Why Antenna Placement on PCBs is Critical for Chip Antenna Radiation Performance
Chip antennas are literally a good fit for size-constrained devices such as wearables and IoT sensors. For example, the Taoglas WLA.04 is just 1.6×0.8 mm, making it ideal for compact devices that need 2.4 GHz connectivity such as BLE, Wi-Fi and ZigBee.
The smaller the device, the smaller the ground plane it can provide. That’s true for every type of antenna, but it’s particularly important to keep in mind when designing PCBs that will have a chip antenna. The PCB ground plane directly affects the chip antenna’s performance and thus the performance of all the applications running on the device.
This relationship highlights why device OEMs should take a holistic view and choose the chip antenna early on in the design process. This enables the designers to use the antenna’s integration guide to reserve the recommended board location. For example, the Taoglas WLA.04’s integration guide recommends placing it mid-point on the long side of the PCB with a minimum ground plane of 90×50 mm.
This approach ensures that the chip antenna will be in the ideal location. Selecting it later on runs the risk of having to redesign the PCB because the only place left for the antenna turns out to be the worst location in terms of ground plane and performance. Shuffling processors, batteries and other components to make room for the antenna can delay the device’s time to market and drive up development costs.
Read on to learn about current excitation and the fundamental role it plays in both chip antenna performance and PCB design.
Small Antennas are Particularly Sensitive About Board Location
Chip antennas are embedded on a high-permittivity substrate (typically ceramic), which can be installed on a PCB, similar to an SMD component. They’re available for a wide variety of bands and technologies, such as Wi-Fi and cellular.
It’s generally recommended to install chip antennas either in the middle of the longer side or in the corner of the PCB, as in the picture below.
Small antennas have limited impedance bandwidth and efficiency and thus are highly sensitive to their placement. If the chip antenna is not installed in the proper location on a PCB, its radiation properties can be greatly reduced. That’s because the PCB boosts the chip antenna’s radiation because currents are excited on the PCB ground plane. If these currents are not properly induced, this boost is not possible.
To understand this effect, it is important to know a bit of antenna theory. The size limitation affects the antenna’s radiation properties and sets a trade-off between size and performance. Radiation efficiency, impedance matching, and frequency bandwidth are the main attributes studied to establish their fundamental limitations. Antenna designers must consider these limitations and achieve a compromise between the dimension of the antenna and its performance.
Electrically small antennas (ESAs) get their name from how they compare to the wavelength (λ) at the frequency bands where they operate. By definition, they satisfy k*a < 1 (where k is the wave number (k=2π/λ) and a is the smallest radius of a sphere containing the antenna).
The following figures illustrate a method for determining the radius (a) of the smallest enclosing sphere. Figure a depicts the calculation of the radius of the sphere of an antenna without a ground plane. Secondly, an antenna installed in a finite ground plane is described in Figure b, and the radius is calculated considering the whole ground plane when it is smaller than λ/4 or the antenna is closer than λ/4 of an edge, which is most of the cases. Finally, when the antenna is located in an infinite ground plane, the theory of images is applied, and the antenna has a size of double its volume, so the radius is duplicated (Figure c).
In most integrated solutions, the ESA is located close to any of the borders, and the whole ground plane should be considered for the calculation of the radius a. This aspect is critical and highlights why chip antennas always lean on the PCB ground plane.
The PCB ground plane highly contributes to the radiated power because the antenna induces currents that flow through it. It can be reinterpreted as an asymmetric dipole where one leg is the chip antenna and the other leg is the PCB ground plane; thus, the placement of the chip antenna on it is crucial.
Understanding Modes
To further understand the radiation properties of a ground plane, we consider the characteristics modes analysis (CMA). The characteristic modes correspond to the natural resonances (or modes) of an arbitrary conductive body analysed in absence of any excitation. After analysing the conductive body resonances, the type and location of the feeding can be properly chosen to excite the desired modes of the object, which in this case is a rectangular PCB ground plane. The goal is to use the chip antenna to excite this PCB current modes and boost the whole setup radiation properties.
There are infinite modes when a structure is analysed, but the ones that are more interesting to excite are the first ones. These resonate at the lowest frequency, like Mode J1, which resonates at a frequency when the longer side of the PCB is λ/2.
In the figure below, the current distribution of the first six modes of a rectangular ground plane is depicted. The closer is the mode to resonance (or has already resonated), the easier it is to excite, and thus is why chip antennas show better performance when installed on bigger ground planes.
To excite these modes, coupling elements must be used — in this case, the chip antennas. There are two kinds of coupling elements due to their different natures: capacitive coupling element (CCE) and inductive coupling element (ICE). The following figure is an example of an ICE and a CCE installed in a 100×150 mm ground plane.
CCEs are resonant or non-resonant elements located close to the platform to create a capacitive effect, as seen in the figure above. They must be located where the electric field of the characteristic mode exhibits a maximum (minimum of current) to excite the mode. CCEs are generally placed on the corner of the metallic platforms due to the number of modes that exhibit a current minimum. From the figure above, if the CCE is placed in the corner of the metallic plate, the first six modes of the metallic plate could be excited.
ICEs are resonant or non-resonant elements that couple via a magnetic field (inductively). Hence, they must be placed where the characteristic mode exhibits a current maximum. They are generally based on loop elements, as seen in Fig. 4. In contrast to CCEs, ICEs are usually placed in the middle of the sides of a metallic structure. Modes J1 and J2 would be mainly excited. Modes J4 and J5 may also be excited if the ICE was moved slightly to the left or right.
With these two kinds of excitations, the ground plane modes are excited, thus enhancing the radiation properties of the antennas. The reason behind the location of the chip antenna depends on whether it is CCE or ICE. Some antennas are recommended to be placed in the corner and others in the middle of the longer side of the PCB.
