This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

CC1101: Multiband antenna, from 315 to 902MHz

Expert 3795 points
Part Number: CC1101
Other Parts Discussed in Thread: CC110L

Dear e2e community,

My customer needs to build a single hardware with one of our sub-G transceiver (CC1101/CC110L), to be able to address multiple bands (315, 433, 868 or 902MHz).

How could we deal with antenna design?

Is there a way through a single PCB or mounted antenna?

Should we use a RF switch?

There is no performance constraint here (data rate nor distance).

Regards,

  • - BOM: The frequency bands you list here require different match and filter components. Correct match is required to get best possible output power/ sensitivity and the various frequency bands require different filtering to pass regulations.
    - Antenna: Antennas typically have a bandwidth up to ~40 MHz and the physical size requirement is different on 315 MHz and 902 MHz. (to list min/ max)

    Is the requirement same PCB or same PCB/ BOM?
  • Hi TER,

    The initial requirement is same PCB/BOM.
    But I can easily understand the challenge of it ; so any suggestion is welcome if it's feasible only on the PCB aspect.

    Regards,
  • If this is just one PCB then it is straight forward but with one BOM it is difficult.

    With DN024 antenna and a dual-band match, 433 MHz, 868-930 MHz can be done. However, 315 MHz will have to be done with another antenna resonance. 

    What is the available size for the antenna ?

    //Richard

  • Hi Richard,

    Thanks for your reply.

    For information, we want to add this RF function to an existing product that doesn't include this function yet. We are studying what size is needed to add the CC1101, the matching network and the antenna(s). The current product is already very integrated (external dimensions are around 50 x 50 x 25mm). So this new function shall be as small as possible (and mainly the antennas). In rough estimation, the dimensions shall not exceed 45 x 38 mm (PCB size with RF hardware + antennas).

    The PCB antenna proposed in the DN024 is too big for our design. However, we don't need a very efficient antenna as our device will be very close to the other RF equipment (couple of meters maximum). Is there any other possibility that match better our requirements?

    Yann
  • Hi,

    The smaller the antenna, the smaller the bandwidth becomes. Since you want to cover from 315 MHz to 902 MHz, this is difficult with a passive antenna structure even with multiple antenna resonances. Suggest to use a general antenna structure that is as large as possible with an active element to switch various antenna matching loads into the antenna match. 

    For electrical small antennas, the antenna bandwidth will be limited especially for the low frequency bands. This will not be a problem for the bandwidth required for the radio but more an issue for production tolerances, body effects etc.

    More information on the limits of the antenna bandwidth can be found in the Harold Wheeler paper.

    Regards,

       Richard

  • Hi Richard,

    Thanks for your support. We are in contact with antenna manufacturers that are working on a wide band antenna 315MHz - 902MHz. If they don't succeed, we will probably split the range into 2 bands (315/433 and 868/902).

    If they succeed to design only one antenna, the question will be the balun. According to the antenna manufacturer, designing a balun that covers 2 frequency octave is something possible. I am not familiar at all with this; how can I calculate the balun by myself? Is there any app note to do it?

    Best Regards,
    Yann
  • Hi,

    One issue with making one design to cover all markets apart from the antenna is the harmonic attenuation. A wide-band balun can be added but the harmonics have to be attenuated and this is tricky when operating at 315 - 433 MHz and the 2nd/3rd harmonics are within the pass-band of the balun.

    There are capacitor switch banks available to vary the capacitance in the balun and LPF. Another method (cheaper) could also include a switch directly after the wide-band balun to switch in two different LPF networks. i.e. 433 (315) MHz LPF and 915 (868) MHz LPF.

    If you use two antennas to cover 315/433 MHz and 868/915 MHz then they can be connected directly to their LPF network. If you use a singular antenna then you will need a switch or diplexer after the LPF.

    Regards,

       Richard 

  • Hi Richard,

     

    Your proposal of using a wide band balun + 2 different LPF networks sounds good. I attached a schematic based on this idea. Could you please have a look on it and check if I well understood.

    Also, can you help to calculate the balun and the 2 x LPF?

    - Balun: L1 to L6 and C5 to C9 on the schematic

    - LPF1 (low band 315-433 MHz): C10, C11 and L7

    - LPF2 (high band 868-902 MHz): C12, C13 and L8

    Kind Regards,

    Yann

    PCB00142AA.pdf

  • Hi,

    L5, L6 and C9 should be placed after the diplexer since this is a part of the LPF.

    The balun network L1-L4, C5-C8 is not a wideband balun.

    This network should be replaced with a wideband balun. Like B0310J50100AHF from Anaren.

    We do not have a reference design with this particular balun so I would keep the component C6 as well in addition to B0310J50100AHF so that the impedance can be tuned for optimal performance.

    Regards,

       Richard

  • Hi Richard,

    I updated the schematic following your advices.

    I don’t know exactly where to place C6. I placed it before the B0310J50100AHF. Can you check? I believed the capacitor C6 were a part of the balun.

    Regards,

    Yann

    PCB00142AA_updated.pdf

  • Hi,
    C6 in the latest schematic is just used as a footprint holder to possibly improve the matching to the wideband balun. Leave this as DNM for the first build and test with various values to determine if a better match can be achieved.
    The general concept of the schematic looks OK.
    Regards,
    Richard