Mixed-Signal Stimulus-Response

• NI LabVIEW
• NI High-Speed Digital I/O
• NI High-Speed Digitizers
• NI High-Speed Sources

A mixed-signal stimulus/response test requires that you have at least two signals: the stimulus signal you send to the device under test (DUT,) and the response signal that you receive from the device under test. These signals can be either analog or digital. If you’re testing an digital-to-analog converter, for example, you would use a digital source and a digitizer, whereas you would use an analog source and a digital input device to test an analog-to-digital converter. Typically a test includes many different measurements and centers around comparing your acquired response to your expected response. To make an accurate comparison, you need two things:

• Software to make the comparison
• Tightly synchronized hardware stimulus and response signals to ensure accuracy

Making this comparison has implications for both hardware and software. In hardware, you need to control the timing between the acquisition and generation modules, so you know which acquisition sample should correspond to which generation sample. Platforms such as PXI and VXI are ideal for such applications where timing and synchronization between instrument modules is critical. These platforms are designed with timing and synchronization resources built-in, so modules can share reference clocks, sample clocks, triggers, and events. Consider a digital-to-analog converter (DAC) test. To minimize error in the comparison of stimulus to response, you would use a start trigger to start the acquisition and generation devices together. These hardware and software resources are built into the PXI and VXI architectures. You would use these same resources if you were using high-speed digital I/O and image acquisition to test a digital display, or if you were using high-speed digital I/O, a high-speed digitizer, and a high-speed source to test a PCB.

Once you have your known good stimulus signal, and the acquired response signal, you need to compare them. You can do this in software with the right analysis tools. In LabVIEW, you can use built-in waveform functions to compare the response to the stimulus, and you can identify which (if any) data points don’t match. You can use similar functions to further analyze the data and make more measurements, such as INL, DNL, SINAD, and others.

Reader Comments | Submit a comment »