Wheelchair Sample Retention and Retesting: Can a Tested Unit Be Reused?

Set the ground rule first: once tested, the unit is no longer the unit you shipped

The question manufacturers ask most often is some version of "can we take this sample back and use it once testing is done?" Here is a straight answer for the trade: as soon as a sample has been through any loaded test, it is no longer equivalent to a production unit. It can be sealed as a retained sample, returned for engineering reference, or torn down for analysis. It should not go back into sales, into a trade show, or into another test submission.

So the real question is not "can we use it". It is three questions. Which class of tests has this unit been through? What role do you want it to play next? And when a supplementary test or a re-check comes up later, do you submit a fresh sample or pull the retained one? All three should be settled before samples ship. Settle them late and the price is one more scheduling round plus one more unit.

Decide each sample's fate by test class

Different tests consume a sample to very different degrees. Sort the programme by how much each test changes the sample's condition and the answer falls out on its own. The table below is the tiering we use routinely on manual and powered wheelchair programmes. Take it straight to your lab and compare notes line by line.

Two rows in that table deserve a second look. The first is "cumulative structural damage, possibly invisible". After the fatigue and impact work is finished, the sample can look completely normal while tubing, welds or bearing internals have already crossed into an irreversible state that neither the eye nor a routine inspection will pick up. The second is "depends on the check result". That verdict cannot be given on feel. It requires a written record of the fastener check, the visual weld inspection and the travel check on moving parts. No record, no return: treat the unit as non-returnable.

Sequence the programme wrong and you have bought another sample

How many tests a single unit can carry is a function of sequence, not of quantity. The general principle is non-destructive before destructive, recoverable before cumulative.

A sound order runs: visual and marking review, then dimensional and mass measurement, then function and operability verification, then the stability work, then driving and braking, and finally the strength, impact and fatigue work that ISO 7176-8 covers. The reason is simple. Once that last block is done, the sample has no structural road left.

The cost of getting the order wrong is immediate and concrete. Take a unit that has already been through fatigue and send it back for static strength, and the data is unusable, because the initial-condition premise of the test is no longer satisfied. A lab exercising professional judgement will not sign out a result like that. At that point you either submit a new sample and re-book the slot, or the whole programme slips. Fixing the sequence at the planning stage costs nothing. Fixing it once the sample is already on the rig costs a re-booked slot plus the price of a new unit.

Size the submission by mutual exclusivity, not by "is that enough"

Sample quantity is not a gut call. You list the mutually exclusive relationships and the number falls out. There are three kinds worth mapping.

First, destructive tests are mutually exclusive with each other. Once a unit has been through one of impact, fatigue or static strength, it should not be asked to satisfy another test's initial-condition requirement.

Second, anything that requires material to be cut out destroys whole-product integrity. Flammability work routinely takes material from the seat cushion, the backrest or the armrest facing. After that the unit is no longer a complete product, and every downstream test that treats the complete product as its object is affected.

Third, tests that change component condition. Once driving, range and battery-related work is done, the battery and tyres are no longer in their initial state, and any later test sensitive to those components has to be reassessed.

On top of those three, keep one spare unit that never goes on a rig. That spare earns its keep in three ways: as a comparison unit if a data dispute arises, as the sealed retained sample, and as a carrier with consistent initial condition when supplementary testing comes up. Plenty of manufacturers cut this unit to save money, then find at supplementary-test time that they have to re-open tooling and re-assemble from scratch, which costs more once you add it up.

A retained sample is not just a spare chair in the warehouse

Retained samples get used for real in three situations.

The first is physical evidence when a report is challenged. A buyer or an overseas customer questions one conclusion, and the lab needs to go back to the retained unit to re-check sample marking, configuration and condition. With no retained unit, the only option is to test again.

The second is the reference object during an on-site audit. Certification bodies and major accounts compare the retained unit against series production for structure, marking and accompanying documentation. Anything that does not line up has to be explained.

The third is the baseline for design changes. Whether a change point affects an existing conclusion is judged against the physical condition before the change. With no baseline, you re-verify the whole thing.

Custody responsibility, retention period and end-of-period disposal should be written into the testing contract rather than left as a verbal understanding. Any disposal of a retained unit before the period ends should carry written confirmation from both sides. That clause is worth a great deal the day a dispute appears.

When a retained sample stops being valid

The row people overlook is the firmware upgrade. The controller programme changed, the physical structure of the sample did not change at all, and yet the conclusions tied to control, braking and speed no longer correspond to the unit in front of you. Manufacturers tend to read this as "we only touched software". On a powered product it is a change to the energy and control path and needs reassessment. Programme sequencing detail for powered products is covered under powered wheelchair testing; where manual products differ is set out under manual wheelchair testing.

When a new sample is mandatory

The decision rule compresses into one sentence: if the change point sits on the load path, the energy path or the user interface, submit a new sample; if it sits only on cosmetic and non-load-bearing parts, a filed change note is enough.

Unpacked, the situations that trigger retesting are these:

• Change of structural-part supplier. New supplier for tubing, castors, bearings or fasteners is a load-path change even when the specification declaration is identical.
• Change or redesign of controller, battery or charger. This is an energy-path change and sits squarely inside the mandatory re-verification band.
• Change to the form of the seating system, armrests or footrests. This is a user-interface change and directly alters centre of gravity and load distribution.
• Change to the declared load range. Change the declaration and the premise of the stability and strength work changes with it.
• A buyer document, tender document or management-system document that requires re-testing on a fixed cadence. This kind of requirement has nothing to do with the product itself; follow the other party's document.

Three traps people actually fall into

Trap one: a post-fatigue unit is sent back to the production floor as a good sample. That chair reaches a trade show or a customer, and if a structural problem later appears, the traceability record points at both the manufacturer and the lab at the same time, with no clean way to apportion responsibility. The safe practice is that every unit which has been through destructive testing gets an irreversible-condition label and never re-enters the usable-parts flow.

Trap two: the retained unit gets borrowed by engineering for modification. This is extremely common in small teams. It goes out on "we just want to open it up and look", and it comes back with several parts swapped. The retention area should be physically separated and locked, with an access log and a named key holder.

Trap three: the submission is described only as "one complete vehicle for testing", with no optional configurations declared. A report can only cover the configuration actually submitted. Trying afterwards to fold a wider seat, a raised backrest and a different battery specification into the same report does not work. State the configuration list completely before samples ship; the mapping between tests and their bases is set out under testing standards and testing services.

On test bases and accreditation scope

The GB/T 18029 series commonly used for domestic submissions and the ISO 7176 series used internationally follow the same thinking in how they are split into parts. Confirm the specific correspondence against the current valid version of the standard text; do not apply it from memory.

One more point worth stating plainly: an accreditation mark only shows that the laboratory holds the relevant technical competence within its accredited scope. It is not a commitment as to market access outcomes in the target market. The same applies to sample retention and retesting arrangements. What they guarantee is traceable data and reviewable conclusions, not a substitute for anyone's market-access judgement. Blurring the two invites misunderstanding on the customer's side.

Pre-submission self-check list

• Has the test list been ordered by degree of destruction?
• Have mutually exclusive tests been split across different samples?
• Is there a spare unit reserved that never goes on a rig?
• Are retention period, custodian and end-of-period disposal written into the contract?
• Have all optional configurations been declared in one pass?
• Has the lab been told in advance about foreseeable change points?
• Does the sample marking position avoid load-bearing surfaces and material-sampling areas?

Work through that list and sample quantity, schedule and budget are effectively locked. Surprises mid-programme become rare.

Need the sample plan settled? Talk to us directly

Sample retention and retesting arrangements are, in the end, a trade: a little patience at the planning stage in exchange for a full round of rework you never have to do. SUNGO Mobility Testing Lab is the dedicated wheelchair and mobility aid testing lab within our group, accredited by CNAS, CMA and IAS (USA), with laboratories in Shanghai and Hefei. If you already have the product structure and configuration list in hand, we can map out test sequence, sample quantity and retention arrangements in one pass before quoting. Technical line +86 132 4819 8029, or request a quote directly.