EMC FAQs

GW Instek's EMC pre-compliance solution focuses on enabling engineers to quickly and easily evaluate and debug product’s EMI in a general working environment. The differences of certified laboratories which focus on certification are as follows: 

  1. Impact of environmental noise: In the certified laboratory, the noise can be well controlled and regularlly examined; however, the environmental noise cannot be effectively suppressed in the general working environment. The conductive interference measurement is mainly the purity of the power supply in the working environment, and the radiated interference is the existing wireless signal in the environment. For suggestions on how to respond, please refer to “How to eliminate the impact of environmental noise when performing Pre-compliance measurement”?"

    Anechoic chamber has good electromagnetic wave isolation and grounding structure

  2. The instrument used by Pre-compliance is a spectrum analyzer, which is functionally different from the EMI receiver used by the certified laboratory.
    Spectrum Analyzer (left) and EMI Receiver (right)

  3. The probe used for the radiated EMI in the debugging stage is different from the antenna measurement used by the certified laboratory; the former is the near-field measurement of the electromagnetic field, and the latter is the far-field measurement. The two are essentially different.
    Near-field measurement of the probe on the PCB surface (left) and far-field measurement of the antenna  10m away (right)
It is normal for the pre-compliance and certified lab test results to be different, so it is important to understand the cause of the difference in order to get a reference from the pre-compliance measurement so as to conduct debugging. The difference between the two can be referred to “What is the difference between the Pre-compliance test and the program used by the certified laboratory?”.
The most important thing is to establish a test reference for pre-compliance after obtaining the differences between the two. For example, pre-compliance test needs to be tightened and reduced a certain standard (such as 4dB reduction) under regulatory standards that is a margin for testing by certified laboratories. After accumulating and establishing such a test reference, the successful measurement rate in the certification laboratory will be greatly improved.
The environmental noise includes conduction and radiation EMI. After connecting the DUT, the EMI measurement is performed before the power of the DUT is turned on. The result obtained is the environmental noise, and then the power of the DUT is turned on to distinguish between ambient noise and DUT EMI signals.
If the EMI generated by environment and DUT is in the same frequency bandwidth and the environmental signal is strong, then the signal is very difficult to distinguish. GW Instek's spectrum analyzer GSP-9330, in this case, uses Topographic mode (Display >> Window Setup >> Topographic or Topographic+Spectrum) to distinguish different signal sources in the same frequency bandwidth. However, the best solution is to measure in an environment with low ambient electromagnetic noise interference.
Topographic mode displays noise hidden in the scan signal (indicated by a red arrow)
In the EMI test, to determine whether the DUT can pass the EMI regulations, the signal must be detected by the detection method of QP Detector and EMI-Avg Detector respectively.
When testing by an EMI receiver in a laboratory, in order to speed up the test, the entire signal is usually extracted via PK+ Detection, and the signal frequency exceeding the limit regulation is captured, and then the narrow frequency or zero frequency is used to find out the QP and AVG values that exceed regulations. As shown in the figure below, the values of QP and AVG obtained in the EMI Report are the results obtained by the software in the background calculation and processing.
The EMI signal detection using the GSP-9330 focuses on the source of possible EMI during the Pre-Compliance phase and the Debug process, so PK+ Detector is usually used at this stage to find all EMI sources and identify larger sources of signals to analyze. If you want to know the value of QP and AVG at the frequency point found by PK+, manually set the Span to narrow or zero frequency and set the detection mode to QP or AVG.
There are three ways to generate EMI test reports for GSP-9330:
  1. Save the measurement result screen on the USB flash drive: Use the USB flash drive to store the measurement results. After the test is completed, open the Peak Table to mark the 10 signals with the highest amplitude, or use the Marker Table to mark up to 6 Markers, and then insert the USB flash drive until the spectrum analyzer identification program is completed, then you can save it. The saved file format is a JPG image file.

    The operation procedures are : first insert the USB flash drive on the GSP-9330 front panel USB slot. When the USB icon appears on the screen, press the Quick Save button to complete the screen storage.

     
    USB slot and icon of the GSP-9330 (left); The screen can be directly saved on the USB flash drive (right)

  2. Store the measurement result data in the USB flash drive and then use the PC software (such as MS Excel) to read it out for processing: store the measurement data with USB flash drive, the file name is tra, and the format of the tra file is Text format, the content is the number, frequency and amplitude of 601 points (0~600 points). You can use the PC software such as MS Excel for subsequent processing. Users can enter the regulatory limit line into MS Excel for displaying at the same time.

    Expressed in MS EXCEL

  3. Use PC software SpectrumShot output report provided by GW Instek: Use GSP-9330 dedicated PC software SpectrumShot to read the test data while in connection, or directly read the tra file in USB flash drive offline to print or output an RTF format file. The software can be freely downloaded from the GW Instek website.
    Reports generated by SpectrumShot
     
    This PC software must be collocated with the NI-VISA driver. Please download it from the NI website.
Yes. The GSP-9330 has built-in EN55022A, EN55022B, EN55015, FCCA and FCCB regulations. If other regulations are required, users can use the User Define function (as shown in the figure below) to edit and store the regulatory limit line on the spectrum analyzer. There is no limit to the storage capacity of the regulatory limit line, as long as the capacity is sufficient to store. However, if you need to call the test regulations, up to 5 sets of limit line regulations can be called at a time.
In addition, you can use the dedicated PC software SpectrumShot for editing and remote control. The software can be downloaded freely from the GW Instek website. The software has built-in EN55011, EN55012, EN55014, EN55015, EN55022, EN55025, EN55032, FCCA and FCCB commonly used regulations. Users can select the items to be tested according to the test requirements, such as conduction and radiation, Class level, product category, Detector level etc., as shown below. If there are other regulatory requirements, users can also edit the new limit line.
This PC software must collocate with the NI VISA driver. Please go to the NI website to download it.
User-defined EMI regulation operating menu PC software SpectrumShot EMI limit line menu)
To display two self-defined Limit Lines, please use the GSP-9330 dedicated PC software SpectrumShot.
This PC software must be collocated with the NI-VISA driver. Please download it from the NI website.
When the EMI signal is too strong, the GSP-9330 will send a warning message to avoid damage to the spectrum analyzer. It is recommended to connect an attenuator before the input terminal.
Yes, at this time, the RBW will switch to the EMI bandwidth according to the current frequency bandwidth.

Frequency

RBW

<30MHz

9kHz

30MHz-1GHz

120kHz

>1GHz

1MHz

After setting, each Trace will simultaneously scan PK+, QP, and AVG according to the settings.
The Correction under EMI Test is used to compensate the signal measurement of the antenna when it is horizontally or vertically polarized; the 3m/10m Correction under Sensor Probe is converted into a 3m/10m anechoic chamber after using the signal measured by the ANT-04 probe. Simulation results. The former is the correction of the antenna far-field measurement, and the latter is the test of the near-field probe simulation in the anechoic chamber.
The digital oscilloscope samples the signal of the DUT and uses FFT technology to obtain the spectrum distribution. Using this method to analyze EMI signals will cause large errors in amplitude measurements due to the following factors:
  1. The bandwidth of the oscilloscope and probe
  2. Input resistance
  3. Sampling rate
  4. The number of bits in the ADC is insufficient
Therefore, in the application of EMI testing, FFT obtained by digital oscilloscope (DSO) cannot replace spectrum analyzer.
The bandwidth of the oscilloscope and probe
The bandwidth indicated by the oscilloscope is 3dB bandwidth. For example, an oscilloscope with a bandwidth of 100MHz will have 3dB (about 30%) distortion when measuring a true 100MHz signal. In order to reduce the measurement distortion caused by the bandwidth, the general choice for the oscilloscope bandwidth selection is 5 times the highest frequency of the DUT signal, for example, measuring a 100 MHz signal, it is recommended to use a 500 MHz digital oscilloscope.
The bandwidth and frequency curves of the oscilloscope and probe
The EMI regulations must be at least 1 GHz for electromagnetic radiation measurements. Passing or failing is often determined by 1 to 2 dB. In order to avoid false judgement caused by high frequency EMI or harmonic signal errors, it is recommended to use an oscilloscope with sufficient bandwidth, which means that the oscilloscope needs a bandwidth of up to 5 GHz in order to maintain consistent amplitude measurement accuracy.
In addition to the oscilloscope, the probe used also has the same bandwidth problem. The active probe has a larger bandwidth, but the price may increase by 5 to 10 times. The bandwidth indicated by the passive probe is that the probe needs to be in the 10:1 position, otherwise the bandwidth will decrease.
Input resistance
If an EMI near-field probe is used to connect the oscilloscope for replacing the oscilloscope probe, reflection will occur in the high-frequency part because the input impedance of the oscilloscope is generally 1MW. Hence, the signal will be reflected back to the probe end instead of entering the oscilloscope. This situation will vary with frequency. If you really want to use it, you must use an oscilloscope with an input impedance of 50W.
The number of bits in the ADC is insufficient
The general oscilloscope's ADC is 8-bit, and the special-purpose DSO will have ADC with more bits. GW Instek's GSP-9330 is 23-bit by patented technology in small signal positions. The difference in the number of these ADC bits will make a huge difference in the results of the measurements.
The lowest range step of the 8-bit ADC, linear scale (left) and logarithmic scale (right)

Assume that the oscilloscope's vertical level has a minimum level of 1mv/div and 10 graticules represent 10mV. The 8-bit ADC is used for sampling measurement. The minimum measurable voltage is,

After converting to dBuV:

This value is already higher than the upper limit value(2) of Class B of the CISPR22 electromagnetic radiation ITE product without considering the antenna factor(1), and since the resolution is 40uV, the following step reading value is 80uV. , after the conversion, the value is 38dBuV, which means that the EMI signal between 32 and 38dBuV cannot be measured.

Note:
(1) See question 5 for antenna factors.
(2) Regulation for CISPR22 electromagnetic radiation ITE products Class B@10m are as follows:

Frequency, MHz

QP limits, dBuV/m

30 ~230

30

230~1000

37

Sampling rate of the oscilloscope ADC

According to Nyquist–Shannon sampling theorem, the sampling rate of a digital oscilloscope must be greater than or equal to twice the highest frequency of the DUT signal. In the analysis of high-frequency EMI signals, it is more important to select an oscilloscope with a bandwidth of 5 times, and then the sampling rate needs to be more than twice the highest frequency of the DUT signal that requires a high-priced oscilloscope.

When GW Instek GSP-9330 is at 10Hz RBW, the base noise is -133dBm when the preamplifier is turned on. That is equivalent to -26dBuV, which measures signals nearly 58dB less than the oscilloscope. Therefore, the difference between the oscilloscope and the spectrum analyzer's ability in measuring small signals is tremendous.

Spectrum analyzer with an input impedance of 50 W can use the GKT-008 probe. However, the functions related to GKT-008 in the EMC Pretest built in GSP-9330 cannot be used.
In terms of hardware architecture, the GKT-008 is designed to be used with a spectrum analyzer. Spectrum analyzer with 50 Ω input impedance can use the GKT-008 probe.
However, the GKT-008 has some very useful extension functions that must be used in conjunction with the EMC Pretest feature of the GSS-9330 spectrum analyzer. These features include:
  1. Far field EMI measurement estimation
  2. Estimation of EMI after the chip's pins are added with traces to the PCB
  3. Estimation of EMI after the communications interface is connected to the cable
  4. After near field probe ANT-04 is connected with TG, simulate signal source for EMS test
Using the near field electric field probe and the magnetic field probe to measure the near field electric field and the magnetic field respectively can only obtain the two components of the electromagnetic energy in the EMI signal. The GKT-008 can directly and effectively sense the actual electromagnetic energy.
The near field electric field probe and the magnetic field probe can only measure two components of the electromagnetic energy of the EMI signal. The actual electromagnetic energy is the vector outer product of the electric field and the magnetic field. Hence, the large near field magnetic field does not mean that it will radiate to become the interference of EMI. It is the same for the electric field. Therefore, measuring the near field electric field and the magnetic field is only a substitute method when the electromagnetic wave energy cannot be directly measured. Moreover, the electric field E and the magnetic field H respectively measured by the spectrum analyzer are difficult for the calculation of the electromagnetic energy vector by using the formula.
Moreover, in the process of EMI debugging, the problem of a magnetic field or an electric field must be considered to solve whether for isolation mechanism, filter circuit or grounding design of the circuit in terms of techniques and methods for solving the problem. The main method is to measure the frequency and amplitude (energy) of the EMI noise so as to confirm which circuit or component produces the interference signal source. The interference can be improved by suppressing the current (such as series resistance) or voltage (such as parallel bypass capacitor).
Using the near field probe of GKT-008 to directly measure the electromagnetic energy can directly measure the actual near field electromagnetic energy. There is no need to measure the electric field and the magnetic field separately by using GKT-008, which becomes a more efficient tool. In some special applications, separate analysis of the electric and magnetic fields is required such as larger power motor rotation systems.
Because the principle of the near field probe is not the same as the working principle of the antenna, there is no antenna factor parameter.
The function of the antenna is to convert the electric field strength ES of the received electromagnetic wave into a voltage , which is measured and displayed by the output port connected to the spectrum analyzer, and the relationship with the antenna factor  is as follows:

is the electric field strength of the signal,is the voltage generated at the output port of the antenna.

In a linear representation,
But in the logarithmic expression,
Therefore, the electric field strength of the electromagnetic wave is equal to the measured reading value (voltage, dBuV) of the spectrum analyzer plus the antenna factor (dB/m). For example, a spectrum analyzer connected to a Biconical antenna has a signal of 30dBuV at 320MHz and an antenna factor of 28dB/m, then the electric field strength is 30dBuV+28dB/m =58dBuV/m.
In the case of the far field of the electromagnetic field described above, the electric field strength Es transmitted in the air is related to the wave impedance Z, which is defined as Z = E / H (the electric field E divided by the magnetic field H). The wave impedance Z of the far field is a constant 377 (120pi) Ohm, which represents that the relationship between the measured value of the electric field and the magnetic field is fixed. The antenna factor can be utilized here.
However, the wave impedance Z of the EMI probe in the near field is not fixed. It is related to the radiation source, the test distance and the probe form. Unlike the fixed far field of 377 ohm, the near field probe does not have the antenna factor.
The SMA Cable model in the GKT-008 is GTL-303 and its specifications are:
  • Insertion Loss: Less than 1.5dB, 0 ~ 3GHz.
  • Return Loss: Less than -18dB, 0 ~ 3GHz.

| Next Page >