Harmonizing Circuit Design and EMC DesignPart 21 EMC Calculation Methods and EMC Simulations (6): Trial Calculation of Radiated Immunity (RI)

Hello! I’m Inagaki, at ROHM.

This 21st article is the sixth dealing with EMC (electromagnetic compatibility) calculation methods and simulations. Here, trial calculations of radiated immunity (RI) are explained. I will be discussing the ALSE (absorber-lined shielded enclosure) method in the ISO 11452-2 standard relating to the electromagnetic compatibility (EMC) characteristics for vehicles; the method is also known as the antenna irradiation method.

An automotive battery, artificial mains network, wire harness, DUT (device under test), etc. is arranged on a reference grounded surface, an antenna at a distance of one meter is used to apply electromagnetic noise (an electric field), and it is determined whether the DUT (device under test) malfunctions or not. Whereas in the ALSE method according to the CISPR25 standard the radiated emission from the wire harness or the like is observed, in the ALSE method stipulated in ISO 11452-2, the radiated immunity of the wire harness or whatever is observed, and so the opposite phenomenon, the effect on the DUT, is being measured.

With respect to calculated predictions, if the environment is such that calculations are possible using the CISPR25 ALSE method, then some setting changes and the like may be needed, but CAD data and other resources can probably be used essentially without modifications. The objects of calculations, in addition to those mentioned above, are antenna signal sources (electric fields), EMC countermeasure components (here, capacitors C), DUTs (LSI models, passive elements), testers to determine standard conformance, and the like.

In this article, we use a combination of electromagnetic field analysis and circuit analysis. As a general outline of the calculations, first the voltage induced on a printed circuit board (PCB) connected to a wire harness is calculated from the electric field intensity of an antenna by using electromagnetic field analysis. Based on the voltage arriving at the DUT (LSI terminals) from the induced voltage, circuit analysis is used to judge whether or not there has been a malfunction. Put simply, electromagnetic field analysis is used for the radiation system (the part of the noise propagating through the air), and circuit analysis is utilized for the conduction system (the noise transmitted in wiring).

Calculations are explained in order. In trial calculations, two processing stages are used: the first stage is IB (malfunction threshold) model extraction, and the second stage is calculated prediction, and both stages are automated using (shell) scripts. In the first-stage IB (malfunction threshold) model extraction, the following calculation procedure is used.

■First Stage: IB (Malfunction Threshold) Model Extraction

1. ① First, a calculation circuit diagram is created from the above-described object for calculation. The idea is to connect the measurement circuit to obtain the calculation circuit. It should be useful to refer to the above for a circuit diagram for electromagnetic field analysis. The circuit diagram for circuit analysis will be essentially the same as that used in the DPI method in the IEC 62132-4 standard (article 19).
2. ② In the trial calculations, the measurement frequency range should be from 200 MHz to 1 GHz, and an electric field intensity limiting value of 200 V/m is desirable. Pass/fail judgment will depend on the measurement results, and so, similarly to the HE method (BCI method) of the ISO 11452-4 standard described in article 20, the pass value is set to 200 V/m and the fail value to 100 V/m, and measured values are used in calculated predictions. The pass/fail measurement/judgment results are converted into numerical values for use in calculations.
3. ③ The electric field intensity (over all measurement frequencies) created in ② is applied as the antenna signal source (electric field) in the electromagnetic field analysis, and the voltage induced on the printed circuit board (PCB) on which the DUT (device under test) is mounted is calculated.
4. ④ Next, the voltage reaching the DUT (LSI terminals) on the printed circuit board (PCB) in the circuit analysis (transient analysis, all measurement frequencies) is calculated, and is used in the IB (malfunction threshold) model. As indicated in the graph below, the wire harness characteristic in which the voltage threshold for malfunction is observed as waves with a constant period is clearly visible. In this stage, the “correlation coefficient of the electric field intensity with the induced voltage” is calculated. Automatically generating a computer model from measured values in this way is useful for shortening calculation times and improving calculation accuracy.

Example of IB (malfunction threshold) model calculation

In calculated predictions in the second stage, the following procedure is used.

■Second stage: Calculated Prediction

1. ⑤ A circuit for calculated prediction for circuit analysis is created. A difference with the circuit for IB (malfunction threshold) model extraction is the addition of a malfunction tester (comparator). The voltage reaching the LSI and the IB (malfunction threshold) are compared by the malfunction tester (comparator).
2. ⑥ First the voltage noise signal source is set to a damped oscillation waveform, and when circuit analysis (transient analysis) is performed, we find that the LSI transitions, for example, from a malfunctioning state to a non-malfunctioning state. This is repeated for all measurement frequencies.
3. ⑦ When a voltage at which malfunctioning occurs is found using the damped oscillation waveforms, next the previously calculated “correlation coefficient of the electric field intensity with the induced voltage” can be used to calculate the electric field intensity at which the malfunction occurred. This is the calculated prediction value that was being sought. By using a correlation coefficient, electromagnetic field analyses can be omitted, shortening the calculation time.
4. ⑧ The graph on the lower left is the result of calculated prediction using the same circuit that was used in IB (malfunction threshold) model extraction prior to EMC countermeasures. Measured values are in good agreement with calculations.
5. ⑨ A capacitor C is added to the EMC countermeasure circuit in circuit analysis, to decrease the voltage noise reaching the DUT (LSI terminals). The calculated prediction result at this time is as indicated in the graph on the bottom right. With calculated values of 200 V/m and higher, the calculated prediction indicates that there is probably conformance to the standard. Upon closer examination, we see that a somewhat more effective EMC countermeasure is needed in the 200 MHz to 220 MHz area. There are no problems at higher frequencies than this.

Thus, even when using an ordinary electromagnetic field analysis tool and circuit analysis tool, if calculation results are saved automatically in a text file (ASCII) and passed in a (shell) script, parameter settings and testing can be automated comparatively easily.

Thank you for your kind attention.