Analysis and Verification of Airborne Instrument Electromagnetic Emission Test

Electromagnetic compatibility is an important factor affecting aircraft performance and safety. With the increase in the number and type of airborne electronic equipment, the requirements for electromagnetic compatibility (EMC) of airborne equipment are increasingly high [1]. Taking airborne equipment as an example, this paper introduces in detail the problem locating method and rectification method of GJB151A standard CE102 and RE102 test exceeding the standard. It analyzes and summarizes the improper design and installation of EMI filters leading to tests exceeding the standard and gives the interference suppression method. Finally, the correctness of the method is verified by the EMC test.


Introduction
With the continuous improvement of airborne platforms, there is more airborne electronic equipment with more complex functions.Different kinds of electronic equipment, such as communication, navigation, radar, and control systems, exist simultaneously, forming a complex electromagnetic environment.A complex electromagnetic environment will affect the normal operation of electronic equipment and even pose a threat to personnel safety [2] .
Different types of electronic devices use the same set of power cables and are installed in the same electromagnetic environment.As a result, devices interfere with each other through power cable coupling and space coupling.High-speed current and voltage transients generated during the switching process of switching power supply produce serious electromagnetic interference [3] .GJB151A standard puts forward higher requirements for military equipment to launch disturbance.Airborne equipment needs to pass the test items stipulated in the GJB151A standard to ensure that the electromagnetic interference between equipment is effectively controlled.

Test setup and requirements
CE102 and RE102 test items are important test items of the GJB151A standard.Among them, the test item CE102 is used to verify that the conducted emission on the EUT power line does not exceed the specified limit value to prevent it from interfering with the work of neighboring equipment through the power coupling path.The RE102 test item is used to detect that the electric field emission from EUT and its related cables, ports, enclosures, panels, etc., does not exceed the specified limit requirements to prevent the impact on adjacent wireless receiving devices and high-sensitivity instruments [4] .The test configuration is shown in Figures 1 and 2. As the tested items are airborne equipment of the Air Force, the limits required by the GJB151A standard are shown in Figures 3 and 4, respectively.The EUT is powered by a linear impedance stabilized network (LISN).

Test and curve
The airborne equipment described in this article contains the following basic characteristics:  The shell of the device is a metal splicing structure, and the internal components are shielded by the shell. The power supply and filter are combined into a power supply filter box, which is not integrated with the equipment.The filter box has a shielding shell. The cables include power cables, RF cables, control lines, etc.The test layout is shown in Figure 5.During the CE102 test of airborne equipment, it was found that the test results exceeded the limit requirements, and the test curve was shown in Figure 6.The test results show that between 10 kHz and 2 MHz, the switching frequency of the switching power supply and its higher harmonics lead to a serious super standard, which exceeds the standard limit of 29.32 dBuV at 53.5 kHz.Since the CE102 test is not passed, it can be judged that the low-frequency band of RE102 will also have high interference, so the RE102 test will be conducted after the problem of CE102 exceeding the standard is solved [5] .Figure 6.CE102 test curve.

Mechanism analysis of CE102
The ideal capacitor has no electrode conductor or dielectric loss, but in practice, the capacitor is a nonideal component whose equivalent series impedance is not zero.All capacitors are composed of RLC circuits [6] .In the actual circuit, the input capacitor needs to withstand the working current and provide the high-frequency pulse current required by the MOSFET of the switching power supply.When the high-frequency current passes through the equivalent series impedance of the capacitor, the voltage drop is generated.The high-frequency voltage ripple will appear at both ends of the input capacitor [7] .The differential noise current ring is shown in Figure 7.Not all switching currents flow through C9.The capacitor places are diverted, and part of the flow is to the power cord.The switch current flowing to the power cord is a differential noise current flowing through LISN [8] .through the parasitic capacitance between the field effect tube and the radiator [7] .Noise injected into the housing flows into V+ and V-at the LISN.The common-mode noise current loop is shown in Figure 9. Controlling the disturbance current in the product does not overpass the 50 Ω resistance of the LISN and the receiver in parallel is the key to solving the excessive conduction disturbance.In Figures 8 and 9, the EMI filter can be used to shunt the conducting disturbance current.The equivalent circuit without a filter is shown in Figure 9 and Figure 11, and the equivalent circuit with a filter is shown in Figure 10 and Figure 12.The insertion loss of differential mode filter is: where IL dm is the filter differential mode insertion loss.Equation 2 is reduced to Equation 3 for convenience of calculation.f cdm is the differential mode cutoff frequency of the filter circuit.Similarly, the common-mode insertion loss of the filter is shown in Equation 4. The performance of a commonmode filter is related not only to the inductance and capacitance selection but also to the installation process of inductance and capacitance.

Mechanism analysis of RE102
The common-mode filter capacitor inside the filter is connected to the metal shell of the filter, and the filter capacitor is grounded by installing the metal shell of the filter directly on the metal chassis.If the shell of the filter is not lapped with the low impedance of the product housing, the common-mode filter capacitor is equivalent to being connected with the housing after passing through the high impedance inductance and cannot play the role of filtering.Its equivalent antenna is a monopole antenna and radiates externally [9] .The equivalent antenna with a poor filter lap is shown in Figure 13.
U G is the noise source voltage driving the power line, Z GN is the ground plane impedance, I N is the noise current flowing through this plane.

Corrective measures and results
In the frequency range of 0.15 ~ 500 kHz, the interference mainly exists in the form of differential mode.In the frequency range of 500 kHz ~ 5 MHz, differential mode and common mode coexist.Above 5 MHz, the form of interference is mainly a common mode [10] .According to the analysis of the test data in Figure 6, it is concluded that the EUT conduction emission test exceeding the standard is mainly in the form of differential mode.Based on the original filter, a differential mode filter circuit is added.The filter circuit diagram is shown in Figure 14, and the test picture of the filter is shown in Figure 15.The insertion loss of the filter circuit is shown in Figure 16.This filtering circuits provide different insertion loss (IL) rates: 60 dB/decade.According to Equation 4, the filtering cutoff frequency of the filter circuit is 1.3 kHz, which meets the requirements.After the paint is removed by sandpaper, the filter and the mounting housing are bonded in a metal-to-metal plane to effectively reduce the contact impedance.
Figure 16.The insertion loss of the filter circuit.
During the rectification process, it was found that the metal plane at the installation position of the filter was treated with non-conductive spray paint.The spray-painting installation face is shown in Figure 17.A conductive tape is used to isolate filter input and output cables, and the effect after reinstallation is shown in Figure 18.

Conclusions
EMI filter is a key circuit component for suppressing conduction emission disturbance and radiation emission disturbance.The following points need to be paid attention to in the selection and installation process.
Due to the different electrical characteristics of different equipment, the commercial shelf products may not meet the actual requirements of specific products.After the production of the prototype, the parameters of the internal components of the filter should be adjusted according to the specific test results.When designing the structure, the filter installation space should be left enough to prevent the filter installation space from being limited after adjusting the parameters.The filter needs to be lapped with the low impedance of the metal shell, and the metal surface needs to ensure the continuity of conduction and cannot be painted.The input and output wires of the filter need to be separated.If conditions are met, shielding and isolation can be carried out to prevent capacitance coupling between wires.

Figure 7 .
Figure 7. Differential mode interference loop.The MOSFET works in the high-speed switching cycle state, and the corresponding du/dt and di/dt will also change in the high-speed cycle.The high-speed changing du/dt charges the parasitic capacitor between the MOSFET and the radiator, and Common-mode disturbance current flows

Figure 14 .
Figure 14.The filter circuit diagram.Figure15.The test picture of the filter.

Figure 15 .
Figure 14.The filter circuit diagram.Figure15.The test picture of the filter.

Figure 18 . 7 Figure 19
Figure 17.Spray painting installation face.Figure18.Effect after reinstallation.After re-testing, the product successfully passed CE102 and RE102 tests.The test curve is shown in Figure19and Figure20.