Methods
How we work
How do we check specs?
- We copy each figure from the maker's own page, spec sheet or brochure, and record the page and the date we read it. If the maker does not publish a figure, we use a retailer's page and say so.
- When two of a maker's pages disagree, we publish both figures and list the conflict on the chair's page.
- We work out battery watt-hours (volts × amp-hours) and compare them with the FAA limit for mobility-aid batteries.
- We search FDA's 510(k) database and recall records for each chair, and record the date we searched.
- Lists on our spec pages group chairs by a stated rule (for example, weight capacity of 400 lb or more) and are sorted by one stated spec. They are not scores.
All of it is in specs.json and a CSV for each category, licensed CC BY 4.0.
Our hands-on test protocol (draft)
Version 0.1, draft. Draft. Test procedures, score anchors and weights will be reviewed by the clinical advisor and the test specialist before the first chair is tested. Anchors and weights are fixed before testing and every later change is logged below and applied to every product at once.
In short
This is the protocol we are setting up for hands-on testing. Each power wheelchair will be bought at retail and put through the same 7 tests, with every trial recorded. A fixed formula will turn those measurements into sub-scores from 0 to 100, and each list will weight the sub-scores for its use. The formulas are published here before any testing and every change will be logged.
What will we measure?
M1: Weight
Based on ISO 7176-5 (mass)
Weigh the chair as delivered, ready to drive, on a calibrated platform scale; then weigh it again with every battery removed. Three readings each; we report the mean.
M2: Range
Adapted from ISO 7176-4 (theoretical distance range)
Fully charge the battery, then drive a fixed, mapped loop (mixed pavement with short ramps) at a steady pace with a 200 lb test load until the chair stops or drops to walking speed. We report the mean distance of repeated runs and the air temperature.
M3: Ramp climbing
Based on ISO 7176-2 (dynamic stability)
Climb and descend test ramps of increasing angle with a 200 lb test load and a spotter. The reported value is the steepest ramp the chair climbed and descended in every trial without tipping, stalling or rolling back.
M4: Curb climbing
Based on ISO 7176-10 (obstacle climbing)
Approach a square-edged curb of increasing height head-on from a short run-up. The reported value is the highest curb climbed in every trial.
M5: Speed and braking
Based on ISO 7176-3 (brakes) and ISO 7176-6 (speed, acceleration, deceleration)
Drive at top speed on a level, dry surface, release the joystick at a marked line and measure the distance to a full stop. We report the longest stopping distance of all trials, and the mean top speed over a timed distance.
M6: Maneuvering and fit
Based on ISO 7176-5 (dimensions and maneuvering space)
Measure the overall width and the smallest turning circle. Drive the chair through a standard 32-inch interior door frame without touching it.
M7: Folding and trunk fit
In-house method
Time a tester folding the chair from ready-to-drive to fully folded (batteries left in unless the maker requires removal). Measure folded dimensions and load the chair into the reference sedan trunk described on this page.
Reference trunk: To be fixed before testing starts (planned: one midsize-sedan trunk and one compact-SUV cargo area, named here with their measured openings).
| Measurement | Unit | Reported as | Better is | Method |
|---|---|---|---|---|
| Weight with battery | lb | Average of trials | Lower | M1 |
| Weight without battery | lb | Average of trials | Lower | M1 |
| Range on test course | mi | Average of trials | Higher | M2 |
| Steepest ramp climbed | ° | Worst trial (lowest) | Higher | M3 |
| Highest curb climbed | in | Worst trial (lowest) | Higher | M4 |
| Braking distance from top speed | ft | Worst trial (highest) | Lower | M5 |
| Top speed on level ground | mph | Average of trials | Higher | M5 |
| Turning radius | in | Worst trial (highest) | Lower | M6 |
| Overall width | in | Average of trials | Lower | M6 |
| Drives through a 32-inch door | pass/fail | Pass only if every trial passes | Pass | M6 |
| Time to fold | s | Median of trials | Lower | M7 |
| Folded length | in | Average of trials | Lower | M7 |
| Folded width | in | Average of trials | Lower | M7 |
| Folded height | in | Average of trials | Lower | M7 |
| Fits the reference sedan trunk | pass/fail | Pass only if every trial passes | Pass | M7 |
How will measurements become scores?
Each measurement gets 0 to 100 points on a straight line between two fixed anchors: the “worst” value earns 0, the “best” value earns 100, and anything beyond the anchors is capped. Pass/fail tests earn 100 or 0. A test that does not apply (for example fold time for a chair that does not fold) earns 0. Anchors do not move when we add a chair, so one chair’s score never changes because another chair was tested.
| Sub-score | Measurement | Share | 0 points at | 100 points at |
|---|---|---|---|---|
| Portability | Weight without battery | 50% | 80 lb | 30 lb |
| Time to fold | 25% | 90 s | 10 s | |
| Fits the reference sedan trunk | 25% | Fail | Pass | |
| Range | Range on test course | 100% | 5 mi | 25 mi |
| Ramps and curbs | Steepest ramp climbed | 50% | 3 ° | 12 ° |
| Highest curb climbed | 50% | 0.5 in | 4 in | |
| Braking | Braking distance from top speed | 100% | 8 ft | 2 ft |
| Maneuvering | Turning radius | 60% | 48 in | 24 in |
| Drives through a 32-inch door | 40% | Fail | Pass |
The score on each list is a weighted sum of sub-scores, rounded to one decimal (halves round up). A chair missing any measurement a list needs is left off that list, and the list says why.
| List | Portability | Range | Ramps and curbs | Braking | Maneuvering |
|---|---|---|---|---|---|
| Everyday / Best overall | 15% | 25% | 15% | 25% | 20% |
| Lightweight & travel | 45% | 20% | 0% | 20% | 15% |
| All-terrain & outdoor | 0% | 20% | 45% | 25% | 10% |
| Heavy-duty & bariatric | 0% | 25% | 20% | 30% | 25% |
What will have to happen before a test result goes live?
- Every scored measurement is recorded, with every trial.
- Our test lead signs off the test log, and our clinical advisor reviews and signs off the results. Our site build refuses to publish a chair without both sign-offs.
- We check FDA 510(k) status (power wheelchairs are wheelchair, powered, FDA product code ITI, Class II, 21 CFR 890.3860), recalls, and Medicare coding. Power wheelchairs are FDA Class II devices that normally need 510(k) clearance. An FDA establishment registration or device listing is not the same as 510(k) clearance.
- We check the battery against airline rules. US airlines follow DOT/FAA rules for lithium batteries on mobility devices: each battery may be up to 300 Wh; a passenger may carry one spare up to 300 Wh or two spares up to 160 Wh each. Always confirm with your airline.
Can I check your numbers?
Yes. Every published spec and its source is in specs.json and in a spreadsheet-friendly file for each category (Power Wheelchairs specs CSV). Test results will be added to products.json. Data licensed CC BY 4.0. Found a mistake? Write to [email protected].
What has changed?
Version 0.1,
- Added the spec library: published maker specs for 15 power wheelchairs, each figure linked to its source, with battery watt-hours and FDA 510(k) lookups. Spec lists group chairs by a stated rule and are not scores.
Version 0.1,
- First draft: 15 measured metrics, 7 test methods, 5 sub-scores with fixed anchors, 4 launch rankings for power wheelchairs.