 |
Home | Applications | Contact Us | |
| Products | Service | Training | ||
|
Home | Applications | Contact Us | |
| Products | Service | Training | ||
|
Linearity As with spectroscopy amplifiers, the linearity of the ADC's response to input signals is an important performance parameter. Two different linearity specifications are required to define the performance of the ADC: the integral nonlinearity, and the differential nonlinearity. Figure 8 demonstrates the measurement of an ADC's integral nonlinearity. Using a precision pulser with adequately low nonlinearity, a calibration curve of channel number versus input pulse amplitude is plotted. A straight line is fitted to this calibration curve using a least-squares fitting method. The integral nonlinearity is specified as the maximum deviation ∆Cmax of the measured calibration curve from the straight line, expressed as a percentage of full scale. It is a measure of the deviation from an ideal, straight-line calibration curve. The differential nonlinearity specifies the non-uniformity of channel widths. For the measurement, a sliding pulser is injected into the ADC input. As the pulse amplitude slowly slides from 0 to 10 V and back to 0 V in repeated cycles, counts are recorded in all channels. In order to reduce the statistical error, the measurement typically takes at least 10 hours to collect sufficient data. If the channel widths are all equal, the counts recorded in each channel will be identical. The differential nonlinearity is computed as the maximum deviation of the counts, in any of the channels, from the average counts in all the channels, expressed as a percentage of the average counts. This is actually a measure of the maximum deviation of channel width from the average channel width, expressed as a percentage of the average channel width.
Figure 8. Measurement of Integral Nonlinearity in an ADC. |
|
