Testing system design components for electromagnetic interference problems - Embedded.com

Testing system design components for electromagnetic interference problems

With the increasing complexity of ICs, test for electromagneticradiation and susceptibility becomes an urgent requirement to meet newdemands in design. Requirements for low EMIand strong anti-EMIcapacity demand further improvements on electronic devices and systemsto meet EMC standards.

Moreover, product engineers must address EMC issues, particularlymeeting threshold requirements and reduced EMI between the systems andother electronic devices. IC integration density doubles almost everyyear, making EMC a key priority in design.

To address the issue, the International Electrotechnical Commission (IEC) released standards IEC 61967, for EME test of ICs withfrequencyfrom 150kHz to 1GHz, and IEC 62132 for EMS test of ICs with frequencyfrom 150kHz to1GHz. Scheduled to be released soon is IEC 62215, acomplementary standard to IEC 62132 that addresses EM disturbance onICs.

EME testing
IEC 61967 requires the use of a transverseelectromagnetic cell (TEMcell) and field probe to measure electromagneticemission (EME). Ituses spectrometers or receivers with frequency range from 150kHz to1GHz. Normally, the ICs to be tested are placed on one circuit board.The specification of the circuit board is shown in Figure 1 .

Figure1. ICs to be tested using TEM cell method are placed on one circuitboard.

A power supply with low RF noise is needed for testing while thetemperature should be at 23°C (±5°C). There can be casesof environmental noise as peripheral circuits are measured. If thebackground noise is lower than the threshold value by more than 6dB, apreamplifier is applied, if needed.

This method measures radiation using a TEM cell, which is anirregular coaxial line. In the middle of the line, a flat chip such asan inner conductor with a quadrate outer conductor connects to thereceiver at one end and the matched load at the other end (Figure 2 ).

Figure2. A flat chip as an inner conductor with a quadrate outer conductorconnects to the receiver at one end and the matched load at the otherend.

A quadrate vent at the top of the outer conductor is placed to fixthe circuit board to be tested. Specifically, the IC will be installedinside the cell, with the wiring side and the peripheral circuitsfacing outside. This way, the radiation emission will mainly come fromthe IC being tested.

High-frequency current produced by the tested chip will flow alongthe wire, while wielding pins and package connections serve as theantenna of radiation emission. If the testing frequency is lower thanthe first-order high-mode frequency, only TEM as the major mode istransmitted. Note that the power of the TEM cell's port has aquantitative relation with the emission power of the disturbancesource.

Hence, the voltage value collected at a curtain area of the TEM cellcan help evaluate the IC's radiation emission.

Figure3. The surface scan method tests the spatial distribution of the IC'ssurface electric and magnetic fields.

The TEM cell method implements surface scan by testing the spatialdistribution of the IC's surface electric field and magnetic field. In Figure 3 , the electronic or themagnetic field probe scans over the IC surface.

The frequency, emission value and spatial position are collectedevery time and processed by the software. Spatial frequencydistribution of the field intensity is displayed in color figures. Theresult's quality relates closely to the precision of the mechanicallocating system and the size of the probes.

Hence, this is applicable to general PCBs and may not necessarilyuse the standard testing circuit boards recommended.

By scanning the electronic magnetic fields over the IC surface, wecan get information about the relative intensity of the EM radiationsource. This method can accurately identify the area on the IC that hastoo much radiation.

Partially-shielded electronic field probes and single-loopminielectronic field probes are recommended in this standard. Bothprobes can be made by 0.5mm semi-rigid coaxial cables.

Radiation measurement
The standard has three methods for radiation measurement: directcoupling, workbench faraday cage (WFC ) and magnetic field probe.

There are two measurement types for the direct coupling method: 1ohm and 15 ohm. The 1 ohm method tests the total disturbance current ofgrounded pins, while the 15 ohm tests the disturbance voltage of outputports.

After going through the chip, RF current will flow to the IC'sgrounded pins, so the measurement against the ground RF current canreflect more precisely the IC electromagnetic disturbance. If youconnect a 1 ohm resistance in serial in the ground loop, you can obtainthe RF current from the loop. Meanwhile, measure the impedance matchingof the testing instruments and the grounding pin.

Figure4. The 150ohm direct coupling method tests the disturbance voltage ofsingle- or multi-output signal lines for impedance matching.

The 150 ohm method can be adopted to test the disturbance voltage ofsingle- or multi-out-put signal lines (Figure4 ). Note that 150 ohm is the statistic average value ofwire-harness common impedance. To match it with a 50 ohm impedancetesting system, apply a matching network.

Meanwhile, the WFC method of the standard measures the conductiondisturbance voltage of power lines and I/O signal lines. Place astandard or an application circuit board with the IC to be tested intothe WFC.

Table1: EMC testing also requires selection of resolution bandwidth forspectrum analyzers and receivers.

All the power and signal lines in and out of the WFC will befiltered. Connect the WFC testing ground port to testing instrumentsand the ground port to be tested to a 50 ohm matching load. Awell-shielded environment reduces the background noise. You can connecta 100 ohm resistance in serial to match the impedance (Figure 5 ).

Figure5. The IEC 61967 WFC method measures the conduction disturbance voltageof power lines and I/O signal lines.

Magnetic probe
The magnetic probe method of the standard evaluates the radiationemission of ICs by testing the current through PCB wires (Figure 6) .

The pins of the chip are connected to the power source or peripherycircuits via wires on the PCB; thus, the magnetic probe can obtain theRF current.

Figure6. The magnetic probe method evaluates the radiation emission of ICs bytesting the current through PCB wires.

Based on the law of electromagneticinduction, the voltage at theprobe's output end is proportional to the RF current of the wires. Aneasy and repeatable method is needed to evaluate the interferenceimmunity of a chip, which is divided into two fields – radiationimmunity and conduction immunity.

This method collects the RF power if ICs fail to work. IEC 62132divides the operation status of an IC into five grades; it will thendetect the continuous and AM waves separately. Similar to ISO11452,peak-level constant AM with depth of 80 percent 1kHz is applied.

Figure7. The TEM cell method can also be used for susceptibility test.

This is different from IEC 61000-4-3. The IEC standard demands thatthe peak power of the modulation signal be 1.8x the carrier wavesignal. However, when the peak level constant is under testing, themodulated signal power, with a modulation depth of 80 percent, is only0.407x the carrier wave signal power.

The TEM cell discussed in IEC 61967 is also suitable for radiatedsusceptibility measurement. One end of the cell is connected to asignal generator and PA, while the other end is connected to a propermatching load. The TEM waves – produced in the cell and similar toremote TEM waves – can be used to measure EMS. Other status monitoringinstruments may be required to monitor the IC status in real-time.

Figure8. In the BCI method, interference power is injected into the singlewire or wire harness of IC pins.

Set of testing
The standard uses three methods for the susceptibility test: bulkcurrent injection (BCI), direct power injection (DPI) and WFC. In theBCI method, interference power is injected into the single wire or wireharness of IC pins. Because of inductive coupling, the cable testedproduces the disturbance current that can be detected by anothercurrent probe. The method derives from the automobile electronicsusceptibility test.

Figure9. DPI test injects capacitive coupling to measure susceptibility.

Compared to BCI that injects inductive coupling, DPI injectscapacitive coupling. RF signals are injected directly to a single pinor a group of pins of the chip. The coupling capacitance prevents DCfrom being directly placed on the output end of AP (Figure 9 ). The WFCmethod in IEC 61967 can also be applied for susceptibility test, butthe receiver is replaced with signal resource and AP (Figure 10).

Figure10. The WFC method in IEC 61967 can also test susceptibility, with thereceiver replaced with signal resource and AP

The shielded structure and good filtering hold RF disturbancesignals inside the cage, thus protecting operation personnel.

Ding Ding is a Product Engineerat Rohde-SchwarzChina Ltd

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.