Developing hardware products is expensive and time-consuming. With wireless products, failing to select the right antenna solution can drive back moths in development if the issue is found too late. In our earlier blog post, we discussed the importance of antenna performance and how antenna measurements are critical to R&D. To run measurements, there needs to be a prototype to measure. This means that an antenna solution has been selected, and a considerable amount of money might have already been spent on creating the prototypes. Unfortunately, in many cases, the selected antenna solution does not work according to expectations. As a result, a large amount of R&D investment become wasted. In order to avoid this risk, there is a solution that removes the need to gamble with your precious R&D investments: antenna simulations.
In this blog post, we are going to showcase two things. First, we will go through how simulation-driven design helps you to reduce R&D costs, followed by a demonstration about the accuracy of antenna simulations.
Simulation-driven antenna design process
The fundamental logic behind the simulation-driven design is the ability to verify all design choices with simulations while product design matures. The antenna design process is divided into four phases: Pre-Study, Concepting, Integration and Verification, all of which are done in close collaboration with the rest of the product development team.
The process starts with first ideas of the product before any mechanical model exists. This phase is usually called pre-study or pre-concepting. What is needed from the product management are the desired size of the device, use case, PCB size, materials and target specification. Antenna engineers will run sets of simulations using simplified models to give input on what is possible and what is not. This information will help the product management to create better specification and business case for the product.
Once the product has target specification, the concepting phase starts. This is when different antenna solutions are tested and developed using simulations with the latest CAD models available from mechanical engineers. Concepting phase continues until one antenna concept has been selected. This requires the maturity or readiness of the product design, meaning that the possible remaining changes of the product would not affect the selection of the antenna concept.
After the antenna concept has been selected, it is optimized for the current product design. Series of simulations are done to test possible effects of mechanic and PCB layout changes. Simulations help to identify unwanted resonances that are either on useful frequencies or have a risk of tuning in the final product. When anticipating these risks early, fixes can be prepared in advance in case they become a problem later. Ground connections and floating can be handled with the help of EMC specialists.
This phase ends with extensive simulation round called virtual antenna testing. Virtual testing is done for a product design that is ready to be released for prototyping. The design is modeled in an extremely detailed way to give a result as accurate as possible.
No matter how accurate the simulations are, verification by measurements is always needed to ensure all the inputs for the simulation such as materials and component models have been correct. It is also required to rule out any human errors in simulation. See more on the blog post where we discuss antenna measurements in more detail.
Comparison between antenna simulation and measurement
Simulation accuracy is a critical foundation of simulation-driven design. Next, we are going to demonstrate the accuracy of antenna simulations provided by Radientum using a commercial chip antenna. The chip is multiband LTE antenna covering frequencies 698-960 MHz and 1710-2690 MHz.
The chip antenna has been modeled and used according to the reference design provided by the antenna supplier.
Picture 1 shows the chip antenna at its reference board in simulation tool. Model details are hidden due to confidentiality. Measurement has been done for the reference board bought from the chip manufacturer. Passive measurement was carried out in an anechoic chamber at Radientum laboratory.
Figure 1 below has both simulated and measured matching levels (S11) while figure 2 shows total efficiency for both cases. In both figures, LTE bands are highlighted with horizontal black lines while light curve shows simulated results and dark curve shows the measured results.
As seen in Figure 1, the two curves match very well with only minor differences. In figure 2, the total antenna efficiency of the simulated result and measured result differs slightly more due to extra losses in real-life measurements. This fact needs to be considered when designing the antenna.
As demonstrated above, professional antenna simulations can predict the final performance with very high accuracy. This was shown with commercial chip antenna as an example but is true for all antennas whether they are off-the-shelf or custom; PCB trace or Chip; Flex or Ceramic.
Simulation-driven antenna design that has been in the past a privilege of multinational corporations and antenna manufacturers is now available for companies and projects of all sizes. Contact our sales team to discuss how antenna simulation can help you!
Disclaimer: The views and opinions expressed in this article are those of the author. It is intended only as a sharing of antenna design knowledge for educational purpose.