What is a Conducted Test?
Conducted Test connects the device under test (DUT) port to a spectrum analyzer or receiver via an RF cable to measure indicators such as output power, occupied bandwidth, adjacent channel leakage, and frequency error under known interface conditions. Compared with radiated testing, the conducted method eliminates antenna pattern variables, making it more suitable for verifying module RF front-end and PA linearity. For IoT customers integrating modules, if a standard antenna port evaluation board is used, a conducted assessment can be done first before moving to full system radiation testing.
Spurious and Harmonics
Apart from the main carrier, transmitters may generate spurious emissions at harmonic frequencies, multiples, and broadband noise. Regulations typically require that spurious power be below limits under specified measurement bandwidth and detection methods. Harmonics are components at integer multiples of the carrier frequency, often caused by PA nonlinearity, and can be improved with filters, power back-off, and predistortion. Switching transients, clock leakage, and digital noise can also cause spikes across wide bands, requiring comprehensive management with PCB stacking and shielding.
Testing Arrangement Key Points
Testing requires calibrated attenuators and low-intermodulation loads to avoid instrument port damage or false signals. The module should operate at target power levels and modulation modes, with voltage and temperature conditions close to production limits. Multi-band products need to be scanned band by band. Note that some certifications allow choosing a compliance path between conducted and radiated methods, but full systems usually still require radiated spurious (RE) verification for enclosure openings and cable radiation.
Failure Troubleshooting Approach
If spurious emissions exceed limits in a certain frequency band, first distinguish whether it is on the PA side or leakage after the filter stage; check whether antenna mismatch causes the PA to enter a nonlinear region. Spurious caused by clock and power noise often appears in clusters at specific frequencies, and a spectrum analyzer near-field probe can be used to locate the radiation source on the board. After changes in mass production such as replacing the crystal oscillator, DC-DC, or shield openings, the conducted curve should be re-tested.
Engineering Practice Supplement
For the practical implementation of "conducted and spurious emissions" in IoT and cellular modules, it is recommended that during the project initiation phase, network constraints (target market frequency bands and phase-out plans, APN and private network strategies, roaming and tariffs, PSM/eDRX power-saving modes) and terminal constraints (antenna and RF layout, power ripple and surges, thermal design and derating, EMC/safety) be written as verifiable items. In the development phase, use weak network simulation, high and low temperature, and voltage deviation tests to reproduce boundary faults, keeping logs and waveforms; in mass production, establish batch and calibration traceability, and device replacements should go through change review. After launch, include connection success rates, reconnection attempts, latency, and power consumption on the operations dashboard, with anomalies traceable to versions and batches.
A pass at the cable port does not mean the housing is quiet
A conducted spurious test means connecting the RF port to a measurement receiver with a calibrated cable and looking, at a known interface, for harmonics, adjacent-channel energy, and broadband noise outside the wanted signal. It removes the uncertainty of the antenna pattern and is suited to judging whether the power amplifier’s linearity and the filters have blocked energy that should not leave. The conditions are that the attenuation and the load match the power, that the device under test runs at the target power and modulation, that voltage and temperature are near the allowed extremes, and that every band to be used is measured. Testing only a middle channel can still leave the edge channels over the limit.
The product also has radiating paths through housing openings, display cables, and the power lead. A conducted pass only says the port is acceptable under cable conditions. It does not replace radiated spurious results with the antenna and housing fitted. Keep both paths in the plan, so a failure shows whether to change the match or to close a gap.
A cluster at fixed frequencies is usually the clock or the supply
Harmonics usually fall on integer multiples of the carrier. Amplifier compression and a worsened match both raise them. First separate energy that is still before the filter from energy that has leaked past it. Clock and switch-mode supply spurs often do not move with the carrier and appear as clusters on the spectrum. Changing the antenna does nothing then. Go back to the stack-up, grounding, and shielding. A near-field probe can find the source on the board, but a probe reading is not itself a limit verdict.
Instrument intermodulation and overload create false signals. Before calling a failure, confirm the attenuator was not removed. After a crystal, a power chip, or a shield opening changes, remeasure the curve. When port reflection rises, the amplifier enters nonlinearity more easily; look at impedance matching first. A long feeder also changes the load the port sees. Record the feeder type when comparing, so two tests are not made on different cables.
A middle channel passing does not mean the edge channels pass
Spurious energy and adjacent-channel leakage move with the carrier frequency. Measuring one channel in the middle of the band can still leave the channels at both ends over the limit, especially when the filter roll-off sits right on the band edge. Every band that will be used should cover at least a low, middle, and high channel, and the power class should be written down. After a crystal or a power chip is changed, look again at the fixed frequencies that do not move with the carrier.