What is passive intermodulation?
In RF systems, passive intermodulation occurs when 2 or more carriers mix at nonlinear points in passive parts, including connectors, cables, antennas and corroded joints. The resulting intermodulation products can fall inside a receive band, raising its noise floor and making weak uplink signals harder for the receiver to distinguish.
The handset hears the site, but the site struggles to hear the handset. That is how a strong downlink can hide a receive problem. When mixing products enter the uplink band, we examine receive sensitivity separately from downlink coverage, even if the coverage plot looks satisfactory. A radio configuration change may help the plot while leaving the physical source right where it was.
Component selection starts with a question we need the carrier inputs to answer: which carriers will operate together, and how will their energy reach the passive hardware? With those inputs, we can use the frequency plan to locate possible unwanted products, then trace the antenna positions and connections to see which nonlinear junctions could be excited and how their energy could return to a receiver.
How mixing creates an unwanted frequency
For two transmit frequencies, f1 and f2, familiar third-order products include 2f1 minus f2 and 2f2 minus f1. These expressions describe possible frequencies, not their measured amplitudes. A frequency coincidence tells us to investigate a receive-band conflict; it cannot tell us whether a particular connection will generate enough energy to impair service.
Take hypothetical tones at 900 MHz and 950 MHz. Working through the third-order arithmetic gives 850 MHz and 1000 MHz; these are teaching values, not a proposed carrier allocation. Now put a protected receiver at the first product's frequency. Neither original tone lies in its receive band, yet the combination warrants review because the mixing product does.
Because real carriers occupy bandwidth, their mixing products can spread across a range of frequencies. We check the affected receive channels across that range; a comparison of center frequencies alone can miss an overlap. The proposed operating configuration also needs to include other tenants' transmitters wherever they can illuminate the same passive structure.
Why a shared site changes the question
What is already transmitting at the site? We ask for that frequency inventory before evaluating a new tenant, because adding a band changes the combinations that can excite an antenna branch. A branch acceptable with the previous carriers may need another look. Equipment ownership tells us who should supply the inputs; the RF interaction still has to be followed across those ownership boundaries.
A neutral-host distributed antenna system makes that coordination especially important because participating carriers share infrastructure while retaining their own operating requirements. We ask who maintains the combined frequency plan and who can authorize a revision. Without that responsibility, a tenant change can bypass the analysis that supported the original design.
Where PIM shows up and why
RF current crosses a mechanical contact surface at a connector, where contamination or poor assembly can make the contact nonlinear. An outside inspection may show a clean connection and still leave a problem to investigate. Our specifications give the installing contractor connection-quality and assembly requirements to work from. Those same requirements give the reviewer a basis for checking the proposed work.
Sources inside the RF path
Follow a coaxial branch from its input to the antenna and there may be several sources to investigate: damaged cable, poorly fitted interfaces and the passive devices selected for that branch. We need component information for both insertion loss and PIM performance, since a low insertion-loss rating tells us about a different property and cannot answer the PIM question.
The substitute fits, but what supports its PIM rating? It may have a different rating, or a matching rating measured under different conditions. That is why we put the exact interface and operating band in the component schedule. Before installation, substitution review must also catch any adapter that would add a joint to the approved RF path.
Sources outside the intended signal path
Even metal with no cable connection to the network can be involved. A corroded joint near an antenna can generate products when transmitted energy reaches it, and those products may couple back into the antenna. Replacing feeder components addresses the connected path. On a rooftop or at an outdoor node, the investigation also has to reach the nearby metal illuminated by the antenna.
The research paper Physical Modelling and Cancellation of External Passive Intermodulation in FDD MIMO, arXiv:2407.21135v2, revised September 26, 2024, models this external mechanism through nonlinear objects illuminated by an antenna system. Its distinction between internal and external sources supports a practical review question: does the investigation include the surroundings, or only the connected RF branch?
A photograph of a corroded railing gives us a place to investigate, provided we also have the proposed antenna orientation. It takes the responsible test specialist's investigation to connect that suspected source to the observed receive-band response and establish whether a repair was effective; the photograph alone cannot establish that the railing caused the measured energy.
These four controls address different parts of the PIM mechanism. We keep all four in the review even when the first third-order calculation shows no obvious overlap, because that calculation alone leaves the component requirements and physical coupling paths unresolved.
| Control | Question to resolve | Design consequence |
|---|---|---|
| Frequency plan check | Which products overlap protected receive bands? | Review carrier combinations and permitted changes |
| Antenna isolation | How can energy reach a nonlinear source or receiver? | Evaluate position and coupling paths |
| Low-PIM parts | Do component ratings match the application? | Specify quality requirements and substitution review |
| Fewer joints | Which interfaces serve a necessary purpose? | Remove avoidable adapters while preserving access |
How passive intermodulation is measured
During a two-tone test, controlled transmit signals are applied and an unwanted intermodulation product is measured. The owner or carrier appoints a qualified RF test specialist to follow its accepted procedure. First, though, we need agreement on what the test covers. A component? An installed branch? Both reports may use the same units while describing different assemblies.
What IEC 62037 establishes
IEC 62037-1:2025, Passive RF and microwave devices, intermodulation level measurement, Part 1: General requirements and measuring methods, Edition 3.0, published February 26, 2025, establishes general methods using two transmitting signals. The owner or carrier sets the applicable acceptance limit and project test conditions; naming the standard alone does not define a complete acceptance requirement.
Under the 2025 edition, the report includes the maximum value measured over the test duration, with applied power kept within the device's capability. Frequency matters too: the edition notes the usefulness of swept or multiple fixed frequencies because PIM depends on frequency, so we look for the method in the report before comparing the result with another measurement.
IEC 62037 addresses measurement rather than long-term product reliability. Weathering or a later equipment change can affect an assembly after it passes, so we retain the frequencies, power and other test conditions with the result. A pass label on its own gives a future reviewer too little information to judge whether the earlier test still describes the installed configuration.
Read the units before comparing results
To compare reports, start with the units and their references. A result in dBm expresses measured power relative to 1 milliwatt; dBc expresses a level relative to a carrier reference. Even when the units match, we still need to reconcile different carrier powers or measurement settings before treating the figures as comparable evidence of PIM performance.
Which product was measured, and at which interface? The owner needs those identified alongside the tone frequencies and applied power. We then check the test direction: energy returning toward the source follows a different path from energy observed beyond the device. That distinction affects what the report can establish about the assembly.
Define the test boundary: Mark the connection point on the RF diagram before testing is assigned. A report attached to the wrong branch identifier cannot establish that the intended antenna path met the carrier's requirement.
Test leads and terminations can contribute residual PIM that obscures the device's response. We ask the owner to include setup verification in the procedure and retain it with the measurement, giving the specialist evidence to separate the installed system's response from anything introduced by the measurement arrangement itself.
On an operating site, the carrier controls the test authorization and any required service isolation. The appointed specialist determines how to execute the procedure safely within the equipment ratings. Our design discussion does not assign those field actions to a drawing reviewer or treat a frequency calculation as an installed-system measurement.
Design controls for multi-carrier sites and DAS
The combined carrier plan comes first in our review. From there, we work through the physical arrangement to find coupling paths we can change or unnecessary sources we can remove, before choosing components. Low-PIM parts address only part of this work. A limitation of our frequency-plan analysis is that it identifies potential products without establishing their measured amplitudes in the installed system. The shared frequency environment still determines which combinations deserve attention once the system is assembled.
Check the frequency plan across tenants
Each participating carrier supplies transmit and receive ranges and the configuration details we need to identify relevant combinations. We keep the approved operating case separate from proposed expansion cases so that, when a band is added later, the next reviewer can see exactly which inputs changed against the baseline used for the earlier design revision.
The carrier's criteria and band arrangement determine whether the calculation needs to include higher-order products. Evaluating only 2f1 minus f2 leaves those other combinations unresolved. Where the spectrum assignments are fixed, we have to carry the remaining risk into decisions about antenna arrangement and component requirements rather than assuming a frequency change will be available.
Our wireless engineering design coordination connects those RF assumptions to the site layout. For broader interference mechanisms beyond passive mixing, our RF engineering guide explains how interference review fits into wireless site design.
Evaluate antenna isolation at the actual position
How much energy couples between the relevant paths? Antenna isolation describes that relationship. More separation can help, but we need the operating bands and antenna arrangement to judge the effect at the site. Nearby reflective structures can alter the path. Distance alone therefore gives us neither a universal spacing rule nor assurance that a relocation will reduce the interaction.
Moving an antenna away from a suspected source may also put its intended users behind an obstruction. Before accepting the new position, we compare the expected isolation improvement with the coverage objective and check whether the antenna can be mounted there. The location has to work for service coverage and for the PIM concern that prompted the move.
For an existing site, macro-cell antenna upgrade design provides the wider context for an antenna replacement. Our PIM review should use the final arrangement, including any approved mounting change, so an earlier analysis does not silently remain attached to geometry that the construction drawings no longer show.
Specify low-PIM parts with comparable conditions
A reviewer needs to compare proposed parts against a stated PIM requirement and the test basis behind it; we put both in the component specification. With only a general low-PIM description, the reviewer still has to establish the rating's conditions, including whether the measurement covers the operating range required for this particular branch.
An alternate component can change a mating interface and leave the installer needing an adapter where the drawing shows a direct connection. We check that interface during submittal review, before the branch is assembled. The review then covers the RF consequences of the resulting connection arrangement and whether the substitute performs the nominal component function.
Review substitutions as assemblies: When an alternate component changes a connector interface, review the resulting connection arrangement. Keeping the nominal component function does not establish equivalent PIM performance for the assembled branch.
Remove unnecessary joints without sacrificing service access
An adapter with no design purpose can go. An interface needed at an approved maintenance boundary stays. In our review, removing unnecessary interfaces reduces the connections whose workmanship and condition must be controlled, but the resulting continuous assembly must remain serviceable. Eliminating a joint solves little if it also prevents the required maintenance access.
Assembly instructions should address cleanliness and the specified tightening method. Cable support also deserves a drawing detail where unsupported weight could load a connector. These controls give the installing contractor a clear basis for workmanship, while the owner's test procedure establishes how the completed RF path will be evaluated.
Our design review cannot predict how every weathered metal contact near an antenna will behave under RF excitation. Even with a consistent frequency study and component schedule, the owner's acceptance plan needs to allow field investigation. That is where the responsible specialist can examine suspected external sources that the design information alone cannot resolve.
Carry the design assumptions into acceptance
At closeout, can we trace the report's branch identifier back to the drawings? A renamed port or an approved component substitution can break that connection unless the owner updates the identification. The measurement itself may be valid. The acceptance reviewer still needs to establish which assembly was tested and whether that is the assembly entering service.
Name the test specialist and the person or organization with authority to accept the result against the carrier's requirement. They may be different organizations. We also need a retest procedure after a failure: replacing a suspected component must lead back to a defined measurement and a recorded result before the corrective work can be treated as effective.
The feeders can stay in place while a new tenant changes the frequencies exciting them. We include that tenant in the review for this reason, and the owner needs to decide whether the affected RF paths require another analysis or measurement before approving the frequency-plan change. The combined operating plan is what lets that decision account for the changed excitation.
Keep the approved configuration with the closeout report. For the next reviewer, we need to preserve whether testing preceded or followed the final antenna relocation, and whether the tested interfaces match the accepted drawings. That history gives troubleshooting a starting point. Present performance still has to be assessed under the conditions that exist now.
Draftech International provides turnkey telecom and power infrastructure, with engineering performed 100% in-house. Our company and partner background identifies the team responsible for delivery. For this PIM discussion, our service role is the stated design scope; the owner or carrier separately appoints the field test specialist and establishes acceptance authority.
Choose the next action for your site
For a multi-carrier owner: Before authorizing an addition, bring the tenant frequency plans together and ask which receive bands are affected. Someone must own the review of changes to that combined configuration. Separate tenant approvals leave the shared infrastructure operator without that answer, even when each tenant has completed its own review.
DAS designers: Finish the RF interface review before issuing the component schedule, specifying low-PIM requirements under identifiable conditions and removing unnecessary joints. Then check that the antenna positions used in the frequency review match the intended layout; a moved symbol on the drawing may require us to revisit the RF assumptions behind it.
Construction and acceptance leads: Get the carrier's procedure and acceptance limit before assigning the specialist. At handoff, tie each result to the installed branch and retain the tested configuration. Does the report cover the boundary shown on the drawing? If the two differ, ask the responsible parties to resolve the mismatch and keep it open until they do.
Draftech's outdoor DAS design addresses these planning concerns through analysis that identifies potential intermodulation products, connector and component quality standards that minimize PIM risk, and multi-carrier frequency plan coordination. To discuss that design scope, send our team the proposed DAS configuration or write to info@draftech.com. Include the carrier inputs you have and identify those still outstanding, so we can work through the unresolved design questions with you.
If the wireless project also needs a qualifying route, our free route engineering offer provides a separate starting point: Draftech engineers the first 20,000 linear feet at no cost, from feasibility and field survey through permit approval. The owner reviews each request before Draftech commits the package, so describe the proposed route when requesting review.

