The conclusion first: delays almost never happen on the test rig
The schedule slipped again, so why? We have run this post-mortem many times and the answer is consistent. Time spent actually running tests on the rig is predictable. Delays come from four things: the configuration list was never frozen, samples and spares were short, the technical documentation had gaps, and the design changed mid-stream. All four sit on the client side. A laboratory can spell out the requirements early, but it cannot push them.
This piece is not about how the tests are run. It covers what each stage between order and report is waiting on, who to call when it stalls, and which delay costs the most. For the scope of complete-wheelchair testing see the testing services overview; what follows is organized by process stage.
A lead time is several segments stitched together, not one number
Break the stages apart first. Most people assume lead time equals test duration; in practice testing is one segment, and the waiting at either end is often longer.
| Stage | What it is actually waiting on | Who can move it | Frequent cause of delay |
|---|---|---|---|
| Requirement confirmation and test plan | Target market, product positioning and configuration list | Client | The configuration list keeps changing and items get re-planned each time |
| Contract and schedule lock-in | Equipment window and staff availability | Both parties | Signature drags and the window goes to another project |
| Samples and documents arriving | Sample count, spares, chargers, accompanying documents, drawings | Client | Complete machines sent without spares, so one breakage stops the line |
| Incoming check and pre-review | Sample condition matching the declared configuration | Laboratory | The machine that arrives is not the configuration on the application |
| Routine performance and safety items | Equipment window; most items can run in parallel | Laboratory | Not enough samples, so everything runs in series |
| Strength, fatigue and durability | The test running to completion under the defined conditions | Time itself | Failure mid-test, repair, then a full rerun |
| EMC and electrical items | Chamber window plus rework round trips | Both parties | One failed attempt adds a rework and retest cycle |
| Nonconformity handling and retest | Design change implemented and new samples delivered | Client | Vague rework plan, repeated attempts |
| Report drafting, review and issue | Complete data and the sign-off process | Laboratory | One missing raw record or one missing rating-label photograph |
There is a pattern in that table: where the row says client, chasing the laboratory achieves nothing; where it says time itself, paying more achieves nothing either. The only compressible part is the segment that can run in parallel, and parallel work depends on having enough samples.
Sample count: one machine will not carry the whole programme
This sits near the top of the delay list. Some items in complete-wheelchair testing are destructive or semi-destructive. Once strength, impact and fatigue are done, the machine is no longer in a condition suitable for appearance, operating force or braking, all of which are sensitive to sample condition.
The right approach is to prepare samples by parallel item group rather than guessing at a number of machines.
- One group for strength, impact and fatigue. Effectively scrap afterwards, and it does not move on to other items.
- One group for stability, braking and drive control, which need the machine intact.
- EMC occupies a sample on its own, and it has to arrive with the full harness, the charger and the accompanying accessories.
Spares matter just as much. Controllers, joystick assemblies, chargers, batteries, wheel sets, cushions and footrests are the parts most likely to be damaged during testing. Without spares on site, one broken part stops the entire line.
Failure case: a single sample is submitted and the frame weld cracks partway through the fatigue direction. The client repairs it and sends it back, but the process condition of the repaired part no longer matches the original sample, so the strength data already generated cannot be carried forward and the sequence restarts. Had a second machine been on site, testing would have resumed the same day and the loss would have been the cost of one sample.
Where documentation gaps bite
Missing documents do not show up immediately. They usually surface during report drafting, when they are most expensive to fix. At minimum, the following should arrive with the samples.
- Instructions for use and accompanying documents, including use limitations, maintenance requirements and warning markings.
- Configuration list and difference statement; multi-configuration families also need a coverage rationale, as covered in the notes under powered wheelchair testing.
- Structural drawings with material and process descriptions, and critical joints clearly identified.
- Electrical schematics, harness routing, grounding and bonding arrangements.
- Battery and charger specifications, including chemistry and protection strategy.
- A list of accompanying accessories and options, marking which are in scope this time.
- Sample rating labels and markings, including how the occupant mass limit and use limitations are worded.
The rating label is the item most often overlooked. What is stated on it drives the loading conditions for a whole set of items, and if the label on the submitted sample is a prototype version that differs from production, the test conditions have to be set up again.
Changing the design mid-stream: the stage sets the order of magnitude
| When the change happens | How it is handled | Cost |
|---|---|---|
| Planning and configuration confirmation | Amend the test plan directly | None |
| Before samples arrive | Re-sequence items and possibly adjust the sample plan | Contained; the window moves slightly |
| During execution | Rule item by item on what is void, according to whether the change touches the load path, the mass in running order or the control logic | Some completed items are void and go back in the queue |
| During report drafting | Treated as a new verification object | New engagement; earlier effort is largely sunk |
A practical rule: if a change touches any one of the load-bearing parts, the mass distribution in running order, or the control and energy system, assume the related completed items are void unless a physical equivalence argument can be made. Do not expect a claim that the change is small to serve as grounds for carrying data forward.
Why strength and durability cannot be pushed
The durability verification in the ISO 7176 series and the corresponding requirements under GB/T 18029 rely on running the process to completion under the defined conditions. Turning up equipment power does not shorten it. This is the segment where time itself is the constraint and there is no expedited route. Three things do help.
First, queue early and lock the window as soon as the plan is confirmed, rather than waiting until everything else is finished. Second, run non-conflicting items in parallel, which again depends on sample count. Third, keep a spare machine so that a mid-test failure can be picked up immediately instead of waiting on a repaired part.
Running export and domestic routes together
GB/T 18029 and the ISO 7176 series overlap in the composition of test items. In principle the overlap can be sampled once, tested once and used on both sides, but this has to be declared at the planning stage. Trying to merge afterwards usually fails because the format of the raw records, the sample condition identification and the way environmental conditions were recorded do not line up, so the work is repeated. The time saved here is generally more than every other optimization combined. Mobility scooters follow the same scheduling logic - see mobility scooter testing.
How to make a lead time predictable
- Freeze the configuration list before the contract is signed, and after that route changes through a change procedure rather than a phone call.
- Prepare samples by item group rather than by a headcount of machines, and align the spares list in advance.
- Supply the documentation in one pass, especially electrical schematics, rating label proofs and the option list.
- Raise export and domestic requirements together at the planning stage and declare what can be shared.
- Leave room for at least one EMC rework round trip; do not plan on passing first time.
- Nominate one contact who can make decisions, so that every question does not go back for a meeting.
Scheduling structures from previous projects are set out under case studies; comparable products distribute across the stages in much the same way.
A note on accreditation and lead time
Clients sometimes ask whether an accredited laboratory is faster. The two are not related. An accreditation mark only demonstrates that the laboratory has the corresponding technical competence within its accredited scope, it is not a commitment regarding market access in the target market, and it does not imply a shorter queue. What actually drives lead time is equipment availability, sample preparation and how complete the documentation is.
Want a schedule you can plan against
SUNGO Mobility Testing Lab is the dedicated wheelchair and mobility aid testing lab within the same group, accredited by CNAS, CMA and IAS (USA), with laboratories in Shanghai and Hefei.
Send over the product positioning, the configuration list, the target markets and your planned milestones. We will map them onto the stage structure above and come back with a segmented schedule that marks which segments can run in parallel, which are hard waits, and what you need to prepare in advance for each. That produces a lead time you can execute against rather than a single vague number. Further background on the lab is available under about us.
Call +86 132 4819 8029, or request a quote.