Plan for compliance from day one
Compliance testing tends to arrive at the end of a hardware project, when the design is finished, the launch date is set, and there is no time left to change anything. A failure at that point means a redesign under pressure.
It does not have to work that way. Most compliance problems are designed in early, and most can be designed out early, at little cost. This article covers when each compliance step belongs in a project, and the design decisions that make certification easier. It is not legal advice; the rules for your product depend on what it is and where it is sold.
When compliance happens
Concept
- Choose your markets
- Module or chip-down radio
Design
- Layout for emissions and immunity
- Room for fixes
- Safety design
Prototype
- Near-field scans
- ESD on the bench
Pre-production
- Pre-compliance at a lab
- Fix what it finds
- Formal testing and certification
Launch
- Labels, manuals, declarations
- Route changes past compliance
ThroughoutCollect test reports, supplier declarations, and the technical file as you go.
Know where you will sell before you design
The markets you sell into decide which rules apply: emissions and radio rules, product safety, restricted substances, and more and more often, cybersecurity. Settle the list in the concept phase, because it shapes the radio approach, the safety design, and the test plan. Compliance by market covers the main approvals in the US, Canada, the EU, the UK, and beyond, and Designing connected products for EU and UK cybersecurity rules covers the security side.
Decisions that make compliance easier
Choose between a radio module and a chip-down design deliberately
There are two ways to add a radio, and the choice shapes the whole certification plan.
| Pre-certified module | Chip-down design | |
|---|---|---|
| Radio certification | Reduced, if integrated as the module’s certification assumes | Full certification in each market |
| Unit cost | Higher | Lower, especially at volume |
| Size | Set by the module | As small as the design allows |
| RF control | Limited to the module’s antenna options | Complete: antenna, matching, and layout |
| Low-level control | Often limited to the vendor’s firmware and interface | Full access to the radio and its firmware |
| Power management | Limited to the modes the module exposes | Tuned to the product: sleep modes, duty cycling, and power rails |
| Engineering effort | Lower | Higher: RF layout, tuning, and test |
A pre-certified module can save a large part of the radio certification effort, because the module already carries its own approvals. That only holds if the module is integrated the way its certification assumed: the same antenna type and gain, the specified layout and keep-outs, and any listed restrictions. Moving the antenna, changing the antenna trace or feed line, putting the module in a metal enclosure, placing the antenna close to the user’s body, or running a second transmitter at the same time can bring back testing you hoped to avoid. In the EU there is no separate modular approval: the finished product is the radio equipment, and the module’s test reports support your own assessment rather than replace it. In every market, the finished product still needs its own emissions testing as a digital device, and its own labeling.
A chip-down design puts the radio chip directly on your own board. It means full radio certification for each market, but it can lower unit cost, save space, and give you complete control over the antenna, RF performance, low-level radio behavior, and power management. Modules often limit the last two to what the vendor’s firmware exposes. With careful RF layout, early measurements, and a test plan built in from the start, it is a routine path to certification. I design chip-down radios regularly and take them through certification.
Modules usually make sense at lower volumes or when time to market matters most. Chip-down designs tend to pay off at higher volumes, in tight enclosures, or when the module options do not fit the product. Module or chip-down: making the radio decision goes deeper.
Treat emissions as a layout problem
Most emissions failures trace back to a few causes:
- Fast signal edges with no reason to be fast.
- Signals routed without a solid return path, especially across plane splits.
- Cables acting as antennas, because nothing filters what leaves the enclosure.
- Switching power supplies laid out without following their datasheet’s layout guidance.
All of these are cheap to get right in the first layout and expensive to fix afterward.
Plan for immunity, not just emissions
Products sold in the EU and UK must also tolerate interference: electrostatic discharge, radio fields, fast transients, and surges on power and signal lines. The US and Canada mostly do not require this for commercial products, but customers and other markets often do. ESD protection on every exposed connector and button, and a plan for where that energy goes, belong in the first schematic.
Leave room for fixes
Reserve space for a few things you hope not to need: series resistors on fast clock and data lines, footprints for filter parts on cables, a shield can over a noisy section, a ferrite on the power input. Unpopulated footprints are nearly free. Adding them after a failed test means a respin.
Design for safety from the start
If the product connects to mains or another hazardous voltage, creepage and clearance distances and insulation shape the board outline and the enclosure. If it holds a large battery or another high-power source, fire protection does too: enclosure materials, spacing from flammable parts, and protective circuits. Both are hard to retrofit. Where you can, use a certified external power supply, which keeps mains voltage out of your product entirely.
Test early, informally
Formal testing at an accredited lab is the last step, not the first. Before then:
- Scan early prototypes with a near-field probe and a spectrum analyzer to find the noisiest areas.
- Book a few hours of pre-compliance time at a test lab once the design is close. It costs a fraction of formal testing and shows where you stand.
- Try ESD on the bench too. A rented or borrowed ESD simulator finds weak connectors and resets long before the lab does.
- Fix what you find, freeze the design, then go to formal testing. Book the lab well ahead; schedules fill up.
After certification, changes to the radio, the antenna, the layout around them, or transmit power can mean retesting. Route every change past whoever owns compliance.
Paperwork is part of the product
Certification also needs documents: test reports, a technical file, a Declaration of Conformity, labels with the right marks and IDs, and required statements in the user manual. Collecting these as the project goes is much easier than reconstructing them at the end. From prototype to production shows where certification fits among the builds. Before you book a lab, the compliance readiness checklist covers what to have ready.
The short version
Decide where you will sell before you design. Choose between a radio module and a chip-down design early, and design the RF for certification either way. Lay out the board with emissions and immunity in mind. Leave room for fixes. Settle safety and the power supply approach early. Test informally before you test formally.
If you are planning a product for sale in several markets and want help mapping the path to certification, get in touch.