Power supply noise testing-EEWORLD

Collect

Power supply noise testing

The distance between the probe's GND and the signal's two detection points is too large. Figure 1: Quantization error of the oscilloscope DC. The oscilloscope has a quantization error. The real-time oscilloscope's ADC is 8 bits, which converts the analog signal into 2 to the 8th power (i.e., 256) quantization levels. When the displayed waveform only occupies a small part of the screen, the quantization interval is increased and the accuracy is reduced. Accurate measurement requires adjusting the oscilloscope's vertical scale (using variable gain when necessary) to make the waveform fill the screen as much as possible and make full use of the ADC's vertical dynamic range. In Figure 1, the vertical scale of the blue waveform signal (C3) is one-fourth of the red waveform (C2). The rising edges of the two waveforms are enlarged (F1=ZOOM(C2), F2=ZOOM(C3)), and then the enlarged waveforms are displayed with a long afterglow. It can be seen that the waveform F1 in the upper right part has more steps (i.e., quantization levels), while the waveform F2 in the lower right part has fewer steps (i.e., fewer quantization levels). If we measure some vertical or horizontal parameters of the two waveforms C2 and C3, we can find that the standard deviation of the statistical value of the measured parameters of the signal C2 that fills the entire screen is smaller than that of the latter, which shows the consistency and accuracy of the measurement results of the former.

Usually, when measuring power supply noise, an active or passive probe is used to probe the power supply pin and ground pin of a chip, and then the oscilloscope is set to long persistence mode, and finally two horizontal cursors are used to measure the peak-to-peak value of the power supply noise. One problem with this method is that the conventional passive probe or active probe has an attenuation factor of 10. After connecting to the oscilloscope, the minimum gear of the vertical scale is 20mV. When the DSP filtering algorithm is not used, the peak-to-peak value of the probe background noise is about 30mV. Taking the 1.8V power supply voltage of DDR2 as an example, if calculated at 5%, the allowable power supply noise is 90mV, and the noise of the probe is close to 1/3 of the signal to be tested. Therefore, it is impossible to accurately test small voltages such as 1.8V/1.5V using a probe with 10 times attenuation. When actually testing 1.8V noise, the vertical scale is usually between 5-10mV/div.

In addition, the distance between the GND and signal detection points of the probe is also very important. When the two points are far apart, there will be many

Figure 2: The EMI noise of the signal current loop on the probe is radiated into the signal loop of the probe (as shown in Figure 2). The waveform observed by the oscilloscope includes other signal components, resulting in erroneous test results. Therefore, the distance between the probe signal and the ground detection point should be reduced as much as possible to reduce the loop area.

Figure 3: Schematic diagram of LeCroy PP066 probe For voltage testing of small power supplies, we recommend a passive transmission line probe with an attenuation factor of 1. When using this type of probe, the minimum scale of the oscilloscope can reach 2mV/div, but its dynamic range is limited, and the adjustable range of the offset is limited to +/-750mV. Therefore, when measuring common 1.5V and 1.8V power supplies, a DC-Block circuit is required before inputting into the oscilloscope.

Figure 3 shows the LeCroy PP066 probe. The distance between the ground and the signal of the probe is adjustable, and the ground pin of the probe can be elastically retracted, which is very convenient to operate. It is connected to the oscilloscope channel through a coaxial cable and a DC isolation module.

You can also strip the coaxial cable and directly solder the signal and ground of the cable to the power and ground of the power supply to be tested. In Figure 4, a section of the coaxial cable with an SMA connector is stripped and soldered to the 1.8V power supply of the DDR2 of the computer motherboard to measure its power supply noise.

Figure 4: Measurement of 1.8V power supply noise of a computer motherboard DDR2

After accurately measuring the waveform of the power supply noise, the peak-to-peak value of the noise can be calculated. If the power supply noise is too large, it is necessary to analyze which frequencies the noise comes from. At this time, it is necessary to perform FFT on the waveform of the power supply noise and convert it into a spectrum for analysis. The length of the signal time in the FFT determines the spectrum resolution after the FFT. In LeCroy oscilloscopes, the industry's largest 128M point FFT is supported, which can accurately locate which frequencies the power supply noise comes from (its spectrum resolution is more than 40 times that of similar instruments).

Figure 5: Measuring a 3.3V power supply noise

Figure 5 shows the noise of the 3.3V power supply of a certain optical module. The frequency of the highest point of the noise spectrum is 311.6KHz. The same 312KHz periodic jitter was found in the jitter test of the 1.25Gbps optical signal output by this optical module. As can be seen in Figure 6, after the periodic jitter of the 1.25G serial signal is decomposed (Pj breakdown menu), the 312KHz periodic jitter is found to be 63.7 picoseconds, and the jitter can also be clearly observed in the eye diagram. This case shows that power supply noise is likely to cause the eye diagram and jitter of some high-speed signals to deteriorate. Spectral analysis of power supply noise can effectively locate the source of noise and guide the direction of debugging.

When using an oscilloscope to measure power supply noise, in order to ensure measurement accuracy, it is necessary to select a sufficient sampling rate and acquisition time.

The recommended sampling rate is above 500MSa/s, so the Nyquist frequency is 250M, and the power supply noise below 250MHz can be measured. For the most popular board-level power integrity analysis, a bandwidth of 250M is sufficient. Noise below this frequency can be filtered using ceramic capacitors, tightly coupled power supply and ground planes on the PCB. Noise above this frequency can only be achieved through decoupling measures at the package and chip level.

The longer the waveform acquisition time is, the smaller the spectrum resolution (i.e. delta f) after conversion to spectrum will be. Usually our switching power supply operates above 10KHz. If the spectrum resolution is to reach 100Hz, at least a 10ms long waveform needs to be acquired. At a sampling rate of 500MSa/s, the oscilloscope needs a storage depth of 500MSa/s * 10 ms = 5M pts.

Reference address：Power supply noise testing

Previous article：CMMB signal generator test principle and application
Next article：The voltage regulator is connected in series with a transformer to increase the output voltage

Popular Resources
Popular amplifiers