Do I Need to Send Accessories and Cables with My Test Samples?

← All guides

Sample boxes arrive incomplete more often than anyone expects. The main unit, wrapped carefully in foam, and nothing else inside. No power adapter, no charging cable, no antenna. A few days later the laboratory emails to ask for them, the project stalls, and someone ships a second parcel pulled off whatever shelf was closest.

Sometimes the opposite happens: the sample arrives with an adapter nobody intends to put in the box, and the laboratory tests whatever was in the parcel. Both cases end the same way. The report describes a configuration you will not be selling, and the gap surfaces months later, in front of a buyer. So the short answer to the question is: send everything that is electrically part of how the product operates, and write down in one page what you are deliberately not sending.

Why accessories count as part of the product

The report you receive is not a report about "a product". It is a report about one tested configuration: a defined unit, a defined firmware, fed from a defined supply, connected with defined cables, operated in defined modes. Every item in that list is part of the evidence. Change one item and the evidence stops describing what you sell.

That is why the accessories matter far more than their price suggests. In EMC work the power adapter is frequently the loudest object in the whole set-up. The cable attached to the product behaves like an antenna. In radio testing the antenna is not an accessory at all - it is the radiating element being measured. In safety work the power supply is part of the electrical assembly being assessed. The item you left on the shelf is part of the measurement whether or not it goes in the box.

Five places where the accessory changes the result

• Conducted and radiated emissions. Mains-port conducted emissions are measured on the power line feeding the product, which is why the CISPR-based multimedia standard family (CISPR 32 / EN 55032) treats the supply as part of the test arrangement. Switching adapters generate broadband noise, and two adapters with the same 12 V output but different internal topology will not put the same spectrum on that line. A substitution can move the result by more than the margin you were counting on.
• Cable behaviour. The standards that govern the set-up expect interconnecting cables to be arranged as they are used in practice, and where the manufacturer supplies no information the laboratory falls back to a default exposed length in the region of one metre. Length, shielding and routing decide where the cable resonances land. A one-metre bench cable and a 1.5 m unshielded production cable are not the same antenna, and radiated emissions will not be the same number.
• Immunity. ESD (IEC 61000-4-2) is applied to accessible conductive parts, while burst, surge and conducted RF (IEC 61000-4-4, -4-5 and -4-6) are injected at the ports - including the DC input and any signal cable. An unknown adapter or a different cable changes what the test actually injects into your electronics.
• Radio. For an intentional radiator, radiated field strength and output power are measured with the antenna the product will use, and an FCC grant names the antenna type and gain. FCC rules generally require the device to be designed so that no antenna other than the one furnished by the responsible party can be used, with limited exceptions such as professional installation or a standard antenna connector - and moving to a higher-gain antenna after testing is not a cosmetic change. On the European side the same logic runs through the RED assessment of the radio interface.
• Safety. A 12 V DC product can sit below the Low Voltage Directive thresholds of 50 to 1000 V AC and 75 to 1500 V DC on its own - but the mains adapter in front of it does not. When the adapter ships with the product, its construction, its thermal behaviour and its fault behaviour sit inside the assessment of the combination.

What to send, and what you can leave out

The list is shorter than most people assume, and it is not "everything in the retail box".

Accessory	Send with samples?	Why it changes the result
AC/DC adapter shipped in the box - every variant	Yes	Defines the conducted emissions test condition, and is part of the safety evaluation of the combination.
Mains lead / plug type per market	Yes, one per variant	Conducted emissions, burst and surge are all applied at the mains port you actually ship.
Charging or data cable included in the box	Yes	Length and shielding shift radiated emissions; it is how the product is used, so it belongs in the set-up.
External antenna	Yes, production type and gain	It is the measured radiating element; the report and the FCC grant apply to that antenna.
Wireless remote, dongle or paired handset	Yes	It is a radio in its own right and becomes part of the emissions arrangement.
Rechargeable battery or user-replaceable pack	Yes, the production cell or pack model	Charging and safety evaluation follow the actual cell; transport documents name the model too.
Cable or adapter you also sell separately	Yes if it contains active electronics	An active accessory is a product in its own right; a plain passive cable normally is not.
Manual, quick-start card, screws, bracket, polishing cloth	No	Not electrically relevant. The manual travels as a document, not as a sample.

One row deserves a second look. A laboratory bench adapter is not a neutral stand-in for a power supply, because it is usually a general-purpose unit with a different switching frequency and a different input filter from the one your supplier will use in volume. If your product needs a supply to run at all and you send none, the lab has three options: use something from its own shelf, ask you to send the right one, or test the product on a DC source and document that. Only the second option produces evidence about your product.

What a missing or substituted accessory actually costs

Four situations that come back often enough to be worth planning around.

The adapter in the box is not the adapter in the report. A project sent samples powered by whatever adapter was in the office that week. Emissions passed with margin, the schedule held, and the BOM later settled on a different supplier's adapter with the same 12 V output and an entirely different internal design. The report listed the office adapter in the auxiliary equipment table, and the buyer's engineer found it on the first read. The fix at that point is a delta evaluation on the new supply - which is cheaper than the conversation about whether the original report was ever valid, but not cheaper than sending the right adapter the first time.

The cable you did not send was doing the radiating. In one repeat case the production cable was 1.5 m and unshielded, while the laboratory's default bench cable was a one-metre shielded assembly. Radiated emissions from a cable-led product shift with exactly those parameters. The outcome is expensive in both directions: either you pass with a cable you do not sell, or you pass in the chamber and fail in the field with the cable you do sell. Sending the production cable costs a few dollars of freight.

The antenna in the photograph is not the antenna in the box. A product with an external antenna arrived with the connector but without the antenna itself, and the laboratory measured radio performance on its own reference antenna. The numbers looked healthy. What they describe, though, is the reference antenna - and when the buyer asks which antenna the report covers, or which gain the radio was approved at, there is no answer in the document. For an intentional radiator this is not a detail; the antenna is the configuration.

The battery never arrived, so the charging circuit was never exercised. A product with a user-replaceable pack was sent as a bare unit and powered from a bench DC supply. The measurement work was fine, but the intended charging behaviour never ran under its real source. When the pack supplier later changed the protection circuit, there was no baseline measurement to compare against, and no way to argue the change was harmless.

There is a newer version of the same mistake. Since 28 December 2024 the common charger requirements of Directive (EU) 2022/2380 have applied to a listed range of radio equipment - mobile phones, tablets, headphones, portable speakers, e-readers and similar - with laptops following from 28 April 2026. USB-C is the required charging receptacle, fast charging is harmonised, and the buyer must be offered the option of the device without a charger, with a pictogram and a charging-capability label carrying the message. If your plan is to ship without an adapter, the tested configuration should contain no adapter either. Sending one "so the lab can power it up" quietly writes an accessory into your technical file that the product will never be sold with.

When we would rather you waited than sent anything

Sometimes adding accessories to the parcel now makes the evidence worse, not better, because it locks in a configuration you cannot commit to.

• The adapter or cable supplier is still being negotiated. If the electrical specification is not frozen - even when the commercial order is not - the report will describe one supplier's adapter while you ship another. Freezing an adapter's part number before the bench slot costs nothing. Finding out afterwards costs a delta test, or a report nobody trusts. The same logic applies to any post-test change: what a component change means for an existing report is worth reading before you substitute rather than after.
• The accessory mix varies by market and the list is not final. Three variants, three mains leads, two cable lengths. Testing the wrong combination wastes bench time, and testing everything is perfectly reasonable - but the variant list has to exist before the slot is booked.
• The product cannot operate without an accessory you do not physically have. A laboratory can document only what it can power and run. If the only working adapter is sitting at the factory, the slot is wasted and the schedule slips by however long the courier takes.
• The accessory is someone else's finished product that you resell. An external power supply is a regulated product in its own right - the Ecodesign requirements for external power supplies (Regulation (EU) 2019/1782) sit alongside the LVD and EMC obligations that apply to it. The starting point is the adapter manufacturer's own evidence, not a test we run on a unit nobody will support afterwards.

Put the configuration on one page

The habit that prevents most of these problems is a one-page configuration list sent with the samples and referenced in the report. Nothing elaborate:

• Product model and variant list, with the firmware version frozen for the test.
• Adapter: manufacturer, part number, input and output rating - or the sentence "no adapter supplied".
• Cable: type, part number and length, for each shipped variant.
• Antenna: type, gain and part number, or "integrated".
• Battery: cell or pack model, where one is supplied.
• Anything sold separately that the sample depends on in order to run.

Without that page the laboratory has to make assumptions, and assumptions land in the report in the form of an auxiliary equipment list - which is precisely the section a buyer's engineer, or a market surveillance officer, reads first. The short version of this whole article: the supply, the cable and the antenna are part of the product, the manual and the mounting screws are not, and anything you deliberately leave out needs a sentence of explanation on paper. The document side of the same handover is covered in what a testing laboratory needs from you, and the accessory discussion starts there rather than after the parcel has landed.

Where we fit in

SZC-LAB runs EMC, radio, safety and battery testing for products that are usually shipped in from overseas, so we see this problem from both sides - the parcel that is missing something and the parcel that contains the wrong something. What we do about it is deliberately unglamorous. Before a project is booked we ask for the configuration list, compare it against what is physically going to arrive, and tell you what is missing while the courier can still fix it. Where an accessory genuinely cannot be obtained, we plan so the gap is documented rather than papered over, and the report states what was tested and what was not. We also keep the tested accessories together with the sample through the project, so that a retest or a delta evaluation runs against the same supply, cable and antenna - a substitution made for convenience in round two produces the same mismatch as a substitution made in round one. That is what turns a report into one that survives a buyer's review in month two instead of being questioned in a meeting. The relevant scope sits across our testing and certification services, and adapter work has its own page under charger and power adapter certification.