How the Programmable Gain Amplifier can be used to optimize Q-21 performance (part 2)

In the previous part of this blog series, we have discussed the use of analog gain and how it compares to using digital gain. This blog will explore the different analog gain modes of the Q-21A230 and for which applications they are most suited.

The QUARTZ Q-21A230 CoaXPress camera, with a resolution of 21 MP at 230 frames per second, is the next generation of Quartz camera and the successor to the Q-12A180 camera. One of many advantages of the Q-21A230 is its ability to switch between different gain modes. This is done by using a programmable gain amplifier (PGA), which can be used to select one of the following analog gains: 1x, 1.5x, 2.2x, and 4.8x. The specification data for each of these analog gain modes is shown in the table beneath.

Table 2: Q-21A230-PA specification for each analog gain mode

The four PGA modes can be divided into two types of measurements, bright and dark field. Figure 1 shows a quick guide for selecting the optimal Q-21A230 gain mode depending on the application type.

Figure 1 Diagram to determine the correct gain mode for your application.

As can be read here, bright field measurements, or shot noise dominated applications, benefit from a high Signal-to-Noise Ratio (SNR). If the signal that needs to be detected is hidden in a bright background, the variance in photons, which is shot noise, disrupts the signal of interest. A high full well reduces the relative effect of shot noise, because it is proportionally smaller to the part of incoming light. The maximum full well and the high full well mode have a respective full well of 33.6 ke- and 22.4 ke-. As can be seen in table (1), the maximum full well mode has the best SNR_max. When the 33.6 ke- full well is not used entirely, e.g. when the video level is always below 80% of full scale, it is more efficient to use the high full well mode.

Dark field applications are often light limited. These applications use scattered light to detect edge or surface defects. If the light is limited, the high sensitivity mode is best suited because it applies the highest analog gain. As described in the previous blog, an increase in analog gain results in a decrease in read noise. Dark field measurements do not require a high full well. Instead, they require a high contrast resolution. The low read noise of 3.3e- of the high sensitivity mode means you can detect details with smaller contrast differences.

If the dark field application is not light limited, the maximum dynamic range mode may be more suitable, as can be seen in the flowchart. This mode has the highest dynamic range, which gives the best contrast between maximum and minimum saturation. When enough light is captured, it is preferred to have more contrast between the edge and dark background to detect the edge as sharp as possible. The high dynamic range also makes this mode best suited to applications that combine dark and bright field measurements.

Conclusion

Often, increasing the analog gain is described as a last resort, which should be avoided if possible due to the decrease in SNR. However, increasing the light source or the exposure time is not always an option. This blog shows that the analog gain modes of the Q-21A230 each have their own characteristics, while all operate at a 21 MP resolution at 230 fps. The application type determines which characteristics are most important. Especially in light limited (dark field) applications, increasing the analog gain might actually be beneficial for the application, as described above.