Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Measurement of Radiated Emissions from Integrated Circuits—TEM/Wideband TEM (GTEM) Cell Method; TEM Cell (150 kHz to 1 GHz), Wideband TEM Cell (150 kHz to 8 GHz)
The SAE J1752/3-2017 standard provides a robust and reproducible method for measuring electromagnetic radiation from integrated circuits (ICs). By integrating the IC test board into the wall of a TEM or wideband TEM (GTEM) cell, this approach eliminates the influence of connecting leads within the cell and ensures controlled coupling conditions. Applicable from 150 kHz up to 1 GHz (TEM cell) or 8 GHz (GTEM cell), this standard is essential for IC EMC characterization in automotive, aerospace, and consumer electronics.
Standard Overview and Key Specifications
SAE J1752/3-2017 (stabilized) defines the measurement procedure, including test conditions, equipment requirements, setup, calibration, and data presentation. The IC under evaluation is mounted on a dedicated test PCB that clamps to a wall port cut into the top or bottom of the cell. The test board becomes part of the cell wall, rather than being placed inside, which fundamentally changes the coupling mechanism compared to traditional TEM cell usage.
The measured RF voltage at the spectrum analyzer port is sensitive to the septum-to-test-board spacing. To ensure reproducibility, the standard specifies a standardized spacing of 45 mm. A conversion factor may allow comparisons between data obtained from cells with different spacings.
| Parameter | TEM Cell | Wideband TEM (GTEM) Cell |
|---|---|---|
| Frequency Range | 150 kHz to 1 GHz | 150 kHz to 8 GHz |
| Septum-to-Wall Spacing | 45 mm (standard) | Average 45 mm over port area |
| Port Configuration | Two 50 Ω ports (one terminated) | Single 50 Ω port |
| Typical Use | Lower frequency IC emissions | Wider bandwidth, higher frequency |
Test Set-Up, Calibration, and Design Insights
The IC test PCB controls the geometry and orientation of the IC relative to the cell. Connecting leads run on the backside of the board, outside the cell, eliminating parasitic coupling. One 50 Ω port of the TEM cell (or the single port of a GTEM cell) connects to a spectrum analyzer or receiver via a preamplifier if needed; the other TEM port is terminated with a 50 Ω load. System gain verification and calibration are critical steps before any measurement.
🛠️ Engineering Design Insight: Because the test board is part of the cell wall, its design and clamping directly affect the measurement. The board must fit flush against the cell port, and all connections must be properly shielded. Use a 50 Ω termination rated for the full frequency range. Verify the system gain with a known source and correct for any losses.
⚠️ Common Mistakes
- Not verifying calibration and system gain before measurements
- Using a test PCB that does not meet specified design guidelines
- Inadequate clamping or poor contact between test board and cell wall
- Operating outside the validated frequency range of the cell
- Ignoring ambient electromagnetic noise during measurement
🔍 Key Consideration: The measured RF voltage is related to the electromagnetic radiation potential of the IC. Any change in septum spacing will affect results, so always document the cell geometry and spacing used.
Frequently Asked Questions
What is the advantage of mounting the test board on the cell wall?
By integrating the board into the wall, the IC is placed at a well-defined position relative to the septum, and connecting leads are kept outside the cell. This reduces measurement uncertainty and improves reproducibility.
Can I use a different TEM cell if my model does not have a 45 mm spacing?
Yes, but results may not be directly comparable. A conversion factor (see the standard) may allow comparisons, but caution is needed when comparing data across different cell geometries.
Why is system gain calibration essential?
The measured emission level depends on the entire RF chain (cables, preamplifier, spectrum analyzer). A system gain check with a known reference signal ensures the reported data are accurate and traceable.
What should I do if ambient noise is present in the test environment?
Ambient (background) noise must be recorded before the IC is powered. If noise levels approach the expected emission levels, use a shielded room or notch filters. The standard requires comparison of ambient and active measurements.
This article is based on SAE J1752/3-2017, a key EMC standard for IC radiated emissions measurement. Always refer to the latest revision of the standard for complete requirements.
