| 2 |
2 |
We have been at the front lines of large mosaics, so we have developed quite |
| 3 |
3 |
a few tools to make life easier for the detector engineer who |
| 4 |
4 |
has to integrate a 500+ channel focal plane system. |
| 5 |
5 |
Voltages are applied by the grasp_pon_ota.c tool. A custom |
| 6 |
6 |
tool should be written for each type of CCD that STARGRASP |
| 7 |
7 |
controls so that the tool can be responsible for enforcing |
| 8 |
8 |
safe voltage limits and power-up sequences for your particular |
| 9 |
9 |
flavor of device. The grasp_pon_ota.c tool does this as |
| 10 |
10 |
defined by MITLL for their type of OTA, specifically CCID58s. |
| 11 |
11 |
We would like to create a graphical tool like "cestlavie" |
| 12 |
12 |
to tune voltages. This tool would do the work we do now with |
| 13 |
13 |
simple shell scripts (sometimes just typed from the command |
| 14 |
14 |
line) that edit the FITS header of your most recent set of |
| 15 |
15 |
FITS files and then re-run grasp_pon_ota. In this way, the |
| 16 |
16 |
grasp_pon_ota tool enforces that the new changed voltages are |
| 17 |
17 |
still within range relative to the other current settings. |
| 18 |
18 |
We have used parameter search loops that work in this way, |
| 19 |
19 |
for instance, changing drain voltages to find the optimum setting. |
| 20 |
20 |
These scripts operate on the entire focal plane at once. |
| 21 |
21 |
We have scripts which produce linearity plots for all cells |
| 22 |
22 |
of all devices in one image with lots of little graphs. |
| 23 |
23 |
At the device level, we have a program which works like a |
| 24 |
24 |
stethoscope, providing audio tones proportional in frequency |
| 25 |
25 |
to the RMS noise in the overscan of a readout at a high rate. |
| 26 |
26 |
(2 outputs if you have stereo ... ooh.) This is useful for |
| 27 |
27 |
all sorts of tuning, including voltage changes, slight clocking |
| 28 |
28 |
pattern changes, and grounding changes. |
| 29 |
29 |
A program called "slotmachine.c" can analyze a multi-dimensional |
| 30 |
30 |
(multi-sampled) STARGRASP FITS file and report the absolute |
| 31 |
31 |
best signal to noise possible by varying the weighting of |
| 32 |
32 |
the samples. This can lead to a better choice of a cestlavie |
| 33 |
33 |
timing pattern, especially if you are generating a relatively |
| 34 |
34 |
slow (>2usec) timing pattern. |
| 35 |
35 |
We make extensive use of X-Rays (from Fe55) to measure gain |
| 36 |
36 |
and charge transfer efficiency. A tool called "otaslop" |
| 37 |
37 |
automatically reports all kinds of statistics for each cell |
| 38 |
38 |
in each OTA when you take an "XRAY" type exposure with STARGRASP. |
| 39 |
39 |
(Can demo this on GPC1, which has an internal XRAY source.) |
