TEST & MEASUREMENT FEATURE
Andreas Grimm, product sales manager Oscilloscopes at Rohde & Schwarz, recommends a few optimised settings for economy class oscilloscopes that can considerably improve the outcome of analyses of an embedded system’s power supply to maintain operation of integrated circuits
DC voltage debugging using an economy class oscilloscope A
stable power supply is key to long-term operation of integrated
circuits. Although this is especially true for high-end circuits based on field programmable gate arrays (FPGA), even lower-speed serial buses can significantly disturb a signal. A quick analysis using an economy class oscilloscope helps to improve system performance substantially. Using a few optimised settings of the oscilloscope can considerably improve the outcome of this analysis. This article introduces methods for optimally analysing effects on an embedded system’s power supply. The analysis, based on an exemplary DC supply voltage of an FPGA with a CAN interface, is performed using the Rohde & Schwarz (R&S) RTB2000 oscilloscope.
OPTIMISED SETTING FOR DC VOLTAGE MEASUREMENTS First, the DC voltage is analysed without any special settings. Figure 1 shows the DC voltage measurement using a passive probe (10:1) connected to the DC voltage. In order to see the signal on the screen, the vertical scaling is set to 1V/Div and a peak-to-peak voltage measurement including statistics is used to determine the ripple. The built-in voltmeter gives a measurement value for the DC voltage level of 4.92V. In this setting, the mean value measured for the ripple is 179.90mV (marked with the red circle with the built-in annotation tool for documentation). Why does the vertical resolution of an oscilloscope play an important role? A quick initial estimation is the theoretical resolution of the oscilloscope in this setting. The R&S RTB2000 uses a 10-bit ADC and therefore has 1,024 decision levels. The vertical setting is 1V/Div, yielding a full range of 10V. Doing the math shows that the theoretical resolution is about 10mV. And although the supply voltage looks flat, the mean ripple derived from over 10,000 measurements is 179.90mV, or in the range of 3.5 per cent of the supply voltage.
Figure 1: Measurement of DC voltage without any optimised
oscilloscope settings.
To improve the accuracy of the measurement, the channel offset is set to the 4.92V level and the sensitivity to 20mV/Div – which increases accuracy by a factor of 50. As shown in Figure 2, the mean value of the peak-to-peak measurement is now 68.28mV. This is about 2.5 times smaller compared with the initial measurement and is much more accurate, the 10-bit ADC resolution in this setting is approximately 0.2mV.
IDENTIFICATION OF DISTURBANCES OF THE DC SUPPLY VOLTAGE The second step is to identify and correlate disturbances coupling into the
DC voltage from other events. Looking at the signal change in Figure 2, it is hard to identify these disturbances, since the time base is not optimally chosen. A common approach is to capture longer time intervals to increase the chance to see coupled events, which are often based on slow signals. A typical source of coupled events for embedded systems comes from the AC/DC converter and may be related to the main supply frequency (50Hz in Europe). In order to identify such patterns, the time base of the oscilloscope should be set to 10ms/Div. In Figure 3, such a configuration is used together with an additional zoom
Figure 2: More accurate measurement results applying the right
vertical settings and advanced frontend technology
>30 INSTRUMENTATION | NOVEMBER 2017 29
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52
