[PIPE2D-44] Find out why amplifiers in assembled CCD are still distinguishable Created: 22/Jul/16 Updated: 24/Sep/16 Resolved: 24/Sep/16 |
|
| Status: | Done |
| Project: | DRP 2-D Pipeline |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Story | Priority: | Major |
| Reporter: | aritter | Assignee: | aritter |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Sprint: | 2014-16 |
| Description |
|
After applying the gain and removing overscan and bias the amplifiers are still distinguishable in the assembled CCD image although the difference in the mean values is small. Possibly the gain values are not correct? |
| Comments |
| Comment by aritter [ 12/Aug/16 ] |
|
The gain values are normally applied during flat-fielding. We don't do flat-fielding in the photometric sense. There is one parameter "applyGains" in BiasTask which is supposed to apply the gains instead of flat-fielding. However, if "applyGains" is set the task fails. After asking Paul Price about that (he called it bitrot) I filed a ticket in LSST Jira (DM-7200). |
| Comment by rhl [ 03/Sep/16 ] |
|
That bug is fixed. Did you change the drp pipeline to apply gains? If so, is this ticket resolved? |
| Comment by aritter [ 09/Sep/16 ] |
|
pipeline has been changed to apply gains although this has made the installation procedure a lot more complicated. After rebasing the bug fix branches for obs_subaru and pipe_drivers before merging them into master a lot more new packages are needed, like ip_isr, log, afw. Maybe we should disable doApplyGains until the next binary release? Otherwise we can close the ticket... |
| Comment by rhl [ 24/Sep/16 ] |
|
I took a look at this, and the feature is really present in the data (science frames and also biases). We need to use real master bias frames, which will require using better inputs than are used in the install.text (or sphinx/user/getting_started.rst) |
| Comment by rhl [ 24/Sep/16 ] |
|
See previous comment |