500Wh stations and daytime top-ups; a 400W panel suits 2kWh+ stations).
Build and weather resistance — published IP rating and deployment format
| Score | Anchor |
|---|---|
| 5 | IP68 |
| 4 | IP67, or IP65 with an integrated adjustable stand |
| 3 | IP65, or no IP rating published with otherwise solid construction |
| 2 | No IP rating and no stand |
| 1 | Known durability complaints in the manufacturer's own warranty terms |
Compatibility
| Score | Anchor |
|---|---|
| 5 | Standard MC4, works across brands with no adapter |
| 4 | Standard connectors, adapter needed for some brands |
| 3 | Native to one brand, adapter required elsewhere |
| 2 | Proprietary connector, adapter not sold |
| 1 | Locked to a single product |
Value — rated watts per dollar at street price
The bands come from the panel catalogue we actually track, not from invented thresholds: 3 sits at the catalogue median, 5 at the best, 1 at the worst, with 4 and 2 spaced evenly between. Across the six panels we list (street prices from dated checks between 2026-06-27 and 2026-07-23), the median is 0.58 W per dollar, the best 0.81, the worst 0.40. That yields:
| Score | Anchor |
|---|---|
| 5 | 0.75 W per dollar or higher |
| 4 | 0.64–0.74 |
| 3 | 0.54–0.63 |
| 2 | 0.45–0.53 |
| 1 | Under 0.45 |
These bands are recomputed when the catalogue or its prices change materially, and the thresholds published here are updated with them.
The factor-by-factor derivation of every published rating — each factor score, the spec it came from, and any adjustment applied — is recorded in our ratings worksheet, so any score can be checked line by line.
One note on recharge speed, because it is the factor manufacturers most often overstate. Several stations advertise an AC charge time that is only reachable with a power supply or cable sold separately, and in at least two cases the advertised figure is arithmetically impossible on the cable in the box. Where we find that, we score the default configuration, show the arithmetic, and say what the upgrade costs you.
Ratings track product merit, not stock status. A product we rate highly may still be unavailable — where that's true, we say so in the review rather than quietly adjusting the score.
What we don't do
We do not currently perform hands-on, in-lab testing of the products we cover, and we will never claim that we do. Our runtime figures are calculated from manufacturer-published specifications and a standard efficiency factor — not measured on a bench under load. Real-world performance varies with temperature, battery age, and exact device draw. Where a number is an estimate, we say so.
We also don't accept payment to feature or rank a product. We earn affiliate commissions when you buy through our links (see our affiliate disclosure), but commissions never influence our editorial picks. When a recommended product is discontinued or unavailable, we say so rather than steering you to a worse option that happens to pay more.
Keeping it current
Hardware and prices move fast in this category. We re-verify pricing and availability regularly and audit our top content on a recurring basis to update prices, flag discontinued or upgraded models, and refresh the year. If you spot something out of date, let us know and we'll fix it.