Uncertainty Analysis for Characterization of a Commercial Real-Time Oscilloscope Using a Calibrated Pulse Standard

Abstract

In this paper, we describe a traceable measurement method for the frequency response of commercial real-time digital oscilloscopes (RTDOs). Since the commercial RTDOs usually use multiple analog digital convertors (ADCs), it is very challenging to characterize each ADC without having access to the internal circuitry. In this study, we use an additional continuous wave (CW) source to match the sampling sequence in a time interleaved ADC (TIADC). We also use a calibrated pulse standard traceable to the National Institute of Standards and Technology (NIST). Since the sampling rate of a single ADC is greatly reduced, we slightly misalign the sampling rate of the pulse and the single ADC and then stack the multiple measured pulses as a single pulse. As a result, the sampling rate can be greatly increased by about 8000 times. The frequency response of the RTDO during testing shows a variation of ±1 dB in amplitude and ±7° in phase between ADCs at the maximum operating frequency. These differences need to be calibrated as they cause systematic errors in the measurement. The 95% confidence interval of the measured frequency response is about 0.2 dB in amplitude and 3° in phase. Finally, the measured amplitude is compared with the swept sine measurement method, which confirms that they agree well within its uncertainties.