3 Common Mistakes When Designing Phased Arrays... And How to Avoid Them

Blog Post created by annamccowan Employee on Feb 13, 2018

Phased Array Systems have been around for decades, mostly confined to the aerospace industry; but with 5G development underway, phased arrays are becoming more common and in demand. In order to successfully design and deploy a product the first time, engineers should know how to avoid costly mistakes by using new techniques and simulation methodologies.


Mistake 1: Not predicting the far field spurious emissions in the simulation.

Whether you have an Aerospace/Defense or a commercial communications system, it must pass the test of a spurious emission mask (SEM). The masks are specific to each application, but the requirement remains the same. In the case of phased array, due to the added spatial dimension, the SEM test is more elaborate and is conducted in an anechoic chamber. The SEM test is conducted over the entire sphere (4π solid angle) for all the desired beam directions in both azimuth and elevation.


This very laborious procedure will be repeated if the spurious emissions are found the first time in the chamber and then have to be corrected in the design and brought back to the chamber. Therefore, if one can predict them upfront in the design cycle, the time spent in the anechoic chamber can be greatly reduced.


Figure 1. Predicting the desired beam directions up front in the design cycle, the time spent in the anechoic chamber can be greatly reduced.


This video goes into detail about how these spurious emissions can be predicted.


Mistake 2: Failing to explore the design thoroughly in the simulation phase.

Why would anybody not explore the design and simulation space?  There can be many reasons, but most likely it's due to the simulation speed and the accuracy of the modeling tool. One troubling behavior of phased arrays is coupling between the elements. The cost of a phased array is directly proportional to the size of the array. It is tempting to reduce the inter-element spacing, but unfortunately, that leads to increased coupling between the elements.


The coupling can happen in a few ways. If the lines in the feed network are close, they get coupled. An element not only transmits but can also potentially receive the energy from the adjacent radiating elements. This appears as the reflected energy back into the element’s input.


A third mechanism can also happen; if the elements are realized on the same substrate, higher order surface modes can be excited, propagated, and ultimately radiated. All these effects might cause loss of directivity in certain directions called blind angles. Unless one models the coupling effects and impedance mismatches accurately and explores the design over all the scan angles, these blind spots cannot be uncovered. A typical simulation shows a mild form of loss of directivity shown in the figure below.


Figure 2. Antenna coupling can cause loss of directivity in certain directions, called blind angles.


Mistake 3: Relying on simple spreadsheet calculations.

It is very popular to use spreadsheets to design RF systems. While it is true that they are readily available and quick to simulate, spreadsheets lack the capability to model and simulate. They cannot model multi-ports, RF mismatches, finite isolation, frequency response, accurate non-linearity, collated or uncollated noise, etc.


These common limitations become severe limitations when you start designing phased arrays. You need insights into all the paths, and phased arrays can have up to 400. You need to look into all the 400 paths to understand the behavior; because in a phased array, you cannot simulate one page and scale it up to 400 paths. You need to consider all of them simultaneously to achieve accurate results.


Figure 3. Phased array system design is much more complex than a single path RF system design. Designers cannot scale up the analysis of a single RF path analysis to a full array.


Under certain conditions, the amplifiers in the array are compressed. Due to this compression, the spurious radiation is violating the SEM as shown in the figure below.


Figure 4. SystemVue helps designers catch spurious radiation that violates the spurious emission mask, and can identify which amplifiers are driven into compression or saturation.


That's because some of the amplifiers in some of the chains are being driven into compression or even saturation. So how do you identify these amplifiers? Modern tools, such as SystemVue, make it easy to identify them and understand where the non-linearity is coming from.



Modern design simulation and modeling technologies will make it easy for engineers to avoid these costly mistakes. Watch Keysight’s latest video, How to Avoid Costly Mistakes in Designing Phased Array Systems, to receive greater insights on how to work around these common errors and download the workspace he uses in the video.




 Start your free trial of SystemVue today!