This post is older than 2 years and might not be relevant anymore
More Info: Consider searching for newer posts

nRF51822QFAA RF range achievable with discrete BALUN and SMD Chip Antenna 2450AT18B100

We are in the process of creating our 2nd pass boards with following components to achieve good RF range>20 meters. We are considering discrete BALUN components and SMD Chip Antenna 2450AT18B100. Has anyone used the above components and achieved a good range>20 meters? Ours is a 2 layer FR4 PCB. Of course we need to perform impedance matching to minimize RF losses and also optimum layout of RF section. Attached is the snapshot of our 2nd pass schematics. image description

  • First of all I would like to start by clarifying one thing. All the components from C15 to C18 are a part of the discrete balun/matching network for the nRF51822. The three components C17, L1 and C18 are also a part of the discrete balun and not a pi-network for the antenna. The only component in your schematic above here that could be used for tuning antenna is the C19. The matching network/balun has four tasks:

    • DC suppl to internal PA.
    • Balun (balanced to un-balanced output)
    • Impedance transformation
    • Filtering

    So the components C15, L3, L2, C16, C17, L1 and C18 all are a part of the matching network/balun and the output is 50 Ohm single ended where C18 meets L1 and C19. Then you would use C19 (and/or additional components to tune the antenna input impedance to match the 50 Ohm out of the balun.

    White paper on tuning matching network/balun. (not upated for nRF51, but principal is the same)

    White paper on tuning antenna.

    For your question related to the discrete balun and antenna. The discrete solution is what most customers use. We strongly recommend customers to follow our(Nordics) reference layout and schematic to ensure good performance with regard to output power and hence also range. At the same time we also want our customers to not have any issues with regulatory requirements around the world. So I don't see at as any problem achieving >20 meters range with the discrete implementation and the antenna you want to use, but it's important to keep antenna placement, ground plane and shape and tuning the antenna in mind when designing. Because one antenna might work really well in one design, it could still require some tuning to work well in a different design.

Related