Changing the Motor, Controller or Upholstery: What to Retest

The short answer: look at which path the change lands in, not at how big it looks

When a wheelchair design changes, the question of whether to retest is not settled by asking whether the change was large or small. It is settled by asking which of the following four paths the change lands in:

• Energy path: the power chain from battery through the control device to the drive motor;
• Control path: command input, logic processing and protective actions;
• Material composition: soft materials and covering components in contact with the user;
• Mass and geometric distribution: whole-product centre of gravity, overall dimensions, assembly relationships.

Whichever path the change lands in is the path whose items are triggered. The advantage of framing it this way is that you do not have to guess item by item — once you can state clearly what was changed and what it affects, the retest scope largely settles itself. The three substitutions that come up most often in this industry are set out below.

The table gives directional judgements based on the usual way the ISO 7176 series and the GB/T 18029 series are organised; the specific items are still to be confirmed against the currently valid version of the standard text itself.

Changing the motor: "it still drives the same" is not the test

Motor changes get underestimated easily. The instinct is that if the parameters are close and the thing drives once it is bolted on, no retesting is needed. That instinct goes wrong in two places.

The current signature has changed, so electromagnetic compatibility has to be redone. The drive motor is one of the principal sources of interference in the product. Any difference in winding arrangement, commutation method or drive current waveform changes the emission and immunity behaviour of the whole product. This content sits in the scope covered by ISO 7176-21, and "it looks the same" does not get you out of it.

The protection trip points have moved, so the control system items have to be redone as well. Overcurrent and overtemperature protection inside the controller were set against the characteristics of the original motor. With a different motor, the current curve under the same load condition is different, and the timing of the protective action moves with it. That content sits in the scope covered by ISO 7176-14.

There is a further layer that gets missed: once motor mass and mounting position change, the whole-product centre of gravity moves with them, and the conclusions from the stability items and some of the whole-product performance items no longer hold. This is the knock-on effect along the mass and geometry path, and it has to be re-checked alongside the rest. In electric wheelchair testing these three almost always trigger together.

Changing the controller: assume the full electrical item set is rescheduled

The controller is the core of the control path. Changing it leaves almost no room for exemption, unless the change is confined to the housing, the colour or the printed legend, with circuitry, firmware and parameter settings all unchanged and traceable evidence available to show it.

It is worth being explicit that a firmware update also counts as a change. Plenty of manufacturers take the view that if the hardware has not moved, nothing has changed — but protection thresholds, acceleration and deceleration curves and fault response logic all live in the firmware, and changing the firmware means changing the control path. A pattern that has played out in this industry: a manufacturer quietly adjusted the acceleration and deceleration curve during production to improve the feel of the product, and at a later verification was required to add electrical and handling items, with the product held up in the channel for a full cycle.

The self-check is direct. Take the old and new controller versions and compare them line by line on input method, output capability, types of protection and their operating conditions, fault indication method, and interface with the charging device. One line that does not match and you treat it as a retest.

Changing upholstery and cushions: resistance to ignition is a hard gate

Upholstery changes look like the lightest of the three categories and are in practice the most rigid. Resistance to ignition requirements for the seating system sit in the scope covered by ISO 7176-16, and the judgement is made on the material itself, independently of the structure.

In practice the traps cluster in three places.

First, the supplier changed but the material designation did not. The same designation from a different manufacturer may carry a different formulation and a different finishing process, and the ignition behaviour need not be the same. The test is whether the material is traceable to the same source, not whether the designation string matches.

Second, only the outer fabric was changed and the foam inside was forgotten. Resistance to ignition is a property of the seating system as an assembly, and a change in any layer — face fabric, liner, filling material — can change the conclusion.

Third, a colour change is treated as no change. Dyes and auxiliaries are part of the formulation, and whether a recoloured version of the same fabric still meets the requirement needs either a statement from the material supplier or fresh verification.

On manual products, cushion and backrest also participate in dimensions and mass distribution, so a change in material thickness carries over into some of those items as well. That comes up fairly often in manual wheelchair testing.

Differential testing or a full retest

Not every change means running the whole item list from the beginning. The test is: is the path affected by the change coupled to any other path?

If the change lands in a single path and does not alter whole-product mass or geometry, differential testing is usually possible — redo the items belonging to that path only and carry the existing conclusions for the rest. If the change affects two or more paths at once, or shifts the centre of gravity or the overall dimensions, there is no shortcut available and the programme has to be rebuilt as if for a new product.

Differential testing is only possible where the sample condition, configuration information and test records from the original report are complete and retrievable. This is exactly why sample management and request form information deserve care at the submission stage — where the original records are incomplete, every subsequent change assessment has to start from scratch. The background to that is in the testing services overview.

Do the assessment at the planning stage

The right moment for a change assessment is before the design is frozen, not after the samples have been built. The same piece of work, done at the planning stage with a bill of materials and a difference statement in hand, costs next to nothing; discovering after the new samples reach the lab that the change has triggered the full electrical item set means new samples, a new place in the queue, a new invoice, and a launch date that moves out by a full cycle.

Here is a self-check you can apply straight away: before every change, write one sentence stating what is being changed and which path it affects. If you cannot write that sentence, the scope of the change is not yet clear in your own mind, and that is the moment to walk through it with the lab rather than build samples first. The relevant item correspondences are set out in the standards and test items overview.

Ask us which items your change triggers

Send us the before-and-after bill of materials, the electrical parameter differences and the material information, and we can give you a view on the retest scope, a conclusion on whether differential testing is available, and the corresponding schedule. SUNGO Mobility Testing Lab is the dedicated wheelchair and mobility aid testing lab within the group, accredited by CNAS, CMA and IAS (USA), with laboratories in Shanghai and Hefei. Technical enquiries: +86 132 4819 8029, or request a quote.