There are lots of hot topics in RF and wireless: 5G, for sure, plus others such a nanocells and MIMO, to cite just a few. Some of these are truly new and didn’t exist or were only a theory until a few years ago; others were better known by other names such as “diversity antennas” which had the concepts of MIMO but was more “static” and generally lacked the dynamic aspects of today’s MIMO.
But there’s one other term that’s getting a lot more attention these days: passive intermodulation, or simply PIM. What is PIM? In brief, it’s a materials-induced nonlinearity in the RF signal path. To quote a Keysight applications note (Reference 1), “PIM is an unwanted signal created by the mixing of two or more RF signals, caused by nonlinearity of the passive components in the RF path such as antennas, cables or connectors. PIM product is the result of high power tones mixing induced by ferromagnetic materials, junctions of dissimilar metals, metal-oxide junctions, contaminated junctions and loose RF connectors.” In simple form, PIM is a manifestation of the classic non-linear mixing situation which is so useful (in most cases) in electronic systems. PIM occurs when two or more carrier signal frequencies are exposed to non-linear mixing, and the resulting signal contains additional, unwanted frequencies which are intermodulation products, Figure 1.
The basics of frequency mixing and resultant intermodulation signals are well known to circuit and system designers. (Image source: Keysight Technologies)
Since PIM may create these intermodulation products at various points along the signal path, it may actually occur after filtering and so can be a major headache. The Keysight note is very clear on this, Figure 2: “Although filtering can reduce unwanted signals generated by power amplifiers in the transmitter path, PIM products from passive components such as antennas, cables or connectors in the RF signal path can not be filtered.” The References at the end delve into PIM with additional perspectives and views on its sources, impact, and implications.
This overview block diagram and frequency bands for communication systems shows how PIM-related signals can “creep into” the system. (Image source: Keysight Technologies)
PIM is causing all sorts of problems with wireless links in some bands, and it will likely become more of an issue as wireless pushes into the tens-of-GHz range of the electromagnetic spectrum. At these frequencies, the intermodulation products due to PIM affect nearby channels and degrade system operation, margins, and SNR. There’s also a cascade effect as the impact of PIM ripples thought the wired circuitry. Further, while PIM has tangible negative effects, its actual magnitude is small and can be hard to identify with instruments, making it difficult to trap and diagnose. An Anritsu note, Reference 2, states “On-site experiments have shown significant decreases in download speeds linked to slight increases in PIM. Drive tests have revealed an approximate 18% drop in download speed when residual PIM level was increased from –125 dBm to –105 dBm. The latter figure is hardly considered poor.” That’s a lot of degradation for a smallish PIM level.
There’s an aspect of historical irony here. PIM as a physical phenomenon is not new; only this manifestation and its consequences are new. The rectifying and non-linear effects of dissimilar metals and their physical interface have been long known as a potential source of problems. There are even anecdotal stories of people who could “receive” and actually hear AM radio stations via their teeth due to some sort of interaction between a filling and a tooth acting as the diode of in the classic crystal-radio design.
To minimize PIM, vendors of passive components such as cables and connectors are working to minimize the underlying sources of PIM to the extent possible, and also devising tests for it, which is not a trivial process. Philosophically, the PIM situation seems to be a way for these passive components to saying “hey, you usually don’t give me a lot of respect and consideration (that is, you don’t…at least at first) in your design, but I’ll show you that I really do matter…and I’ll be tough to analyze and model, too!”
If you look at the ads in publications which are focused on RF and microwave-system design – admittedly an imprecise metric of where the interests of the design audience lie – you’ll see that a very high fraction of them are for basic, cables, interconnects, and cable assemblies, usually citing their flatness, stability in impedance and phase-related performance despite temperature shifts, low PIM, matched performance, and other virtues. I think what the vendors in these ads are saying is “we help you avoid problems.” This in contrast to ads for active devices which tend to be about letting you do more: get more power, work across wider bandwidth, achieve more efficiency, do so with increased linearity, and so on. I think this tells you something about engineering reality.
Still, if you can’t count on your basic passive components to do you no harm, what can you count on? It just isn’t fair, you might say. But life (and design) aren’t necessarily fair, and PIM is another one of those things that designers of GHz-class products will have to get used to working with and minimizing through careful components selection and other techniques. As they say, “get used to it.”
Have you had PIM issues in any of your designs? Do you expect it is a now-latent issue that may yet hit you when you are not looking, or cause you to spend hours trying to track down elusive sources of performance degradation?
- Keysight Technologies, “Innovative Passive Intermodulation (PIM) and S-parameter Measurement Solution with the ENA”
- Anritsu, “Passive Intermodulation (PIM)”
- L-com/Infinite Electronics International, Inc., “Passive Intermodulation Explained: What is PIM?”
- Pasternack/Infinite Electronics International, Inc. “Passive Intermodulation (PIM) Explained”
- Analog Devices, “Passive Intermodulation (PIM) Effects in Base Stations: Understanding the Challenges and Solutions”
What Analog's 'Imperfections' Taught Me
RF Shielding: All or Nothing at All?
Sophisticated Cables: Not Just for RF