Abstract: Sensitive SDR and EW receivers can fail RE102 for reasons that have nothing to do with the RF front end and everything to do with clocks, converters, seams, and cable terminations. This post explains what RE102 is actually measuring, why limit tailoring and receiver-band notches matter, and a set of practical fixes that hold up in the chamber—without wrecking noise figure or spurious performance.
If you build tactical SDRs or EW receivers, RE102 mitigation becomes a design feature, not a test-time patch. The uncomfortable truth is that the most “sensitive” receiver is often the one you accidentally create with your own wiring harness: any unintended radiator inside the enclosure can couple straight into antenna ports, LNAs, and wideband digitisers. When RE102 fails, it’s rarely a single smoking gun—it’s a stack-up of small leakage paths that add up at exactly the wrong frequencies.
RE102 is the MIL-STD-461 radiated emissions requirement for electric-field emissions from enclosures and interconnecting cables (not permanently mounted antennas). For modern receive systems with GHz-class ADCs and brutal instantaneous bandwidth, the margin between “good enough” and “fails in the notch” can be a seam, a connector backshell, or a DC/DC edge rate you didn’t think anyone could see.
What RE102 is really policing (and why receivers care)
At its core, RE102 is asking a simple question: “How much E-field energy is your equipment throwing into free space across 10 kHz to 18 GHz?” The answer is measured at a defined distance (commonly 1 metre in many setups) with antennas appropriate to each band, using specified detector and bandwidth rules.
For sensitive receivers, RE102 is not just about platform-level coexistence—it is about self-jamming. Your own radiated emissions can re-enter through:
- Front-door coupling (radiated energy entering via the antenna aperture),
- Back-door coupling (through I/O, power leads, control harnesses, and chassis seams), and
- Re-radiation (cables behaving like efficient antennas when common-mode current is present).
In practice, the worst RE102 failures for receiver-heavy payloads are narrowband peaks: harmonics of clocks, switcher fundamentals, SerDes lanes, FPGA fabric noise, and local oscillators, all made visible by a cable or enclosure feature that provides just enough antenna efficiency.
RE102 mitigation starts with defining the victim receiver (not the EUT)
One of the most useful shifts in recent guidance is the renewed emphasis on tailoring RE102 to what you are actually protecting. Programmes increasingly specify receive-band notches (for example around common GNSS bands and other protected links), because a flat “one-size” limit can hide the real risk: a modest emission peak landing inside a high-gain, narrowband receiver can do far more damage than a larger emission elsewhere.
Two practical industry insights worth bringing into your process:
- Bandwidth matters in the notches. Current practice increasingly aligns measurement bandwidth in each notch to simulate the protected receiver bandwidth (a point echoed in recent EMC community training referencing GEVS guidance). If your victim receiver is 200 kHz wide, measuring with a much wider RBW can misrepresent risk—or, conversely, create test artefacts and wasted debug cycles.
- MIL-STD-461G test complexity has gone up for platforms with transmitters. Compared with earlier revisions, transmitter-active considerations and the use of filtering to avoid measurement receiver saturation are now a real-world issue. If your receiver is collocated with high-power transmit functions, the “test in standby” assumption is no longer a safe mental model when planning compliance and debug.
The upshot: before you touch copper tape, write down which receivers you must protect, their tuned bands, instantaneous bandwidth, and what “degradation” means (desense, intermod, false detections, DF bearing error). That becomes your RE102 mitigation target—not an abstract line on a plot.
Typical RE102 emitters inside tactical SDR/EW payloads
Most RE102 problems in modern receiver systems are born in the digital and power domains. The usual suspects:
- Switch-mode power supplies: fundamental plus harmonics; fast di/dt loops; transformer leakage; noisy rectifiers.
- High-speed clocks and synthesiser spurs: reference oscillators, PLL phase detector products, and their harmonics landing in sensitive bands.
- High-speed digital interfaces: Ethernet/SerDes/PCIe lanes leaking common-mode energy into chassis and harness.
- Wideband digitisers and direct-sampling ADC architectures: the industry push for wider instantaneous bandwidth and real-time streaming (including multi-GS/s class approaches) reduces analogue conversion stages—but it also concentrates brutal digital activity close to your RF ground references and enclosure walls.
- Cable harness resonance: a “quiet” emission becomes a loud radiator when cable length and termination create a nice common-mode antenna at the failing frequency.
A telling pattern: teams focus on the RF front end (LNA, preselector, LO) yet the chamber plot screams at 100–800 MHz clock harmonics or at a DC/DC harmonic sitting right in a comms receive notch. RE102 failures are often system engineering failures, not RF circuit failures.
Practical RE102 mitigation fixes that survive test house reality
There is no single silver bullet, but there is a reliable order of operations. The goal is to reduce source energy, block coupling paths, and avoid creating new radiators.
1) Kill the source before you build a better Faraday cage
Start with the switching and clocking fundamentals:
- Slow the edges where you can. Gate resistors, snubbers, controlled slew-rate drivers, and damping on fast lines often buy more than heroic shielding.
- Fix power loop geometry. Tighten the hot loop of each converter; move the highest di/dt nodes away from seams and connector fields; ensure return paths are short and contiguous.
- Spread-spectrum clocks (with caution). They can reduce peak emissions but may raise the noise floor across a wider band—sometimes a bad trade for EW receivers hunting for weak signals.
2) Treat cables as antennas unless proven otherwise
RE102 commonly fails because of common-mode current on external cables. Robust fixes include:
- 360° shield termination at the enclosure entry (proper backshell/banding, not a pigtail).
- Connector selection that supports bonding (and doesn’t sabotage you with long ground fingers).
- Common-mode chokes and ferrites placed with intent—close to the emission source or at the egress point, depending on where the mode conversion occurs.
- Feedthrough capacitors / filtered connectors for low-frequency lines where you can tolerate the capacitance.
If you do nothing else, do not let a braided cable shield “float” inside the enclosure and then expect the chassis to behave. Termination quality is often the difference between passing and failing by 10 dB.
3) Control apertures, seams, and panel currents
Enclosures leak through the details:
- Seams and lids: use conductive gaskets with the right compression, and ensure paint/finishes don’t insulate the contact points.
- Apertures: honeycomb vents, waveguide-below-cutoff techniques, or simply moving vents away from the noisiest internal zones.
- Chassis current management: ensure your internal grounds don’t force high-frequency return currents across panel joints or around cut-outs.
Be wary of “local fixes” such as a small shield can over a converter if the can simply capacitively drives the lid and turns the entire enclosure into the radiator. Bonding strategy matters.
4) Partition the receiver like you mean it
Receiver sensitivity is won by isolation and lost by convenience:
- RF compartmentalisation (RF, digital, power) with controlled interconnects between them.
- Filtering at partition boundaries so noise doesn’t hitch a ride across the box on a control line.
- Clock hygiene: keep references and LO routing physically and electrically distant from high-current switching nodes.
Integration gotchas: cables, notches, and platform coupling
A receiver that passes RE102 on the bench can fail in the vehicle or aircraft because integration changes the coupling network. A few realities to plan for:
- Harness reroutes change resonance. A 20 cm difference can move a peak into a protected notch.
- Bonding to platform structure is part of the RF design. If the chassis bond is inductive, high-frequency current finds another way—often your cable shields.
- Notches should reflect actual protected bands. If the platform has GNSS, SATCOM, or specialised tactical links, expect tighter limits (and measure with bandwidth representative of the victim receiver where required).
In other words: RE102 mitigation is a system-level discipline. The earlier you treat mechanical, harness, and EMC as one design problem, the fewer surprises you’ll buy in the chamber.
How Novocomms Space & Defence helps de-risk RE102 on sensitive receivers
At Novocomms Space & Defence, we support programmes where receivers are both the mission and the liability: tactical SDRs, EW/ESM payloads, SATCOM terminals, and mixed-signal RF systems that must remain reliable under harsh environmental and EMC constraints. Our teams work across antenna/RF design, rugged integration, and practical compliance thinking—because RE102 failures rarely respect organisational boundaries.
Typical engagements include:
- Architecture reviews to identify likely RE102 emitters and coupling paths before metal is cut.
- Receiver-band coexistence planning, including advice on notch strategy and integration risk areas.
- Rugged RF and antenna system design where enclosure, filtering, grounding, and RF performance are developed together rather than traded off at the last minute.
Most importantly, we approach RE102 the way the chamber does: as an energy-and-coupling problem. That mindset tends to save time, weight, and rework—precisely what tactical programmes don’t have.
Conclusion
Passing RE102 with a sensitive receiver is entirely achievable, but it demands discipline: tailor the requirement to the real victim receivers, control measurement assumptions in notches, reduce emissions at the source, and treat cables and seams as first-class RF structures. The receiver you’re protecting is unforgiving; your mitigation strategy needs to be equally deliberate.
If you’re fighting an RE102 peak that keeps reappearing—or you want to design it out before first build—contact Novocomms Space & Defence via https://novocomms.space/contact-us/.