Signal receiving equipment for EMI testing is typically EMI receiver or EMI spectrum analyzer, which receives signals coming in through the antenna, probe or cable. Regardless of the EMI receiver or EMI spectrum analyzer, both must meet the following regulations:

- ±2dB absolute amplitude accuracy
- Filter (-6dB) that meets CISPR requirements, as shown in the table below:

Frequency band	Filter (6dB)
9k ~ 150kHz	200Hz
150kHz ~ 30MHz	9kHz
30MHz ~ 1GHz	120kHz
1GHz ~ 3GHz	1MHz

- Support Detector for EMI: Max Peak (PK+), Quasi-Peak (QP), EMI Average (EMI-AVG), RMS Average (RMS-AVG)

Because spectrum analyzer and EMI receiver are narrow-band sweep receivers, they only receive energy in a certain frequency range at a certain time. Transient interference of electrostatic discharge is a kind of pulse interference, and its spectral range is very wide, but the time is very short. When an instrument produces transient interference, it is only a small part of its total energy that cannot represent the actual interference situation.

CISPR 16 defines the regulations for EMI test equipment, as shown in the following table, which is the Receiver or spectrum analyzer for EMI measurement. The regulation characteristics must meet the requirements of the parameters. It mainly defines which Detector to be used in the measurement bandwidths and the corresponding time constants.

For an explanation of the various Detector differences, please refer to the video “Peak, Average and QP detectors” on YouTube.

GSP-9330 cannot perform editing, but the GSP-9330's dedicated PC software SpectrumShot can perform editing.

This PC software must be collocated with the NI VISA driver. Please download it from the NI website.

The frequency band for automotive electronics testing in the conducted section is from 10 kHz to 108 MHz and the test must be collocated with a dedicated DC LISN. GW Instek GLN-5040A is a LISN that uses AC power and cannot be used for conducted EMI measurement of automotive electronics. However, GSP-9330 can conduct the above-mentioned test while collocating with a DC LISN of other brands.

1. A power supply cable with a magnetic ring can suppress conducted or radiated EMI by the characteristics of the magnetic ring.
2. The better the isolation, the better the power supply wire can suppress the EMI from radiating out of the power supply to form radiated EMI, but the cost is relatively higher.

Common mode and differential mode noise can theoretically be analyzed in conducted and radiated EMI. However, when it comes to the actual noise suppression technology, it is more meaningful to discuss the conducted noise. It is also possible to seek reference countermeasures through measurement. The following figure shows the common circuit for conducted EMI suppression in power supply. CY, LCM is a common mode noise filtering network, and CX is a differential mode noise filtering network.

If we can measure the common mode component and the differential mode component of the conducted noise separately, we know that we should adjust the common mode filter CY, LCM, or the differential mode filter CX to reduce the noise. The following is a brief description.

and in the figure represent the currents on the two power lines. In theory, we can treat and as the sum of the common mode and differential mode noise currents, namely:

and , is common mode current, is differential mode current.

Furthermore, we found that and can be obtained from

Comparing the two equations, as long as the signal of one of the power supplies can be added to the other power line signal in the forward and reverse directions, the sum and the difference of and can be obtained separately, and the and can be measured. Therefore, a and coupler/adder can be designed at the power supply terminal of the DUT to obtain and .

There is an approximation, which uses a current probe with a spectrum analyzer to measure the sum of the two power lines, as shown in the following figures.

In Figure (a), run two power cables through a current probe, connect the probe with a spectrum analyzer to obtain . Run one of the power cables around the probe (b) to make the current direction opposite to the direction as shown in. (a). By so doing, the can be measured from the spectrum analyzer, and the results are shown below. It can be seen that the common mode noise component is higher than that of the differential mode, and we know that the common mode filter and need to be adjusted to reduce the common mode noise component. Although this is not an accurate measurement, it is an approximation that is worthy of reference.

Common mode and differential mode noise

If users concern big signal fluctuation while conducting EMI testing, especially the spectrum analyzer might be damaged when using PR-01 to directly contact AC circuit, use GPL-5010 to protect GSP-9330, as shown in the following diagram.

GPL-5010 operates from 9 kHz to 200 MHz with an attenuation of 10 dB to 12 dB. Detailed specifications can be found in GPL-5010 user manual, which can be downloaded from the website.

Yes. But not to choose Switching Mode AC Source to avoid the mixing of PWM signals and try to choose CE certified.  The noise of AC Source will be smaller.

If the test capacity is greater than GIT-5060 requirement, two methods are recommended: first, replace GIT-5060 with AC Source, but it is better to use the Linear Mode and it has passed the CE specification. Second, use EMI power filter. This method requires special EMI manufacturers to construct because site evaluation and further specification evaluation are required to meet your needs.