Smart Card Manufacturing Notes
Published: November 12, 2026 | ZOWINDA Application Engineering
Most card test programmes are a list of tests somebody inherited, run in whatever order the lab has free capacity. The tests themselves are usually correct. What is missing is the link between a test and the failure it is supposed to prevent, which is why a card can pass everything on the list and still fail in a wallet, a turnstile or a passport gate. A programme is not a list; it is an order, with a reason for each step.
Electrical tests ask whether the card still works as a resonant circuit: resonance position, bandwidth, load modulation depth and the field at which the chip starts. They are fast, they are the cheapest screening available, and they are the only category that can be run on every card rather than on a sample. A programme that starts anywhere else is spending money before it has spent the cheapest money.
Mechanical tests ask whether the card still works after the abuse a card receives. Bending, torsion, peel strength, surface pressure and flatness each target a specific interface: the antenna-to-body bond, the module-to-cavity bond, the cover-to-core laminate, the printed layer. They are slow, destructive and sample-based, which is precisely why they must be paired with an electrical re-test afterwards. A mechanical result on its own tells you the card survived a fixture; it does not tell you the card still reads.
Functional and environmental tests ask whether the credential still behaves as a credential: transaction sequences, cryptographic responses, behaviour after temperature and humidity cycling. These are the closest proxy for the field, and they are also the slowest, so the whole art of a programme is deciding what runs on every card, what runs on a sample, and what runs only at qualification.

Figure: mechanical stress applied at a defined interface, followed by an electrical re-test, is what turns a destructive test into evidence. Equipment: ICC Backside Spot Pressure Tester.
Electrical measurement on every card catches tuning drift and process excursions while the batch is still in the building, at a cost per card that no mechanical test can approach.
Pressure, peel, bend and torsion results are only useful when they name the interface that failed. A pass or fail on a whole card is a data point; an interface is a corrective action.
Field failures are dominated by cards that were electrically fine when shipped and degraded afterwards. Post-stress electrical data is the closest laboratory proxy for that.
One hundred percent electrical plus sampled destructive testing is a defensible programme. Sampled electrical testing plus sampled destructive testing is a programme that finds problems late.
Test methods and pass limits come from the standard and scheme you are qualified against. What this matrix fixes is the order and the pairing, which is where most programmes lose the value of tests they are already paying for.
The value of the matrix is not that it adds tests. It is that it stops tests from being read in isolation. A resonance reading means something different before and after a bend; a peel result means something different when it is paired with a post-stress range measurement. Once the pairing is fixed, the programme produces evidence that survives contact with a customer, and each new failure mode arrives with the test that should have caught it already identified. The companion notes on antenna resonance and on inlay sheet yield explain the two measurements this programme depends on most.
Because the mechanical result only says the card survived the fixture. Most degradation from bending, torsion or pressure shows up first as a resonance shift or a higher starting field, long before the card stops working entirely. Re-testing converts a destructive test into a trend.
No. Electrical tests are sensitive to tuning and to some bond degradation, but they cannot see a laminate that will delaminate after fifty bend cycles. They are a screening layer, not a substitute.
Enough to see the failure mode you are worried about, taken from the same population the electrical screening passed. The number matters less than the rule that every destructive result is followed by an electrical one.
More equipment in this range is listed on the ZOWINDA CardMachine product page, and related application notes appear in our Industry News section.
Related reading:
Contact us for a tailored solution and quotation on inlay production, card testing and personalisation equipment.
Email: info@zowinda.com
WhatsApp: +86 186 2085 0485
We are ready to answer your questions.