Passive Intermodulation (PIM) has been studied in great detail for more than 75 years, so it’s reasonable to assume it should no longer be a major concern. Unfortunately, PIM is as big a problem today as it has always been. Just ask someone who has faced its effects. This might seem odd considering that test equipment to find the source of PIM and the means for its mitigation have been known for years. But as wireless communications enter a new era, it brings more opportunities for PIM to appear and greater difficulties in dealing with it. To understand why, it’s necessary to delve into the domain of intermodulation distortion (IMD) in general and PIM in particular.
PIM is a complex phenomenon and unique among the many types of IMD because it can arise from an astounding variety of sources, many of which are well beyond a designer’s ability to mitigate them. In fact, many times they have nothing whatsoever to do with the equipment itself but rather how it interacts with its environment.
IMD can only be created by active (non-linear) devices such as semiconductors that are powered and can generate and amplify an input signal. In contrast, passive (linear) devices such as cables, connectors, and many other common RF and microwave components can modify signals, but they cannot generate or amplify them, so theoretically a passive device should not be able to generate PIM.
But under certain circumstances, it can, as a result of the interaction of dissimilar metals, oxidation, damaged or improperly torqued connections, fatigue, cold solder joints, corrosion, and other factors. This, in effect, allows the device to become active in the sense that it can generate signals, and when these signals (harmonics and spurious emissions) fall within the passband of a receiver, they will desensitize it, reducing its ability to discriminate between signals and noise, and in the worst cases, render it unable to properly function.
Making matters worse is the fact that passive RF and microwave devices are not the only sources of PIM. Other sources include the presence of metals beyond the device and perhaps even outside the cabinet or equipment shelter (Figure 1), and electro-thermal conductivity modulation that appears in resonant structures such as filters and antennas that is different from other types of PIM because it does not require dissimilar metals, junctions, or contacts. If all this sounds more than a little disturbing, consider that at the microscopic level, researchers are still delving deeper into the causes of this deceptively complex phenomenon, some of which remain unknown.
PIM arises when two or more signals of relatively high levels of RF energy mix together and, like a mixer, produce sum and difference signals that are related to the fundamental (desired) signal. They are intermodulation products that fall into orders (that represent the number of times the IMD product is multiplied. The most important of these are third-order IMD products that are closest to the fundamental signal and the strongest, while higher orders are weaker but, in some cases, can still present problems.
That said, as the orders become higher, the signal can become wider, so the orders present in a 50 MHz signal, for example, can cause interference over a bandwidth of up to 150 MHz. As channel bandwidths increase to hundreds of megahertz or more, there are thus more opportunities for emissions of the fifth- and seventh-order to wreak havoc.
When PIM occurs beyond the device itself, it becomes a vexing problem because it opens the door to an almost unlimited number of unlikely sources with dirty connections or corroded parts ranging from vent pipes to fences, metal roof panels, rusty screws and bolts, and so on. In such an uncontrolled and changing environment, it may not even be possible to find the PIM source because it is located beyond the reach (or jurisdiction) of the tower owner or operator.
But, it gets worse because PIM can occur in corroded or rusty metal a remarkable distance away from a receiver, but can still able to degrade it depending on the strength of the intended signal’s radiated RF energy. It should be noted, however, that while it is generally assumed that only the mixing of strong signals can produce objectionable levels of PIM, studies have shown that in certain circumstances signals of much lower signal strength can also cause it. In addition, a directional antenna increases effective radiated power by tens of decibels, so even a relatively low-power transmitter when combined with a Yagi, reflector, or another type of directional radiator can deliver enough energy to enable the mechanism that causes PIM.
Summing this up, while the ideal recourse in the case of a component would be to replace it or rectify the problem by creating better connections and removing corrosion and other contaminants, this is not a task for the faint of heart. A typical base station and its tower have hundreds of potential offenders, like cables, connectors, power dividers, duplexers and diplexers, and others, one or more of which can cause the problem. And in cases where PIM is created by corrosion or rust in metals that are unrelated to the system itself, the source can often be extremely difficult and even impossible to find.
The problem of PIM has become so severe in the last 15 years that all major manufacturers of test equipment and several others have created comprehensive portable instruments that can detect and locate the sources of PIM, and in some cases analyze them. These instruments use one of several methods for this purpose, the latest of which is spectral analysis over CPRI, in which analysis of I/Q data on the fiber interface between the baseband unit and remote radio head makes it possible to identify the characteristics of PIM.
This method has the advantage of being frequency agnostic and can be performed either at a base station or virtually anywhere via the CPRI link and can also measure individual carrier signals without affecting operation. However, none of these measurement approaches are perfect. For example, CPRI and noise-rise monitoring work best when PIM signals are strong and still require other tests to confirm their results. The much more common RF-based approach requires the instrument to be employed at the base station and at various places on the tower, although it can precisely determine where the source is located.
PIM and Antennas
While most of the attention to antennas focuses on the elaborate phased array and other approaches required to make use of the millimeterWave spectrum, the vast majority of antennas today are employed for traditional applications such as land mobile and public safety radio, SCADA links, distributed antenna systems (DAS), residential and commercial satellite uplinks, and receiving broadband Internet access. Without careful construction techniques and the use of proper materials, these antennas along with their mounting hardware and supporting structures can cause PIM to appear.
Fortunately, antenna manufacturers are well aware of PIM, and they avoid the use of materials such as ferrites, nickel and nickel plating, and some steels that when exposed to reversing magnetic fields generate PIM. A good example of such antennas is the HG74204UPR-NF omnidirectional antenna from L-com that is designed to minimize PIM and is fabricated using only those materials that are unlikely to cause its effects (Figure 2). It can be specified to cover 698 to 960 MHz, 1710 to 2700 MHz, and 3400 to 4200 MHz with vertical polarization and PIM performance of at least -150 dBc. The antenna can be deployed either indoors or outdoors and delivers up to 4 dBi of gain with VSWR of 1.8:1 and power handling ability of 50 W.
The antenna elements are only one part of the equation because they’re connected to cables via connectors, so the entire assembly must be considered rather than simply the antenna itself. The point at which the connector on the antenna attaches to the cable is a likely PIM contributor because it is subject to the vagaries of hostile operating environments along with movement caused by winds, icing, and often the effects of salt spray. As a result, the center and outer conductors of cables are typically soldered rather than connected by other methods and jacketing is typically copper rather than braid because the latter is more susceptible to PIM.
And unlike components inside an enclosure, antennas in service for many years will experience the effects of aging. While it was once common to examine these components only when a reduction in performance had occurred or by periodic inspection, densely populated co-located sites make this approach no longer acceptable. This, in combination with the higher-order modulation techniques that deliver extremely high data rates and are intolerant of interference, support the need for more strenuous evaluation.
Perhaps the most important date in the life of an antenna is when it is initially deployed, when cleanliness and secure connections are essential, because even the tiniest of metal particles, scratches, and rough handling will, if not immediately, eventually create the conditions conducive to the generation of PIM.
Anyone charged with the responsibility of procuring antennas must rely on specifications on the datasheet and equally important, the reputation of the manufacturer. The latter becomes even more valuable because it is generally impractical for the wireless carrier, tower owner, or installer to verify that the specifications provided are accurate by evaluating it in an anechoic chamber. And even then, these measurements do not take into consideration what the antenna will experience in service, so although manufacturers must perform these tests, it’s questionable whether they offer much to those who use them.
Overcoming the perils of PIM has been a tremendous challenge for many years, but today it is more difficult because of the vast changes that are taking place not only to deploy 5G but also to bolster the existing 4G infrastructure. And while this article covered many aspects of this challenge, the mechanism by which PIM is created in many other highly technical aspects requires greater knowledge. Fortunately, there are many sources available on the Web that cover everything from basic principles to testing that will be invaluable to anyone involved with reducing PIM in components as well as their installation on towers and other antenna locations, and the effects of seemingly innocuous metal structures.