Close-up of a dark blue circuit board with a chip at its centre
5 min read

Module or Chip-Down? Choosing a Wireless Strategy for Your Hardware Product

Author:

Richard Beranek

Almost every connected product we work on reaches the same fork in the road. The firmware is coming together, the enclosure is taking shape, and someone looks at the bill of materials and asks: why are we paying for a module when the chip inside it costs less on its own?

It’s a fair question. A pre-certified wireless module bundles the radio chip, flash memory, crystal, RF matching network and antenna into one shielded part. Going “chip-down” means placing that radio chip directly on your own board and designing everything around it yourself. On paper the saving per unit looks obvious. In practice, the decision is about much more than the BOM.


What a module actually gives you

The most valuable thing a module brings isn’t on the BOM at all: it’s certification. Modules from established vendors come with modular radio approvals, so the radio portion of your product is already covered. You still need EMC and safety testing on the finished device, but you skip radio testing for each radio on the board.

You also inherit a proven RF design. The antenna, matching network and layout have already been tuned by the vendor, so your product gets known, repeatable wireless performance without an RF engineer on the team.


What going chip-down really involves

Placing the bare chip is the easy part. Around it, you now own:

  • External flash memory, a crystal and supporting passives, each sourced, placed and verified.

  • An RF matching network and antenna, laid out with controlled impedance and tuned on your actual board, inside your actual enclosure.

  • Radio certification for every radio on the board. Wi-Fi plus Bluetooth, or dual-band Wi-Fi, means more test scope and more cost.

  • More engineering time and more risk at every board revision, since any change near the antenna can shift RF performance.

Once you add those parts back, the per-unit saving shrinks. Module pricing also benefits from the vendor’s huge volumes, so beating it with discrete parts is harder than it looks.


Where the break-even sits

Chip-down trades a lower unit cost for a one-time bill: certification, RF design and tuning, and extra validation. You only come out ahead once you’ve shipped enough units for the small per-board saving to pay that bill back.

A common rule of thumb puts the break-even around 10,000 units. For multi-radio, multi-antenna products, we often see it land at two to three times that. If your first production runs are in the hundreds or low thousands, the module is almost always the cheaper product, even though it’s the more expensive part.


So which should you start with?

Your volume decides.

If you know you’ll hit those volumes, start chip-down. You pay for certification and RF work once, keep the savings on every unit after that, and avoid redesigning the board later.

If you’re not sure yet, start with the module. You get to market faster, with less risk and less up-front spend, and you learn what your real volumes look like. If they take off, moving to chip-down means a new PCB revision and a new round of radio certification, not a full product redesign. The enclosure, firmware architecture and user experience largely carry over.

To keep that door open, plan the board with the switch in mind. Keep the area around the module and antenna clear, and pick a module built on a chip you could later place directly. A future chip-down revision then becomes a focused change rather than a re-layout.


The takeaway

Don’t optimize the BOM before you have the volume to justify it. Match your wireless approach to the production volume you can actually commit to, and design the board so you can change your mind later.

If you’re weighing this decision for your own product, the Brash team is happy to walk through the numbers with you.