Methods of Setting Up an Electromagnetic Compatibility Laboratory

Products must comply with EMC (electromagnetic compatibility) requirements. European regulations stipulate that products that violate the Electromagnetic Compatibility Directive (89/336/EEC) will face high fines. Therefore, businesses are paying more and more attention to the electromagnetic compatibility of products. In order to reduce costs, an EMC laboratory should be established according to the needs and size of the company. This article introduces the advantages, disadvantages and methods of building internal EMC test equipment.

The minimum requirements for setting up an EMC laboratory depend on the needs and financial situation of the company. Usually, companies will try to save money (in terms of equipment and human resources) and minimize risks. But one thing must be understood, there is no such thing as "low cost, low risk and low EMC technology" in the world. Engineers must master a high degree of skills to design low-cost equipment with comparable performance. The lower the precision of the product, the higher the risk.

When setting up an in-house EMC lab, no matter how large or small, some minimum guidelines must be followed. First, the room or area where the EMC lab is built must be clean, free of extraneous items, and completely dedicated to EMC measurements. Whenever conditions permit, a ground reference plane made of metal and reliably connected to the earth is absolutely necessary; if conditions do not allow (such as the room is not on the first floor), at least a protective ground system should be connected. All metal objects in the lab must be reliably grounded or removed. The power supply system must be "cleaned" (properly connected to a line filter somewhere before the power supply enters the EMC lab).

Configuration of EMC test equipment

1) Conducted emission test --- requires a spectrum analyzer (or EMI receiver), cables and LISN (Line Impedance Stabilization Network, handmade or purchased), if possible, a shielded room (at least a shielded tent) and an insulated table 80cm above the ground.

2) Radiated emission test - requires the same spectrum analyzer or EMI receiver, an antenna, cables and OATS (open area test site or (semi) anechoic chamber); absorbing clamps made or purchased for measuring interference power.

3) Harmonic testing (and flicker testing) - If you want to do full compliance testing, you need dedicated equipment (dedicated harmonic analyzer); but if you just want to evaluate, a portable harmonic analyzer or even an oscilloscope capable of FFT evaluation is sufficient.

4) ESD (Electrostatic Discharge) Immunity Test - Only an ESD gun can reliably evaluate the results of this test.

5) Radiated electromagnetic field immunity testing - requires similar equipment to the radiated emissions test, in addition to a signal generator, amplifier , attenuator, field strength meter, and possibly a computer.

6) Conducted disturbance immunity test---The equipment required is similar to 1) and 5), plus CND (heterogeneous coupling decoupling network), but no antenna is required.

7) Electrical fast transient (EFT/Burst) immunity test.

8) Surge immunity test.

9) Power frequency magnetic field immunity test.

10) Voltage dips, short interruptions and voltage variations immunity test.

The last four tests, 7) through 10), require specialized equipment that is available from a variety of vendors.

It is worth noting that manufacturers and distributors of measuring equipment usually provide packages and/or complete sets for performing different tests, as well as instructions and even training on how to perform these tests in accordance with the latest standards. Please check with your nearest local sales representative about the performance and functionality of the equipment. In most cases, the equipment has pre-set test procedures according to the requirements of the standard, so please read the instruction manual first.

Internal pre-compatibility testing

There is no definition for pre-compliance testing, but at the very least it is reasonable to assume that the testing must be performed under conditions as close as possible to those required by the standard.

1) Conducted emission test

Conducted emissions are probably the most difficult problem that electronics manufacturers have encountered over the years, so this article will discuss this area first. The test setup for conducted emissions is shown in Figure 1.

This device is built according to CISPR 22 (EN 55022), and the equipment used must comply with the requirements of CISPR 16-1. The device mainly includes: EUT (equipment under test), if it is a desktop, it must be placed on an insulated table 80cm above the ground; auxiliary equipment (peripherals), connected as normal, unused inputs and outputs must be properly terminated, and excess cables must be shortened or wound into a coil with a diameter of 30~40cm. The spectrum analyzer (or EMI receiver) must have a resolution bandwidth (RBW) of 9kHz in the frequency range of 0.15~30 MHz. The measurement process is described in detail in CISPR 16-1 and the product specification standard. If performed correctly, the results will not be much different from the tests of third-party laboratories.

2) Interference power test

The interference power test (30~300 MHz) method is somewhat similar to 1), but the signal will come from the absorbing clamp (not the LISN). CISPR 16-1 and the product series standards describe the test setup in detail. In summary, the absorbing clamp is placed around the cable under test (trunk, DC, audio/video) and moved along a 6m long line to find the maximum emission value at each frequency point.

The radiated emission test device is shown in Figure 2.

Although this test may seem more complicated than the previous one, it is not particularly difficult. As can be seen from the figure, the EUT is placed on a turntable on an insulated table, 80 cm above the ground, so that the maximum emission value can be found by rotating the EUT during the test. The antenna is mounted on an antenna pole and can be moved between 1 and 4 meters, also for the purpose of finding the maximum emission value. The distance between the EUT and the center of the antenna (marked on it) is 3m or 10m. The receiver device is also made in accordance with CISPR 16-1, and its resolution bandwidth must be 120kHz in the reception range of 30~1000 MHz. The receiver may have a setting for radiated emission measurements.

For all the measurements above, it is important to note that there are some correction factors that must be taken into account when evaluating the results. Firstly, for all measurement setups, ±4dB is the accepted uncertainty interval, as stated in Annex L of CISPR 16-1. Secondly, cable attenuation and connector attenuation must also be considered. But there are more factors that must be considered.

For conducted emissions, the impedance deviation and tolerance of the LISN must be considered. However, its maximum error is only 2-4 dB. The situation is different for the absorbing clamp, whose attenuation is between 14-22 dB, with an average of 17 dB.

The factor in the radiated emission test is the largest, with an NSA (normalized position attenuation) of -24 to 24 dB. In this case, no approximation can be made, and the antenna factor must be used when performing the test. In addition, based on design practice experience, an additional 6 dB design margin should be reserved before sending the product to a third-party laboratory for testing.

3) Harmonics and flicker test

There are no environmental requirements for harmonic and flicker testing. Simply connect the EUT to the power inlet of the harmonic analyzer and perform the test according to the manufacturer's instructions and the requirements of the standard. Again, the test equipment will include some existing settings, but engineers must ensure that these test settings meet the standard requirements for their products. If other methods (such as portable power harmonic analyzers) are used for evaluation, please read the standard requirements carefully before evaluating the measurement results.

4) ESD immunity test

ESD immunity testing may not be very important for large devices. However, in today's era of miniaturization of various products, ESD testing has become one of the "key" EMC tests for most devices, such as portable calculators, MP3 and MD players, USB memory sticks, audio equipment, etc. The test setup is shown in Figure 3.

As can be seen from the figure, the EUT is still placed on an insulating table, located on the HCP (horizontal coupling plane, made of a metal conductive material), and isolated from it by an insulating antistatic pad. The VCP (vertical coupling plane) and HCP are connected to the ground reference plane respectively, and a 470 kΩ resistor is used at each connection end. For each side of the EUT and the VCP, HCP, and each metal surface on the EUT that can be touched by hand, a sharp tip is used for contact discharge (direct discharge), usually 5 times for each polarity. For all plastic parts of the chassis, a round tip is used for air discharge (indirect discharge).

5) Radiated electromagnetic field immunity test

The test setup for radiated electromagnetic field immunity is very similar to the radiated emission test, but in this test, the signal generator and power amplifier will feed the antenna to generate a "uniform electromagnetic field" (±6dB) near the EUT (3V/m or 10V/m in the frequency range 80~1000 MHz, AM, 1kHz, 80% modulation depth). It should be noted that the frequency range is different for different products.

6) Conducted disturbance immunity test

The purpose of the conducted disturbance immunity test is to establish a 3V level (effective value, 150 kHz~230 MHz, AM, 1kHz, 80% modulation depth) at the EUT port input. The signal generator and power amplifier must provide enough power so that the CDN can couple the signal to the line under test. Since test items 3), 7), 8), 9), and 10) use highly specialized equipment, if these equipment are available in the laboratory, engineers do not need to do much, just connect the EUT correctly, and the most important task is to monitor the working mode of the EUT.

How to perform near field testing

As mentioned in the introduction to this article, near-field testing is very suitable for the product development stage. At this stage, standard test methods may give accurate results, but they cannot show the source of the problem. When selecting components, some controller chips have 40dB lower radiation than others, or have higher immunity. Even after the product is developed and the compliance test fails, standard test methods can hardly give any information about the source of the problem. At the printed board level, engineers use near-field test probes to make measurements, and may also use defect detectors, etc. On the other hand, it must be understood that near-field test probes can (almost) not give any information about the conducted or radiated levels of the device, and the error is 20 to 40dB. But the near-field test probe can guarantee one thing: every time it is used, the measurement results are always better than the various measurements mentioned above. In order to get a rough idea of ​​whether the product can pass the EMC test with the near-field test probe, it is necessary to try it several times on samples with known results.

Figures 4(a) and (b) show some examples of H-field probes, E-field probes, and a pin probe. They have the advantage of being easy to make and fairly cheap to buy. They all use 50Ω cables and are connected to a (cheap) spectrum analyzer.

Near field probes are used to pick up both components of the electromagnetic field. Although there are some very sensitive electromagnetic field strength meters on the market, the near field strength of the electromagnetic field is not easy to measure. They cannot give any information about the frequency content of the radiated noise, but can easily point out the "problem component". Near field probes can also give frequency content information when connected to a spectrum analyzer.

A magnetic field probe provides an output voltage that is proportional to the strength of the magnetic radio frequency (RF) field. This probe makes it easy to find the source of RF in a circuit. However, the strength of the magnetic field varies rapidly with distance (cubic). In addition, the orientation of the probe is critical, as the magnetic field must be perpendicular to the magnetic loop. As mentioned earlier, the probe will not give much quantitative information, but for a certain component (IC, switching transistor, etc.), the voltage output of the probe will increase as the probe gets closer to the component. Even if there are many components around, the designer can easily identify the source of noise by studying the schematic. If the engineer decides to replace the component, he can easily measure the result of the replacement, which makes it possible to choose reliable components at the beginning.

Pin probes allow the identification of noise voltage directly on IC pins or thin wires on PCBs. It is also convenient to judge the effect of the filter, although it is only a qualitative judgment. Pin probes can make contact measurements before and after the filter and observe their effects. However, engineers must correctly estimate the contact capacitance and choose probes with small capacitance (no more than 10pF).

The E-field probe can pick up the common mode voltage as well as the desired signal. Common mode voltage is a significant source of radiated voltage. Additionally, the H-field probe cannot pick up the E-field. Using all three probes can prove to be beneficial and time-saving.

Conclusion

This article describes the advantages and disadvantages of having in-house EMC test equipment. Overall, there is no disadvantage to having EMC capabilities, except for the cost of equipment, space occupied, and possible personnel training. In any case, EMC technology is essential in product development, and companies must invest in this area in some way (such as relying on the services of a third-party consulting company), and the cost of ignoring it will be high.