Use ASI-Console CRISP control “Step 1” to turn on the LED with sufficient brightness that it is easily observable with the camera.
Photo shows LED mask focused in center of Camera – also the substrate is in focus. This shows the nominal correct placement and tilt of the LED.
If you focus the microscope higher or lower, Figure 2 & 3, the mask image morphs into a blur. In one direction, the LED die comes into focus. When the splitting mirror is correctly adjusted, the bulk of the light will translate laterally (vertically in the photos).
1)Click Step 2 “Dither” state. Adjust the main CRISP adjusting Thumb Screw for maximum ERR number. 2)Click Step 3 “Set Gain” and wait a few seconds for it to finish. 3)Re-focus microscope on sample and set the Z-axis controller position to Zero. (HERE Z=0) 4)Be sure NA setting is correct for the objective and click “GRAPH” button.
CRISP graph of the focus function for a “deep” scan +/- 60um. Operating position is where the curve crosses the vertical line at Z=0. The horizontal line shows the position where the two detectors have the same signal level (zero for the vertical axis).
CRISP graph for normal-depth scan. 20X NA 0.75 objective on silicon Good operation point with +/-27um operating range and near zero offset.
5)Choose and operating point and adjust the main CRISP adjusting Thumbscrew to get there. If you only adjust for maximum ERR number, then the point you choose will be at the steepest part of the focus curve. Sometimes this can be very close to where the curve makes a sudden change in slope.
This figure shows the problem of choosing just maximum ERR to determine the operating point. The center vertical red line is at roughly the maximum value of the focus curve slope. If the operating point was at the center red line, then the system could correct for errors over a range of errors of about 7um in one direction and about 45um in the other direction. This is not robust. To operate at the center red line, the system would need to include a significant OFFSET because one detector will have much more light on it than the other.
Better is to move the operating point to where there is approximately zero offset but on a portion of the curve with a lower slope. Lower slope means that the ERR number in the STEP 2 “Dither” operation will be smaller than the peak value. You should be able to “move down the curve” by adjusting the Thumbscrew for slightly lower ERR number and then repeating the STEP 3 “Set Gain” operation. You should observe after the “Set Gain” that the calculate OFFSET reduces as well.
6)Repeat adjusting the Thumbscrew to lower slightly the ERR number and also gradually bring the OFFSET number close to zero. Repeat the GRAPH function and look for something like Figure 6.