Why Does My Waveform Degrade When I Change the Vertical Position in VirtualBench-Scope?Primary Software: Other NI Software>>VirtualBenchPrimary Software Version: 2.1 Primary Software Fixed Version: N/A Secondary Software: N/A
Problem: I'm reading 0.6 V peak-peak sine wave in VirtualBench-Scope with 0.0 V vertical position and 0.1 Volts/Div, however when I move the Vertical Position of the wave form up or down 0.2 V my signal becomes noisy or looks worse. Isn't this just adding a constant offset in software? Why does VirutalBench-Scope do this, when my regular oscilloscope doesn't? Solution: As you may have noticed in VirtualBench-Scope there is not a specific setting that allows you to set the input range of the board you are using, as in VirtualBench-DMM or VirtualBench-Logger. VirtualBench-Scope uses the values that correspond to the top and bottom of the graph to determine the input limits, so that the you don't have to continually change them. These top and bottom values are determined by the Volts/Div and V. Position settings in the following manner: If we have 0.0 V for V. Position, the values that correspond to the top and bottom of the graph will be 4 times the Volts/Div setting (because there are 4 divisions on either side of 0.0 V). So if we have 0.25 Volts/Div our input limits would be +/- 1.0 V; if we use 1.0 Volts/Div this would increase our input limits +/- 4 V. These input limits are passed to a function in the NI-DAQ data acquisition (DAQ) driver software that then uses the actual input range for the board that comes closest to fitting the input limits we pass it. In the above two cases this would be +/- 1.0 V and +/- 5.0 V, respectively (typical ranges for most DAQ boards). If we leave the Volts/Div fixed and change the V. Position the following will happen: If we use 0.25 Volts/Div as our fixed value, and a V. Position of 0.25 V (one full division), our topmost value of our graph becomes + 0.75 V and the bottom-most value of the graph becomes -1.25 V. If we moved our V. Position down to where it was 2 divisions from the bottom, then these values would be +1.5 V and -0.5 V. If we passed either of these sets of values to the function that actually configures the board, it's best fit would have to be at least +/- the largest absolute value, depending on the board this could be +/- 1.25 (only first case), +/- 2.0 +/- 2.5, or +/- 5.0 V. With only changing the Volts/Div there is no noticeable noise increase to the trace, because you typically make the trace smaller on the screen. If you are making it larger, then the signal itself must be small to fit on the screen and decreasing the Volts/Div actually helps it. However, when you modify the V. Position enough to cause the board to need to go to another range you will notice a degraded signal, because the same number of points used to digitize your signal are being distributed across a larger range, and you typically have the signal taking up most of the screen. Why not just use a constant software offset?? If the input signals are always going to be symmetric about 0.0 V, a constant offset in software could be added instead, but a more generic approach is needed, so that signals that are not symmetric about 0.0 V could be displayed as large as possible and not be clipped. My regular oscilloscope doesn't have this behavior, why does VirtualBench-Scope? A couple of things that regular "box" oscilloscopes could do to account for this issue may be to use a Digital to Analog converter to add constant voltage to the signal or use an input range that is always greater than the displayed range so that the "noise" isn't as noticeable. The first item is not possible with National Instruments DAQ boards and the second would place limitations on the software. To get the best results with your board, check the user's manual for the valid input ranges and if you need to adjust the V. Position, then make sure that the top and bottom values of your scale don't exceed the next largest range. Related Links: Attachments:
Report Date: 06/08/1999 Last Updated: 12/08/2004 Document ID: 1M7FJ8TS |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
